Datasets:

taln-ls2n

/

schutz

like 2

Follow

TALN research group at LS2N lab 11

J_Mol_Model-4-1-2386531

Application of the PM6 method to modeling the solid state

The applicability of the recently developed PM6 method for modeling various properties of a wide range of organic and inorganic crystalline solids has been investigated. Although the geometries of most systems examined were reproduced with good accuracy, severe errors were found in the predicted structures of a small number of solids. The origin of these errors was investigated, and a strategy for improving the method proposed. Introduction The semiempirical method PM6 [1] was designed primarily for the investigation of molecular species of biochemical interest. That is, the objective of parameter optimization was to reproduce the properties of molecules. When other semiempirical methods, e.g., MNDO [2, 3], AM1 [4], and PM3 [5, 6], were developed, initial reports indicated that they were significantly more accurate than earlier methods. But later, when each new method was used for modeling species that were significantly different from those used in the training set, average errors typically increased quite significantly. This unfortunate result was a natural consequence of the way in which semiempirical method development was done: if, during a survey, a systematic fault was identified, then the training set would be modified in such a way as to correct the fault. The close relationship between the survey set and the training set meant that, by its nature, properties of species in the survey set were reproduced with a higher accuracy than those of species not in the survey set. During the development of PM6, efforts were made to minimize the potential for this increase in error. Among these were the construction and use of very large survey and training sets. In contrast to previous methods in which the training set was a subset of the survey set, during the development of PM6 the training set was a superset of the survey set. No solids were used in either the training set or the survey set while PM6 was being developed because inclusion of even one solid in the parameter optimization would have made the whole process extremely slow, which in turn would have precluded optimization of the parameters in any reasonable time. Because of this, solids were excluded from the parameterization, and therefore they form an ideal, clearly defined set of systems for testing the applicability of PM6 to species that were not used in the development of the method. Theory There are several problems associated with solid-state calculations that do not exist when discrete molecules are modeled, all of which are related to the fact that there are an infinite number of interacting atoms. The most obvious consequence of this is that the electric potential experienced by each of these atoms is the result of the contributions of an infinitely large number of electrostatic terms arising from the partial charges of all the other atoms. Another implication is that the number of one-electron wavefunctions contributing to the density matrix during the solution of the self-consistent field (SCF) equations is also infinite. Various techniques have been developed for solving these problems. Thus, in all solid-state methods, the assumption is made that the wavefunctions exhibit a perfect periodicity; this assumption is formalized in the Born–von Kármán [7] periodic boundary conditions. The electrostatic contribution or Madelung integral can be solved analytically using the Ewald sum [8]. In this procedure, an elegant mixture of real and reciprocal space contributions is used in the evaluation of the potential. To assist in the solution of the SCF equations, the near infinite number of occupied wavefunctions contributing to the density matrix is replaced by an integration over the Brillouin zone. In turn, this integration is approximated by a Simpson’s rule technique involving a weighted sampling of points within the zone. Several complete procedures have been developed for modeling solids using semiempirical methods. One of these, the MOSOL program [9], used sampling of the Brillouin Zone but, because it used complex mathematics, it was impractical for application to anything more complicated than simple binary solids. If the unit cell used is sufficiently large, then, instead of sampling the Brillouin Zone using a regular mesh of points, only one point need be used and, if the point chosen is at the origin of k-space, i.e. the Γ point, then complex mathematics can be avoided entirely. This is the basis for the large unit cell [10] or cluster [11] approximation. More recently, Gale [12] has addressed the problem of solving the Ewald sum when neglect of diatomic differential overlap (NDDO) methods are used, and developed a technique that would allow the Madelung integral to be solved more rapidly. In turn, this has allowed the structures and energies of some crystalline oxides, such as corundum and some of the polymorphs of silica, and of ice, to be modeled. During the development of a procedure to allow PM6 to be applied to solids, various deficiencies and limitations were found in earlier procedures. Some of these, and the resulting modifications that had to be made in order to allow PM6 to be used for modeling solids, will now be described. NDDO error The NDDO methods pioneered by Dewar and Thiel use the Dewar-Sabelli [13–15]-Klopman [16] (DSK) approximation, Eq. 1, which is equivalent to the Ohno approximation [17], for the two-electron two center integral γAB involving atoms A and B separated by a distance RAB. The DSK approximation has the correct behavior at the extremes. That is, it converges to the exact point-charge expression as the interatomic distance becomes very large, and also converges to the exact two-electron, one-center term, GA, when the interatomic separation becomes zero. Additionally, it has good behavior at chemical bonding distances. Over the past 40 years, the DSK approximation has proven very successful in NDDO models when applied to both discrete species and to polymers. Surprisingly, in its unmodified form, the DSK approximation gives rise to an infinite error when applied correctly to any non-elemental crystalline solid. This error arises from the fact that the one-center two-electron integrals differ from element to element. The origin of the error can be understood by considering the potential at an atom A in a simple binary solid, AB, arising from all atoms on the surface of a spherical shell of radius R; in such a solid, if the charge on atom A is Q, then the charge on atoms of type B would be −Q. When R becomes large enough, the fraction of atoms of type A and B at that distance will be the same, and the resulting electric potential, V, at atom A could then be represented by Eq. 2. A Taylor series expansion of this function shows that V is proportional to the reciprocal of the distance. For all values of R greater than about 10 Å this potential is clearly very small. In solids, however, the potential at an atom is the result of the summed electric fields of all such shells, out to infinity. For convenience, this sum can be replaced by an integral, Eq. 3. The value of this integral is infinity, which means that, if the DSK approximation is used and the integration is done correctly, the potential experienced by an atom of type A arising from the electrostatic contributions of all other atoms would then be either plus infinity or minus infinity, depending on the sign of its partial charge. This is an obviously unphysical result. This catastrophe can readily be avoided by modifying the DSK approximation to ensure that it converges to the point charge expression for large values of R. The simplest modification would be to ensure that, as the interatomic separation increased, a smooth transition is made from the DSK equation to the exact point charge equation. Several trial functions were examined and, from these, a Gaussian function was selected as having the best characteristics; this function is shown in Eq. 4. Below 5 Å, the unmodified DSK equation would be used; at larger distances, Eq. 4 would be used. This function is well-behaved in that it is single-valued and has finite first and second derivatives. Electrostatic interaction Evaluating the electrostatic potential at an atom in a crystal involves summing interactions from all surrounding atoms, and since there are, for all practical purposes, an infinite number of these, the direct sum must be replaced by a tractable alternative. The simplest and most efficient method of evaluating the electrostatic potential arising from an infinite lattice of point charges is the Ewald sum [8]. In this summation, the contribution to the potential is divided into two terms, a real-space and a reciprocal- or Fourier-space term. When an appropriate error function is used, the Ewald sum is both accurate and readily evaluated, and is the method of choice when the model used represents the electrostatic potential as the sum of contributions from classical point charges. Gale successfully applied a modified version of the Ewald sum [12] in evaluating the electrostatic potential used in MNDO, AM1, and PM3 solid state calculations. In all NDDO [18, 19] methods, including PM6, the electrostatic contribution to the potential at an atom arising from the charges on distant atoms can be represented by the classical point charge equation, at small distances by the DSK, and at intermediate distances by the modified DSK approximation. Gale [12] noted that a modification must be made to the potential in order for the Ewald summation to be used in an NDDO method. This change requires the point-charge contribution to the potential of each atom that arises from all nearby atoms to be replaced by the exact NDDO contribution. Derivatives of the energy with respect to geometry require all potential functions to be continuous, but if corrections of the type just described were made, the resulting function would obviously be discontinuous, and further corrections would be needed. So, although the Ewald sum is aesthetically attractive, its practical implementation would necessarily involve aesthetically unattractive corrections. An alternative to the Ewald sum would be to modify the way in which the electrostatic sum is evaluated. In this approach, use is made of the fact that an integer number of interacting unit cells are used in a solid state calculation. If the DSK equation, either unmodified or modified as in Eq. 4, is used, then the potential at any given atom arising from the direct summation of the NDDO electrostatic terms from all the other atoms would contain artifacts reflecting the asymmetric environment. In other words, the presence of boundary effects introduces spurious terms into the potential. If these terms were not eliminated, they would have a perturbative effect on the optimized geometry that would severely compromise the validity of the results. A method for removing these spurious effects was developed that involves modifying the distance term in the DSK approximation. The potential experienced by each atom in a solid that arises from the partial charges on other atoms falls off rapidly with increasing distance. This is a natural result of the fact that the net charge arising from all atoms in a spherical shell must rapidly converge to zero as the radius increases. An implication of this is that, for large radii, the precise value of the radius used in evaluating the potential is unimportant. Conversely, when the radius is small, and there are relatively few atoms, the potential arising from the associated partial charges is large. In that case, the value of the interatomic separation used is of great importance. This behavior can be used as the basis for modifying the electrostatic sum. At large distances, because the electrostatic effect of the distant atoms is small, the value of the interatomic distance used in calculating the potential can be different from the actual value, and, in fact, can be set to any arbitrary large fixed value. That is, all potentials arising from distant atoms can be treated as if their partial charges were moved in to the surface of a sphere of fixed radius. A result of this is that the gradient or force arising from a charge that was initially outside the sphere would be exactly zero: any potential motion of the central atom in response to the presence of a charge on the surface of the sphere would be accompanied by a simultaneous motion of that charge. A consequence of this is that the gradient of the potential arising from a charge on the surface of the sphere is precisely zero. This modification of the effective interatomic distance (EID) used in evaluating the electrostatic potential completely eliminates all directional effects, in particular all artifacts arising from the use of a finite number of interacting unit cells. If no further modifications to the EID were made, then there would be a discontinuity in the gradient arising from the presence of the sphere. The gradient arising from a partial charge just inside the sphere would be finite, but if that charge were to move just outside the sphere, its gradient would now become zero, and there would be a discontinuity. The presence of such discontinuities would then preclude the gradients being used in subsequent operations such as geometry optimization and calculation of vibrational frequencies. To avoid them, the EID must be further modified to ensure that the gradient arising from an atom near the surface of the sphere drops smoothly to zero as the atom approaches the surface of the sphere. This is most simply accomplished by reducing the EID of an atom as it approaches the surface of the sphere. To summarize: the value of the EID is set to a constant for all atoms separated by a large distance, is set less than the actual distance for intermediate distances, and is equal to the actual distance when the interatomic separation is small. A simple function that satisfies these criteria can be defined using three domains, as shown in Fig. 1. Fig. 1Truncation approximation for Madelung integral. C = 30 Å For atoms that are at a distance less than some predefined value, 2/3C, the exact DSK approximation is used. Between 2/3C and 4/3C, the EID to be used in Eq. 4 would be reduced as shown in Eq. 5. , and at distances greater than 4/3C the value of the EID would be a constant C. The effect of these changes when applied to an example set of charges is illustrated in Fig. 2, with the original charges shown in black, and the locations of the charges that would be used in evaluating the electrostatic potential shown in green. Fig. 2Effect of truncation on apparent position of charges. For a set of charges (black), the position used in evaluating the electrostatic interactions is shown in green Provided enough unit cells are used to ensure that all atoms within the sphere of radius 4/3C are present, the effect of this modification is to remove any directional influence, specifically surface effects, arising from the presence or absence of distant atoms. As with the unmodified DSK equation, the potential arising from an atom at any distance is single-valued and its first derivative is finite. An integer number of unit cells is always used in the evaluation of the electrostatic potential; therefore, the net charge on the surface of the sphere of radius 4/3C precisely counterbalances the sum of all the charges within the sphere, regardless of how many unit cells are outside the sphere. This is a natural and necessary consequence of the requirement that unit cells in a solid must have a zero charge. The electrostatic potential is, of course, dependent on the value of C. With increasing values of C, the potential converges rapidly to a constant, but also as C increases the number of unit cells that need to be used increases rapidly. The value of C was set to 30 Å, this being the best compromise between computational effort and numerical stability. Unlike the Ewald summation, this modified DSK approximation can be used directly in evaluating the electrostatic potential. The new approximation is relatively simple in that the use of error functions and reciprocal space terms are avoided. Solids with unpaired electrons Many solids, particularly those containing transition metals, have unpaired electrons. Of the two standard methods available for modeling such systems when only molecules are involved, unrestricted Hartree Fock (UHF) and restricted Hartree Fock followed by configuration interaction (RHF-CI), only UHF is suitable for modeling solids. The use of RHF-CI methods is precluded because of the very large active space involved. For example, consider the garnet uvarovite, calcium chromium silicate, Ca3CrIII2Si3O12.. Each chromium ion in this mineral has three unpaired d electrons. The unit cell contains eight formula units, so, if the RHF-CI procedure was used, at a minimum the active space would need to include all 80 molecular orbitals of predominantly d character. Even if the reasonable assumption was made that all the unpaired electrons had the same spin, the number of microstates involved would still be very large, , and evaluation of the gradients for the resulting non-degenerate state would be prohibitively slow. For solids in which the ions with unpaired electrons are well separated, that is, where the ions are electronically isolated from each other, the assumption can be made that the spin-state of one ion will not interact significantly with the spin-state of any adjacent ion. In addition, if the atom in question is a transition metal ion, then the spin state can usually be inferred from its environment. In the case of uvarovite, each chromium ion is in an almost octahedral environment (the exact symmetry is S6), being surrounded by six oxygen atoms, so the three d electrons would be in a t2g manifold, and would therefore be unpaired. If the Hund’s rule assumption is made that the spin state is a maximum, then each chromium atom would be in a local 4A2g state, and any Jahn-Teller tendency to geometric distortion to a lower symmetry would be avoided. This assumption can be formalized in the calculation when a UHF method is used by defining the difference between the number of electrons of α and β spin to correspond to the maximum possible spin state of the entire system. For uvarovite, the unit cell would then be defined as having a spin of Ms = 24, and therefore would have 48 more α than β electrons. Applications Organic compounds Data sets were constructed for each organic compound, with, in each case, the starting geometry being the X-ray structure: i.e., the observed geometry. In contrast to molecular calculations, where internal coordinates are normally used, in the work reported here Cartesian coordinates were used exclusively. An attempt was made initially to use internal coordinates, but the numerical instabilities associated with the geometric gradients at the interfaces of the unit cells rendered their use impractical; no such difficulties were encountered when Cartesian coordinates were used. Each cluster consisted of between 100 and 200 atoms, and geometries were converged until the gradient norm had dropped below 5 kcal mol−1 Å−1, this corresponding to an uncertainty in the optimized geometry of about 0.001 Å. All unit cell parameters were optimized, as were the coordinates of all atoms within the unit cell. Unless indicated otherwise, symmetry was not used to accelerate the optimization. All calculations were done using MOPAC2007 [20] on a 3.6 GHz Pentium computer, and each geometry optimization took between 20 min and 1 day, with most taking about 1 h. No problems were encountered in any of the optimizations. With the possible exception of polymers, crystalline organic compounds consist of discrete molecules held together by relatively weak forces. As PM6 has been shown [1] to reproduce bond lengths and angles of simple organic compounds with useful accuracy, in this work attention was focused on the prediction of the structures of entire molecules and on the forces and energies arising from intermolecular interactions. A useful measure of accuracy of prediction of molecular structure is the root-mean-square (RMS) difference between the calculated and reference geometries of a single molecule or ion in a crystal. This quantity differs from the geometric quantities reported earlier [1] in that it measures the accuracy of prediction of the overall structure of a molecule, not just the accuracy of prediction of individual bond lengths and angles. It is possible for only relatively small distortions to exist in individual angles and, at the same time, for the overall structure to be severely in error. The RMS error is therefore complementary to the errors in individual bond lengths and angles. In order to probe the suitability of PM6 for modeling organic solids, compounds were selected that illustrate a wide range of common intermolecular interactions, the most important of these being, in order of the energies involved: ionic, hydrogen bonding, and π-stacking. Densities Another useful measure of the accuracy of prediction of organic and inorganic solids is the density. In most cases when the density is accurately reproduced the internal structure of the unit cell is also accurately predicted. This is not an infallible rule, in that it is possible for the density to be predicted with good accuracy and, at the same time, the unit cell structure to be significantly distorted. This rare occurrence can usually be detected by distortions of the unit cell parameters. No cases were found where the unit cell parameters were predicted with good accuracy and, at the same time, significant errors existed in the internal structure of the unit cell. A comparison of PM6 and X-ray unit cell parameters for 124 organic solids is presented in Table 1. In this table, the unit cell used was often different from that reported in the literature, particularly so in hexagonal crystals, that is, crystals in which the interface angles are 90°, 90°, and 60°. Unit cells were chosen that would maximize the size of sphere that could be contained in a given cluster; to this end, most hexagonal unit cells were replaced by equivalent orthorhombic unit cells. Predicted densities were reproduced with good accuracy, the average unsigned error in density being 6.9%, with the bulk of this error arising from errors in the calculated intermolecular distances. Although most systems optimized with only small changes in the geometry, in three instances quantitative changes occurred. Table 1Calculated and X-ray structural parameters for organic compounds RMS errorPM6X-rayDensity (g/cm3)dabcαβγabcαβγ(Z Z)-N N′-Dimethylurea (NIJHUJ)0.119.694.3611.3890.992.396.410.354.5711.4090.090.0102.71.23 (1.11)1, 4 Dioxocine 6-carboxylic acid chloride (CUWWIA)0.2914.6712.303.9089.691.789.215.1512.493.8190.087.790.01.63 (1.59)1, 3, 5-Triaminobenzene, 1, 3, 5-trinitrobenzene (NIBZAM)0.0514.696.8413.8277.590.389.915.086.9814.0676.590.090.01.65 (1.55)1, 3-bis((Pyrid-2-ylamino)carbonyl)adamantane α-ketoglutaric acid (RIZWUS)0.149.9713.6610.7292.0113.167.210.5313.7310.7391.8114.767.81.42 (1.35)1,4-diazabicyclo[2.2.2]octane azelaic acid (UNEGEZ)0.5030.368.335.9391.189.191.328.158.606.9890.093.090.01.33 (1.18)1-diazonia-4-azabicyclo[2.2.2]octane glutarate (UNEFIC)0.2910.5611.6710.3989.483.191.410.2812.4610.6890.066.190.01.42 (1.30)2-(2-(3-Carboxypyridyl))-4-isopropyl-4-methyl-5-oxo-imidazole (JAZCOC01)0.1016.0710.747.20188.786.585.516.3110.787.1785.690.090.01.40 (1.38)2, 4, 4, 6, 6-Pentachloro-2-(piperidyl)cyclotriphosphazene (POTKEO)0.238.1222.5017.2089.590.190.18.3222.0117.2690.090.090.01.68 (1.67)2, 4, 6-Tribromoaniline (BRANIL)0.094.2013.5912.9990.090.090.04.2614.6213.4490.090.090.02.96 (2.62)2, 4, 6-Trinitro-N-methyl-N-nitroaniline (Tetryl) (MTNANL)0.3011.377.2715.9690.472.390.410.617.3714.1390.084.990.01.52 (1.73)2-Pyridone (PYRIDO04)0.035.775.9312.7590.290.490.15.605.7913.5690.090.090.01.45 (1.43)4, 5-bis(Dimethylamino)-1,8-dihydroxynaphthalene (RISBIE)0.0610.3313.758.91107.790.390.110.6513.779.32107.090.090.01.36 (1.25)4-Aminobenzoic acid 4-nitroaniline (RILJEB)0.0530.908.664.8695.089.591.231.258.644.8793.790.090.01.41 (1.39)4-Fluoro-2-(phosphonomethyl)benzenesulfonic acid monohydrate (KIXQIR)0.178.499.717.60105.8108.699.48.499.257.92104.8110.097.41.74 (1.75)4-Hydroxybenzoic acid isonicotinamide (VAKTOR)5.4822.299.3790.390.091.96.0720.679.4090.090.095.31.51 (1.47)5H-Dibenz(b,f)azepine-5-carboxamide saccharin solvate (UNEZAO)0.1712.2311.467.1669.684.682.712.6810.457.5175.485.783.61.50 (1.46)5-Methyl-2-((2-nitrophenyl)amino)-3-thiophenecarbonitrile (QAXMEH)0.2318.5015.964.2488.789.688.918.6916.403.9586.290.090.01.38 (1.43)5-Methyl-2-((2-nitrophenyl)amino)-3-thiophenecarbonitrile (QAXMEH01)0.1116.428.328.5790.089.689.716.418.548.5088.290.090.01.47 (1.47)5-Methyl-2-((2-nitrophenyl)amino)-3-thiophenecarbonitrile (QAXMEH02)0.117.4811.907.91104.2117.078.67.4911.917.79104.5116.477.801.42 (1.44)5-Methyl-2-((2-nitrophenyl)amino)-3-thiophenecarbonitrile (QAXMEH03)0.178.0211.3613.6690.690.4106.57.9811.6813.3290.090.0104.81.44 (1.44)5-Methyl-2-((2-nitrophenyl)amino)-3-thiophenecarbonitrile (QAXMEH04)0.2010.9612.054.3789.690.573.011.2512.324.5990.291.871.21.56 (1.43)5-Methyl-2-((2-nitrophenyl)amino)-3-thiophenecarbonitrile (QAXMEH05)0.1213.2922.577.9190.690.190.513.1822.808.0290.090.090.01.45 (1.43)5-Methyl-2-((2-nitrophenyl)amino)-3-thiophenecarbonitrile (QAXMEH12)0.177.9312.1612.2590.890.9102.88.2312.3111.8290.090.0102.51.50 (1.47)6-(Pyran-4-one-2-yl)-1,4-dioxocine (CUWWEW)0.197.6910.9110.9690.774.490.67.9110.8011.5890.070.490.01.53 (1.46)9-(2-Hydroxyethyl)adenine (FABFUJ)0.117.188.3413.9690.290.089.47.348.2713.5790.090.090.01.42 (1.45)Acetophenone (ACETPH)0.048.718.459.7590.958.590.68.568.6810.2690.059.090.01.31 (1.22)Acetylacetone (LIWPIQ01)0.047.854.1115.4892.590.189.88.464.1516.0390.090.090.01.33 (1.18)Acetylcholine chloride (ACHOLC01)0.1715.295.949.6290.0 89.390.215.326.309.8990.090.090.01.38 (1.26)Adenosine-3″-phosphate dihydrate (ADPOSD)0.355.9111.6210.5588.690.090.26.3411.909.9487.890.090.01.76 (1.70)Adenosine-5″-diphosphate dihydrate (HMADPH)0.2918.506.768.8991.789.393.018.446.899.2090.087.590.01.75 (1.66)α Glycine0.105.095.5010.7790.190.069.65.105.4611.9790.090.068.31.77 (1.61)α Resorcinol0.055.609.5910.3889.789.990.15.609.5310.5390.090.090.01.31 (1.30)Anthracene (ANTCEN14)0.0210.367.326.8489.9109.489.811.176.028.5590.0124.690.01.21 (1.25)Aqua-tris(2-((dimethylamino)methyl)phenyl)borane (QEVYEV)0.1619.818.4114.9989.975.189.620.358.5715.2390.072.390.01.19 (1.13)Aspartic acid (LASPRT)0.205.136.387.7090.0106.290.05.146.987.6290.099.890.01.83 (1.64)Barium oxalate oxalic acid dihydrate (BAHOXH11)4.7510.6813.0968.990.190.15.4112.4614.4563.890.090.03.77 (2.67)Benzene (BENZEN)0.019.276.797.1590.190.289.89.556.927.4490.090.090.01.15 (1.06)Berberine sulfate (CISREB)0.1519.917.4327.5197.6117.390.720.377.4427.4397.7116.285.51.61 (1.56)β Glycine0.085.346.095.1490.065.789.85.386.275.0890.066.890.01.64 (1.59)β Resorcinol0.048.814.9912.3489.988.990.07.815.4312.6290.090.090.01.35 (1.37)bis(1,4-Diazoniabicyclo(2.2.2)octane) bis(1-aza-4-azoniabicyclo(2.2.2)octane) tetrakis(tribromide) dibromide[26] (DAHGUO)15.3920.6012.81103.789.990.016.1721.3713.23101.090.090.02.65 (2.33)bis(3,5-Dinitro-4-methylbenzoic acid) 1,2-bis(4-pyridyl)ethane (LAPTUS)0.1812.0116.117.2775.489.794.112.7715.777.2773.191.193.41.56 (1.51)bis(Pyridinium) oxalate oxalic acid (DEFCUM)4.1211.397.77102.199.891.04.0011.328.44102.097.388.61.60 (1.52)bis(Urea) oxalic acid (UROXAL01)12.347.005.1384.289.890.012.376.885.0583.690.090.01.58 (1.64)Bromo-tris(2-dimethylaminoethyl)amine-manganese(ii) (DAEAMN)0.2711.8511.8511.8390.290.389.712.2212.2212.2290.090.090.01.78 (1.62)Calcium acetate chloride pentahydrate (CALCLA)0.0413.2611.206.66119.189.788.613.7211.516.82116.790.090.01.73 (1.55)Camphor (UGAHUF)0.068.4526.777.3290.891.189.98.9327.047.3890.090.090.01.22 (1.14)Cholesteryl acetate (CHOLAD04)0.1216.989.2116.2887.971.788.417.629.2216.5290.072.890.01.18 (1.11)cis Copper Diglycinate (CUGLYM02)0.195.3312.6010.5489.990.189.95.1913.5510.6990.090.090.02.16 (2.03)Citric acid (CITRAC10)0.3111.925.3812.1889.5115.289.611.475.6212.8190.2111.590.01.81 (1.66)Coronene (CORONE)0.0316.024.799.8890.168.890.016.124.7010.1090.069.190.01.41 (1.40)Cyclohexane (CYCHEX)0.028.126.4111.2390.0107.190.08.206.4411.2390.0108.890.01.00 (1.00)Cyclotrimethylene-trinitramine (RDX)0.2211.2411.4814.3289.888.489.910.7111.5713.1890.090.090.01.60 (1.81)Cystine (LCYSTI10)0.425.3454.255.3788.459.989.25.4256.285.4290.060.090.01.78 (1.67)Cytosine (CYTOSM03)0.0410.217.097.7987.290.488.49.827.527.7379.590.090.01.52 (1.53)Dinicotinic acid (DINICA10)0.186.7111.099.6889.7111.990.16.5911.159.7090.0107.890.01.66 (1.64)Dipyridinium bis(hydrogen-oxalate) oxalic acid (DUVLUB)27.718.307.3590.083.890.026.848.917.4590.087.190.01.69 (1.60)Disodium adenosine-triphosphate trihydrate (ADENTP)0.766.2420.5831.3790.091.488.37.0720.8830.4590.090.090.02.00 (1.79)DL-2-Amino-4-phosphonobutyric acid monohydrate (CAXDIO01)0.118.114.8319.8987.6102.390.68.404.9520.6090.0100.490.01.76 (1.58)γ Glycine0.075.526.938.94114.450.990.75.487.048.92113.252.190.01.65 (1.59)Glucose (GLUCSA)0.1910.274.2015.1290.090.090.010.364.9714.8490.090.090.01.84 (1.57)Guanine0.063.6216.849.8888.285.996.23.5516.349.6990.090.090.01.68 (1.79)8-Azaguanine (AZGUAN01)0.063.6711.8116.3889.6102.789.73.5611.4416.4790.095.190.01.46 (1.51)Hydroquinone bis(aniline) (HIBFUT)5.038.2118.4276.890.090.05.398.2218.7477.590.090.01.33 (1.22)Isonicotinamide (EHOWIH)0.089.985.8010.1190.079.389.910.035.7310.1790.081.890.01.41 (1.40)Isonicotinamide 3-hydroxybenzoic acid (LUNMEM)20.705.1622.1989.996.290.020.875.1422.4290.097.790.01.47 (1.45)Keggin pentakis(Tetraethylammonium) bis(meso-tetraphenyl-porphyrinato-zinc) tetraconta-oxo-silicon-dodeca-molybdenum bromide(PIJFUJ)15.1432.0114.9890.190.390.115.2232.2815.2290.090.090.01.88 (1.83)l-Alanine0.065.9311.675.8690.089.989.95.7912.265.9390.090.090.01.46 (1.41)Leucine (LEUCIN02)0.1313.855.249.4489.282.490.214.525.309.5690.085.890.01.28 (1.19)Lithium acetate (Li(CH3COO) 2(H2O))0.0211.236.387.1598.394.488.210.886.626.8290.090.090.01.34 (1.38)l-Lysine monohydrochloride dihydrate (LYSCLH)0.1112.355.7614.3881.690.489.813.325.8814.9881.290.090.01.44 (1.25)Malonic acid bis(isonicotinamide) (ULAWEJ)14.6514.6812.2565.881.868.815.6814.8611.9467.385.665.71.55 (1.49)m-Aminophenol (MAMPOL02)0.028.246.1211.3390.089.589.98.286.1011.2390.090.090.01.27 (1.28)m-Aminopyridine (AMIPYR)0.036.305.6815.2390.189.5111.46.195.7115.3090.090.0110.51.23 (1.24)m-Cresol N,N,N′,N′-tetraisopropyloxamide (DUGRIG)6.6226.7211.8893.489.889.86.8627.1912.0896.190.289.81.15 (1.07)m-Diaminobenzene, 3,5-dinitro-1-cyanobenzene (REDDEJ)13.0710.399.8584.089.590.413.2610.459.8886.790.090.01.51 (1.46)Methionine (LMETON02)0.115.1816.409.3185.393.883.35.2014.839.4981.090.090.01.27 (1.37)Methyl (+)-(1α,2β,8α,9α,10β)-2-chloro-4-aza-3-oxatetracyclo(8.4.0.02,9.04,8) tetradecane-9-carboxylate oxalic acid monohydrate (HUZKOC)0.156.1918.358.1196.5101.681.76.6117.898.07102.399.883.21.47 (1.43)m-Hydroxybenzoic acid (BIDLOP)0.0323.804.975.0490.074.390.023.894.945.4990.074.390.01.60 (1.47)tetrakis(2-Carboxypyridinium) octacyano-molybdenum(iv) (PYCMOA) (4[C6H6NO2] [MoIV(CN)8]4-)0.0917.9610.469.7093.391.489.218.5110.159.1789.890.090.01.46 (1.54)m-Toluidine (FANDOO)0.089.135.7923.1291.780.989.48.765.8124.8790.079.990.01.18 (1.14)Natroxalate (Na2(C2O4))0.584.239.295.3890.190.286.53.4510.385.2490.090.090.02.11 (2.38)N, N′-(6 6″-dimethylbiphenyl-2 2″-diyl) bis(methylamine) (ENIWIH)0.0710.6711.7710.1590.190.190.011.0912.0210.5790.090.090.01.25 (1.13)N, N-2 6-Tetramethyl-4-nitroaniline (FOCVOI)0.127.8516.177.6790.795.688.97.6418.087.6490.090.090.01.33 (1.22)N, N-Dimethylaniline (DMAFBZ01)0.087.0913.208.7182.4101.295.46.7812.338.2680.574.092.01.29 (1.57)Nicotinamide adenine dinucleotide tetrahydrate (CEVYEH11)0.258.898.2011.3383.269.7103.28.848.5911.1989.470.4103.91.66 (1.58)N-Methylurea (MEUREA)0.096.856.648.0890.191.290.36.926.988.4890.090.090.01.34 (1.20)N, N-Dimethylbenzamide (ODOTOQ)0.086.5116.197.4990.090.690.86.6316.307.6590.090.090.01.26 (1.20)N, N-Dimethylurea (WIFKEB)0.099.278.496.0690.489.6109.39.278.606.0490.090.0108.81.30 (1.29)o-Aminophenol (AMPHOM02)0.057.508.1019.7090.090.590.07.857.2519.7590.090.090.01.21 (1.29)o-Aminopyridine (AMPYRD)0.047.415.6511.6290.085.989.97.595.6711.7190.084.590.01.29 (1.25)o-Diaminobenzene (BAGFIY)0.047.727.579.9689.982.590.07.727.5410.3290.080.090.01.25 (1.21)o-Dimethoxybenzene (TUKGEL)0.0612.949.355.5890.189.889.913.359.925.5390.090.090.01.36 (1.25)Oxalic acid (OXALAC06)0.036.916.088.1689.989.990.06.566.097.8590.090.090.01.74 (1.91)Oxalic acid dihydrate (OXACDH26)0.044.9211.453.4489.790.0106.96.0911.933.4789.790.0106.92.26 (1.73)Oxytocin 1XY10001 [23]0.4422.068.8926.7390.699.188.423.049.0427.2790.0102.290.01.43 (1.33)p-Aminophenol (AMPHOL01)0.037.9512.904.7890.190.190.18.1812.955.2690.090.090.01.48 (1.30)p-Aminopyridine (AMPYRE)0.045.4811.927.1989.690.391.05.5712.127.3290.090.090.01.33 (1.26)p-Chloroaniline (CLANIC05)0.038.377.308.9890.089.790.28.597.249.1990.090.090.01.54 (1.48)p-Diaminobenzene (PENDAM)0.038.2923.085.9489.990.093.58.3722.955.9790.090.093.61.27 (1.26)Phenanthrene (PHENAN08)0.049.165.738.5690.280.989.99.446.148.4490.082.090.01.34 (1.22)Phenolphthalein (NIMDAO)0.0911.2414.5819.0190.089.889.511.3914.8219.2790.090.090.01.34 (1.28)Phloroglucinol (PHGLOL)0.0312.719.354.8789.889.689.912.569.374.8390.090.090.01.45 (1.47)p-Hydroxybenzoic acid (JOZZIH)0.046.2110.0518.4689.885.790.16.3410.4618.5190.086.890.01.60 (1.50)Picric acid (PICRAC)0.159.63 19.499.3888.990.489.89.7019.139.2590.090.090.01.73 (1.77)Potassium hydrogen acetate0.013.916.7725.0090.389.882.14.017.1923.8890.090.081.71.60 (1.54)1-Dimethylamino-8-dimethylammonionaphthalene saccharin dihydrate (AJOHUC)0.098.819.5424.5090.299.690.19.259.1724.9390.095.890.01.42 (1.37)Salicylaldoxime (SALOXM)0.0712.915.4510.0890.365.489.813.605.0810.4190.067.190.01.41 (1.38)Sodium acetate Na(CH3COO)·3(H2O) (NAACET01)11.6511.3915.4088.978.891.710.4010.4712.3590.068.390.00.90 (1.45)Sodium hydrogen acetate16.3316.3316.3389.989.990.015.9215.9215.9290.090.090.01.30 (1.40)Sucrose (SUCROS01)0.2410.368.597.5090.0103.090.010.828.687.7290.0103.090.01.75 (1.61)Tetramethylammonium dihydrogen phosphate monohydrate (FIJHEL)0.028.018.0712.32 90.190.098.18.528.4412.9090.090.099.11.59 (1.37)Tetramethylurea (TIDBIR)0.168.696.0011.7392.193.191.69.976.2610.6390.093.090.01.27 (1.17)Thymine (THYMIN01)0.056.796.9611.7178.189.889.66.856.7812.8975.190.090.01.55 (1.45)trans Copper digycinate (IWUBEH)0.329.5327.275.3289.092.388.59.4427.935.0690.090.090.01.85 (1.91)Trinitrotoluene (ZZZMUC01)0.115.4322.6314.7290.191.490.16.0820.0214.9990.090.090.01.67 (1.66)tris(Acetylacetonato) titanium(iv) perchlorate (TIACPC)0.258.9711.739.7891.083.592.38.6811.749.8988.584.392.01.45 (1.47)Tryptophan (TRYPTC)0.1315.265.176.9089.378.388.914.675.307.4590.081.290.01.50 (1.40)Tyrosine (LTYROS10)6.8520.395.9089.990.090.06.91 21.125.8390.090.090.01.46 (1.41)Urea (UREAXX13)0.034.675.585.5890.090.090.04.695.575.5790.090.090.01.37 (1.37)Urea nitrate (UREANT02)7.818.209.7089.952.890.17.508.209.5490.055.890.01.65 (1.69)Weddellite (Ca(C2O4)·2(H2O)7.2311.7211.7289.789.989.97.3612.3712.3790.090.090.02.20 (1.94)Whewellite (Ca(C2O4)·(H2O) (CALOXM03)6.049.7714.5789.990.5107.76.2910.1214.5890.090.0109.52.37 (2.22)tris(2,2,6,6-tetramethylheptane-3,5-dionato)-yttrium(iii) (HAHTOZ01)0.289.578.8918.9589.686.490.210.639.9817.8790.090.090.01.32 (1.12)dX-ray densities in parenthesisValues of a, b, c in Ångstroms; α, β, γ in degrees. Reference crystal structures from the Cambridge Structural Database [21]. RMS Root-mean-square In the solid state, individual molecules of oxalic acid in oxalic acid dihydrate [Cambridge Structural Database [21] (CSD) entry OXACDH26] exist as the neutral species. PM6 incorrectly predicts them to be fully ionized, as oxalate, [C2O4]2-, plus two hydronium ions, [H3O]+. This change was accompanied by a very large increase in density, from 1.73 to 2.23 g/cc, a direct consequence of going from a neutral to an ionic species. A related system is barium oxalate oxalic acid dihydrate (CSD entry BAHOXH11) in which there exist polymeric chains of oxalate groups connected by bridging protons. As with oxalic acid, in the optimized PM6 geometry the proton is abstracted by the water molecule to give oxalate groups and hydronium, resulting in an increase in density of almost 40%. Sodium acetate trihydrate is an ionic solid consisting of sodium ions surrounded by an acetate group and five water molecules, four of which form bridges between pairs of sodium ions. PM6 completely fails to predict the observed structure: the distance between the sodium ions increases considerably, effectively destroying any tendency of the water molecules to form bridges. When these three solids were removed from consideration, the average unsigned error (AUE) in density decreased to 6.1%. The most common intermolecular interaction is hydrogen bonding, which PM6 predicts to be too short by about 0.1 Å, with the result that the average signed error in calculated densities of organic compounds is too high by 3.9%. When a systematic correction to the density was made, the AUE in density decreased still further to 4.8%. Heats of formation Table 2 presents a comparison of experimental and calculated heats of formation of organic solids. The largest difference occurs with 2, 4, 6-tribromoaniline, which is predicted to be too stable by 44.5 kcal/mol. The optimized structure revealed an unrealistically short intermolecular Br–N distance of 2.14 Å, indicating that the Br–N core-core repulsion was severely underestimated. Examination of the values of the PM6 parameters for the Br–N core-core interaction revealed that the Voityuk interaction would be negligible at chemical bonding distances. This error in PM6 can be attributed to the absence of appropriate reference data in the training set, a fault that could be readily corrected in future work. Table 2Comparison of calculated and experimental heats of formation of organic compounds (kcal/mol)PM6ReferenceaDifference(Z Z)-N, N′-Dimethylurea (NIJHUJ)−62.0−76.314.32, 4, 6-Tribromoaniline (BRANIL)−30.713.8−44.5α Glycine−122.9−126.13.2α Resorcinol−79.3−88.08.7Anthracene26.430.0−3.6β Glycine−121.8∼−126.1∼4.3Camphor−66.3−76.310.0Citric acid−348.7−369.020.3Cyclotrimethylene-trinitramine (RDX)−5.918.9−24.8Cystine−237.9−246.88.9γ Glycine−120.8∼−126.1∼5.3l-Alanine−132.0−134.12.1Leucine−142.7−152.39.6m-Aminophenol (MAMPOL02)−37.6−47.910.3m-Aminopyridine (AMIPYR)23.214.48.8m-hydroxybenzoic acid−123.7−142.018.4o-Aminophenol (AMPHOM02)−38.2−48.19.9o-Aminopyridine (AMPYRD)21.49.412.0o-Diaminobenzene (BAGFIY)12.99.33.6Oxalic acid−175.5−198.422.9p-Aminophenol (AMPHOL01)−40.7−46.45.7p-Aminopyridine (AMPYRE)19.010.09.0p-Chloroaniline (CLANIC05)2.5−8.010.5p-Diaminobenzene (PENDAM)5.110.1−5.1Phenanthrene24.126.2−2.1p-Hydroxybenzoic acid−125.9−145.019.2Picric acid−43.4−52.18.7Salicylaldoxime−27.0−43.916.9Sucrose−535.8−532.0−3.8Trinitrotoluene−5.7−15.19.4Tyrosine−150.7−163.713.0Urea−65.7−79.613.9aReference values taken from the CRC Handbook [32] Heats of sublimation The heat of sublimation is a measure of the intermolecular interaction energy. In some cases sublimation is accompanied by large geometric and electronic changes. For example, the simple amino acids exist as the Zwitterion in the crystal phase, but in the gas phase they are unionized. Representative values for calculated and observed heats of sublimation are presented in Table 3. Determining the accuracy of PM6 for the prediction of heats of sublimation is made difficult by the acknowledged unreliability of many experimental measurements. Thus in the reference compendium of sublimation enthalpies [22], the authors indicate that the reported value for aspartic acid, 22.9 ± 1.0 kcal/mol, was likely unreliable. Table 3Comparison of calculated and experimental heats of sublimation Reference [22]PM6Difference(Z Z)-N, N′-Dimethylurea (NIJHUJ)22.115.9−6.2α Glycine32.629.7−2.9Anthracene23.833.19.3Aspartic acid22.937.814.9Benzene10.03.2−6.8Camphor12.46.5−5.9Cyclohexane9.02.1−7.0Guanine44.525.2−19.3l-Alanine31.732.60.9Leucine36.028.5−7.5m-Aminophenol (MAMPOL02)23.611.8−11.8m-Aminopyridine (AMIPYR)19.39.5−9.8Methionine32.027.2−4.8m-Hydroxybenzoic acid29.515.3−14.3N-Methylurea22.617.0−5.6N, N-Dimethylbenzamide21.410.0−11.4o-Aminophenol (AMPHOM02)22.317.6−4.7o-Aminopyridine (AMPYRD)18.39.2−9.1o-Diaminobenzene (BAGFIY)20.49.1−11.3Oxalic acid22.319.2−3.1p-Aminophenol (AMPHOL01)24.219.3−4.9p-Aminopyridine (AMPYRE)20.810.4−10.4p-Chloroaniline (CLANIC05)21.78.7−13.0p-Diaminobenzene (PENDAM)22.016.0−6.1Phenanthrene22.030.28.2p-Hydroxybenzoic acid29.514.5−15.1Picric acid25.114.6−10.5Trinitrotoluene26.914.7−12.2Tyrosine24.133.79.6Urea21.717.5−4.2 Biomolecules The primary objective in developing PM6 was to more accurately model systems of biochemical interest. The applicability of PM6 to the study of crystals of biochemical importance was therefore of interest. Oligopeptides The X-ray structures of many small polypeptides have been determined and are readily available in the Protein Data Bank [23]. One representative entry in this collection, 1XY1, is the nonapeptide deamino-oxytocin, Cys-Tyr-Ile-Glu-Asp-Cys-Pro-Leu-Gly. The structure of this hormone had been refined to a resolution of slightly better than 1.1 Å [24]. Each deamino-oxytocin molecule contains a disulfide bridge between atoms S1 and S6, and two strong hydrogen bonds between N2 and O5, and N5 and O2. The unit cell contains four deamino-oxytocin molecules related by a pseudo C2 operation, that is, there are two inequivalent polypeptide molecules and 26 water molecules. The coordinates of the hydrogen atoms in the peptide were given, but not those of the hydrogen atoms on the water molecules. That water of crystallization exists implies that peptide–water hydrogen bonds also exist. Because the positions of the hydrogen atoms on the water molecules were not given in the X-ray structure, an estimate of the locations of the 52 hydrogen atoms had to be made before the geometry could be optimized. For this operation, the “ice rules” were used: each oxygen atom in a water molecule was involved in forming two hydrogen bonds and each hydrogen atom formed one hydrogen bond. Of necessity, some of these bonds involved atoms on the peptide. Several candidate structures, each of which satisfied these conditions, were constructed, and the positions of all hydrogen atoms optimized while those of the other atoms were held constant. An incidental benefit of this operation was that any potential errors in the X-ray positions of some hydrogen atoms, in particular the apparently faulty location reported for H147, were automatically corrected. After the positions of the hydrogen atoms were optimized, an unconstrained optimization on the unit cell was carried out. This involved the optimization of the Cartesian positions of all the atoms in the unit cell of formula (C43H65N11O12S2)4·26(H2O), and the unit cell dimensions, i.e., the simultaneous optimization of 1,839 coordinates. Each initial geometry optimized to give a different final structure. That is, the optimized geometry was very sensitive to the choice of initial locations of the hydrogen atoms assigned to the water molecules. As a result, it was not possible to unambiguously define an optimized PM6 structure; however, all the fully optimized structures were within a few kcal/mol of each other, so one structure was chosen arbitrarily and used in the following analysis. The optimized PM6 unit cell dimensions are shown in Table 1. The optimized PM6 structure of the entire unit cell had an RMS error of 0.61 Å, and an RMS error of 0.44 Å for a single molecule of deamino-oxytocin. Deviations from the pseudo-C2 symmetry were small, and were very sensitive to the initial choice of hydrogen bonds; it is likely that the time-average would be exactly pseudo-C2. All the weak intra-chain bonds were preserved: PM6 predicted the disulfide bridges, S1–S6 to be 2.06 Å and 2.05 Å compared with the X-ray values of 2.08 Å and 1.95 Å, and the hydrogen bonds between N2 and O5 to be 1.98 Å versus the X-ray value of 1.93 Å, and N5 and O2 1.98 Å, compared to the X-ray, 1.90 Å. Acetylcholine One of the simplest of the important biochemicals is the neurotransmitter acetylcholine, [CH3COOCH2CH2N(CH3)3]+ (ACh). In one form, this ion exists in the solid state as the chloride (CSD entry ACHOLC01). PM6 reproduced the structure of ACh in this salt with good accuracy, the RMS error for a single ACh being only 0.21 Å. Part of this error can be attributed to the C–H and N–H bond-lengths from the X-ray structure being about 0.2 Å too small; when only the heavy atoms are used in the comparison, the RMS error decreased to 0.13 Å. An estimate of the effect of the crystal-packing forces can be obtained by comparing the PM6 predicted structure of the gas-phase ion with that found in the crystal; when this was done the RMS difference increased to 0.54 Å. This is unequivocal evidence that inclusion of crystal-packing forces is essential in order to reproduce the observed structure. Adenosine diphosphate Adenosine diphosphate (ADP) is an important intermediate in energy transfer in biochemistry. ADP contains an pyrophosphate group, P2O7, which, in the tris(hydroxymethyl)-methylammonium dihydrate salt (CSD entry HMADPH), is described as being doubly ionized, with the counterions being the adenine and the quaternary ammonium ion. Hydrogen atoms were added to the structure given in the CSD and, in a preliminary calculation, their positions were optimized using PM6, the rest of the geometry being fixed at the X-ray structure. One of the hydrogen atoms in the resulting geometry was located between an oxygen of the terminal PO4 group on one ADP ion and a nitrogen atom on an adenine in an adjacent ADP ion. Because these oxygen and nitrogen atoms were only 2.57 Å apart, it is likely that a bridging hydrogen bond exists between them in the crystal. The geometry of the entire system was then optimized. This resulted in an insignificant increase in the N–O distance to 2.58 Å, and gave an RMS error for a single ADP ion of 0.41 Å. This suggests that a better description of the pyrophosphate moiety would be that it is singly ionized and that it is part of a two molecule ADP–adenine system connected by a bridging hydrogen. As with acetylcholine, the structure of ADP depends strongly on its crystal environment: when the geometry of a single ATP ion was optimized using PM6, the structure distorted dramatically and the resulting RMS error relative to the X-ray structure increased to 2.08 Å. Adenosine triphosphate The nucleotide adenosine-triphosphate (ATP) is a source of energy for many biochemical reactions and, as such, its structure and properties are of considerable interest. Only one simple compound of ATP was found in the CSD, disodium adenosine-triphosphate trihydrate, ADENTP. Its structure was very badly reproduced when PM6 was used, with the RMS error of the ATP molecule being 1.18 Å. Examination of the unit cell revealed that the sodium ions had formed spurious weak bonds to nearby hydrogen atoms, and that this was partially responsible for the distortion of the ATP ion. This was confirmed when potassium was used in place of sodium and the RMS error dropped to 0.70 Å. Because of this and other results, there is convincing evidence that PM6 sodium has severe problems when solid state systems are being modeled, and it is highly probable that part of the error in prediction of the structure of ADENTP can be attributed to faults in sodium parameters. Nicotinamide adenine dinucleotide In contrast to ATP, the structure of nicotinamide adenine dinucleotide (NAD) tetrahydrate (CSD entry CEVYEH11) was reproduced with good accuracy, the RMS error being only 0.33 Å. Like ATP, NAD contains a polyphosphate group. Hydrogen bonding Because of the importance of hydrogen bonding in biochemistry, a range of types of hydrogen bond were examined. Most of the important hydrogen bonds in biochemistry involve a proton positioned between either two oxygen atoms, two nitrogen atoms, or an oxygen and a nitrogen atom, the more exotic bonds, such as those involving halogen ions, while interesting, being of secondary importance. Individual types of hydrogen bonds O–H–O A common example of hydrogen bonds is provided by simple organic compounds that contain hydroxyl groups. Examples of such systems are sucrose, aspartic acid, citric acid, glucose, oxalic acid, and the hydroxybenzoic acids. In all such simple hydrogen bonded systems, the geometry predicted by PM6 was similar to that observed. Of more interest are those hydrogen bonds that occur in systems in which relatively large charges are involved.The tendency for a proton to form a bridging structure between two oxygen atoms in organic acids was investigated. Such a situation occurs in the polymeric solid potassium hydrogen acetate, where a single proton is shared between two acetate groups, the whole assembly, two acetates plus the proton, having a formal charge of −1. In catena-((μ4-acetato)-(μ2-acetic acid)-potassium) (CSD entry KHACET02), this structure has the geometry O–O′: 2.49 Å, O–H: 1.09 Å and O′–H 1.40 Å, the three atoms forming a straight line. For this solid, PM6 predicts the following: O–O′: 2.52 Å, O–H: 1.22 Å and O′–H 1.30 Å, with the O–H–O angle being 178°.Solid 4-fluoro-2-(phosphonomethyl)benzenesulfonic acid monohydrate (CSD entry KIXQIR) exists as the Zwitterion. In this system, the sulfonic acid group donates a proton to the phosphono group, giving –[SO3]– and its counterion –[P(OH)3]. One of the hydrogen atoms of the phosphono group then forms a strong hydrogen bond with the nearby water molecule. PM6 incorrectly predicts this bridging hydrogen to be nearer to the water than to the phosphono group. Where the X-ray structure has the PO⋯H distance of 1.05 Å, PM6 predicts 1.48 Å; the corresponding distances for the H⋯OH2 are 1.37 Å, observed, and 1.10 Å, calculated.Another interesting hydrogen bond exists in crystalline acetylacetone (CSD entry LIWPIQ01). In this system the X-ray structure shows that the hydrogen atom involved in hydrogen bonding is disordered over two equivalent positions within a single molecule, with the result that the observed bond lengths and angles are symmetric about the central C–H unit. As expected, when the geometry is optimized using PM6, the extra symmetry is destroyed. This is a consequence of the requirement that, in a quantum chemical calculation, each atom must be in a defined position. However, although the calculated structure was of lower symmetry, when the optimized geometries of the two halves of the molecule were averaged, the result was in good agreement with the X-ray structure: where PM6 predicted the C–O bond to be 1.25 Å and 1.35 Å, the X-ray structure gave 1.28 Å, and where PM6 gave the C2–C3 distance as 1.36 Å and 1.44 Å, the observed value is 1.40 Å. The structure of the hydrogen bond was also well reproduced, with the calculated O–O distance being 2.62 Å compared with the observed 2.54 Å, albeit the calculated O–H bond length, 1.06 Å was much greater than the reported value of 0.92 Å. N–H–N In 1-dimethylamino-8-dimethylammonionaphthalene saccharin dihydrate (CSD entry AJOHUC), saccharine donates a proton to the “proton sponge” 1,8-bis(dimethylamino)-naphthalene to form an ionic crystal. The reported structure has that proton asymmetrically positioned between the two nitrogen atoms: N–N′: 2.56, N–H: 1.35, N′–H: 1.26 Å. The optimized PM6 structure gives N–N′: 2.68, N–H: 1.71, N′–H: 1.11. PM6 thus both underestimates the bridging power of the proton and exaggerates the asymmetry of the bond.A closely related species, 4,5-bis(dimethylamino)-1,8-dihydroxynaphthalene, exists as the Zwitterion in the solid (CSD entry RISBIE). In this system, the geometry of the N–H–N′ structure is symmetric, N–H: 1.27 Å, N–N′: 2.57 Å, and the O–H–O structure is unsymmetric, O–H: 1.00 Å, O–O′: 2.45 Å. PM6 predicts both the N–H–N′ (N–H: 1.12 Å, N–N′: 2.68 Å) and the O–H–O′ structures (O–H: 1.09 Å, O–O′: 2.49 Å) to be unsymmetric. In the gas phase, 4,5-bis(dimethylamino)-1,8-dihydroxynaphthalene would most likely exist as the neutral species; a B3LYP/6–31G(d) calculation predicts the energy of the Zwitterionic form to be 0.6 kcal/mol above that of the neutral form; however, PM6 incorrectly predicts that the Zwitterion should be 15.8 kcal/mol more stable than the neutral form.In 2005, an even stronger proton sponge, 1,8-bis(hexamethyltriaminophosphazenyl)naphthalene, HMPN, was reported [25]. The effect of steric crowding in HMPN arising from the –N = P(N(Me2))3 groups distorts the naphthalene skeleton so that the reported C1–C9–C10–C5 torsion angle, Fig. 3, is 173.9°. The fully optimized PM6 crystal structure predicted this angle to be 170.3°. For the gas-phase structure, B3LYP/6–31G* predicted the torsion angle to be 172.8° [25] while PM6 gave an angle of 165.8°, indicating that the PM6 model was producing either a less rigid naphthalene structure or a greater steric repulsion energy. Fig. 31,8-bis(hexamethyltriaminophosphazenyl)naphthaleneAn additional example of the importance of crystal packing forces is provided by the accuracy of prediction of the structure of HMPN in the gas and solid state phases. Excluding hydrogen atoms, the RMS difference between the B3LYP gas-phase structure and the X-ray structure was 0.286 Å, while for the PM6 crystal structure the equivalent difference was 0.267 Å, and for the PM6 gas-phase structure the RMS difference was 0.677 Å. That is, the B3LYP gas-phase structure was a significantly better fit to the observed crystal structure than that given by PM6, but when crystal forces were included in the PM6 calculation, the PM6 gave a slightly better fit than the B3LYP result.The energetics involved in the sublimation process for HMPN can be modeled in three stages. In the solid phase, PM6 gives a ΔHf of −47.8 kcal/mol for HMPN. Using the optimized geometry for the crystal form, PM6 gives a ΔHf of −17.9 kcal/mol for HMPN in the gas phase. When the geometry is allowed to relax, the ΔHf of the optimized gas-phase geometry decreased to −27.9 kcal/mol. From this it follows that crystal packing forces distort the geometry of HMPN so that its energy increases by 10 kcal/mol. This increase is, however, more than offset by the intermolecular stabilization energy of 29.9 kcal/mol resulting in a net sublimation energy of 19.9 kcal/mol.Because HMPN has a very high proton affinity, it is interesting to speculate about the minimum energy structure of gas-phase 4,5-bis(hexamethyltriaminophosphazenyl)-1,8-dihydroxynaphthalene. As with 4,5-bis(dimethylamino)-1,8-dihydroxynaphthalene, PM6 predicts this system to exist as the Zwitterion in the gas phase, yet, like 4,5-bis(dimethylamino)-1,8-dihydroxynaphthalene, a BLYP/6–31G* calculation suggests that the neutral form is the more stable form, but only by 0.3 kcal/mol. That is, the possibility exists that 4,5-bis(hexamethyltriaminophosphazenyl)-1,8-dihydroxynaphthalene might form a stable gas-phase Zwitterion, but the probability of this being the case is low.An unusual complex resulting from the reaction of elemental bromine and 1, 4-diazabicyclo[2.2.2]octane (DABCO), (DABCO)4⋅Br14 has been reported [26] to contain an almost linear structure of three DABCO groups and seven bromine atoms (Fig. 4) with the bromine atoms split into three fragments of , , and Br−, resulting in a formal charge of −3. From the X-ray structure (CSD entry DAHGUO), the authors concluded that partial protonation had occurred, that the complex of three DABCO units included four protons, and that the remaining isolated DABCO moiety was doubly protonated. Fig. 4Detail of bis(1,4-Diazoniabicyclo(2.2.2)octane) bis(1-aza-4-azoniabicyclo(2.2.2)octane) tetrakis(tribromide) dibromideWhen the structure suggested by the authors was optimized using PM6, only very small changes occurred in the geometry, as shown in Table 4; all the essential features proposed by the original authors were reproduced, and are therefore confirmed. These included the conjectures that, in the set of three DABCO units, the central DABCO was doubly ionized, that the terminal DABCOs were singly ionized, and that the isolated DABCO was doubly charged. The two tribromide ions had charges of −0.98 and −1.00, with the central atom in each group carrying only an insignificant charge, in accordance with the postulated existence of a tribromide anion, and the isolated bromide ion had a partial charge of −0.78, again in accordance with the proposed structure. Table 4Interatomic distances and charges in bis(1,4-Diazoniabicyclo(2.2.2)octane) bis(1-aza-4-azoniabicyclo(2.2.2)octane) tetrakis(tribromide) dibromideInteratomic distancesChargesX-rayPM6PM6N1-N22.692.70DABCO1+1.01N2-H1.16DABCO2+1.68N3-H1.16DABCO3+1.01N3-N42.662.70DABCO4+1.81N5-Br13.172.97Br1−0.78Br1-Br23.603.00Br2−0.54Br2-Br32.512.48Br3−0.05Br3-Br42.592.50Br4−0.39Br4-Br53.252.98Br5−0.45Br5-Br62.452.47Br6−0.09Br6-Br72.692.53Br7−0.47 N–H–O The simple amino acids exist as Zwitterions in the solid state. In this form, an –NH3+ group on one ion electrostatically interacts strongly with a –COO– ion on a neighboring ion. In all cases examined, the structure of the resulting salt was reproduced with good accuracy. This included the structures of three polymorphs of glycine, α, β, and γ, a set of systems in which any differences in heats of formation could arise only from the different intermolecular interactions. PM6 predicted all three polymorphs to have heats of formation within 2 kcal/mol of each other, albeit not in the order observed experimentally. Of more interest are those cases where the energy of interaction is smaller and, consequently, the geometry of the N–H–O system would be more sensitive to the environment. A good example of such a system is methyl (+ −)-(1α,2β,8α,9α,10β)-2-chloro-4-aza-3-oxatetracyclo(8.4.0.02,9.04,8) tetradecane-9-carboxylate oxalic acid monohydrate (CSD entry HUZKOC), in which a molecule of oxalic acid hydrogen bonds to a neutral nitrogen on the large organic fragment. In this system, PM6 predicts the N–O distance to be slightly too large, 2.61 Å compared to the reported 2.57 Å.The existence of a uniquely short heteroatom separation in a hydrogen-bonded compound, 2-(2-(3-carboxypyridyl))-4-isopropyl-4-methyl-5-oxo-imidazole, Fig. 5, was reported in 1989 [27]. In this system (CSD entry JAZCOC01), the pyridine and imidazole rings are almost coplanar, a condition essential to the formation of the strong hydrogen bond. The optimized gas-phase geometry predicted by AM1 and PM3, and now PM6, all have large twist angles, 47, 37, and 40°, respectively. In the solid state, however, PM6 predicts the rings to be coplanar, and the N–O distance to be 2.50 Å, which is very close to the reported 2.47 Å. In the observed crystal structure the hydrogen atom is asymmetrically positioned, with the O-H and N-H distances being 1.16 Å and 1.32 Å, and the O–H–N angle being 170°. PM6 also predicts the hydrogen to be asymmetrically positioned, but in the opposite sense: the PM6 O–H and N–H distances being 1.46 and 1.12 Å, and the O–H–N angle to be 150°. Fig. 52-(2-(3-Carboxypyridyl))-4-isopropyl-4-methyl-5-oxo-imidazoleThe tendency of PM6 to incorrectly favor the Zwitterionic form of a hydroxy–amine over the neutral was investigated by modeling the three isomers of hydroxyaniline. All three isomers exist in the solid state as the neutral form; however, on optimizing the geometry using PM6, the ortho and para forms spontaneously transitioned to the Zwitterion. Meta-hydroxyaniline optimized to the neutral form, but when the calculation was repeated, starting with the Zwitterionic form, the structure optimized to the Zwitterionic form, and in that form was 1.1 kcal/mol lower in energy than the neutral form. As expected, the density increased from 1.27 to 1.34 on going from the neutral to the Zwitterion. PM6 thus has a definite and demonstrable error in its exaggerated tendency to form Zwitterions.DABCO forms several very unusual hydrogen-bonded systems, among which are 1,4-diazabicyclo[2.2.2]octane azelaic acid (CSD entry UNEGEZ), where each DABCO forms two different types of bridging N–H–O bonds to adjacent azelaic acid molecules. PM6 predicts one of these to have the structure N–O: 2.53(2.55), N–H: 1.29(1.21), and O–H: 1.27(1.40), (X-ray structures in parentheses) and the other to be N–O: 2.75(2.61), N–H: 1.75(1.50), and O–H: 1.06(1.11). The X-ray structure shows the existence of a polymeric chain of alternating azelaic acid and DABCO units, from which it follows that the different geometries of the two N–H–O structures can only be attributed to crystal packing forces. A closely related compound is 1-diazonia-4-azabicyclo[2.2.2]octane glutarate (CSD entry UNEFIC), in which a similar structure exists but now with the bridging hydrogen being nearer to the nitrogen than to the oxygen. In UNECIF, the N–H distance predicted by PM6 was 1.10 (0.95) Å and the O–H 1.65 (1.76) Å. π−π stacking π−π stacking occurs in the polycyclic aromatic hydrocarbons where it is the result of Van der Waals (VDW) interactions between the rings. In general, VDW interactions are weaker than hydrogen bonding interactions, and, historically, have been the hardest to model using semiempirical methods. Thus, when the default single determinant wavefunction is used, VDW terms are completely absent, and, in order to mimic the effects of VDW attraction, the normal procedure is for modifications to be made to the core–core interaction. An estimate of how accurately PM6 can reproduce the VDW interaction is provided by anthracene (CSD entry ANTCEN14), benzene (CSD entry BENZEN), and coronene (CSD entry CORONE). In anthracene, the molecules are stacked in a staggered arrangement. While PM6 reproduces the density with good accuracy, the optimized structure predicts the parallel sheets of anthracene molecules to be separated by 5.37 Å rather than the observed 2.83 Å, and pairs of anthracene molecules forming “T” structures rather than the observed “V” configuration. Conversely, both benzene and coronene crystallize with a perfect herringbone packing, and this structure is reproduced with very good accuracy by PM6, the calculated and observed inter-plane distance being essentially identical. Very weak interactions In small saturated hydrocarbons the strongest intermolecular interaction energy arises from instantaneous correlation or VDW forces. A consequence of this is that such compounds are, in general, highly volatile and most have very low melting points. Another characteristic of compounds of this type is that intermolecular separations are typically very large: in cyclohexane the smallest intermolecular separation is about 2.6 Å. When the structure of cyclohexane was optimized from the experimental structure using PM6, the final and X-ray structures agreed almost exactly. However, the heat of formation of each cyclohexane decreased from an initial +65.8 kcal/mol for the starting X-ray structure to the final −29.6 kcal/mol for the optimized PM6 structure, reflecting the relaxation of the X-ray C–H bond lengths, which are normally too short. However, the PM6 ΔHf for isolated cyclohexane is predicted to be −27.5 kcal/mol, from which it follows that the energy of interaction amounted to only 2.1 kcal/mol, much less than the reported 9.0 kcal/mol. The implication is that, although PM6 was able to reproduce the observed crystal structure, the magnitude of the VDW interaction was grossly underestimated. In solid methionine (CSD entry LMETON02) the molecules form double layers with the hydrophobic end, –CH2–S–CH3, on the outside. The X-ray structure shows that these layers are separated by about 2.21 Å; PM6 predicts the inter-layer separation to be much larger, 2.83 Å. Further investigation of the properties of sulfur predicted by PM6, see below, indicated the lack of any sulfur–sulfur VDW attraction. This deficiency is likely responsible for the unrealistically large inter-layer separation. Most acetylacetonato transition metal complexes, such as tris(acetylacetonato) titanium(iv) perchlorate (CSD entry TIACPC), adopt an almost octahedral coordination of the oxygen atoms around the central metal ion. An exception is the hexamethyl acetylacetone complex of yttrium(iii), tris(2,2,6,6-tetramethylheptane-3,5-dionato)-yttrium(iii) (CSD entry HAHTOZ01), where the oxygen atoms form a trigonal prism. This structure was reproduced by PM6. The deviation from the conventional octahedral structure cannot be attributed to crystal packing forces—when the geometry of the isolated complex ion was optimized using PM6, the trigonal prism structure was retained. However, when the geometry of the unsubstituted complex, tris(acetylacetonato) yttrium(iii), was optimized, the expected D3d structure resulted, suggesting that the likely driving force was steric crowding arising from the tertiary butyl groups. Polymorphs 5-Methyl-2-((2-nitrophenyl)amino)-3-thiophenecarbonitrile is unique in the CSD in that there are seven distinct polymorphs [28]; because three of these are red, orange, and yellow, this chemical is commonly referred to as ROY. To test the suitability of PM6 for modeling polymorphs, the structure of each of the polymorphs of ROY was optimized using PM6, starting with the X-ray geometry. In every case, the optimized PM6 structure was qualitatively the same as the X-ray structure; that is, the crystal packing arrangement was preserved. Within each unit cell, the structures of the individual molecules were reproduced with good accuracy (Table 1). Although the calculated heats of formation of the X-ray structures spanned a range of 31 kcal/mol, the optimized PM6 structures all had similar heats of formation, spanning a range of 4.3 kcal/mol, as would be expected for polymorphs. Co-crystals Designing crystal structures for active pharmaceutical ingredients presents an important challenge to crystal engineers. One promising avenue of research involves designing co-crystals, so determining the suitability of PM6 as a tool for this work is of obvious interest. The structures of several co-crystals were optimized using PM6; all the resulting geometries were in good agreement with the structures found in the CSD. In all of the systems investigated the two components were held together by hydrogen bonds. Thus in the bis-urea–oxalic acid co-crystal (Fig. 6, CSD entry UROXAL01), each oxalic acid forms two strong O⋯H⋯O hydrogen bonds to the keto groups of the neighboring urea molecules, this resulting in a tri-molecular unit. In turn, these units pack together in the crystal using weaker N⋯H⋯O hydrogen bonds. PM6 predicts that the co-crystal would be significantly more stable than its two separate precursors: the predicted ΔHf of the co-crystal is −313.4 kcal/mol, while the sum of the ΔHf of the precursors, Table 2, is −306.9 kcal/mol. Fig. 6Crystal structure of the co-crystal of urea and oxalic acid Another example of such a co-crystal is provided by isonicotinamide 3-hydroxybenzoic acid, Fig. 7 (CSD entry LUNMEM), in which each hydroxybenzoic acid forms hydrogen bonds to three adjacent isonicotinamide molecules in a complicated three-dimensional structure. As with the previous co-crystal, PM6 predicts LUNMEM to be more stable than its precursors but, in this case, only by 0.1 kcal/mol. Fig. 7Co-crystal of isonicotinamide and 3-hydroxybenzoic acid showing PM6 hydrogen-bond lengths (X-ray in parenthesis) Metal-containing species Many organic compounds that contain elements of Groups I or II are ionic salts. A good example is calcium oxalate, which forms three well characterized solids: whewellite, (Ca(C2O4)·(H2O)), weddellite, (Ca(C2O4)·2(H2O)), and caoxite, (Ca(C2O4)·3(H2O)), at least one of which the author involuntarily prepared in vivo, the separation of which from the surrounding organic material was accompanied by severe algia. The structures of all three minerals are reproduced with good accuracy. Anhydrous disodium oxalate exists as the mineral natroxalate. In contrast to the calcium oxalates, the structure predicted by PM6 was completely different to that observed: within each oxalate dianion, one carboxylate group was rotated by ∼90° to give an approximately D2d structure. More seriously, some sodium–sodium distances became unrealistically short, 1.23 Å instead of the observed 3.30 Å. This specific error is attributable to the faulty PM6 values of the Na–Na core–core parameters. Solids containing metal complex ions can be regarded as salts: that is, as molecular metal complexes, cations or anions, plus counterions. An example is bromo-tris(2-dimethylaminoethyl)amine-manganese(ii) (CSD entry DAEAMN), where the metal complex consists of the neutral tris(2-dimethylaminoethyl)amine that chelates a manganese dication, the whole complex behaving like a large dication. Two bromide counterions are present for each such complex ion in order to maintain electroneutrality. PM6 predicts the single N–Mn distance to be 2.09 Å versus 2.19 Å in the X-ray structure, and the three N–Mn to be 1.96 Å versus 2.27 Å in the X-ray. Inorganic compounds In general, most inorganic solids differ from crystalline organic compounds in that they do not involve discrete molecules. Instead, they exist as extended covalently or ionically bound infinite systems. A consequence of this is that identification of simple structural units in inorganic solids is often either difficult or impossible. A more subtle consequence is that the band-structure of inorganic solids is usually more complicated than those of organic solids: for the same reciprocal distance, bands arising from inorganic solids generally have a much greater curvature than those for organic solids, this being a consequence of the strong bonds that extend throughout such solids. Conversely, in most organic solids there is an intermolecular gap that effectively confines molecular orbitals to individual molecules. This means that the band-structure of most organic compounds consists of relatively flat bands. In the cluster method, the Γ point represents the entire Brillouin zone so that a much larger cluster must be used when inorganic solids that are not composed of discrete molecules are modeled. In practice this means that the cluster used has to be large enough to contain a sphere of radius 10–12 Å, in contrast to the 7–8 Å used in modeling organic solids. A survey of the applicability of PM6 to a wide range of inorganic solids was carried out, using structural reference data obtained from the American Mineralogist Crystal Structure Database [29]. As with organic crystals, the starting geometry was the X-ray structure. But, in contrast to most organic solids, many inorganic solids had to be modified before a PM6 calculation could be started. Semiempirical calculations require that a definite structure be used. However, some minerals, such as forsterite, (FeII,Mg)2SiO4, are of variable composition, with one or more sites occupied by the two types of metal atoms at random. Before a calculation on such a system can be performed, all variable atoms have to be replaced by definite atoms. In the case of forsterite, all iron atoms were replaced by magnesium atoms. In a few solids there was still disorder in the lattice although the formula was stoichiometric. In spinel, for example, the formula is MgAl2O4 but, in the observed crystal structure, a small fraction of the aluminum sites are replaced by magnesium atoms, and vice versa. Of necessity, the model used in the PM6 calculation was idealized so that sites that were 80% magnesium were made 100% magnesium, and sites that were 90% aluminum were made 100% aluminum. Although this idealized structure does not occur naturally it could be considered a good approximation to the observed structure. Some minerals that contain hydrogen atoms did not have the positions of these atoms reported. In those cases a preliminary calculation was carried out in which the hydrogen atoms were positioned in likely sites, and then the positions of those atoms optimized. During this operation, the positions all the other atoms were fixed at the experimental values. The results can conveniently be partitioned according to the types of solids involved and sequenced in order of complexity. In most cases, the calculated structure was similar to the X-ray structure in that no bonds were made or broken as a result of geometry optimization. Unit cell parameters for those solids where the calculated PM6 structure inside the unit cell was substantially similar to the reference are presented in Table 5 and Table 6, and heats of formation are given in Table 7. Table 5Calculated and X-ray structural parameters for inorganic compounds. Values of a, b, c are in Ångstroms; α, β, γ in degreesPM6X-rayDensity (g/cm3)dabcαβγabcαβγElementsDiamond (C)3.593.593.5990.090.090.03.573.573.5790.090.090.03.44 (3.52)Glitter (C)e2.602.605.9690.190.190.02.602.605.9090.090.090.02.98 (3.08)Graphite (C)2.472.4787.1789.690.060.02.462.466.7090.090.060.02.12 (2.28)Silicon (Si)5.115.115.1190.090.090.05.435.435.4390.090.090.02.78 (2.33)Sulfur (S)11.7714.0126.8190.490.289.810.4312.8324.3690.090.090.01.54 (2.09)Group I HalidesCesium bromide (CsBr)4.354.354.3590.090.090.04.304.304.3090.090.090.04.29 (4.43)Cesium chloride (CsCl)4.174.184.1890.389.889.74.124.124.1290.090.090.03.84 (3.99)Cesium fluoride (CsF)6.026.026.0290.090.090.06.016.016.0190.090.090.04.62 (4.64)Cesium iodide (CsI)4.734.734.7390.090.090.04.574.574.5790.090.090.04.07 (4.51)Potassium bromide (KBr)6.796.756.7587.690.090.06.616.616.6190.090.090.02.55 (2.74)Potassium chloride (KCl)6.126.126.1290.090.090.06.296.296.2990.090.090.02.16 (1.99)Potassium fluoride (KF)5.035.035.0390.090.090.05.385.385.3890.090.090.03.02 (2.48)Potassium iodide (KI)7.177.177.1790.190.090.07.077.077.0790.090.090.02.99 (3.12)Lithium bromide (LiBr)5.495.485.4890.090.090.05.505.505.5090.090.090.03.50 (3.46)Lithium chloride (LiCl)5.135.135.1390.090.090.05.145.145.1490.090.090.02.09 (2.07)Lithium fluoride (LiF)4.034.044.0590.089.990.04.034.034.0390.090.090.02.62 (2.64)Lithium iodide (LiI)6.046.036.0490.090.090.06.036.036.0390.090.090.04.04 (4.06)Sodium bromide (NaBr)6.276.206.2085.990.090.06.046.046.0490.090.090.02.84 (3.10)Sodium chloride (NaCl)6.246.246.2490.090.090.05.605.605.6090.090.090.01.60 (2.21)Sodium fluoride (NaF)4.794.794.7990.290.189.84.614.614.6190.090.090.02.54 (2.85)Sodium iodide (NaI)7.027.027.0290.090.090.06.476.476.4790.090.090.02.87 (3.67)Rubidium bromide (RbBr)6.876.886.8890.390.090.06.906.906.9090.090.090.03.38 (3.35)Rubidium chloride (RbCl)6.706.686.6889.190.090.06.636.636.6390.090.090.02.69 (2.76)Rubidium fluoride (RbF)5.865.865.8690.090.090.06.016.016.0190.090.090.03.45 (3.20)Rubidium iodide (RbI)7.847.767.7683.390.090.07.357.357.3590.090.090.03.01 (3.55)HalidesCalcium iodide (CaI2)6.214.484.48119.990.090.16.964.484.48120.090.090.04.51 (4.04)Carnallite (KMg(H2O)6Cl3)15.8622.3110.2473.192.895.916.1222.479.5590.090.090.01.61 (1.60)Chloromagnesite (MgCl2)19.403.753.75119.989.190.917.593.603.60107.990.090.02.00 (2.41)Cryolite (Na3AlF6)5.678.035.6990.190.089.95.467.805.6190.090.090.22.69 (2.92)Fluorite (CaF2)5.495.455.4890.090.090.05.465.435.4690.090.090.03.17 (3.20)Frankdicksonite (BaF2)6.226.226.2390.090.090.06.206.206.2090.090.090.04.83 (4.89)Hydrophilite(CaCl2)4.286.426.3889.990.090.04.206.436.2490.090.090.02.10 (2.19)Magnesium dibromide (MgBr2)5.834.024.02119.191.688.36.263.813.81120.090.090.03.71 (3.88)Magnesium diiodide (MgI2)6.284.274.24118.290.689.66.884.144.14120.090.090.04.61 (4.52)Na3(TiCl6)10.237.357.0090.087.189.89.817.096.6890.089.790.02.08 (2.36)Neighborite (NaMgF3)8.148.138.1390.090.090.17.677.677.6791.490.090.02.57 (3.07)Sal ammoniac (NH4Cl)4.074.073.6890.089.973.53.863.863.8690.090.090.01.52 (1.54)Sellaite (MgF2)3.074.624.6290.090.090.03.054.634.6390.090.090.03.16 (3.17)Oxidesα Crystobalite (SiO2)5.365.367.4789.990.189.64.984.986.9590.090.090.01.86 (2.32)α Quartz (SiO2)5.195.185.7490.090.0120.04.914.915.4190.090.0120.02.24 (2.65)Anatase (TiO2)10.293.923.9290.190.289.69.523.783.7890.090.090.03.36 (3.90)Arkelite (ZrO2)6.094.915.5090.089.990.15.135.135.1390.090.090.04.97 (6.06)Barium oxide (BaO)5.645.645.6489.890.090.05.635.635.6390.090.090.05.69 (5.72)β Quartz (SiO2)5.195.185.7490.090.060.05.005.005.4690.090.060.02.24 (2.54)β Tridymite (SiO2)5.345.348.7290.090.0119.95.055.058.2790.090.0120.01.86 (2.18)Brookite (TiO2)5.425.679.6190.090.490.05.145.459.1790.090.090.03.59 (4.13)Cassiterite (SnO2)4.944.963.4990.090.089.94.744.743.1990.090.090.05.86 (7.00)Chabazite (SiO2)9.559.569.5693.993.993.89.409.409.4094.394.394.31.38 (1.46)Coesite (SiO2)7.4312.857.4990.060.890.07.1712.377.1490.059.790.02.56 (2.92)Claudetite (As2O3)14.005.435.0689.885.881.812.874.545.2586.290.090.03.46 (4.29)Corundum (Al2O3) (a = b = c,α = β = γ)5.135.125.1256.256.256.25.135.135.1355.355.355.33.91 (3.99)Corundum (Al2O3) (a = b ≠ c,α = β ≠ γ)4.834.8312.9190.090.0120.04.764.7612.9990.090.0120.03.91 (3.99)Ice-I (Ice-Ih)7.454.367.0990.590.389.37.834.527.3890.090.090.01.04 (0.92)Ice-II7.2212.265.9189.9106.591.07.7812.986.2490.0105.590.01.43 (1.18)Ice-III6.696.506.4689.989.590.76.736.736.7390.090.090.01.28 (1.18)Ice-V7.029.448.8070.190.692.67.5310.359.2070.790.089.11.53 (1.24)Ice-VI5.385.945.9090.989.589.05.706.186.1890.090.090.01.59 (1.37)Ice-VIII6.444.364.3690.090.590.16.414.454.4590.090.090.01.96 (1.89)Ice-X (0Gpa)2.902.902.9090.090.090.02.732.732.7390.090.090.02.46 (2.96)Ice-X (62Gpa)2.722.722.7290.090.090.02.732.732.7390.090.090.02.98 (2.96)Ice-XI7.624.267.1590.190.289.97.834.527.3890.090.090.01.03 (0.92)Ice-XIII9.777.058.4690.569.590.610.297.479.2490.070.390.01.54 (1.25)Ice-XIV3.838.077.8490.390.090.34.088.358.1490.090.090.01.48 (1.29)Keggin Dodecatungstophosphoric acid hexahydrate H3(PO4@W12O36).6(H2O)13.0113.4313.0090.492.289.712.5112.5112.5190.090.090.04.37 (5.07)Lime (CaO)4.954.954.9590.090.190.04.824.824.8190.090.090.03.07 (3.34)Mordenite (SiO2)18.6821.007.7590.290.090.018.1320.497.5290.090.090.01.58 (1.71)Lead oxide (PbO)3.834.743.8389.990.089.93.985.023.9890.090.090.010.68 (9.34)Periclase (MgO)4.294.294.2990.090.090.04.224.224.2290.090.090.03.40 (3.57)Perovskite (CaTiO3)5.385.547.8489.789.990.15.385.447.6490.090.090.03.87 (4.04)Plattnerite (PbO2) (Rutile structure)3.504.664.7389.890.090.03.394.964.9690.090.090.010.31 (9.55)Rutile (TiO2)4.844.843.1290.090.090.04.594.592.9690.090.090.03.71 (4.25)Stishovite (SiO2)2.824.384.3890.090.090.02.674.184.1890.090.090.03.70 (4.28)Zincite (ZnO)3.333.335.2690.090.060.33.253.255.2090.090.060.05.33 (5.68)SpinelsChrysoberyl ((BeO)(Al2O3))4.589.585.7290.390.090.04.409.335.4490.090.090.03.36 (3.78)Spinel (MgAl2O4)8.3116.6316.6390.290.090.08.2116.4116.4190.090.090.03.29 (3.42)BoratesBoracite (MgO)5(MgCl2)(B2O3)78.548.5912.1190.090.090.08.558.5512.0990.090.090.02.93 (2.95)Borax (Na2B4O7·10(H2O))12.3311.2716.20105.193.783.012.1910.7411.8990.073.490.01.18 (1.70)Parahilgardite (Ca2(B5O9)Cl(H2O))17.196.396.18119.699.884.617.506.496.31119.2100.484.02.84 (2.69)Ulexite (NaCaB5O9·8(H2O))6.7713.628.8897.4108.2103.36.6812.878.82110.0109.190.41.82 (1.96)CarbonatesAragonite (CaCO3)4.557.806.0890.090.090.14.967.975.7490.090.090.03.08 (2.93)Azurite ((CuO)3(CO2)2(H2O))5.735.129.5091.090.190.05.855.0110.3492.490.090.04.10 (3.78)Calcite (CaCO3)6.326.326.3245.245.245.26.366.366.3646.146.146.12.87 (2.73)Cerussite (PbCO3)5.846.43 7.0989.391.489.35.186.148.4990.090.090.06.66 (6.57)Dawsonite ((Na2O)(Al2O3)(CO2)2(H2O)2)6.735.985.7590.090.290.06.715.215.5890.090.090.02.07 (2.45)Dolomite (CaMg(CO3)2)4.744.7415.3290.089.9120.04.814.8116.0290.090.0120.03.08 (2.86)Huntite (CaMg3 (CO3)4)9.409.357.6790.590.259.89.509.507.8290.090.060.03.02 (2.88)Ikaite ((CaO)(CO2)(H2O)6)8.727.8510.8191.371.390.38.878.2311.0290.069.890.01.97 (1.83)Kalicinite(KHCO3)19.634.164.1690.977.890.215.195.633.7190.075.590.02.00 (2.17)Magnesite (MgCO3)14.914.577.9590.090.290.015.024.638.0390.090.090.03.11 (3.01)Natrite (Na2CO3)9.936.815.6690.290.0102.08.916.045.2490.090.0101.31.88 (2.55)Thermonatrite (Na2CO3(H2O))5.8511.817.8190.089.391.65.2610.726.4790.090.090.01.53 (2.26)Otavite (CdCO3)5.564.7115.5990.088.990.94.934.2716.2790.090.090.04.21 (5.02)Smithsonite (ZnCO3)4.834.8314.4689.590.0120.04.654.6515.0390.090.0120.04.28 (4.44)Strontianite (SrCO3)6.598.655.0290.190.090.06.008.365.0990.090.090.03.42 (3.84)Witherite (BaCO3)4.997.268.4989.989.989.95.316.438.9090.090.090.04.26 (4.32)Zabuyelite (Li2CO3)5.268.806.2367.590.090.04.978.366.2065.290.090.01.84 (2.10)Group IVMoissanite (SiC)5.005.363.0990.090.090.05.055.333.0890.090.090.03.22 (3.22)Silicon carbide (SiC)4.394.394.3990.090.090.04.354.354.3590.090.090.03.16 (3.24)Groups III-VAluminum antimonide (AlSb)5.945.945.9490.090.090.06.146.146.1490.090.090.04.71 (4.28)Aluminum arsenide (AlAs)5.915.915.9190.090.090.05.665.665.6690.090.090.03.27 (3.73)Aluminum nitride (AlN)4.404.404.4090.090.090.04.374.374.3790.090.090.03.20 (3.25)Aluminum phosphide (AlP)5.495.495.4990.090.090.05.425.425.4290.190.190.12.33 (2.42)Boron Nitride (BN)3.643.643.6490.090.090.03.613.613.6190.090.090.03.42 (3.49)Gallium arsenide (GaAs)5.665.665.6690.090.090.05.655.655.6590.090.090.05.31 (5.32)Indium arsenide (InAs)5.975.975.9790.090.090.06.066.066.0690.090.090.05.93 (5.67)Group VICadmium telluride (CdTe)5.585.585.5890.090.090.06.366.366.3690.090.090.09.17 (6.20)Coloradoite (HgTe)6.616.616.6190.090.090.06.326.326.3290.090.090.07.56 (8.63)Lead selenide (PbSe)4.344.344.3590.090.090.06.176.176.1790.090.090.023.18 (8.10)Lead telluride (PbTe)7.877.877.8790.090.090.07.107.107.1090.090.090.04.56 (6.22)Stibnite (Sb2S3)10.674.0511.4090.087.990.011.293.8311.2190.090.090.04.58 (4.66)Stilleite (ZnSe)5.955.955.9590.090.090.05.545.545.5490.090.090.04.56 (5.65)HydridesBH3NH34.824.245.2594.490.190.14.994.895.3990.090.090.00.96 (0.78)HydroxidesBrucite (Mg(OH)2)3.203.214.3089.890.0120.13.143.144.7790.090.0120.02.53 (2.38)Diaspore (AlO(OH))4.179.753.0090.089.989.84.409.432.8590.090.090.03.26 (3.38)Gibbsite (Al(OH)3)9.0410.384.4590.190.090.09.7410.164.3490.085.590.02.48 (2.42)MnO(OH)7.006.2313.4590.288.489.85.295.8013.3189.789.690.02.99 (4.29)Orthoboric acid (B(OH)3)10.977.267.26120.183.4100.46.357.047.02119.878.587.50.84 (1.56)NitratesAmmonium nitrate (NH4NO3)8.8213.8510.7695.889.472.19.4912.2310.7690.089.991.31.71 (1.70)Niter (KNO3)8.965.386.5990.689.189.59.165.416.4390.090.090.02.11 (2.11)Nitratine (NaNO3)18.665.434.0890.196.290.516.835.074.3990.090.090.02.06 (2.26)PhosphatesBobierrite (Mg3(PO4)2(H2O)8)10.4626.994.5889.4104.393.010.0727.934.6790.0105.090.02.16 (2.13)Calcium phosphate (Ca3(PO4)2)5.5319.975.3488.756.990.35.2518.675.2590.060.090.03.13 (3.47)Fluorapatite (Ca5(PO4)3F)9.289.296.8789.890.059.69.379.376.8890.090.060.03.28 (3.20)Dorfmanite (Na2HPO4·2(H2O))8.4911.7715.9089.990.290.16.6010.3616.8790.090.090.01.48 (2.05)Lithiophosphate (Li3PO4)5.496.7511.6290.090.190.14.936.1210.5390.090.090.01.79 (2.42)Newberyite (MgHPO4·3(H2O))9.8710.0610.3290.090.090.410.0210.6810.2090.090.090.02.26 (2.12)Phosphammite ((NH4)2HPO4)7.756.7410.9189.966.490.18.036.7011.0490.066.690.01.68 (1.61)Potassium dihydrogen phosphate (KH2PO4)7.697.577.5988.091.088.76.977.457.4590.090.090.02.05 (2.33)Potassium oxovanadium(IV) phosphate(V) (KVPO5)7.667.718.8190.289.890.07.617.318.4190.090.090.02.56 (2.85)Sodalite (AlPO4)8.328.338.6089.490.590.49.049.049.0490.090.090.02.04 (1.64)Wagnerite (Mg2PO4F)12.1112.429.6590.074.690.011.9612.689.6490.071.790.03.09 (3.11)SulfatesAlum (KAl(SO4)2·12(H2O))12.3112.3512.2789.389.890.512.1812.1812.1990.090.090.01.69 (1.74)Aluminite (Al2SO6 9(H2O))6.9015.3311.2990.2108.090.07.4415.5811.7090.0110.290.02.01 (1.80)Alunite (KAl3(SO4)2(OH)6)7.017.0017.3889.989.760.26.966.9617.3590.090.060.02.79 (2.84)Ammonium sulfate (2(NH4)SO4)10.995.757.2991.090.190.210.645.997.7890.090.090.01.90 (1.77)Anglesite (PbSO4)7.406.337.0092.491.495.16.968.485.4090.090.090.06.18 (6.32)Anhydrite (CaSO4)6.798.397.0987.190.090.26.996.257.0090.090.090.02.24 (2.96)Arcanite (K2SO4)7.559.626.3889.990.190.27.4810.075.7690.090.090.02.50 (2.67)Barite (BaSO4)7.445.379.3490.189.990.07.155.468.8890.090.090.04.15 (4.47)Celestine (SrSO4)8.435.676.9390.189.891.18.395.366.8990.090.090.03.68 (3.94)Epsomite (MgSO4 (H2O)7)11.8211.816.1989.990.090.111.8912.016.8690.090.090.01.89 (1.67)Glauberite (CaNaSO4)8.4110.5910.1850.590.190.18.318.5310.1376.790.090.02.64 (2.78)Gypsum (CaSO4·2H2O)5.5814.426.7190.2106.190.25.6715.106.4990.0118.590.02.21 (2.34)Hanksite (KCl)2(Na2O)22(SO3)18(CO2)411.6511.7621.8788.090.5119.210.4710.4621.1990.090.0120.01.99 (2.59)Hexahydrite (MgSO4·6(H2O))24.306.929.6589.878.290.024.417.2110.1190.081.790.01.91 (1.72)Langbeinite (K2Ca2(SO4)3)10.7010.3810.5889.389.090.210.4310.4310.4390.090.090.02.52 (2.62)Leonite (K2Mg(SO4)2·4(H2O))10.738.9913.0489.995.690.09.859.4711.7890.084.790.01.95 (2.23)Mercallite (KHSO4)9.8517.858.71590.290.090.69.8018.968.4190.090.090.02.36 (2.32)Thenardite (Na2SO4)6.3413.2811.3190.291.190.25.8712.309.8390.090.090.01.98 (2.66)Zinkosite (ZnSO4)4.936.368.8590.186.690.14.776.758.6090.090.090.03.87 (3.87)SulfidesCinnabar (HgS)4.314.319.8790.090.060.04.154.159.5190.090.060.07.30 (8.17)Galena (PbS)5.875.875.8790.090.090.05.945.945.9490.090.090.07.86 (7.60)Greenockite (CdS)6.707.154.1390.089.989.96.757.164.1390.090.090.04.85 (4.80)Hawleyite (CdS)3.843.843.8490.090.090.05.835.835.8390.090.090.05.02 (4.84)Metacinnabar (HgS)6.296.296.2990.090.090.05.855.855.8590.090.090.06.21 (7.71)Niningerite (MgS)5.205.205.2090.090.090.05.205.205.2090.090.090.02.66 (2.66)Sphalerite (ZnS)5.415.415.4190.090.090.05.435.435.4390.090.090.04.08 (4.04)Wurtzite (ZnS)3.843.846.3090.090.090.03.813.816.2390.090.060.04.03 (4.12)Vanadates, tungstates, chromates, molybdatesDescloizite (PbZn(VO4)(OH))8.366.309.3594.788.990.87.606.079.4490.090.090.05.48 (6.17)Ferberite (FeWO4)5.005.354.6489.995.489.94.955.704.7390.090.090.08.17 (7.55)Phoenicochroite (Pb2CrO5)5.677.7415.6742.191.888.25.677.1414.0064.890.090.07.90 (7.07)Scheelite (CaWO4)5.455.7314.4391.087.892.45.245.2411.3890.090.090.04.25 (6.12)Tarapacaite (K2CrO4)10.576.057.7089.989.890.310.395.927.6690.090.090.02.62 (2.74)Vanadinite (Pb5(VO4)3Cl)7.7811.9716.7487.492.391.67.3410.3317.8990.090.090.06.04 (6.93)Wulfenite (PbMoO4)5.6212.495.5990.390.089.25.4712.185.4790.090.090.06.22 (6.69)Zincochromite (ZnCr2O4)8.408.408.4090.090.090.08.338.338.3390.090.090.05.24 (5.36)Zirconium tungstate (ZrW2O8)9.439.499.5090.289.789.89.189.189.1890.090.090.04.59 (5.04)dX-ray densities in parenthesiseSee [31]Table 6Calculated and X-ray structural parameters for silicates. Values of a, b, c are in Ångstroms; α, β, γ in degreesPM6X-rayDensitydabcαβγabcαβγCyclosilicateBenitoite (BaTiSi3O9)6.7611.609.8290.090.390.06.6011.439.7190.090.090.03.57 (3.75)Beryl (Be3Al2Si6O18)19.0919.069.0290.090.0120.018.4318.439.2490.090.090.02.51 (2.63)Dravite (NaMgTi2Al6(BO3)3(Si6O18)O4)16.1416.137.3989.989.960.215.9515.957.2190.090.060.02.99 (3.14)InosilicateCummingtonite (Mg7Si8O23·(H2O))9.0618.385.6790.0100.090.09.5118.195.3390.0101.990.02.79 (2.87)Diopside (MgCaSi2O6)5.369.119.2290.076.390.05.258.909.7590.074.490.03.29 (3.28)Enstatite (MgSiO3)5.388.9617.6990.189.990.05.188.8218.2390.090.090.03.13 (3.20)Jadeite (NaAlSi2O6)8.689.085.7290.1106.290.19.478.615.2490.0107.690.03.10 (3.30)Johannsenite (CaMnSi2O6)9.795.269.7975.690.290.19.165.299.9874.590.090.03.36 (3.52)Kosmochlor (Si2CrNaO6)5.699.3910.0790.076.290.25.278.729.5890.072.690.02.89 (3.59)Spodumene (LiAlSi2O6)9.526.118.0464.086.287.48.395.229.4669.990.090.02.95 (3.18)Tremolite (Ca2Mg5Si8O22(OH)2)9.3018.835.5890.177.490.09.8418.055.2790.075.390.02.83 (2.98)Wollastonite (CaSiO3)9.1611.127.63100.0102.182.710.1211.077.3199.5100.583.43.10 (2.93)NesosilicateÅkermanite (Ca2MgSi2O7)7.917.734.7790.087.290.17.847.845.0190.090.090.03.11 (2.94)Andalusite (Al2SiO5)6.137.965.9097.185.987.95.567.907.8090.090.090.03.78 (3.14)Chromium silicate (Cr2SiO4)12.5211.385.7789.890.289.09.5911.175.7090.090.090.03.17 (4.26)Euclase ((BeO)2(Al2O3)(SiO2)2(H2O))4.8315.124.4190.099.990.04.7814.324.6390.0100.390.03.03 (3.09)Forsterite (Mg2SiO4)4.776.1010.2390.189.990.14.766.0010.2290.090.090.03.14 (3.20)Garnet (Al2O3(CaO)3(SiO2)3)11.7511.7511.7590.090.090.011.8611.8511.8590.090.090.03.69 (3.59)Kyanite (Al2SiO5)7.965.775.6582.184.978.37.857.135.5778.990.074.04.29 (3.67)Pyrope ((Al2O3)(MgO)3(SiO2)3)11.3611.3711.3689.889.789.811.5511.5511.5590.090.090.03.65 (3.48)Sillimanite (Al2SiO5)12.497.365.7389.890.090.011.557.687.4990.090.090.04.09 (3.24)Titanite (CaTiSiO5)7.447.008.4389.989.6113.97.076.578.7290.090.0113.93.24 (3.52)Topaz (Al2SiO4F2)7.114.188.9589.789.189.69.334.428.3990.090.090.04.60 (3.54)Uvarovite (Ca3Cr2Si3O12)12.0712.0612.0689.990.190.012.0012.0012.0090.090.090.03.79 (3.85)Willemite (Zn2SiO4)8.948.938.95108.8108.7108.78.628.628.63107.9107.9107.93.93 (4.26)Zircon (ZrSiO4)6.096.088.5390.690.190.36.616.615.9890.090.090.03.85 (4.66)PhyllosilicateApophyllite ((Na2O)(CaO)8(SiO2)16(H2O)15)9.559.8114.9192.185.989.08.978.9715.7790.090.090.02.08 (2.28)Chlorite ((MgO)10(Al2O3)3(SiO2)4(H2O)9)5.459.3914.1689.397.289.25.249.0714.2990.097.090.02.57 (2.64)Kanemite (Na2O)(SiO2)4(H2O)75.1420.707.6090.490.091.14.9520.507.2890.090.090.01.76 (1.93)Kaolinite (Al2Si2O5(OH)4)6.599.655.3487.577.888.67.408.945.1689.975.188.32.59 (2.60)Lithium disilicate (Li2Si2O5)5.3714.786.4690.090.090.04.7814.655.6890.090.090.01.95 (2.50)Prehnite (Ca2Al2Si3O10(OH)2)4.365.8818.4390.488.889.84.655.4818.4990.090.090.02.90 (2.91)Pyrophyllite (AlSi2O5(OH))9.435.6320.7099.690.190.08.905.1618.64100.890.090.02.21 (2.85)Talc (Mg3Si4O10(OH)2)5.579.619.7991.181.290.25.299.179.4690.481.389.92.43 (2.78)SorosilicateBertrandite ((BeO)4(SiO2)2(H2O)16.264.928.7590.190.090.015.274.578.7190.090.090.02.26 (2.60)Danburite (CaB2Si2O8)8.708.198.2490.090.090.08.048.757.7390.090.090.02.78 (3.00)Hardystonite (Ca2ZnSi2O7)8.048.044.8286.486.989.87.837.835.0190.090.090.03.36 (3.39)Hemimorphite (Zn4Si2O7)(OH)2·(H2O))5.4010.618.3290.189.990.05.1210.738.3790.090.090.03.36 (3.48)Lawsonite (CaAl2Si2O7(OH)2·(H2O))7.566.3313.3490.190.090.08.805.8513.1490.090.090.03.27 (3.09)Rosenhahnite (H2Ca3Si3O10)9.376.706.8288.479.2105.09.486.816.9682.984.1108.63.01 (2.90)Tanzanite (Ca2Al3(SiO4)(Si2O7)O(OH))15.615.529.9590.090.090.116.225.5410.0390.090.090.03.52 (3.35)Thortveitite (Sc2Si2O7)7.444.468.9889.989.773.46.654.698.6290.090.077.83.00 (3.27)TectosilicateAlbite (NaAlSi3O8)8.7612.697.0489.665.090.28.1512.877.1193.163.589.72.46 (2.62)Mordenite (NaAlSi11O24)18.6721.007.7590.290.090.018.1320.497.5290.090.090.01.58 (1.71)Nepheline (KNa3Al4Si4O16)10.1510.298.2690.090.061.110.0510.058.3890.090.060.02.57 (2.64)Paracelsian (SrGa2Si2O8)7.059.379.3890.190.290.18.409.489.0090.089.390.04.40 (3.81)Sodalite (Na4Al3Si3O12Cl)8.918.908.9090.190.190.08.888.888.8890.090.090.02.28 (2.30)Thomsonite (NaCa2Al5Si5O206H2O)13.6913.5712.7480.775.884.213.2513.0613.1090.090.090.02.37 (2.36)dX-ray densities in parenthesisTable 7Comparison of calculated and experimental heats of formation of inorganic compounds (kcal/mol) PM6ReferenceaDifferenceα Quartz (SiO2)−189.4−217.728.3α Crystobalite (SiO2)−193.2β Tridymite (SiO2)−193.2β Quartz (SiO2)−189.4Chabazite (SiO2)−190.7Coesite (SiO2)−187.2Mordenite (SiO2)−191.2Stishovite (SiO2)−160.7Aluminum nitride (AlN)−42.7−76.033.3Aluminum phosphide (AlP)−26.5−39.813.3Ammonium chloride (NH4Cl)−70.0−75.15.1Ammonium nitrate (NH4NO3)−88.0−87.4−0.6Ammonium sulfate ((NH4) 2SO4)−251.5−282.230.7Anatase (TiO2)−229.3−224.4−4.9Andalusite (Al2SiO5)−665.2−619.5−45.7Anglesite (PbSO4)−221.9−219.9−2.0Anhydrite (CaSO4)−275.3−342.967.6Aragonite (CaCO3)−269.0−288.619.6Arcanite (K2SO4)−377.5−343.6−33.9Arkelite (ZrO2)−286.4−263.0−23.4Barium oxide (BaO)−182.2−131.0−51.2Barite (BaSO4)−383.8−352.1−31.7Boron (B)−8.20.0−8.2Boron Nitride (BN)−75.5−60.8−14.7Brookite (TiO2)−229.8Brucite (Mg(OH)2)−176.6−221.044.4Calcite (CaCO3)−271.7−288.616.9Calcium Iodide (CaI2)−108.8−127.518.7Cassiterite (SnO2)−59.8−138.178.3Celestine (SrSO4)−297.4−291.6−5.8Chloromagnesite (MgCl2)−153.9−153.3−0.6Chrysoberyl ((BeO)(Al2O3))−490.6−549.959.3Cinnabar (HgS)−53.3−13.9−39.4Coloradoite (HgTe)−20.4−10.0−10.4Corundum (Al2O3)−370.9−400.529.6Cryolite (Na3AlF6)−871.0−792.8−78.2Enstatite (MgSiO3)−324.2−370.246.0Fluorite (CaF2)−207.1−293.085.9Forsterite (Mg2SiO4)−452.9−520.367.4Frankdicksonite (BaF2)−305.3−288.5−16.8Galena (PbS)−24.8−24.0−0.8Gallium arsenide (GaAs)−35.5−17.0−18.5Graphite (C)1.30.01.3Greenockite (CdS)−85.1−38.7−46.4Hawleyite (CdS)−85.0−38.7−46.3Hydrophilite(CaCl2)−151.4−190.138.7Indium arsenide (InAs)−17.6−14.0−3.6Lead selenide (PbSe)−124.7−24.6−100.1Lead telluride (PbTe)6.2−16.923.1Lime (CaO)−116.2−151.835.7Magnesite (MgCO3)−224.8−265.740.9Magnesium dibromide (MgBr2)−116.9−125.38.4Magnesium diiodide (MgI2)−67.2−87.019.8Sodium sulfate (Na2SO4)−372.5−331.5−41.0Orthoboric acid (B(OH)3)−238.3−261.723.4Lead oxide (PbO)−85.9−52.3−33.6Lead dioxide (PbO2) (Rutile structure)−21.6−66.344.7Periclase (MgO)−96.9−143.746.8Rutile (TiO2)−230.1−225.6−4.5Sellaite (MgF2)−234.9−268.733.8Silicon (Si)−13.50.0−13.5Silicon carbide (SiC)−37.6−15.6−22.0Smithsonite (ZnCO3)−175.7−194.318.6Soda Niter (NaNO3)−159.6−111.8−47.8Sphalerite (ZnS)−40.4−49.28.8Spinel (MgAl2O4)−493.7−549.555.8Sulfur (S)1.50.01.5Willemite (Zn2SiO4)−375.9−391.215.3Wurtzite (ZnS)−39.4−46.06.6Zincite (ZnO)−84.6−83.8−0.8Zinkosite (ZnSO4)−211.0−234.924.0Zircon (ZrSiO4)−461.0−529.968.9aReference values taken from the CRC Handbook [32] Elements Three allotropes of carbon were modeled: diamond, consisting of pure covalent sp3 bonds, graphite, with sp2 bonds in-plane and π-stacking or VDW forces between planes, and “glitter”, a hypothetical structure composed of 1,4 cyclohexadiene fragments [30]. The simple C–C bond in diamond is reproduced with good accuracy, 1.57 Å versus 1.54 Å observed, as is the in-plane bond in graphite, 1.43 Å versus 1.42 Å. This last result was unexpected in light of the zero band gap of graphite, in that severe difficulties in achieving a SCF are often encountered when systems with very small band-gaps are modeled; graphite appears to be a unique exception to this. Graphite layers are held together by weak VDW forces, which are poorly represented in all NDDO methods. One consequence of this is that the predicted interlayer distance, 3.56 Å, is significantly larger than the observed 3.35 Å. Intermediate between graphite and diamond is the hypothetical glitter, a tetragonal allotrope of carbon composed of single and double bonds. A PM6 calculation predicted that the structure would be essentially the same as that given by DFT methods [31]. The standard state of carbon is graphite, and therefore by definition the value of its ΔHf is 0.0 kcal/mole-atom. PM6 predicts the ΔHf of graphite to be 1.29 kcal/mole-atom, in good agreement with the reference value. The higher energy allotrope, diamond, has a reported heat of formation of 0.45 kcal/mole-atom [32]; for diamond, PM6 predicts the ΔHf to be 1.34 kcal/mole-atom, i.e., 0.05 kcal/mole-atom above graphite. Glitter is a hypothetical allotrope, but is predicted by PM6 to be unlikely to be formed under equilibrium conditions: both PM6 and DFT predict the ΔHf to be large and positive +9.1 kcal/mole-atom and +11.8 kcal/mole-atom [33], respectively. Silicon also crystallizes in the diamond lattice, but the calculated silicon–silicon distance, 2.22 Å, is significantly less than the observed value of 2.35 Å, and while the experimental ΔHf is, by definition, zero, the calculated heat of formation is −16.6 kcal/mole-atom. Sulfur forms eight-membered rings, with 16 rings per unit cell. Because the unit cell is so large, and because there is a distinct insulating gap between each ring, the approximation that Γ represents the entire Brillouin zone is valid even when only a single unit cell is used. Within each ring, the sulfur–sulfur distance is 2.04 Å, in perfect agreement with the 2.04 Å observed, but the inter-ring distance is badly predicted, resulting in a calculated density of 1.54 g/cc, considerably less than the observed 2.06 g/cc. This lack of inter-ring interaction is the likely cause of the calculated heat of formation being 1.45 kcal/mole-atom, rather than being nearer to the reference 0.0 kcal/mole-atom. Halides One of the simplest sets of inorganic solids are the alkali metal halides. Most of these crystallize with the rock-salt lattice, the exceptions being cesium chloride, bromide, and iodide, which crystallize in the body-centered cubic or cesium chloride structure. Because of their high symmetry there is only one geometric variable, so for this group of solids, symmetry was used to accelerate the geometry optimization. Geometries were optimized for both the rock-salt and cesium chloride structures for each salt. With the exception of rubidium fluoride, the calculated heats of formation (Table 8) of the isolated molecule and of the crystal are reproduced with good accuracy, although PM6 did a very poor job of predicting which of the two crystal forms was the more stable. Table 8Comparison of calculated and experimental heats of formation of alkali metal halides (kcal/mol)SaltMoleculeCrystalPM6Reference [1]PM6 (NaCl)PM6 (CsCl)Experimental [32]Lithium fluoride−81.5−86.0−138.2−139.7−147.4Lithium chloride−46.8−53.9−107.4−108.2−97.6Lithium bromide−36.8−38.1−89.6−87.3−83.9Lithium iodide−19.4−16.1−72.0−73.2−64.6Sodium fluoride−69.6−64.0−123.2−125.7−137.5Sodium chloride−43.4−49.4−87.7−86.8−98.3Sodium bromide−34.2−37.6−99.4−101.7−86.4Sodium iodide−19.0−19.8−75.2−75.4−68.8Potassium fluoride−78.1−74.9−185.7−183.5−135.9Potassium chloride−51.2−53.4−112.0−118.3−104.4Potassium bromide−43.0−44.3−99.8−113.2−94.1Potassium iodide−30.0−30.0−82.5−93.9−78.4Rubidium fluoride−79.2−101.3−152.8−111.3−133.3Rubidium chloride−54.7−61.3−117.9−108.2−104.1Rubidium bromide−43.7−50.8−107.7−103.5−94.3Rubidium iodide−32.1−24.0−66.4−69.6−79.8Cesium fluoride−85.2−81.9−155.9−115.3−132.6Cesium chloride−57.4−62.9−128.4−130.2−105.8Cesium bromide−50.0−55.0−106.0−117.6−97.0Cesium iodide−36.3−43.5−86.0−98.4−82.8 Elements of Group II form two main groups of halides: those with 8 coordinate metal atoms, such as fluorite, CaF2, and frankdicksonite, BaF2, and the six-coordinate layer structures of the chlormagnesite, MgCl2, type. Other structures include the rutile lattice, as in sellaite, MgF2, and the distorted rutile lattice, hydrophilite, CaCl2. For all alkaline earth halides investigated, the calculated structure had the same lattice type as that in the starting geometry. A more complex halide is carnallite, KMgCl3·6(H2O), composed of isolated potassium and chloride ions, and magnesium ions that are octahedrally coordinated by water molecules. For the X-ray structure, PM6 predicted the charge on the coordination complex ion [Mg(H2O)6]2+ to be +1.89, on the potassium ion, +0.74, and on the chloride, −0.87. Within the complex ion, the charge on magnesium was +0.79, and, on average, each water molecule had a charge of +0.183. The complex ion thus behaves like an extremely ionic Group II element; the net charge being much greater than that on any Group II metal ion in any halide. Trisodium hexachlorotitanate crystallizes in a cryolite-like structure [34]. Each titanium ion has one unpaired electron and although the Ti–Cl bond is highly covalent, the smallest distance between the [TiCl6]3- complex ions is more than 3.6 Å; therefore, from an electronic perspective, the unpaired electrons could be regarded as isolated. To verify the validity of this assumption, the geometry of solid Na3TiCl6 was optimized using the UHF Hamiltonian. A cluster of 16 formula units was used, this representing 8 unit cells. In one calculation, the magnetic component of spin, Ms, was set to zero, and in the other it was set to the maximum value for 16 unpaired electrons, i.e., Ms = 8. Both optimized geometries were essentially the same, geometric differences were negligible, and the two heats of formation were −459.2 kcal/mol and −462.0 kcal/mol, respectively. Therefore. the assumption that the unpaired electrons are electronically isolated, at least when the focus of interest is energetics and structure, is justified. Oxides SiO2 PM6 predicts the Si–O distance in silicon dioxide to be too large by 3–5%, with the result that the predicted densities of the polymorphs of silica were all underestimated. This is best seen in the very dense stishovite, where the predicted density, 3.70, is about 14% less than the observed 4.28. Stishovite has a rutile structure, so the positions of the atoms in the unit cell can be defined using two bond lengths and one angle. PM6 predicts the Si–O–Si angle correctly, 97.8° compared to the observed 98.8°, but overestimates the Si–O bond-lengths, 1.88 Å and 1.87 Å compared to the 1.76 Å and 1.81 Å observed. In α-quartz the predicted Si–O distance, 1.65 Å, is also larger than that observed, 1.61 Å, but, in addition, the Si–O–Si angle opens from the observed 143.7° to 157.2°. This additional decrease in density results in the predicted density of α-quartz being too low by 15.5%. The error in the Si–O–Si angle reaches its maximum in α-crystobalite where PM6 predicts it to be 180.0° instead of the observed 146.8°. This results in a large error in density, of −20%. Linear Si–O–Si systems are not unknown in nature: in the high temperature form of tridymite, hexagonal β-tridymite [35] and in thortveitite, (Sc,Y)2Si2O7 [36], these atoms form a straight line, and, as expected, this angle is also precisely reproduced by PM6.In addition to the dense phases of silica there are open lattice structures which, of their nature, necessarily have very low densities. Two examples are mordenite and chabazite. Calculated and X-ray structures in these systems are shown in Figs. 8 and  9. Chabazite has the lowest density of any silica polymorph, 1.46. PM6 reproduces these systems but, as with the higher density polymorphs, it underestimates the density, but in this case by about 5%. Fig. 8Part of the unit cell of Mordenite, SiO2. Left X-ray structure, right optimized PM6 mirror image structureFig. 9Chabazite, SiO2Left X-ray structure, right optimized PM6 structure H2O Water forms several polymorphs. In Ice-I or Ice-Ih, the common polymorph of ice, oxygen atoms form a hexagonal lattice and each atom has two hydrogen atoms strongly covalently bound to it and two hydrogen atoms at a larger distance. The locations of the hydrogen atoms can be specified by application of the “ice rules”. However, even when that is done correctly, the positions of the hydrogen atoms are not unambiguously defined. That is, although the X-ray structure shows that the oxygen atoms are located precisely on the vertices of the hexagonal lattice, there is statistical disorder in the positions of the hydrogen atoms—between each pair of adjacent oxygen atoms there are two potential positions that a hydrogen atom can occupy, and in the observed structure each position has a 50% occupancy. This type of disorder must be resolved before a solid state calculation can be carried out, so in Ice-Ih, an arbitrary distribution of hydrogen atoms consistent with the ice rules was used.Geometry optimization of the Ice-Ih structure gave a recognizable hexagonal lattice, as expected, but as a result of the asymmetric distribution of hydrogen atoms each oxygen atom was slightly displaced relative to the ideal lattice.All the ices, except Ice-IV and Ice-VII, were modeled successfully. The simple unit cell of Ice-IV contains 16 water molecules but, in order to satisfy the ice rules and have the hydrogen atoms disordered, a large cluster would be needed. Assigning hydrogen atoms in such a cluster in order to achieve both randomness and compliance with the ice rules would be a non-trivial task. However, Ice-VII is the high-temperature (above 5°C) modification of Ice-VIII and differs from Ice-VIII only in that the positions of the hydrogen atoms are disordered. As such, Ice-VII is trivially different from Ice-VIII.With the exception of Ice-X, each polymorph optimized to the correct structure. That is, the structure of each calculated polymorph was qualitatively similar to that of the observed polymorph. Ice-X is unique in that the oxygen atoms form a body-centered cubic lattice with hydrogen atoms equidistant between half of the adjacent pairs of oxygen atoms: all oxygen atoms are symmetrically tetrahedrally coordinated by hydrogen atoms. Unconstrained optimization of this structure resulted in a disordered, essentially amorphous, solid. When symmetry constraints were imposed, the geometry of Ice-X optimized to give an O-H distance of 1.214 Å. Ice-X is stable at pressures above 62 GPa and, at that pressure, PM6 predicts the O–H distance to be 1.179 Å, in good agreement with the X-ray value [37] of 1.181 Å. At a much higher pressure, about 480 GPa, PM6 predicts that the antifluorite structure would become the preferred polymorph; in this structure each hydrogen bonds to four oxygen atoms and each oxygen bonds to eight hydrogen atoms. Al2O3 An attempt [12] by Gale to use the MNDO, AM1, and PM3 methods to predict the structure of corundum met with only limited success. The failure to reproduce the observed structure was attributed to deficiencies, mainly the lack of d-orbitals, in the basic semiempirical methods. To a large degree these deficiencies have been corrected in PM6, as can be seen by comparing the structures predicted by the various methods with the known X-ray structure (Table 9). Table 9Calculated and observed structure and ΔHf of α-corundumExperimentalMNDOaAM1aPM3aPM6a (Å)4.764.854.695.284.83c (Å)12.9913.1212.3315.4712.91ΔHf (kcal/mol)−400.5−266.5−241.9−236.4−370.9aResults for MNDO, AM1, and PM3 taken from Gale [12] TiO2 Within each of the polymorphs of titanium dioxide, all the titanium atoms are in the same approximately octahedral environment, surrounded by six oxygen atoms. This is illustrated in Fig. 10 for rutile, the commonest polymorph. The structures of all three polymorphs were qualitatively reproduced by PM6, but the Ti–O bond-length was over-estimated by 5%, as shown in Table 10 for rutile, resulting in the predicted density being too low by about 16%. Fig. 10Unit cell of rutile (TiO2). Crossed-eyes stereo viewTable 10Interatomic distances in rutileDistancePM6X-rayTi-O2.0221.981Ti-O’2.0801.948Ti-Ti3.1162.959 B(OH)3 Orthoboric acid, B(OH)3, forms a layer structure in which individual molecules within each layer are held together by hydrogen bonds, and the layers are held together by VDW forces and very weak electrostatic forces. Although the layer structure is reproduced with good accuracy, the predicted inter-layer distance, 11.0 Å, is much greater than the observed 6.35 Å. The lack of any significant inter-layer interaction can be attributed to the absence of a VDW core–core term in boron. It is likely that if such a term were added, the correct interlayer interaction would be reproduced.During the survey of organic solids, the structure of the very complicated organometallic pentakis(tetraethylammonium) bis(meso-tetraphenyl-porphyrinato-zinc) tetraconta-oxo-silicon-dodeca-molybdenum bromide (CSD entry PIJFUJ) was accurately reproduced. In contrast, problems had been encountered with various other complicated solids, particularly those involving protonated species. To determine the ability of PM6 to model complex ionic systems, a solid containing an archetypal proton ion, Zundel’s cation [38], [H5O2]+, was used. This ion had been identified [39] in Keggin’s [40] pentahydrate, dodecatungstophosphoric acid hexahydrate. X-ray analysis indicated that the [H5O2]+ ion has D2h symmetry and that the O–O separation is 2.38 Å. However, a PM6 calculation on the isolated ion predicted its structure to be C2h with an O–O separation of 2.51 Å. This was not unexpected: ab-initio calculations [41, 42] indicate that the ion should have C2 symmetry with the Cs structure being only about 0.4 kcal/mol higher in energy, and that the D2h structure should be significantly higher in energy. When D2h symmetry was used, PM6 predicted that the separation would decrease to 2.45 Å, and the energy would increase by 7.8 kcal/mol. Disorder had been reported in Keggin’s pentahydrate, with Zundel’s ion assuming two different orientations, each having a 50% occupancy. After resolving the disorder, an unconstrained geometry optimization was performed. Examination of the optimized structure revealed that the Zundel ion had been destroyed and that the proton had migrated to the polyoxometalate. This phenomenon, the neutralization of a polyanion, had also been observed in tetrakis(2-carboxypyridinium) octacyano-molybdenum(iv), 4[C6H6NO2]+ [MoIV(CN)8]4-, CSD entry PYCMOA, where PM6 predicted that protons on the pyridinium ions would migrate to the metal complex. Other AB-type solids A number of solids of the type AB are formed from elements of Group III and V, while others involve elements of Group IV, and still more involve elements of Group VI and heavy elements. Some of these occur naturally, such as wurtzite and sphalerite (zinc sulfide), and coloradoite (mercury telluride), while others are formed synthetically, often by chemical vapor deposition methods. In most of these materials, each atom of type A is tetrahedrally coordinated to four atoms of type B. This results in two types of packing, best exemplified by the two polymorphs of zinc sulfide. For convenience all compounds of this type will be grouped together. Many of these compounds are semiconductors, and have small band-gaps between the occupied and virtual orbitals. A consequence of this is that the cluster used must be very large in order to minimize errors arising from the cluster approximation. In addition, it was anticipated that the narrow band-gap would give rise to difficulties in solving the self-consistent field equations—such difficulties had frequently occurred when molecules that had small HOMO–LUMO gaps were being studied. Surprisingly, the SCF equations were solved using default options, albeit more iterations than normal were needed. Symmetry was used to accelerate the geometry optimization of those solids that had the sphalerite structure; such solids have only one adjustable parameter. In most cases, the calculated density for the optimized structure was close to that expected, the exceptions being cadmium telluride, which PM6 predicts to be too dense, and coloradoite, where the density was predicted to be too low. Carbonates, nitrates, and borates Of the simple carbonate minerals that crystallize in the calcite lattice (Fig. 11), PM6 correctly reproduces the structures of magnesite, MgCO3, smithsonite, ZnCO3, otavite, CdCO3, and calcite, CaCO3 itself (Table 11). PM6 was unable to reproduce the structures of two other carbonates: rhodochrosite, MnCO3, and siderite, FeCO3. The UHF method was used in modeling both of these minerals because they contain transition metal ions that have unpaired electrons. In the case of rhodochrosite, each metal ion has five d-electrons in an approximately octahedral environment. Because the crystal field of the carbonate ions is too weak to cause a large separation of the t2g and eg levels, MnII would be in the high-spin configuration. This implies that the local electronic structure would be approximately 6A1g, and that there would be no angular terms arising from the d-electrons. That is, a MnII ion would be expected to behave like a Group II ion. The failure of PM6 to reproduce the calcite structure for rhodochrosite indicates a fault in the parameterization of manganese, most likely attributable to errors in the core-core terms. Fig. 11Unit cell of calcite. Crossed-eyes stereo viewTable 11Interatomic distances in calciteDistancePM6X-rayCa-O2.2972.389C-O1.2891.210Ca-Ca3.9894.036 The other common polymorph of calcite is aragonite. Other minerals that have the aragonite structure are cerussite, PbCO3, strontianite, SrCO3, and witherite, BaCO3. Of these, PM6 predicts correctly the structures of aragonite, strontianite, and witherite, albeit some carbonate groups were rotated slightly about their C3 axis. In cerussite, the carbonate groups were tilted out of plane, giving rise to a herringbone-like structure. The densities of all Group I carbonates are predicted by PM6 to be too low. This includes the simple carbonates natrite, Na2CO3, and zabuyelite, Li2CO3, potassium hydrogen carbonate, kalicinite, and sodium carbonate monohydrate, thermonatrite. Nitrates PM6 correctly reproduces the structures of ammonium nitrate and potassium nitrate (niter), but the predicted structure of sodium nitrate, nitratine, is completely incorrect; as with other sodium compounds, the predicted Na–Na distances are unrealistically small. Borates Several borates were examined, including the simple sodium salt, borax. Like other solids that contain a large amount of sodium, its optimized PM6 structure was severely in error. With that single exception, the structures of the borates were reproduced with good accuracy. Almost all minerals have some symmetry in their unit cell, the most well known exception being the naturally-occurring calcium chloroborate, parahilgardite, Ca2(B5O9)Cl(H2O). PM6 was able to reproduce the unit cell of this almost unique mineral with good accuracy, despite the fact that the unit cell was completely devoid of any symmetry (Fig. 12). This system was unique among the solids studied in that, unlike all the organic compounds, there were no readily identifiable discrete molecular fragments, and, unlike the other inorganic solids, it is devoid of any elements of symmetry. Fig. 12Unit cell of parahilgardite. Crossed-eyes stereo view Molybdates, tungstates chromates vanadates, sulfates, and phosphates This group of solids is characterized by the presence of tetrahedral oxyanions, e.g., [SO4]2-, [PO4]3-, [MoO4]2-, [WO4]2-, and [CrO4]2-, with counterions ranging from the highly ionic Group I and Group I-like ions, e.g., [NH4]+, to softer cations such as Pb2+ and Zn2+. With the exception of descloizite, PbZn(VO4)(OH), the internal structures of all systems modeled were quantitatively reproduced. In descloizite, the distance between lead and zinc ions was predicted to be unrealistically small, which, in turn, resulted in severe distortions to the positions of the other entities in the unit cell. There are diatomic core–core repulsion parameters in PM6 for Pb and Zn, but clearly the magnitude of the term was too small. As with other errors of this type, this fault could readily be corrected by re-parameterizing the core–core terms only. Charges on the ions are significantly reduced from the formal values. Thus, in fluorapatite (Fig. 13, Table 12), calcium ions have charges of +1.35 and 1.29, depending on environment, fluoride is −0.77, and each phosphate group has a net charge of −1.95, all indicating a high degree of covalence in the bonds. Fig. 13Unit cell of fluorapatite. Crossed-eyes stereo viewTable 12Interatomic distances in fluorapatite. Ca1 Ca on a vertex, Ca2 Ca on an edge, Ca3 Ca attached to FDistancePM6X-rayCa1-Ca15.4155.408Ca1-Ca23.4503.456Ca1-O2.3052.338Ca3-O2.3862.375Ca3-F2.3302.361P-O1.5801.585 Strong hydrogen bonding can occur in some sulfates and phosphates, which can be illustrated by salts in which the acid is partially neutralized, such as compounds of the type MHSO4, MH2PO4, and M2HPO4, where M is a Group I element. A common structure in systems of this type consists of a hydrogen atom positioned between two oxygen atoms on two different acid groups. The positions of the protons as determined by X-ray analysis are of limited reliability, so for the purposes of comparison the oxygen–oxygen distances were used. In mercallite, potassium hydrogen sulfate (Fig. 14), PM6 predicts the oxygen–oxygen distance (Table 13) in perfect agreement with the X-ray structure. PM6 also predicts the hydrogen atoms to be asymmetrically positioned between the two oxygen atoms, this also being in accord with the published X-ray structure. Potassium dihydrogen phosphate is more complicated in that each phosphate forms four hydrogen bonds to four different phosphate groups (Fig. 15). Again, PM6 predicts (Table 14) the oxygen–oxygen distance with good accuracy, and predicts the hydrogen atoms to be asymmetrically positioned; however, in both the hydrogen sulfate and hydrogen phosphate PM6 underestimates the degree of asymmetry. Fig. 14Structure of hydrogen sulfate in KHSO4. Top X-ray structure, bottom PM6 structureFig. 15Detail of structure of dihydrogen phosphate in KH2PO4 (upper pair) and in (CH3)4NH2PO4. Left X-ray structures on left, right PM6 structureTable 13Interatomic distances in mercallite (KHSO4)DistanceX-rayPM6O(H) ⋯O2.632.63O–H0.731.04S–S4.404.60Table 14Interatomic distances in hydrogen phosphate in KH2PO4DistanceX-rayPM6O(H)⋯O2.492.53O–H1.071.20O–H1.431.33P–P4.114.26 The potassium ion is relatively small in comparison to a phosphate ion, so in KH2PO4 the phosphate ions can form a three-dimensional lattice. If a larger cation, e.g., tetramethylammonium, is used, the three-dimensional phosphate lattice becomes energetically unfavorable relative to a one-dimensional chain of phosphates. This structure was also reproduced accurately. PM6 reproduces the structures of most of the phosphates, but, as expected, in the case of sodium phosphate the predicted structure was qualitatively incorrect. Silicates Silicates form the largest and most complicated group of minerals and exhibit a very wide range of properties. From a chemical perspective, however, they are relatively simple: most silicates involve SiO4 tetrahedra in various degrees of polymerization, and in turn these tetrahedra interact with fully oxidized metal ions. In addition, for any given empirical formula, naturally occurring minerals represent the most stable, or almost the most stable, structure. Some structures might be quite complicated; nevertheless, because they formed under equilibrium conditions, they are near or at the energy minimum, and, for any given empirical formula, an accurate computational model should not be able to generate a polymorph that is significantly more stable than any that occurs naturally. This is in contrast to organic chemistry, where high energy forms of the same empirical formula occur frequently: for example, dimethyl ether is the high energy isomer of C2H6O, with ethanol being the low energy form. For the purposes of this study, the silicates can be divided into families based on the degree of aggregation of the silicate units. Isolated SiO4 tetrahedra: nesosilicates The simplest of the silicates are the nesosilicates, which contain isolated SiO4 groups. With the exception of the three polymorphs of Al2SiO5, kyanite, andalusite, and sillimanite, the structures of the nesosilicates are reproduced with good accuracy. In the aluminum silicates there are two distinct anions, an [SiO4]4- ion and an isolated oxygen ion, [O]2-. PM6 incorrectly predicts these ions to coalesce to form a SiO5 moiety. Double and triple tetrahedra: sorosilicates All the sorosilicates examined were reproduced with good accuracy. In most of the minerals that contained the [Si2O7]6- moiety the Si-O-Si system is bent. As expected, PM6 reproduced the bent structure. The only exception was thortveitite, an important source of scandium, which exists in nature as (Sc,Y)2Si2O7, but which, for the purposes of this work, was simplified to Sc2Si2O7. In the crystal structure of this mineral [36], the [Si2O7]6- moiety has C2h symmetry; consequently the Si–O–Si angle is precisely 180°. When the structure of thortveitite was optimized using PM6 the symmetry did not change: the predicted Si–O–Si angle was also 180°. Chains: inosilicates With the exception of spodumene, where PM6 predicted the aluminum-silicon distances to be too short, the structures of all the chain silicates were reproduced with good accuracy. As with the aluminum silicates, the fault in spodumene can be attributed to errors in the Al–Si core–core parameters. In the case of cummingtonite, Mg7Si8O23·(H2O), hydrogen atoms were added before the structure was optimized. Cyclosilicates Three cyclosilicates were examined: benitoite, BaTiSi3O9, beryl, Be3Al2Si6O18, and dravite, NaMgTi2Al6(BO3)3(Si6O18)O4. All were reproduced with good accuracy, as illustrated by the bond lengths and angles (Table 15) in the simplest of these, beryl (Fig. 16). Dravite has a particularly complicated structure, involving seven different elements, in which the positions of the cations Na+, Mg2+, Al3+, and Ti4+ and the anions [O]2-, [BO3]3- and [Si6O18]12- are determined by roughly equal contributions of covalent and ionic terms. Fig. 16Unit cell of beryl. Crossed-eyes stereo viewTable 15Interatomic distances and angles in berylGeometric quantityX-rayPM6Al-Be2.7542.656Al-O1.9281.906Si-O(Al)1.6551.610Si-O(Si)1.6111.607Si-O-Si168.1165.3Al-O-Si128.6137.1 Sheets: phyllosilicates The structures of the phyllosilicates are reproduced with modest accuracy. In talc (Fig. 17) for example, in-plane distances are reproduced well, but the inter-plane separation (Table 16) is only poorly modeled. This fault can be attributed to the absence of strong inter-plane interactions: any errors in the weak long-range interatomic terms would give rise to disproportionately large errors in inter-plane separations. Fig. 17Unit cell of talc. Crossed-eyes stereo viewTable 16Interatomic distances in talcDistancePM6X-rayMg-O(H)2.0602.062Mg-O(Si)2.0842.080Si-O(Mg)1.6291.624Si-O(Si)1.6741.622O–Oa3.7623.095aDistance between layersThe mineral chlorite has a variable stoichiometry, (Fe, Mg, Al)6(Si, Al)4O10(OH)8, which does not allow a simple resolution into a fixed stoichiometry. In it, the presence of main-group metals of different oxidation states (MgII and AlIII) in the cationic part of the formula requires that, in order to maintain electrical neutrality, for any given ratio of magnesium to aluminum, there should be a corresponding composition in the anionic part. In nature, this condition would be satisfied by the random distribution of magnesium, aluminum, and silicon atoms in the lattice. Computationally, however, such a distribution is currently not possible, and fixed stoichiometries are required. In the extremes, these would be Mg6Si4O10(OH)8 and Mg2-Al8O10(OH)8. However, the lattice structures of these two extremes would likely be so different to that of chlorite that comparison between any computed result for one of these extremes and that of chlorite would be meaningless. In order to obtain a meaningful result, an unusually large cluster was used, in which the ratios of the elements was as near to that of chlorite as possible. Frameworks: tectosilicates With the exception of thomsonite, NaCa2Al5Si5O20·6H2O, PM6 predicted the internal structures of the tectosilicates accurately. As with the other aluminosilicates, in thomsonite the predicted aluminum-silicon distance was unrealistically short. Discussion With a few exceptions, the geometries of individual organic molecules and ions and their packing arrangement in the crystal lattice were reproduced with good accuracy. In the original PM6 article it was shown that the average error in predicted bond-lengths in organic compounds was about 2–3%. The effect of intermolecular forces on the geometries of the component molecules is likely to be small, so by implication the accuracy of prediction of the geometries of those molecules would be similar to that reported in the earlier article. By far the largest structural error in organic solids involves intermolecular separations. The values of these are determined by several forces, from the weak VDW and π-stacking attractions, in cyclohexane and coronene, through simple hydrogen bonding of the type found in sucrose, to very strong intermolecular interactions, usually ionic, best typified by the Zwitterions, such as the simple amino acids. Errors in intermolecular separations could be determined by a direct comparison of calculated and reference structures. However, because errors in intermolecular separation have a direct impact on the unit cell dimensions, and consequently on the density, a convenient and reliable estimate of the accuracy of prediction of intermolecular separations of molecules in the unit cell can be obtained from a comparison of predicted and X-ray densities. The internal structures of three organic solids were incorrectly predicted using PM6; for all other solids the internal packing was qualitatively correct. Two of the faulty solids involved the oxalate and water moieties. From analysis of other systems known to contain neutral species, but which PM6 predicts to be ionic, it appears that PM6 unrealistically favors the formation of the hydronium ion. The fault in sodium acetate trihydrate appears to arise from faults in sodium core–core parameters, in that similar faults were found in several other inorganic sodium-containing compounds. Interestingly, the fault was not present in the closely related species sodium hydrogen acetate, a system chosen because of the presence of the hydrogen-bonded structure, [Ac2H]–. When the three solids that were badly predicted were removed from consideration, the average error in density decreased to 6.1%, with most of this attributable to the 2% average error in intermolecular separations. This error was not evenly distributed within the organic solids in that when interaction energies were large, as in the ionic solids, in particular Zwitterions, PM6 reproduced the intermolecular separations with good accuracy, but when the interaction energies were small, as in species where only VDW forces are involved, such as methionine, errors in intermolecular separations were often quite large. The structures of most of the inorganic solids were predicted with useful accuracy, with the highest accuracy being exhibited by minerals in which metal atoms interact with oxygen, and the oxygen then interacts with a main-group element. Most of the silicates, phosphates, and sulfates fall into this group. One reason for this can be attributed to the procedure used in developing PM6, in that a large quantity of reference data for systems that had metal-oxygen or main-group oxygen bonds was used during the parameterization. This naturally resulted in increased emphasis being placed on the structure and thermochemistry of oxygen-containing systems. On the other hand, inorganic systems exhibit a much wider range of types of interaction than those found in organic chemistry, and this makes the general application of PM6 to inorganic solids more difficult. An implication of the fact that PM6 uses the Voityuk diatomic core-core term is that parameters must be present for each pair of elements in a system, unless the pair of elements is separated by a distance sufficiently large that there would be no significant core-core interaction. Only a limited number of types of interactions were surveyed in this work; even within that number, several instances were found in which the PM6 values of the Voityuk parameters were severely in error and gave rise to results that were nonsense. Because of this, great care should be exercised in determining the suitability of PM6 for modeling solids that include diatomic interactions of types not found in any of the species reported here. In general, PM6 reproduced the X-ray densities of both organic and inorganic crystals with useful accuracy, Fig. 18, the R2 value being 0.94 and the least squares fit was ρPM6 = 0.936ρX-ray+0.121. The average unsigned error in the predicted densities of inorganic solids, 9.3%, is higher than that of organic solids, 6.9%, although the average signed error was −2.5% compared to the equivalent +3.9% for organic solids. Fig. 18Comparison of calculated and X-ray densities Systems that are badly predicted Of the systems whose properties were badly predicted, three types could be identified. In the first group, illustrated by lead selenide and 2, 4, 6-tribromoaniline (TBA), the origin of the error could be traced to the values of individual diatomic parameters used in PM6. These were either incorrect, as in lead selenide, where the PM6 value for Voityuk’s lead–selenium core–core repulsion was much too small, and in TBA where the bromine–nitrogen parameters were either physically unrealistic or absent. Errors of this type could be easily corrected by carrying out a small parameterization operation involving only the faulty parameters and using a training set consisting of examples of the two atoms in close proximity. The second type of error was found in some solids for which no equivalent error was found in the isolated molecule. This is best illustrated by the Group I halides, where PM6 predicted the lowest energy structure to be either rock salt or cesium chloride, almost at random. During the development of PM6, only small representatives of I-VII species were used. Apparently, these systems were too small to allow the lattice properties to be accurately characterized. Errors of this type were therefore not immediately obvious, and only became apparent when full solid-state calculations were done. The implication is that future parameterizations should include solids in the training set. The best solids for use in parameterization would be those that were badly predicted by PM6. Addition of solids to the parameterization would be unlikely to cause a significant increase in error in the prediction of molecular properties. Although the use of solids in parameterization is impractical at present, it is likely that increases in computer power will make such calculations possible in the not-too-distant future. The third type of error is specific to organic compounds, where PM6 predicts some ions to be too stable. Thus the aminophenols are predicted to exist in the solid state as the Zwitterions rather than as the neutral species. The hydrate of oxalic acid was predicted to exist as oxalate and hydronium ions, a prediction completely in variance with the X-ray structure. No instances were found where PM6 underestimated the stability of ions relative to their neutral counterparts. That is, the error was completely systematic and unequivocally indicated a fault in the parameterization. The origin of this error most likely lies in the set of atomic electronic parameters, rather than in the diatomic core–core parameters, and as such could only be corrected by a re-parameterization. There was no evidence that any faults were due to an underlying defect in the set of approximations. Accuracy of geometry vs hardness In general, the structures of solids that are mechanically extremely hard, seven or more on Moh’s scale, are predicted with good accuracy, whereas the geometries of many softer solids, such as most organic species, the transition metal carbonates, and various layer silicates, particularly the micas, are predicted with significantly less accuracy. This inverse relationship of mechanical hardness and computational accuracy can be rationalized by consideration of the interatomic forces involved. In hard solids, all atoms are connected to adjacent atoms by strong covalent bonds, and, of their nature, those bonds have large force constants. In general, semiempirical methods predict such geometries with good accuracy. Any tendency to deviation from the expected geometry would result in a large energy penalty, therefore distortions in predicted geometries are likely to be small. Conversely, in mechanically weak solids some atoms interact only weakly with adjacent atoms. Thus the layers in the micas and in boric acid are held together by low energy electrostatic and VDW forces. Much effort was expended during the development of PM6 in attempting to accurately model hydrogen bonds and other weak interactions because of their importance in biochemical systems. This work has apparently been of limited success, and, as a result, in solids that are mechanically weak very small changes in energy are often associated with large geometric distortions. The nature of this inverse relationship can be used in future method development, where, currently, the use of solids in parameterization is precluded because of the heavy computational effort required. At the present time, however, solids can be used in surveys. There, the hardness-accuracy relationship can be used in the choice of solids to be used, in that potential errors in any new method are more likely to be detected in a survey of mechanically weak rather than very hard solids. Crystal packing automatically considered In order to accurately reproduce molecular structures determined from X-ray analysis, theoretical methods must not only be able to accurately reproduce gas-phase geometries, but must also be able to include crystal packing forces. When these forces are small, the gas-phase structure is a good approximation to that of the solid, but the converse is also true, in that, in systems where there are large intermolecular interactions, particularly strong ionic forces, a solid state calculation is essential in order to accurately reproduce such structures. This was most evident in the case of HMPN-H(+), where the gas-phase B3LYP geometry is a significantly worse match to the observed crystal structure than that generated by the PM6 solid state calculation, despite the intrinsically higher accuracy of B3LYP. In this system, it is obvious that the packing forces arising from the electrostatic interaction of the [PF6]– ion with the HMPN-H(+) have significant geometric consequences. Conclusions The newly developed PM6 method has been demonstrated to reproduce the geometries and heats of formation of many solids, both organic and inorganic, with useful accuracy, although in a few cases the geometries predicted by PM6 were severely in error. Most of these errors can be attributed to incorrect values of the Voityuk core–core parameters. Three examples of this type of error were identified: the Pb–Sb, Br–N, and Pb–Zn interactions all had serious errors. Where diatomic parameters were either missing or improperly optimized, as a result of a fault in the training set used, a simple correction to the method can be made. This would involve only a re-optimization of the faulty diatomic parameters, and would not alter the performance of PM6 when applied to other systems. One general error was found in the treatment of organic compounds: many compounds that exist in the solid state as neutral species were incorrectly predicted by PM6 to be ionized. Examples include the aminophenols and compounds involving oxalic acid and water, such as oxalic acid dihydrate and barium oxalate oxalic acid dihydrate. In contrast to the errors in the values of the Voityuk parameters, errors of this type involve electronic properties and can only be corrected by a complete re-parameterization. The error was not detected during the development of PM6 because solids were not used. Many solids, particularly those that are mechanically weak, can be used as sensitive detectors of potential faults in new methods, and as such they should be used for quality control in future method development work. Electronic supplementary material Below is the link to the electronic supplementary material. ESM 1 Archive files of all systems, both organic and inorganic, described in this article are provided as separate files in a ZIP collection. Each file contains a complete Z-matrix of the geometry optimized using PM6, together with all keywords used in the calculation. The archive files are suitable for running using MOPAC2007 (PDF 32.4 MB)

Mar_Biotechnol_(NY)-4-1-2263118

DNA Microarrays for Identifying Fishes

In many cases marine organisms and especially their diverse developmental stages are difficult to identify by morphological characters. DNA-based identification methods offer an analytically powerful addition or even an alternative. In this study, a DNA microarray has been developed to be able to investigate its potential as a tool for the identification of fish species from European seas based on mitochondrial 16S rDNA sequences. Eleven commercially important fish species were selected for a first prototype. Oligonucleotide probes were designed based on the 16S rDNA sequences obtained from 230 individuals of 27 fish species. In addition, more than 1200 sequences of 380 species served as sequence background against which the specificity of the probes was tested in silico. Single target hybridisations with Cy5-labelled, PCR-amplified 16S rDNA fragments from each of the 11 species on microarrays containing the complete set of probes confirmed their suitability. True-positive, fluorescence signals obtained were at least one order of magnitude stronger than false-positive cross-hybridisations. Single nontarget hybridisations resulted in cross-hybridisation signals at approximately 27% of the cases tested, but all of them were at least one order of magnitude lower than true-positive signals. This study demonstrates that the 16S rDNA gene is suitable for designing oligonucleotide probes, which can be used to differentiate 11 fish species. These data are a solid basis for the second step to create a “Fish Chip” for approximately 50 fish species relevant in marine environmental and fisheries research, as well as control of fisheries products. Introduction Compared with terrestrial ecosystems, little is known about marine biodiversity and changes in species richness and ecosystem function. This is mainly because of sampling difficulties and problems in taxonomy. Many marine organisms and especially their diverse developmental stages, such as (1) eggs and larvae of fishes and invertebrates, (2) zoo- and phytoplankton, and (3) benthic invertebrates, are cumbersome and difficult to identify by morphological characters. Classical microscopy methods are time-consuming and require a high degree of taxonomic expertise, which is currently falling short. In many cases identifying a species is the major bottleneck in marine biodiversity and ecosystem research, hampering the necessary monitoring of marine biodiversity. As an example, a review of 138 studies on invertebrate diversity in European seas showed that approximately one-third of the specimens could not have been identified to species level (Schander and Willassen 2005). DNA-based identification methods are meanwhile established (Barlow and Tzotzos 1995) as powerful tools, exhibiting an unprecedented accuracy because of their inherently highest possible resolution, which can reach even the level of single base changes in a whole genome. Using these methods, the following marine animals have been investigated: eggs, larvae, and adults of fishes (Rocha-Olivares 1998; Noell et al. 2001; Fox et al. 2005; Ward et al. 2005), planktonic copepods (Bucklin et al. 1999), invertebrate larvae (Garland and Zimmer 2002; Barber and Boyce 2006), and prey in the gut content or feces of penguins, whales, and fishes (Jarman et al. 2002; Saitho et al. 2003). Sequences of the small subunit of the rRNA gene are used as a standard method for identifying microbial organisms (Ludwig et al. 2004), and a fragment of the mitochondrial cytochrome oxidase I gene is in use as “DNA-barcode” for the identification of metazoans (Hebert et al. 2002; Ward et al. 2005). As shown above, a growing number of recently published studies are using molecular genetic identification methods. Nevertheless, their application is still restricted mainly because of methodical problems and because special knowledge and experience in molecular genetics is required. This is especially true if DNA-based identification is performed by using microarray platforms that are error-prone and difficult to quantify (Shi et al. 2006). Whereas most of the methods presently in use, such as PCR-based DNA amplification followed by sequencing techniques, allow to handle only single or a few species at the same time, DNA microarrays are believed to have the potential of identifying hundreds of species in parallel and to differentiate them against an even larger number of related species. Commonly, microarrays are glass microscope slides on which oligonucleotide probes are spotted that are complementary to the DNA target sequences to be analysed (Relógio et al. 2002, Pirrung 2002, Dufva 2005). The DNA target, which is usually fluorophore-labelled during PCR amplification, hybridises with the oligonucleotide probe on the microarray and can be detected after washing steps by its label. Whereas applying of DNA microarrays for gene expression has already reached the routine level of high throughput systems (Blohm and Guiseppi-Elie 2001, Hoheisel 2006), they have been only recently applied for the identification of organisms, such as microbes (Wang et al. 2002; Call et al. 2003; Korimbocus et al. 2005; Loy and Bodrossy 2005), animals (Pfunder et al. 2004), and plants (Rønning et al. 2005). In terms of identifying marine organisms, microarrays have been used for bacteria (Peplies et al. 2003; Peplies et al. 2004) and phytoplankton (Metfies and Medlin 2004; Metfies et al. 2005; Godhe et al. 2007) Other DNA-hybridisation methods for the identification of higher marine organisms, such as invertebrate larvae (Goffredi et al. 2006), copepods (Kiesling et al. 2002), and larvae of fishes (Rosel and Kocher 2002), have been applied, but microarrays have not been used for this kind of studies. Other applications of microarrays in research on marine organisms are gene expression analysis (Williams et al. 2003; Lidie et al. 2005; Wang et al. 2006; Cohen et al. 2007; Jenny et al. 2007) and genotyping in population genetics (Moriya et al. 2004; Moriya et al. 2007). One of the methodical limitations for using microarrays is the design of species specific probes. On the one hand, oligonucleotide probes designed in silico do not always exhibit the experimental hybridisation properties they were selected for and must be empirically tested. On the other hand, the molecular marker must have highly selective characteristics, such as low intraspecific and a high interspecific variation. One of the most frequently used markers in phylogenetics of fishes is the mitochondrial 16S rRNA gene. This gene has a well-characterized secondary structure (Meyer 1993; Ortí et al. 1996) and especially the loop regions exhibit many insertions, deletions, and substitutions forming highly variable molecular features, which usually allow the design of highly specific probes. This is underlined by a study on lionfishes (Kochzius et al. 2003), which showed that individuals of one species exhibit identical 16S rDNA haplotypes even though they were sampled at sites thousands of kilometres apart, but clear differences could be detected between closely related lionfish species. In this study, the development of a DNA microarray is described, demonstrating the suitability of the 16S rRNA gene for designing oligonucleotides as microarray probes to differentiate at least 11 fish species from European seas. Based on these data, a “Fish Chip” for approximately 50 fish species is under construction to support the identification of eggs and larval stages from species that are otherwise difficult to identify, and of adult or processed fishes in fisheries industry. Material and Methods Sampling and DNA Extraction To consider possible intraspecies sequence variations, fishes were collected in five different regions of the European seas: North Sea, Bay of Biscay, and Western, Central, as well as Eastern Mediterranean (Fig. 1, Table 1). All together 267 individual samples from 79 fish species were investigated. In addition sequences from the EMBL sequence data base were included. Voucher specimens and tissue samples have been preserved in absolute ethanol and stored at 4°C or were frozen at -20°C. DNA was extracted from gill filaments with the Agowa mag midi DNA isolation kit (AGOWA, Berlin, Germany) or from muscle tissue with the DNeasy tissue kit (Qiagen, Hilden, Germany) according to the instructions of the manufacturer. Figure 1Map with the sampling areas; NS North Sea; BB Bay of Biscay; WM Western Mediterranean; CM Central Mediterranean; EM Eastern Mediterranean.Table 1Number of sequences per fish species and sampling regionSpeciesFamilyOrderNorth SeaBay of BiscayWestern MediterraneanCentral MediterraneanEastern MediterraneanEMBL sequence data base and other projectsTotalAccession number of EMBL data base sequencesTarget species Boops boopsSparidaePerciformes3418AF247396 Engraulis encrasicolusEngraulidaeClupeiformes3216 Helicolenus dactylopterusSebastidaeScorpaeniformes316111AY538975 Lophius budegassaLophiidaeLophiiformes369 Pagellus acarneSparidaePerciformes3339 Scomber scombrusScombridaePerciformes3126AB120717, AF055615 Scophthalmus rhombusScophthalmidaePleuronectiformes33118AY359665 Serranus cabrillaSerranidaePerciformes347 Sparus aurataSparidaePerciformes3216AF247432 Trachurus trachurusCarangidaePerciformes125210AB108498, AB096007 Trigla lyraTriglidaeScorpaeniformes336Additional species Chelidonichthys lucernusTriglidaeScorpaeniformes13610 Diplodus sargusSparidaePerciformes246 Gadus morhuaGadidaeGadiformes1236X99772, NC_002081, AY850363 Merluccius merlucciusMerlucciidaeGadiformes233614 Mullus surmuletusMullidaePerciformes252211 Mullus barbatusMullidaePerciformes257 Pagellus erythrinusSparidaePerciformes12710 Platichthys flesusPleuronectidaePleuronectiformes42410AY359670, AB125255, AY157320, AF113180 Pleuronectes platessaPleuronectidaePleuronectiformes426AY359673, AY157328 Psetta maximaScophthalmidaePleuronectiformes134 Sardina pilchardusClupeidaeClupeiformes33410 Scorpaena notataScorpaenidaeScorpaeniformes5611 Scorpaena porcusScorpaenidaeScorpaeniformes3317 Serranus hepatusSerranidaePerciformes2237 Solea soleaSoleidaePleuronectiformes332311AB125247, AF488442, AF112845 Zeus faberZeidaeZeiformes332614NC_003190, AF488474, AF221894-AF221896, AP002941 Σ 230Accession numbers are given for sequences obtained from EMBL sequence databaseTaxonomy according to FishBase (2007) Polymerase Chain Reaction and Sequencing A fragment of approximately 1380 bp length from the mitochondrial 16S rRNA gene was amplified with the primer 16fiF140 (5′-CGY AAG GGA AHG CTG AAA-3′), which has a single-base modification compared with Palumbi et al. (1991, unpublished manuscript) as well as with the newly designed primer 16fiR1524 (5′-CCG GTC TGA ACT CAG ATC ACG TAG-3′). Polymerase chain reaction (PCR) reactions with a total volume of 15 μl contained 1.5 μl 10 X reaction buffer, 1.5 μl dNTPs (10 mM), 0.05 μl of each primer (100 pmol/μl), 5 μl DNA-extract, 0.3 μl Teg polymerase (3 U/μl; Prokaria, Reykjavik, Iceland), and 6.6 μl deionized water. Thermal profile began at 94°C for 4 min, followed by 35 cycles of 94°C (30 s), 54°C (30 s), 72°C (90 s), with a final step of 7 min at 72°C. PCR products were purified by using the ExoSAP-IT for PCR clean-up (GE Healthcare, Uppsala, Sweden). The newly designed sequencing primer 16fiseq1463 (5′-TGC ACC ATT AGG ATG TCC TGA TCC AAC-3′) was used to sequence one strand of the amplified fragment using the BigDye Terminator Cycle Sequencing Kit (ver. 3.1, PE Biosystems, Foster City, USA). The sequencing reactions were run in an ABI Prism 3730 automated DNA Analyzer (Applied Biosystems, Foster City, USA) according to the manufacturer’s instructions. Sequence Analysis and Oligonucleotide Probe Design In the framework of this project, 267 sequences have been acquired and 944 sequences have been obtained from EMBL sequence database and other projects, representing approximately 380 species of fish from European seas. Probes were designed based on 230 sequences obtained from 27 species (Table 1). A multiple alignment of these 230 sequences was performed with the programme Clustal W (Thompson et al. 1994) as implemented in BioEdit (version 7.0.4.1; Hall 1999) to ensure that all sequences represent a homologous fragment of the 16S rDNA. Before probe design gaps have been removed from each sequence. A computer program developed by the bioinformatics group of the Centre for Applied Gene Sensor Technology (CAG) and the Zentrum für Technomathematik (ZeTeM), both at University of Bremen, was used to design species-specific oligonucleotide probes, which ideally cover all sequences of one species and do not match any other species (Nölte 2002). The following criteria have been considered: (1) length of 23 to 27 bp, (2) melting temperature (Tm) of 81 to 85°C based on the unified model (SantaLucia 1998), (3) GC content of 52% to 54%, (4) secondary structure of the oligonucleotides and the target sequence, (5) possible dimer formation, and (6) the energy content of a bond between the probe and the target sequence. Minimal free energy (mfe) structures are computed by using RNAfold (Hofacker et al. 1994). Probes exhibiting strong secondary structures or binding to a region of the target with such a strong secondary structure were not used. If more than one probe qualified for a species, the one with the highest binding energy between probe and target was chosen. Already during the design-phase the selected oligonucleotide probes were tested in silico against 1211 background sequences from approximately 380 fish species. Preparation of DNA Microarrays and Hybridisation Experiments Aminosilane (3-aminopropyltrimethoxysilane)-coated glass slides were used with a PDITC-linker (1,4-phenylendiisothiocyanate) from Asper Biotech (Tartu, Estonia). Oligonucleotide probes (Thermo Hybaid, Ulm, Germany) with a 5′-amino-C6-modification were spotted in 150 mM Na3PO4 buffer (pH 8.5) at a concentration of 30 μM using a spotting robot based on a modified version of the contactless TopSpot® technology. The spotted volume of this oligonucleotides solution was 200 pl, producing a spot diameter of approximately 220 μm. Each probe was spotted in four replicates per block. An array contained five blocks and three arrays were spotted on one microarray slide (Fig. 2). After spotting, the microarrays were incubated for 16 h in a wet chamber to ensure efficient covalent binding of the oligonucleotides. Finally, the microarrays were shrink-wrapped under a nitrogen atmosphere and were storable at 4°C for up to 6 months. Fig. 2Layout of the microarray DNA for hybridisation experiments was amplified and labelled with 5′-Cy5-modified primers. The primers 16sar-L (3′-CGC CTG TTT AAC AAA AAC AT-3′) and 16sbr-H (5′-CCG GTT TGA ACT CAG ATC ACG T-3′) amplify a fragment of approximately 600 bp length from the mitochondrial 16S rRNA gene (Palumbi et al. 1991, unpublished manuscript). PCR reactions with a volume of 100 μl contained 10 μl 10 X reaction buffer, 4 μl dNTPs (5 mM), 2 μl of each primer (10 μM), 2 μl DNA-extract, 0.4 μl Taq polymerase (5 U/μl), 2 μl BSA (20 mg/ml), and 77.6 μl deionized water. The PCR thermal profile began at 95°C for 2 min, followed by 35 cycles of 95°C (30 s), 54°C (45 s), 72°C (60 s), followed by a final step of 10 min at 72°C. The Cy5-labelled PCR amplified DNA was purified using the QIAquick PCR Purification Kit (QIAGEN, Hilden, Germany). Hybridisation experiments were performed with 11 target and 14 nontarget fish species. As nontarget species fishes were chosen that are closely related to the target species and from which false-positive signals could be expected according to the in silico specificity tests performed during the design phase of the probes (Table 2). Table 2Nontarget species tested in hybridisation experimentsSpeciesFamilyOrderDentex dentexSparidaePerciformesDiplodus vulgarisSparidaePerciformesGadus morhuaGadidaeGadiformesMelanogrammus aeglefinusGadidaeGadiformesMerlangius merlangusGadidaeGadiformesMerluccius merlucciusGadidaeGadiformesMicromesistius poutassouGadidaeGadiformesMullus surmuletusMullidaePerciformesPollachius pollachiusGadidaeGadiformesPollachius virensGadidaeGadiformesPsetta maximaScophthalmidaePleuronectiformesSerranus hepatusSerranidaePerciformesTrachurus mediterraneusCarangidaePerciformesTrachuru picturatusCarangidaePerciformesTaxonomy according to FishBase (2007) A positive control at a concentration of 1 nM (5′-CGT GTG AGT CGA TGG ATC ATA-3′; 5′-Cy3-labelled) and 10 nM of the purified Cy5-labelled PCR product were hybridized to the microarray in a volume of 65 μl using GeneFrames® (ABgene House, Epsom, UK), which were applied to the microarray slides according to the manufacturer’s instruction (Fig. 2). Hybridisation was conducted at 50°C in a hybridisation oven. After 2 h hybridisation time the GeneFrames® were removed and the microarrays were washed 5 minutes each with 2 × SSC (sodium chloride trisodium citrate) buffer containing 0.05% SDS (sodium dodecyl sulphate), 1 × SSC containing 0.05% SDS, and 1 × SSC. Finally the microarrays were dried in a centrifuge at 2000 rpm for 3 minutes. Measurement of Fluorescence Signals and Data Analysis Hybridisation signals were measured using an Axon 4000B fluorescence microarray scanner at 635 nm (Cy5) as well as at 528 nm (Cy3). The fluorescence signal analysis was conducted with the software GenePix 4.1 (Axon, Union City, USA). The fluorescence signals of each probe were measured and the arithmetic mean was calculated. However, data were removed from the analysis if the spots showed artefacts caused during the spotting process (e.g., inhomogeneous spots documented by a monitoring camera during spotting) or experimental artefacts (e.g., air bubbles). Background noise was corrected by subtracting the arithmetic mean of the negative control measurement from the arithmetic mean of the spot measurements. Negative values were set to zero. Results Probes for 11 commercially important fish species have been designed (Table 3). Positions of the oligonucleotide probes in the 16S rDNA fragment used for probe design are given in Fig. 3. Following the nomenclature of Ortí et al. (1996), the binding sites of the microarray probes for Engraulis encrasicolus, Sparus aurata, and Trigla lyra are located in the variable region j, whereas the probes for Boops boops, Helicolenus dactylopterus, Lophius budegassa, Pagellus acarne, Scomber scombrus, Scophthalmus rhombus, Serranus cabrilla, and Trachurus trachurus bind to the variable region l. Table 3Oligonucleotide probes for the identification of fish species from European seasSpecies nameProbe nameProbe sequence (5′-3′), 5′-amino-C6-modifiedLength (bp)Tm (°C)GC (%)Oligo mfeDimer mfeSpecificity (in silico)Boops boopsBooboo_315GCACCACACTCCTAAACCCAAGA2382.6452≥0−0.07speciesEngraulis encrasicolusEngenc_213CAAGTCCTAAATACCCGCAGCCT2382.4952≥0−0.17speciesHelicolenus dactylopterusHeldac_317ACCCCTCCTACAATTAAGAGCCG2381.8452≥0−0.22speciesLophius budegassaLopbud_312AACACCCTTCCTATCACCCAGAGCTAC2784.3952≥0−0.2genusPagellus acarnePagaca_317TACTACACTCCCACATCCGAGAGC2482.7754≥0−0.89speciesScomber scombrusScosco_321CAACTACTCCTACAGTCAAGAGCCACC2782.9152≥0−0.43speciesScophthalmus rhombusScorho_322CCCCTTAACTCCTCGAAGCAAGA2381.8852≥0−0.37speciesSerranus cabrillaSercab_313CCATTTTCCTACAACCCAGAGCGAC2582.7452≥0−0.18speciesSparus aurataSpaaur_201AGAACAGCTCACGTCAAACACCC2383.0252≥0−0.5speciesTrachurus trachurusTratra_333TTCCTCTCCTCCCACAAGCAAGA2383.6252≥0−0.15genusTrigla lyraTrilyr_232AAGACCGAACCAAATGAGCCCTG2383.1652≥0−0.17familyThe number in the probe name indicates the binding site in the 16S rDNA sequenceOligo mfe minimal free energy of the secondary structure of the oligonucleotide; Dimer mfe minimal free energy of the dimer of two identical oligonucleotide moleculesValues for mfe are given in kcal/molFig. 3Alignment (5′ > 3′) of representative 16S rDNA sequences from the target species with binding sites (light grey) of probes (5′ > 3′; probes hybridise to the reverse complementary target strand). Double stranded (dark grey) and single stranded regions (grey) of the secondary structure are indicated in the reference sequence of Pygoplites nattereri (Ortí et al. 1996; Accession number: U33590) All single target hybridisations of the Cy5-labelled 16S rDNA fragment gave true-positive fluorescence signals for the corresponding probe (Fig. 4), but the hybridisation efficiency is very different. The signal intensity of the weakest probe-target pair gave approximately 1,000 arbitrary units and the strongest almost 30,000 (Table 4). Under the experimental conditions selected, single target hybridisations with B. boops, E. encrasicolus, H. dactylopterus, L. budegassa, S. rhombus, S. aurata, T. trachurus, and T. lyra did not show any false-positive signals. Eight probes gave under these conditions very weak false-positive signals with P. acarne, S. scombrus, and S. cabrilla, representing approximately 7% of all possible cross-hybridisation reactions. The values of these false-positive signals were several orders of magnitude lower than the true positive-signals. Figure 4Signals (background subtracted from absolute signal) of single target and nontarget hybridisations. White bars represent true-positive signals; false-positive signals are shown as grey bars. Numbers at the basis of the bars indicate the amount of measured spots. The number of hybridisations is given in brackets after target and nontarget names. Replication and absolute signal intensities (± standard deviation) of hybridized targets to the corresponding probe are given in Table 4.Table 4Target hybridisationsHybridized targetsNo. of hybridisationsMeasurements of specific probesMeasured probes/absolute no. of probesMean absolute fluorescence signal in arbitrary unitsStandard deviationBoops boops221/402991±1491Engraulis encrasicolus220/401659±962Helicolenus dactylopterus220/403502±912Lophius budegassa240/403450±1515Pagellus acarne227/403727±1270Scophthalmus rhombus115/201528±269Scomber scombrus240/4027827±5330Serranus cabrilla240/4010814±4396Sparus aurata240/40963±227Trachurus trachurus120/202015±880Trigla lyra235/402343±560 Single nontarget hybridisations showed 48 very weak false-positive signals, representing 31% of 154 possible cross-hybridisations. These false-positive signals were at least one order of magnitude weaker than the true-positive signal. Merlangius merlangus and Merluccius merluccius did not gave any false-positive signal, whereas Dentex dentex, Diplodus vulgaris, Gadus morhua, Melanogrammus aeglefinus, Micromesistius poutassou, Mullus surmuletus, Pollachius pollachius, Pollachius virens, Psetta maxima, Serranus hepatus, Trachurus mediterraneus, and Trachurus picturatus showed two to eight cross-hybridisations. Considering all possible cross-hybridisations, only 18% showed usually very weak false-positive signals. Discussion The results show that the “Fish Chip” described enables the identification of 11 commercially important fish species from European seas in certain experimental limits. These limits are primarily given by the fact that the fluorescence signal intensities of true-positive hybridisation signals were heterogeneous. This phenomenon is commonly encountered in DNA microarray experiments (Peplies et al. 2003; Warsen et al. 2004; Korimbocus et al. 2005; Rønning et al. 2005; Tobler et al. 2006) and can probably be overcome only by an extreme methodical effort (Shi et al. 2006). The problems encountered when single-colour microarray experiments need to be quantified are severe and in part not yet solved, because complex parameters are influencing the results. These are the sequence dependent hybridisation efficiency, specifically steric hindrance, secondary structures (Southern et al. 1999), and the relative position of the fluorescent label at the target (Zhang et al. 2005). Sometimes the duplex formation can be favoured by using spacers, which obviously give the captures a greater degree of freedom from their neighbouring molecules and from the surface (Southern et al. 1999), leading to an enhancement of signal intensity with increasing spacer length (Peplies et al. 2003). An important criterion of probe design was the base composition (Southern et al. 1999) and therefore all but one capture oligonucleotides were designed to have a GC content of 52% and 54% for the exception (Table 3). The different hybridisation efficiencies of the captures and the varying sensitivity for the different fishes is still a severe disadvantage, because it can hamper the estimation of a small amount of fishes of one species in the presence of a larger number of individuals of another species in a mixed sample. These limits are presently under investigation with fish eggs and other biological material. Because all oligonucleotide probes bind to the variable regions j and l of the 16S rRNA gene, which represent large single-stranded loops, the secondary structure is unlikely to be a factor contributing to the partially low sensitivity. Also, the POL effect, a phenomenon decreasing the fluorescence signal with increasing distance between the probe binding site and the label on the target (Zhang et al. 2005) seems not to be important in this case. Most of the probes used in this study hybridise to the variable region l of the 16S rDNA and their distance from the fluorescence label is more or less identical. Although cross-hybridisations occur, true-positive signals could clearly be differentiated from false-positive signals because of their generally higher signal. Most false-positive signals occurred when the 16S rDNA fragment of nontarget species was hybridised on the microarray. Testing of nontarget species is seemingly important in studies using DNA microarrays for the identification of organisms, because even if a comprehensive sequence background is utilised for probe design and even if extensive in silico testing has been performed, the specificity of oligonucleotide probes has to be evaluated with closely related species and specifically those that show cross-hybridisations in silico. This study shows that the 16S rRNA gene of fishes is suitable to design oligonucleotide probes that are able to differentiate eleven fish species from European seas by single target hybridisation on a microarray. Such a “Fish Chip” can hopefully be applied in marine environmental and fisheries research, as well as in fisheries and food control if the uneven hybridisation signal intensities of the different probe-target pairs can be improved or compensated. The correct identification of fish eggs and larvae is crucial for fish stock assessment based on ichthyoplankton surveys. Genetic identification has shown that the majority of eggs in the Irish Sea, wrongly believed to be from cod, were actually from whiting, leading to an overestimation of cod stocks (Fox et al. 2005). A study on food fish in the United States revealed that three-quarters of fish sold as “red snapper” were mislabelled and belonged to other species (Marko et al. 2004), a situation that needs better analytical tools to be changed. The European Union (EU) also has strict regulations for seafood labelling, which must include, for example, the species name (EU Council Regulation No 104/2000; EU Commission Regulation No 2065/2001). Approximately 420 species of fish are sold in the German market alone, making a reliable identification method urgently necessary to protect the customer. DNA microarrays might have the potential to fulfill these requirements. Recent efforts in compiling sequences of fishes, such as the European projects “FishTrace” http://www.fishtrace.org) and “Fish & Chips” (http://www.fish-and-chips.uni-bremen.de; Kochzius et al. 2007), as well as the international “Fish Barcode of Life Initiative” http://www.fishbol.org; Ward et al. 2005), will provide the necessary sequence background for the design of species specific oligonucleotide probes for the development of DNA microarrays for the identification of fishes.

Neurochem_Res-4-1-2295255

Increased Nitric Oxide Production and GFAP Expression in the Brains of Influenza A/NWS Virus Infected Mice

The cause of influenza to the brain was investigated using the A/NWS/33 influenza virus infected BALB/c mouse model. NOS-2 mRNA levels in the infected mouse brain was greater than in control mice in all brain regions examined, particularly in the olfactory bulb and hippocampus by 1 day p.i. On the contrary, no differences in NOS-1 or NOS-3 mRNA levels were found between infected and control mice. There was also a marked increase in the levels of metabolites of nitric oxide in the olfactory bulb and hippocampus. Immunohistochemistry showed positive staining for anti-NOS-2 primarily in the hippocampus of infected mice. Further, anti-NOS-2 and GFAP staining was mostly found around capillary blood vessels of the hippocampus starting early in the course of the disease. These results indicate that the NWS enhances the activation of astrocytes and NOS-2 expression which in turn enhances NO production and the expansion of capillary blood vessels. Introduction Influenza associated encephalopathy (IE) is one of the most severe neurological diseases amongst children, mostly under the age of 5. It is marked by rapid progression and the mortality rate is estimated to range from 27 to 44% [1]. Recently, we reported that production of the biological free radical gas, nitric oxide (NO), in the serum and cerebrospinal fluid (CSF) of patients with IE was increased in the initial stages and correlated with disease severity [2, 3]. Yokota et al. hypothesized that the influenza virus activates the limbic system followed by glial cell activation which in turn produces NO and inflammatory cytokines which flood the brain and cause a cytokine storm [4]. Recent in vitro studies by Imanishi et al. showed that a mouse lung-adapted strain of influenza virus could induce inducible NO synthase (NOS-2) mRNA expression and NOS-2-mediated NO production in murine macrophage cell [5]. Moreover, Ino et al. studied five autopsies cases of IE patients. They reported that four out of the five patients showed degenerated astrocytic spines in the cortex, by glial fibrillary acidic protein (GFAP) staining. They suspected that the IE was largely due to these altered glial cells (Ino et al. unpublished). Taken together, it is likely that IE is related to both the influence of NO and glial cells changes. Nevertheless, there have been no reports on a relationship between NOS-2 and glial cells within the brain after infection with the influenza virus. Thus, the aim of these studies is to better understand the clinical state of IE by focusing on NO and astrocytes in the central nervous system (CNS) after influenza virus infection. By using the A/NWS/33 influenza virus infected BALB/c model mouse, we were able to demonstrate that influenza virus infection leads to an up regulation of NOS-2 and astrocytes, mostly around capillary blood vessels of the hippocampus and olfactory bulb, starting at an early stage of the disease. Although IE has been reported for more than a decade, a therapy for IE has not yet been defined. We expect that our results will provide evidence which will aid in the development of a novel IE therapy. Experimental procedures Virus Influenza A/NWS virus, which is a mouse brain adapted type of human influenza A/NWS/33 virus was used. (A kind gift from Dr. K. Hayashi, Toyama University, Japan.) Experimental animals In total, 82 male BALB/c mice (5-week-old) were purchased from Japan Charles River Co. and given free access to food and water (Oriental Yeast, Co., Tokyo) and acclimatized for at least a week before the experiment. Anesthetics Each mouse was anesthetized by inhalation of ether before being inoculated intranasally with 50 μl NWS, 5.0 × 107 plaque-forming units (PFU)/ml (5.0 × 105 PFU/mouse). For negative control mice, the same amounts of 0.1 M phosphate-buffered saline (PBS) were intranasally inoculated. The mice were examined at 1, 3 and 6 days postinfection (p.i.). Body weight Body weight changes were monitored daily. Euthanasia All mice were decapitated whilst under ether inhalational anesthesia. Each brain was divided into eight regions (frontal cortex, occipital cortex, cerebellum, medullar oblongata, hippocampus, corpus striatum, thalamus/hypothalamus and olfactory bulb). Then each section was immediately frozen with liquid nitrogen. NO2− and NO3− measurements by HPLC-UV method Both NO2 and NO3 (NOx) levels in each brain region were measured by HPLC-UV system (ENO-10, EICOM) which was reported by us [3]. Briefly, thawed brain regions were deproteinized with a ten times volume of methanol and then centrifuged (12,000 r.p.m. for 20 min at 4°C) to avoid an occlusion in the polystyrene polymer column used in this system, by protein or other macromolecules. Detection was measured at 540 nm (absorption) and the concentrations were calculated from the area under the curve (AUC) of NO2 and NO3 (Power Chrome, EICOM, Kyoto) by Griess reaction. RNA extraction RNA was extracted from each part of the brain by using a QIA amp RNA extraction kit (QIAGEN K.K., Tokyo), according to the protocol suggested by the manufacturer. cDNA synthesis cDNA was synthesized from purified RNA by using the Transcriptor First Strand cDNA Synthesis Kit (Roche, Switzerland). Random hexamer primer was used and the protocol was as follows; after annealing for 10 min at 25°C, and incubating 30 min at 55°C, the reaction was heated to 85°C for 5 min and chilled on ice. Real time reverse transcriptase-polymerase chain reaction (RT-PCR) Messenger RNA levels of NOS-1, NOS-2 and NOS-3 were quantified by using a Light Cycler Fast Start DNA Master HybProbe (Roche). This was done according to the following protocol; Pre-incubation was 95°C for 30 s. This was followed by 40 cycles of denaturation at 95°C for 10 s, annealing at 62°C for 15 s, amplification at 72°C for 8 s for 45 cycles. Cycles were followed by 30 s 40°C cooling program. The primers and probes were designed by the Light Cycler primer and probe set (Roche Diagnostics Applied Science, Tokyo). NOS-2, NOS-3 and NOS-1 primer sequences were designed on GenBank database NM_0100927, NM_008713, and NM_008712 respectively. All data were normalized by dividing with the corresponding GAPDH mRNA from the same sample. Immunohistochemistry Anesthetized animals were perfused intracardially with 20 ml of 0.1 M PBS, followed by 30 ml of 4% paraformaldehyde in 0.1 M phosphate buffer. The brains were removed, postfixed overnight and cryoprotected in 30% sucrose, processed through graded alcohols and embedded in paraffin. Then, they were cut as 6 μm thick coronal sections. They were incubated with 0.3% H2O2 for 20 min. The sections were then rinsed with tris-buffered saline, pH 7.6 (TBS). To identify reactive astrocytes, we stained with Anti-GFAP (rabbit polyclonal, DAKO, Denmark). Sections were incubated over night at a 1:2,000 dilution. The NOS-2 stains were incubated for 4 h at a 1:1,500 dilution (Anti-iNOS/NOSII, rabbit polyclonal IgG;Upstate Biotechnology, Lake Placid, NY). After the first staining, each section was incubated with the secondary antibody for 30 min (Nichirei Histofine Simple Stain, anti-rabbit IgG) at room temperature after TBS rinsing. Labeling was visualized by incubating with a solution containing 2%, 3,3′-diaminobenzidine, 0.1 M imidazole and 30% H2O2. All sections were counterstained with Mayer’s hematoxylin (SIGMA, MI). Statistical analysis Fisher’s exact test was used when comparing body weights. All results obtained from the real time RT-PCR were subjected to one- or two-way analysis of variance (ANOVA) and differences among the means were analyzed by two-way ANOVA followed by a Newman–Keuls range test at the 0.05 significance level. Results are expressed as mean ± standard error. Results Body weight Figure 1 shows the change in body weight of the non-infected (PBS) and infected (Flu) mice over day. The mice infected with virus showed a decrease in weight and ruffled hair from 3 days p.i. and by 6 days p.i. and there was a significant decrease in the body weight of the infected mice as compared to the non-infected mice (P < 0.05). Prominent neurological signs in the infected mice did not appear during the experimental period and none of the mice died during the experiment, although spontaneous locomotor activity was reduced in the infected mice. Fig. 1Changes in body weight of A/NWS/33 influenza virus infected BALB/c model mice after virus infection. Each result shows the mean ± SE value of eight experiments. () non-infected (PBS treatment) mice group, () infected (Flu) mice group Viral titers At one day p.i., 120 PFU/50 mg virus was detected in the infected mouse brain and 136 × 104 PFU/100 mg virus was detected in the infected mouse lung. Real-time RT-PCR for NOS-2, NOS-3 and NOS-1 Figure 2 shows the NOS-2/GAPDH and NOS-3/GAPDH ratios as measured in the olfactory bulb and hippocampus of infected mouse brains as compared to the non-infected mice on days 1, 3 and 6 p.i. Due to the scattered results obtained in both the infected and non-infected mice, data on the NOS-1/GAPDH ratio is not reported. The increases in the NOS-2/GAPDH ratio as seen in the olfactory bulb(OB) and hippocampus(HIP) at day 1 and 3 p.i., were obvious as shown in Fig. 2a and b. However, no changes were seen in the levels of NOS-3/GAPDH in the same regions (Fig. 2a’, b’). Fig. 2NOS-2/GAPDH and NOS-3/GAPDH ratios in the olfactory bulb and hippocampus of non-infected and infected mice on days 1, 3 and 6 p.i. The open column shows the mean and S.E. value obtained from nine non-infected mice and the closed column shows the mean and S.E. value obtained from eight infected mice. *Significant difference between non-infected mice and infected mice. *P < 0.05. (a and a’): olfactory bulb, (b and b’): hippocampus Changes in brain NO levels between non-infected and infected mice Figure 3 shows that NO levels in the eight brain regions examined, either with or without NWS virus infection, changes in a day-by-day manner after virus infection. The levels of NO2 and NO3 were measured separately, as shown in the Materials and methods. In the isolated brain regions, NO3 levels were more than 40-fold higher than NO2 levels, and the final data are shown as the total value (NO) of NO2 plus NO3. At day 1 p.i., the NO levels detected in olfactory bulb (OB), hippocampus (HIP) and frontal cortex (FC) of the infected mice (Flu) were significantly increased as compared to the levels in the same brain regions of the non-infected (PBS) mice (P < 0.05). Further, at day 6 p.i., the increased levels of NO in the olfactory bulb (OB) and the frontal cortex (FC) returned to the levels seen in non-infected mice. However, the levels of NO in the hippocampus (HIP) were significantly increased (P < 0.05) at all the time points examined. Fig. 3Day to day changes in NO levels in the eight brain regions either with or without NWS virus infection. The open column shows the mean and S.E. value obtained from eight non-infected mice and the closed column shows the mean and S.E. value obtained from seven infected mice. *Significant difference between non-infected mice and infected mice. *P < 0.05. OB: olfactory bulb, HIP: hippocampus, St: striatum, Hypo/Thal: hypothalamus and thalamus, MO: medulla oblongata, CE: cerebellum, FC: frontal cortex, and OC: occipital cortex Immunohistochemistry Figure 4 shows the results of anti-NOS-2 staining and GFAP staining in the hippocampus of non-infected (PBS) and infected mice (Flu) at 1 (Fig. 4a) and 6 (Fig. 4b) days p.i. in the same magnification each. The brown-yellow staining was recognized as positive staining for anti-NOS-2. We found positive staining in the hippocampus, especially in the area surrounding capillary blood vessels of the infected mice at 1 day p.i. (Fig. 4a, Day 1, Infected). By 6 days p.i., positive staining had spread to the striatum, hippocampus and cortex of the infected mice and positive staining were seen in the areas surrounding capillary vessels (Fig. 4b, Day 6, Infected). Add that infected mice showed an increase in GFAP staining on days 1 much more so than that seen in non-infected mice. Positive staining was most notable around hippocampal blood vessels in infected mice. Additionally, most of the blood vessels in hippocampi of infected mice were markedly expanded as compared to those seen in the non-infected mice. Fig. 4Anti-NOS-2 immunohistochemistry and GFAP immunohistochemistry in the hippocampus at 1 (Fig. 4a) and 6 (Fig. 4b) days p.i Discussion Since our aim was to see the effect to the brain after influenza infection, this experiment was undertaken by using the A/NWS/33 virus because it is proven by Li et al that the NWS/33 strain can replicate in the brain of adult mice after intracerebral injection, add that the virus usually fails to spread beyond the respiratory epithelium after internasal instillation [6]. During the experiment, we observed that the NWS virus infected mice lose significantly more body weight at 6 days p.i. (Fig. 1) in agreement with previous studies [5, 7]. Since bodyweight changes are an indicator of the overall condition of the mice, this bear out that the mice was infected to the virus. More over, we have checked that RT-PCR for A/NWS/33 virus were positive in all brain regions expect medullar oblongata on day 1 p.i. (data not shown). The reason why we determined the amounts of influenza virus in this experiment is as followed. During the past experiment, we had experienced that the mice die within 3 days p.i. when inoculating any more than the virus we used this time and when we use fewer viruses, there were hardly any difference in the non-infected and infected behavioral change. Our results show that an enhancement NO biosynthesis was induced in the NWS virus infected mouse brain. Over-production of NO was substantiated by determining an increase in the expression of NOS-2 mRNA as shown in Fig. 2, and increased immuno-histochemical staining for anti-NOS-2 as shown in Fig. 4. At 1 day p.i., the amount of NOS-2 mRNA within the infected mouse brain was greater than that seen in control mice in all brain regions studied. Above all, the olfactory bulb showed the greatest increase. On the basis of these results, we compared NOS-2 mRNA expression in the olfactory bulb on days 1, 3, and 6 p.i., respectively. This showed that the NOS-2 mRNA ratio was greater in the infected mouse brain at all experimental times studied (Fig. 2). On the contrary, there was no difference in NOS-3 or NOS-1 mRNA levels between infected and control mice (data not shown). These results are important since these three isoforms of NOS, NOS-2 the inducible form of NOS, is capable of producing a continuous flux of NO and a large amount of NO generated by NOS-2 over a sustained period is implicated in the pathogenesis of various disorders and with the response to infectious organisms [8]. In regards to the immuno-histochemical analysis, positive staining for anti-NOS-2 was mostly detected in the hippocampus of infected mouse brains, whereas hardly any staining was seen in control brains (Fig. 4). Furthermore, positive staining had spread to the hippocampus and striatum by 6 days p.i., which correlated with the transition of NO levels. NOS-2 has been shown to play a role in the defense against bacteria, parasites and viruses. However, it seems that NO induced by the influenza virus has a dual role. Rimmelzwaan et al. reported an inhibitory effect of NO on virus growth [9], whereas Akaike et al. reported that NO played a pathological role in the influenza virus-induced pneumonia in mice [10]. The fact that the influenza virus can induce NO has been reported previously [2, 3]. Yao et al. reported that the mRNA’s of both NOS-2 and NOS increased in the brain after the intranasal infection with influenza A virus of newborn mice [11]. Further, Imanishi et al. pointed out that a mouse lung-adapted strain of influenza virus could induce NOS-2 mRNA expression and NOS-2-mediated NO production in murine macrophage cells [5]. Our study has demonstrated that the NWS virus can induce NOS-2 expression in the infected mouse brain, especially in the olfactory bulb and hippocampus. These results may relate to the fact that intranasally administered NWS virus can invade the brain through the olfactory pathway as has been reported previously [12–16]. Recently, Yossi et al. reported that astrocytes are capable of rapid and massive NOS-2-dependent NO production in brains of normal mice in situ [17]. According to the report by Ino et al. on the brain autopsy of patients who had died of IE, four out of five IE patients showed degenerated astrocytic spines in the cortex by GFAP staining. They suspected that the IE was largely due to altered glial cells (Ino et al. unpublished). Takahashi et al. also made a statement on the autopsied case of a 2-year-old girl with encephalopathy associated with influenza A virus infection and according to their report; reactive astrocytes were seen in the vicinity of blood vessel in the brain but astrocytosis was not seen anywhere, including the areas where virus antigen-positive neurons were abundant [18]. Additionally, Yokota et al. hypothesized that the influenza virus activates the limbic system followed by the activation of glial cells which produce NO and other inflammatory cytokines which together flood the brain and cause a cytokine storm [4]. Astrocytes are the major glial cell in the CNS and have important physiological properties in terms of CNS homeostasis. Astrocytes also influence the formation and maintenance of the blood–brain barrier (BBB). The importance of astrocytes is demonstrated by a study from Bush et al. According to their report, an astrocyte ablation led to failure of BBB repair and vasogenic edema [19]. There have been several reports related to glial cells and NO [17, 20, 21]. However, there has not been a report considering the relationship between influenza virus infection, NOS-2 and astrocytes together. By determining an increase in GFAP staining as shown in Fig. 4, our results show that astrocytes are activated in the infected mouse brain and especially in hippocampus. Interestingly, we noticed that both the increased NOS-2 and GFAP staining were mostly found around capillary blood vessels of the hippocampus starting at an early stage. Our results suggest that since NOS-2 and astrocytes became activated around capillary blood vessels after viral infection, this may lead to brain capillary blood vessels break down. However, we were unable to determine whether the NOS-2 measured is derived from vascular endothelial cells or macrophages which will be an issue for future experiments. Although IE has been mostly reported in Asian countries, influenza infection occurs in epidemics in both the northern and southern hemispheres and is a pandemic disease throughout the year. Recently, influenza virus which is resistant to anti-influenza drugs has been shown to emerge and there is real concern in regards to a new type of influenza pandemic [22]. However, no standardized therapy for the treatment of IE has yet been defined. In this paper, we suggest that the influenza virus causes the activation of astrocytes and NOS-2 in the brain. Thus it is likely that our results will add supportive evidence for the development of a novel IE therapy.

Photosynth_Res-3-1-2117334

Time sequence of the damage to the acceptor and donor sides of photosystem II by UV-B radiation as evaluated by chlorophyll a fluorescence

The effects of ultraviolet-B (UV-B) radiation on photosystem II (PS II) were studied in leaves of Chenopodium album. After the treatment with UV-B the damage was estimated using chlorophyll a fluorescence techniques. Measurements of modulated fluorescence using a pulse amplitude modulated fluorometer revealed that the efficiency of photosystem II decreased both with increasing time of UV-B radiation and with increasing intensity of the UV-B. Fluorescence induction rise curves were analyzed using a mechanistic model of energy trapping. It appears that the damage by UV-B radiation occurs first at the acceptor side of photosystem II, and only later at the donor side. Introduction The solar spectrum that reaches the earth’s surface includes wavelengths in the visible or photosynthetically active range (PAR, 400–700 nm), in the ultraviolet-A (UV-A, 320–400 nm), ultraviolet-B (UV-B, 280–320 nm) and in the ultraviolet-C (UV-C, 200–280 nm) ranges. While UV-C is totally absorbed by the atmosphere, ozone is the principal atmospheric attenuator of UV-B radiation (Cockell and Horneck 2001). The depletion of ozone by the emission into the atmosphere of man-made chlorine- and bromine-containing compounds has been correlated with an increase in the background level of UV-B radiation. UV-B radiation can affect a multitude of physiological and morphological plant processes that ultimately can lead to inhibition of growth and reproduction (Jansen et al. 1998). Potential molecular targets for direct UV-B damage via photomodification or photosensitization include nucleotides, proteins, lipids and pigments (Jordan 2002). Besides direct macromolecular damage, UV-B can also induce specific signal transduction pathways and changes in gene expression that stimulates acclimation and repair processes (Hollósy 2002; Brosché and Strid 2003). The photosynthetic process can be affected by UV-B radiation at different levels, including alterations in plant and leaf morphology that decreases light interception (Jansen 2002), changes in stomatal function that limit the availability of CO2, changes in photosynthetic pigments (Strid and Porra 1992), on the expression of photosynthetic genes and on enzymes of the carbon fixation pathway. However, it is the effects of UV-B radiation on light harvesting and primary photochemical reactions of photosynthetic membranes, particularly on the Photosystem II (PS II) reaction center, which has attracted much attention and study (Vass 1997; Vass et al. 2005). PS II is one of the reaction centers that, together with the cytochrome b6f complex, the photosystem I (PS I) reaction center, and the ATP-synthase, forms the electron transport chain that drives energy transduction in the thylakoid membranes of oxygenic eukaryotes. The photochemical core of PS II is formed by the D1/D2 heterodimer, where the redox electron carriers and cofactors of electron transport are bound. Light absorbed by the antenna system of PS II induces the excitation of a special reaction center chlorophyll (P680), which is photo-oxidized on the first electron transfer reaction of PS II, with a pheophytin molecule (Pheo) acting as the primary electron acceptor and the formation of a radical pair state (P680+Pheo−). This radical pair state is “stabilized” at the reducing (acceptor) side of PS II by the electron transfer from Pheo− to a primary D2-bound quinone electron acceptor (QA) and then to the secondary D1-bound quinone acceptor (QB), which, upon accumulation of two reducing equivalents, and becoming protonated to plastoquinol, dissociates from the reaction center. At the oxidizing (donor) side of PS II, P680+ is reduced by a redox-active tyrosine residue of D1 (Yz) which acts as an electron transfer intermediate between P680+ and the oxygen-evolving complex (OEC), the metalloenzyme system composed of a cluster of four Mn ions and inorganic cofactors (Ca2+ and Cl−) that oxidizes water into molecular oxygen. Accordingly, the consecutive photochemical formation of the primary radical pair P680+Pheo− by light absorption and the reduction of P680+ forming the radical YZ+ drives the sequential four-step oxidation of the OEC inducing the S-state transitions and ultimately the splitting of the water molecule. For detailed reviews on oxygen evolution and PS II structure and function see, Diner and Babcock (1996), Xiong et al. (1998), and Van Rensen and Curwiel (2000). From the above brief description of the PS II components and the potential molecular targets of UV-B radiation, it is clear that several components of PS II could be directly affected, as actually has been observed (Vass et al. 1996). Several studies, most of them in vitro, have shown that UV-B can inhibit PS II electron transport by damage to the quinone electron acceptors’ redox function (Rodrigues et al. 2006), to Yz function, to the OEC, and to the D1 protein. It should be noted that the time-sequence of events that ultimately leads to inactivation of PS II function is not completely clear, as it might involve a direct destruction or impairment of the absorbing molecule, modification of the protein environment to which these redox components are bound as well as a process of energy transfer by the sensitizer species to the damaged site (Vass 1997). The time sequence of damage to acceptor and donor sides of PS II, respectively has never been measured in a single experiment. In this work, we have studied the effect of UV-B radiation on the function of PS II in leaves of Chenopodium album using both steady-state (light-adapted) fluorescence and the time-resolved fast Chl-a fluorescence induction curve. For a review on Chl-a fluorescence, see Govindjee (1995). The results were analyzed based on a mechanistic model of energy trapping and electron transport of PS II (Vredenberg 2000). It appears that damage to the acceptor side of PS II occurs first, and that only later the donor side becomes affected. Materials and methods Plant growth Plants of Chenopodium album L. were grown as described earlier (Rodrigues et al. 2006). After around 28 days the plants were transferred to a growth cabinet with a constant temperature of 20°C, 70% relative humidity and a PAR level of 60 μmol m−2 s−1. The plants were watered daily and after 10–14 days the sampling of leaves for the UV exposure treatments started. Several different batches of plants were used and no significant difference could be detected when comparing results from experiments replicated with plants with similar ages but from different batches. UV-B radiation treatments The plant material used for the exposure to UV light was either fully developed detached leaves or leaf discs (0.6 cm2) floating on petri-dishes filled with demineralized water with the abaxial leaf surface exposed to the radiation. The control samples were either kept in the darkness or exposed to white light (PAR). In some experiments leaves were vacuum infiltrated with 3-(3,4-dichlorophenyl)-1,1-dimethylurea (DCMU) in 0.5% ethanol before being exposed to UV-B, with the controls infiltrated with a 0.5% ethanol solution. UV-B irradiations were performed using a Vilber Lourmat lamp with filter (VL-115M, Marne-la-Vallée, France). This lamp has a peak emission at 312 nm. UV-B irradiance was measured with an optometer (United Detector Technology Inc, Baltimore, USA) equipped with a probe specific for UV radiation. Chlorophyll a fluorescence measurements Measurements of Chl-a fluorescence were performed at room temperature on leaf samples that had been dark adapted for 1 h after the exposure to UV-B radiation, unless otherwise stated. Steady-state Chl-a fluorescence was measured with a pulse-amplitude fluorometer (PAM 101–103, Walz Inc., Effeltrich, Germany). A weak modulated measuring light (1.6 kHz, 650 nm) was used for the determination of the minimal fluorescence level Fo, followed by a 1 s pulse of saturating light (5,500 μmol m−2 s−1) for the measurement of the maximum fluorescence level Fm. The leaf was then irradiated with a fixed or increasing level of actinic PAR and, at each level, we measured the steady state light-adapted fluorescence (Fo′), the maximum fluorescence of a light-adapted state (Fm′) induced by a saturating light pulse superimposed on the actinic light, and after removing the actinic light and the application during 10 s of far-red light (6 μmol m−2 s−1, 720 nm), the minimum fluorescence of a light-adapted state (Fo′). During the saturating pulses the modulation of the measuring light was automatically changed to 100 kHz. Several fluorescence parameters were calculated from the measured fluorescence levels (Schreiber et al. 1986; Genty et al. 1989; Van Kooten and Snel 1990): the potential photochemical efficiency (quantum yield) of PS II, expressed as the ratio Fv/Fm with Fv = Fm − Fo; the coefficient of photochemical quenching, qP, given by (Fm′ − Fs)/Fv′ with Fv′ = Fm′ − Fo′; the efficiency of excitation energy capture by open PS II reaction centers, Fv′/Fm′; and non-photochemical quenching, qN, given by 1 − (Fm′ − Fo′)/(Fm − Fo) (Walters and Horton 1991). The fast Chl-a fluorescence rise was measured with the Plant Efficiency Analyser (PEA, Hansatech Ltd, Norfolk, UK). Leaf samples were excited with a saturating pulse of red light (peak at 650 nm, 3,500 μmol m−2 s−1) and the fluorescence signal recorded at an acquisition interval of 10 μs in the first 2 ms of excitation (for more details on this instrument see Strasser et al. (1995, 2000): a model describing discrete steps of excitation energy and electron transfer associated with PS II was published by Zhu et al. (2005). Due to the limiting response time of the fluorescence detector in the time region below 50 μs, the Fo level was taken as the model-calculated (see below) fluorescence level at 10 μs. This procedure gave similar results to the method of extrapolation as proposed by Vredenberg (2000). The fast Chl-a fluorescence rise is plotted in a log-time scale either as the absolute measured values or as the relative variable fluorescence (rFv), calculated as (Fm − Ft)/(Fm − Fo), with Ft as the fluorescence level at any time. Modeling and interpretation of fast Chlorophyll a fluorescence rise The fast Chl-a fluorescence rise curve was modeled according to Vredenberg et al. (2001). The mathematical formulation is based on a three-state model of energy trapping and fluorescence (Vredenberg 2000) and equates the polyphasic fluorescence rise upon a saturating pulse of light to a multi-state transfer pattern of the PS II reaction center. Results and discussion Exposure of the leaves to UV-B radiation caused an increase of Fo and a decrease of Fm (Fig. 1B), resulting in a decrease of the Fv/Fm value (Fig. 1A). Using a lower intensity of UV-B radiation on grape leaves, Pfündel (2003) also found a decrease of Fm. However, in his experiments, the Fo first decreased with time, and only increased after about 3 h. The increase of the Fo indicates an increase in the amount of QB-nonreducing centers; the decrease of the Fv/Fm value indicates that UV-B has a strong damaging effect on the functioning of PS II. Fig. 1Effect of UV-B radiation of variable duration on Fo, Fm, and the Fv/Fm ratio in leaf discs. The UV-B radiation level was 720 (■), or 150 (◯) μmol m−2 s−1, respectively (PAR level was zero); control samples were kept in darkness. Each value is the mean of four measurements and the bars indicate ±S.D.; bars not shown are within the symbols Illumination with white light during 6 h decreased the photochemical quenching qP from 1.0 at the lowest intensity to a value of 0.58 at 656 μmol m−2 s−1 (Fig. 2). Curwiel et al. (1993), using the same plant material, reported a decrease to about 0.80 after illumination at about 500 μmol m−2 s−1. Up to 100 μmol m−2 s−1, irradiation with UV-B has a stronger effect on qP than white light; at 200 μmol m−2 s−1 and higher, the effect of UV-B is smaller. The efficiency of excitation energy capture by open PS II reaction centers, Fv′/Fm′, was already decreased strongly after 6 h at 5.4 μmol m−2 s−1 of UV-B, while the effect of white light is much smaller; comparable results were found for the quantum yield of PS II given by (Fm′ − Fs)/Fm′. The non-photochemical qN increased from 0 at the lowest intensities to 0.51 (white light) and 0.41 (UV-B) at the highest irradiation levels. Curwiel et al. (1993) reported a value of about 0.6 after illumination with white light at about 550 μmol m−2 s−1. Fig. 2Effects of 6 h irradiation, at various levels, of detached leaves on some fluorescence parameters. CO is control; control is white light (PAR); UV-B is given by the Vilber lamp (zero PAR). Each value is the average of 4 measurements and the bars indicate ±S.D.; bars not shown are within the symbols Measurements of Chl-a fluorescence transients of intact leaves with the PEA fluorometer are presented in Fig. 3. In Fig. 3A the effects of 0, 1, 3, and 6 h irradiation with the Vilber UV-B lamp at 150 μmol m−2 s−1 are illustrated. Also in these measurements, the most obvious effects of UV-B are an increase of the Fo level and a decrease in Fm. After infiltration with water (plus 0.5% ethanol) the effects of UV-B appear the same (compare in Fig. 3B, □ with ■). Compared with infiltration with water, infiltration with DCMU (●) causes a much higher Fo level, and a faster increase to the J-P level; irradiation with UV-B after DCMU-infiltration (◯) causes a still higher Fo level and a slower increase to the J-P level. Comparable results were found using another PS II-inhibiting herbicide, atrazine (not shown). The higher Fo level indicates an increased amount of QB-nonreducing centers. Apart from the Fo level, the differences between the curves before and after UV-B treatment are much smaller in the presence of DCMU. This may be related with the finding by Jansen et al. (1993) that PS II-inhibitors like DCMU, which replace QB from its niche on the D1 protein, inhibit UV-B-driven D1 degradation. Fig. 3Fluorescence induction curves after various treatments of leaves. (A) Fluorescence transients after 0, 60, 180, and 360 min irradiation with UV-B light at 150 μmol m−2 s−1. ( PAR is zero). (B) ■, After infiltration with 0.5% ethanol (CO is control), no UV-B; ●, after infiltration with 10−5 M DCMU in 0.5% ethanol, no UV-B; □, after infiltration with 0.5% ethanol and 6 h irradiation with 150 μmol m−2 s−1 UV-B light; ◯, after infiltration with 10−5 M DCMU in 0.5% ethanol and 6 h irradiation with UV-B. Each curve is the average of three measurements To study the kinetics of the various parts of the transients, induction curves were normalized (Fig. 4). While the level of the J transient increases a little after 60 min UV-B, it decreases clearly after 180 and 360 min UV-B; the kinetics of the O-J increase become lower after UV-B irradiation. In all cases the level of the I transient decreases and the kinetics of the J-(I)P rise decreases. From curves like those in Fig. 4, rate constants of reactions involved in the photochemical trapping in the reaction center of PS II were calculated using the Three-State Trapping Model of PS II (Vredenberg 2000; Vredenberg et al. 2001). The rate constants for QA− oxidation by QB (kAB1) decreased very fast, both at 150 and 720 μmol m−2 s−1; it is down to about 50% of control already after 15 min UV-B irradiation (Fig. 5A). The rate constant for QA− oxidation by QB− (kAB2) was down to 5% of control, already after 15 min UV-B (not shown). This indicates that UV-B irradiation has a fast action on the reducing side of PS II. The rate constants for Yz+ reduction by the OEC in S1 and S2 (k1 and k2 in Fig. 5B and C, respectively), were also affected, but much slower. From this we conclude that the damage by UV-B starts at the reducing side, and affects the donor side later. In Fig. 5D it is illustrated that kL (excitation rate) increases by UV-B irradiation. This indicates that UV-B radiation decreases the antenna size, or that a lower number of open reaction centers is available after UV-B treatment. Fig. 4Relative variable fluorescence measured after illumination of leaves during various times (T, in min) with UV-B light at 150 (A) or 720 μmol m−2 s−1 (B); PAR is zero. CO is control. Each curve is the average of three (A) or four (B) measurementsFig. 5Rate constants of some reactions involved in the photochemical trapping in the reaction center of PS II, calculated using the Three-State Trapping Model of PS II (Vredenberg 2000). Curves like those presented in Fig. 5 were taken as a source for the calculations; these curves were measured after various times of irradiation with the Vilber lamp at 150 μmol m−2 s−1 (○) or 720 μmol m−2 s−1 (■); PAR was zero. Average of four measurements and the bars indicate ±S.D.; bars not shown are within the symbols As may be concluded from the results presented in Fig. 5, UV-B radiation affects rate constants of reactions at both the acceptor and donor sides of PS II. With respect to the acceptor side, Rodrigues et al. (2006) reported that QA− is a photosensitizer for UV-B radiation; absorption of UV-B by this semiquinone radical initiates reactions leading to damage to PS II. The group of Vass, most often using isolated plant preparations, has reported several times on the effect of UV-B radiation on the donor side of PS II. In a recent paper (Szilard et al. 2006), they describe a model where UV-B-induced inhibition of water oxidation is caused either by direct absorption within the catalytic manganese cluster or by damaging intermediates of the water oxidation process. Vass et al. (1996) measured the effects of UV-B radiation on the OEC, the QAFe2+ complex, the redox-active tyrosines, and the D1 protein in isolated spinach PS II membrane particles. While all these functions were affected, the OEC appeared to be the most sensitive; the authors concluded that the primary damage by UV-B occurs at the OEC, and that modification and/or inactivation of tyrosine and the quinone acceptor complex are subsequent events. Our work confirms that UV-B radiation affects both acceptor and donor sides of PS II. However, our analysis leads to the conclusion that the acceptor side of PS II is affected first and the donor side later. This different conclusion may be caused by the different material and methods of the two works. Vass et al. (1996) used PS II particle membranes, while in our work intact leaves were measured. Furthermore, Vass et al. (1996) measured the effect of UV-B separately on the various PS II functions, while we calculated the effects of UV-B from one single measurement.

Tuberculosis_(Edinb)-1-5-1913930

The role of GlnD in ammonia assimilation in Mycobacterium tuberculosis

Summary Introduction Mycobacterium tuberculosis is responsible for the largest number of human deaths from a single bacterial agent. Nearly two million people die from tuberculosis each year and more than eight million are newly infected.1 A better understanding of the basic metabolism of this pathogen could lead to new strategies for eradication. Although progress has been made in understanding some of the nutritional requirements of this organism both in vitro and in vivo, in particular its carbon source acquisition,2,3 little is known about nitrogen metabolism. M. tuberculosis has four enzymes with glutamine synthetase (GS) activity (GlnA1-4).4,5 Of these, GlnA1, GlnA3 and GlnA4 synthesise l-glutamine, whereas GlnA2 synthesises the d-glutamine and d-isoglutamine required for cell wall biosynthesis.5 The major GS, GlnA1, is expressed to a high level and is exported.6 The role of the secreted enzyme is not clear, but it has been suggested that it may play a role in pH modulation, although the biosynthetic reaction does require a source of ATP. GS catalyses the production of glutamine from glutamate and ammonia. Since this reaction requires ATP, it needs to be strictly regulated in the presence of excess ammonia to conserve both energy and glutamate pools in the cell. GS activity can be controlled by several mechanisms, including feedback inhibition, transcriptional control of gene expression and by post-translational modification. In Escherichia coli a regulatory cascade of three proteins, GlnD, PII and GlnE, is involved in the latter mechanism (Fig. 1). GlnD is a uridylyl transferase which modifies the PII protein. PII in turn controls the activity of GlnE. GlnE is an adenylyl transferase which controls the interconversion of GS and GS-AMP. Transfer of the AMP moiety to the GS enzyme reduces its glutamine synthetic activity. GlnE is also able to deadenylylate GS with the predominant reaction being determined by interaction with the PII protein; PII promotes the adenylylation reaction, whereas PII-UMP promotes the deadenylylation reaction. In this way the cells can rapidly control GS biosynthetic activity in response to ammonia availability. An M. tuberculosis GlnA1 mutant is auxotrophic for glutamine, and attenuated in macrophages and guinea pigs,7 suggesting that the assimilation of ammonia via this pathway is required in vivo. We have previously shown that, in contrast to other bacteria including closely related organisms such as Streptomyces coelicolor, GlnE is an essential gene in M. tuberculosis.8 Thus the control of GS activity seems critical to normal growth. Here we show that glnD is not essential and that although it has an effect on GS activity in the cells, it is not required for virulence. Materials and methods Culture M. tuberculosis H37Rv (ATCC25618) was grown in Middlebrook 7H9 plus 10% v/v OADC supplement (Becton Dickinson) and 0.05% w/v Tween 80, Middlebrook 7H10 agar with 10% v/v OADC supplement or TSM media (1.5 g/L K2HPO4, 0.5 g/L KH2PO4, 0.5 g/L MgSO4, 0.5 mg/L CaCl2, 0.1 mg/L ZnSO4, 0.1 mg/L CuSO4 and 50 mg/L ferric chloride) supplemented with 10% v/v OADC and 0.05% w/v Tween 80.9 Working pH 7.2. For TSM-high ammonia, 30 mM (NH4)2SO4 was added; for TSM-low ammonia, 0.1 mM (NH4)2SO4 was added, l-amino acids (alanine asparagine, glutamine and glutamate) were added to 3 mM. Growth curves were obtained in 12 mm diameter borosilicate tubes with 4–5 ml media and stirring at 250 rpm with an 8 mm flea. Hygromycin was used at 100 μg/ml and kanamycin at 20 μg/ml. Quantitative RT-PCR Probes and primers were designed for quantitative PCR for sigA (endogenous control), amt, glnB and glnD using the software Primer Express (Table 1). cDNA was synthesised from RNA using RT and random hexamer primers using AMV reverse transcriptase. PCR was carried out in a Taqman 7900 using a standard PCR master mix. For sigA, amt, glnB and glnD, the primer pairs were SigA-R and SigA-F, Amt-R and Amt-F, GlnB-R and GlnB-F, and GlnD-F and GlnD-R, respectively, and the probes used were SigA-T, Amt-T, GlnB-T and GlnD-T. The primer and probe concentrations were first optimised. The optimal primer concentration was 300 nM for all four genes, the probe concentration was 100 nM for sigA and glnB, 125 nm for glnD and 200 nm for amt. In order to measure relative gene expression levels, standard curves for each primer-probe set were generated using genomic DNA. CT values were converted into the equivalent of ng using the standard curve. Control reactions without RT were used to confirm that there was no significant contaminating genomic DNA present. CT values for genomic DNA were converted to ng and subtracted from the plus RT values. In order to standardise the samples to ensure that equal amounts of cDNA were used, each value was standardised to sigA to generate unit-less values. At least three independent RNA samples were assayed in triplicate for each gene. Construction of glnD mutant We used our previous method for generating delivery vectors with a marker cassette.10 The delivery vector was constructed by amplifying two regions flanking glnD such that an in-frame deletion was engineered and cloning them into p1NIL. Primer pairs gap7 CACAACGGATACCACAAC and gap8 CGTCAATGCTGTTGCTGC, and gap5 CAAGACCTGGGGAGACGC and gap6 CAGTTTGTCGGTGCCCTC were used to amplify the upstream and downstream regions and the PCR products were cloned into pGEM EasyT (Promega). The two regions were then excised as KpnI-EcoRI and EcoRI–HindIII fragments respectively and cloned into the KpnI-HindIII sites of p1NIL thereby deleting 1.7 kbp of the glnD gene. The marker cassette from pGOAL19 (hyg, lacZ, sacB) was introduced as a PacI fragment to generate the final delivery vector pKOD3. A deletion mutant was constructed according to our previous method.10 Briefly, we treated the vector DNA with UV and electroporated M. tuberculosis to generate a single cross-over strain.11 This strain was streaked out without antibiotics and double cross-overs selected and screened on 2% w/v sucrose and 50 μg/ml X-gal. White colonies were patch tested for hygromycin and kanamycin sensitivity and then screened by PCR for the deletion gene. Potential mutants were confirmed by Southern blotting using XhoI digested genomic DNA and hybridising to a probe derived from the upstream flanking region. GS assays Cell-free extracts were generated using the MiniBeadBeater.12 Culture filtrates were prepared using 0.2 μM filters and the filtrates were concentrated using Centricon 20 units (Amicon). Total GS activity was assayed using the transferase assay.13 GS activity is given in nmol per minute per mg of total protein. Virulence assays THP-1 cells were maintained in culture, treated with PMA to induce differentiation, washed and then infected as described.14 A total of 5×105 macrophages were infected at MOIs of 1:50 and 1:5 bacteria to macrophage. Extra-cellular bacteria were removed by washing several times. Determination of the initial inoculum was assessed by plating serial dilutions and the number of intra-cellular bacteria was monitored over 7 days. Mice were infected with approximately 106 viable mycobacteria in 200 μl of pyrogen-free saline via a lateral tail vein. Where appropriate, infected mice were killed by cervical dislocation in accordance with humane endpoint protocols under the Animals Scientific Procedures Act, 1986 (UK). Results Expression of glnD and nitrogen regulation We are interested in the regulatory cascade that controls the activity of GS by post-translational modification. We previously demonstrated that GlnE, an adenylyl transferase which modifies GS, is essential in M. tuberculosis.8,15 The activity of GlnE is modified by the PII protein (encoded by GlnK or GlnB) and PII is controlled by the GlnD protein (Fig. 1). In order to gain a better understanding of this pathway, we extended our work to look at the other members of the cascade. The glnD gene is arranged in an apparent operon with two other genes, amt and glnB (Fig. 3A). The start and stop codons of amt and glnB overlap, but there is a 60 bp gap between glnB and glnD which could theoretically contain a promoter. We wanted to determine if the genes in this region are controlled by ammonia availability. We used RT-quantitative PCR to look at the expression levels of the three genes relative to sigA (Fig. 2). Amt and glnB were up-regulated three-fold and glnD was up-regulated two-fold in low-ammonia medium as compared to high-ammonia medium. The lesser induction of glnD is not unexpected as it is often seen with genes that are at the 3′ end of the operon. Alternatively, glnD could be independently expressed from a promoter located in the intergenic region. Construction of glnD mutant In order to characterise the role of GlnD in the regulatory cascade which controls GS activity, we constructed a deletion mutant. An in-frame deletion of the gene was made in the vector p1NIL and the gene cassette from pGOAL19 containing the hyg, lacZ, sacB genes was inserted. A two-step homologous recombination process was used to generate the mutant.10 The in-frame deletion and expected genotype was confirmed by Southern blotting (Fig. 3). Out of 32 double cross-overs screened, 12 were mutants. One strain (Tame 69) was selected for further study. We analysed the ability of the glnDΔ strain to grow in various nitrogen sources, since it has been shown that mutations in the control of GS activity can lead to deleterious effects on growth. We measured growth of the deletion mutant Tame 69 with various nitrogen sources. The mutant was able to use all the nitrogen sources and there was no difference in the growth rates from the wild-type strain (data not shown).9 Thus the mutant is not compromised in its ability to utilise any of these substrates. Control of total GS activity In other bacteria, GlnD is involved in the post-translational modification cascade which ultimately controls GS activity by modulating its adenylylation state. It has also been shown to play a role in regulation of nitrogen-controlled genes in Corynebacteria.16 Therefore, we investigated total GS activity of the mutant grown with ammonia or glutamate as nitrogen sources (Fig. 4) to determine if deletion of glnD affected the activity of GS. The majority of GS is found in the culture filtrate in M. tuberculosis6 and our data confirmed that we had higher activity in this fraction than in the cell-free extracts for the wild-type strain. Interestingly, we found that total GS activity was higher in ammonia-rich conditions (30 mM ammonium sulphate) as compared to either low ammonia (0.1 mM) or 7H9 (glutamate) in the wild-type strain. This was in agreement with RT-qPCR data for glnA1 which showed increased expression in high-ammonia medium (Pashley and Parish, unpublished data), although glnA2 expression was unchanged.9 There was no significant difference in the total GS activity between the mutant and wild-type in the cell-free extracts. However, in culture filtrates the situation was very different. We found a large reduction in total GS activity in the mutant strain grown in all media except low ammonia (where the level of activity was at its lowest). This was particularly pronounced in ammonia-rich conditions, where the wild-type activity was at its highest. Virulence in macrophages and SCID mice M. tuberculosis glnA1 mutants are glutamine auxotrophs and showed a reduced ability to multiply in macrophages and guinea pigs.7 Since extra-cellular GS activity was reduced in the glnD mutant, we determined whether GlnD played any role in intra-cellular survival. We used two measures of this: (1) the ability of the mutant to grow in macrophages and (2) its ability to cause disease in mice. For the macrophage infection assay we used the human macrophage-like THP-1 cell line. Infections were carried out at two different MOI (Fig. 5). The results showed that there was no difference between the wild-type and mutant strain, indicating that disruption of GlnD function had no effect on intra-cellular survival. We also tested virulence in the SCID mouse model to detect any strong attenuation profile (Fig. 5). Again, the mutant strain behaved just as the wild-type and was not attenuated. Discussion We have shown that expression of glnB and amt is controlled in response to ammonia levels. The up-regulation of these genes in low-ammonia conditions is in agreement with previous findings in other related bacteria. For example, in Corynebacterium glutamicum, these genes are transcribed as an operon under the control of nitrogen availability; the operon is switched off in nitrogen-rich media and is turned on during nitrogen starvation.17,18 Our previous results indicate that at least one member of the GS regulatory cascade is essential, as we have been unable to construct GlnE mutants in M. tuberculosis.8,15 Our current hypothesis is that GlnE is required to inactivate GlnA1 because the latter is expressed at such a high level. If it were all enzymatically active, the intra-cellular levels of glutamate and/or ATP would quickly be depleted. In agreement with this we have shown that the adenylylation function is essential, but the deadenylylation function is not (Pashley et al., unpublished). Deletion of GlnD would result in a lack of uridylyation of the PII protein. Since unmodified PII promotes the adenylylation activity of GlnE, inactivation of GS should still occur in the mutant and therefore deletion of GlnD is possible. However, our hypothesis predicts that PII is likely to be essential since it directly modulates GlnE activity. This concurs with our preliminary evidence that we are unable to make a glnB deletion mutant, in contrast to the relative ease of constructing a glnD mutant (Pashley and Parish, unpublished data). Although the glnD mutant was viable, there were substantial changes in the level of GS activity. Normally, a large amount of GS is exported from the cell during growth.6 Export of GS has been linked to high levels of expression, rather than any specific export mechanism.19 If glnD deletion leads to reduced expression of glnA1, this would explain why only extra-cellular levels are depleted. The observation that total GS activity in the culture filtrates was markedly reduced in the glnD mutant is an intriguing one. It is possible that overall expression of GlnA1 is markedly reduced and therefore little of it will be exported to the outside. Alternatively, GlnA1 production could be completely abolished. The enzyme assay we used does not distinguish between the four M. tuberculosis GS enzymes, so GS activity measured could arise from GlnA2-4. However, this seems unlikely, since it has already been demonstrated that deletion of the glnA1 gene leads to glutamine auxotrophy, implying that all the biosynthetic activity in the cell is GlnA1.7 It is likely that the GlnD regulatory cascade only controls activity of GlnA1, although we cannot rule out the possibility that GlnA3 activity would also be modulated. However, GlnA2 and GlnA4 will not be controlled by this cascade (since they cannot be modified by GlnE). Thus GS activity arising from enzymes other than GlnA1 cannot be rule out. Further work to determine the levels of expression of each GS would help to further elucidate the mechanism of control of GS activity in the whole cell. However, the fact that GlnA1 is the major GS in M. tuberculosis confirms the importance of this post-transcriptional regulatory control system. Although extra-cellular GS was markedly reduced in the glnD mutant, the strain was still fully virulent in macrophages and SCID mice. This was a surprising observation, since it has already been demonstrated that GlnA1 is required for virulence.7 Our data indicate that GS is not required at high levels for virulence in the SCID mouse and that the presence of the lower level of intra-cellular GS is sufficient for bacterial survival. It is possible that a defect may be seen in the glnD mutant under conditions in which intra-cellular multiplication is seen or in activated, rather than resting macrophages. However, since we have not measured extra-cellular GS directly during infection, we cannot exclude the possibility that extra-cellular GlnA1 is found in the glnD mutant in vivo. Therefore, the attenuation of GlnA1 mutants most likely results from their auxotrophic nature and the lack of availability of glutamine in the host environment. In conclusion, we have shown that glnD is not an essential gene, but that it is required for the normal expression and activity of GS in culture filtrates of M. tuberculosis. Future work to address the question of how deletion of glnD results in altered expression of GS by determining the mechanism of regulation for ammonia-regulated genes is underway.

Ann_Biomed_Eng-2-2-1705544

Proteoglycan Breakdown of Meniscal Explants Following Dynamic Compression Using a Novel Bioreactor

Motivated by our interest in examining meniscal mechanotransduction processes, we report on the validation of a new tissue engineering bioreactor. This paper describes the design and performance capabilities of a tissue engineering bioreactor for cyclic compression of meniscal explants. We showed that the system maintains a tissue culture environment equivalent to that provided by conventional incubators and that its strain output was uniform and reproducible. The system incorporates a linear actuator and load cell aligned together in a frame that is contained within an incubator and allows for large loads and small displacements. A plunger with six Teflon-filled Delrin compression rods is attached to the actuator compressing up to six tissue explants simultaneously and with even pressure. The bioreactor system was used to study proteoglycan (PG) breakdown in porcine meniscal explants following various input loading tests (0–20% strain, 0–0.1 MPa). The greatest PG breakdown was measured following 20% compressive strain. These strain and stress levels have been shown to correspond to partial meniscectomy. Thus, these data suggest that removing 30–60% of meniscal tissue will result in the breakdown of meniscal tissue proteoglycans. INTRODUCTION Mechanical loading of the meniscus plays a crucial role in the metabolic activity of fibrochondrocytes.7,11,12,15 Previous studies have demonstrated that increased load on meniscal tissue leads to an increase in proteoglycan and collagen levels,16 whereas unloading of meniscal tissue causes a decrease in aggrecan expression, collagen formation and cell growth.2,3 It is not fully understood how biomechanical and biochemical events interact to produce changes in the extracellular matrix. This lack in knowledge is, in part, due to the difficulties associated with performing real time meniscal loading experiments in vivo. Recreating the physiological forces in vitro using tissue explants while measuring the biological response provides one method for observing the effect of mechanical stress on the meniscus;9,15 however, the majority of commercially available bioreactors may not be suitable for application to meniscal loading studies. The Biopress system (The Biopress system, Flexcell International, Hillsborough, NC, USA) uses air pressure applied to a flexible bottom under each well. The Biopress has been used to apply pressures of 0.1 MPa in previous studies on meniscal explants,7,15 noting strain levels of approximately 10%. Because pressures up to 10 MPa and strains ranging from 2% to 20% are seen in the meniscus13,18 this system may not generate high enough pressures to mimic the full range of meniscal strains thought to occur during active loading of healthy and damaged tissue in vivo. Stresses and strains are approximately 5–10%, and 3 MPa, respectively, in the normal healthy meniscus, but these strain levels increase with a partial meniscectomy to approximately 20% and 8 MPa.18 Another biaxial tissue-loading device, previously used to compress articular cartilage explants, is able to create a maximum 400 N axial force on as many as 12 explants at once, however, is limited to 100 μm of motion.8 Previous studies have shown that following knee trauma, such as ACL transection, prior to any articular cartilage damage, there are signs of meniscal tissue degeneration.10 We have previously shown that following partial meniscectomy, the remaining meniscal tissue is subjected to an altered loading state.18 This altered loading state may cause a direct breakdown of matrix components such as proteoglycans (PG), or indirectly affect matrix production by induction of catabolic biomolecules. The objectives of this study were (1) to design and build a practical, cost-effective device for applying homogeneous strains to tissue explants, and (2) to utilize the system to overload and underload meniscal tissue and measure the biochemical output. In this communication, we show that the ensuing device is a simple biocompatible design that applies accurate and reproducible strains and is made of components that can be sterilized. We utilized the bioreactor to apply both load and displacement controlled dynamic compression tests. Dynamic strain compression tests showed PG breakdown following overloading of meniscal tissue. No significant changes in concentration of PG released to the conditioned media was measured following various levels of dynamic compressive stress. A comparison between stress controlled dynamic compression versus strain controlled dynamic compression was also made. MATERIALS AND METHODS Design of Bioreactor The frame consists of two 2.54 cm thick parallel aluminum plates separated by 2.54 cm diameter aluminum support rods (Fig. 1). Centered on the bottom plate is a load cell that is attached to a six well dish. The system is driven by a belt-driven linear actuator utilizing a control package by Animatics (Smartmotor 1720, Ultramotion, Mattituck, NY, USA). The actuator has a maximum stroke length of 5 cm and can thrust to 2225 N. It also has a maximum speed of 50 cm/s with bi-directional repeatability of ±0.00762 mm and a unidirectional repeatability of ±0.00254 mm. Motor control was achieved by using the SmartMotor Interface (SMI). This language allows the motion of the actuator to be controlled by the signal generated by the load cell or by the displacement of the actuator. Displacement resolution for the actuator is 0.4 μm. FIGURE 1.A 2-dimensional view of the assembly of the plunger, dish, and cap. The test frame is composed of two aluminum plates supported by aluminum rods. The actuator is positioned in a centered hole in the top plate and tighten into alignment with an adjustable collar. A strain gage load cell (Model 1210AF, Interface, Scottsdale, AZ) with a load capacity of 1334 N (sensitivity of ∼1.3 N) was utilized. For tests that require loads near or above 1334 N, a dimensionally identical load cell with 8896 N (sensitivity ∼2.17 N) (Interface, Scottsdale, AZ, USA) capacity is interchangeable with the current load cell. This design feature of interchangeable load cells allows for a larger range of loads to be accurately measured. A 2100 series signal conditioner (Vishay Instruments, Raleigh, NC, USA) amplifies the load cell signal to produce a 5 V signal at the maximum load. The load cell is centered on a 2.54 cm thick aluminum plate that is the base of the system frame. A threaded stud leaving the load cell connects to the aluminum dish via a quick disconnect pin. The dish has six 10 mm deep wells equally spaced in a circular orientation. Teflon-filled Delrin compression rods (diameter = 8 mm) for each well are press fit into a plunger which attaches to the actuator via a quick disconnect pin. To ensure only one plunger/dish orientation and to keep the compression rods centered in each well, the plunger also features two press-fit aluminum pins that slide into matching holes in the dish. An aluminum cap rests on the shoulder of the dish and houses a linear bearing that is press fit into the cap. Along the resting edge of the cap, four shallow grooves were machined to allow carbon dioxide supply to the explants during testing. The linear bearing allows the plunger to move up and down within the cap and restricts the plunger to vertical motion (Fig. 2). The entire frame is small enough that it can be placed in a commercially available CO2 incubator to maintain physiological conditions (Model 5015, VWR, West Chester, PA, USA). FIGURE 2.A 2-dimensional drawing of the test frame shows the side and top view of the test system. The linear actuator is attached to the plunger using a quick-disconnect pin. The dish is attached to the load cell in the same manner. The cap improves alignment of the plunger by utilizing a linear bearing. Accuracy Evaluation of the System Ultra-low pressure film (Sensor Products Inc., East Hanover, NJ, USA) was used to measure well pressure during compression. First, the repeatability of the pressure film was tested by loading the film (seven times) between flat platens in a tensile testing machine (Model 8872, Instron Corp., Canton, MA, USA) to a 70 N target load, corresponding to a pressure of 0.477 MPa for the given indentor size. Calibration of the pressure film was also done using the tensile testing machine and included loading pieces of pressure film ranging from 0.2 to 1.64 MPa. Pressure film analysis was completed using commercial software (Scion Image, National Institute of Standards and Technology, Gaithersburg, MD, USA) to measure the density of the pressure film samples. Film was compressed between the platens and a piece of rubber similar to the rubber used for testing well pressure. To determine well pressure in the bioreactor, a machined plate was set on top of the dish with a 3 mm thick piece of uniform rubber. Pressure film was placed on top of the rubber and the plunger was lowered near the surface of the film. Four tests were conducted, each to the same pressure (0.477 MPa) to determine the repeatability of the bioreactor in load control. The film from the bioreactor was analyzed and density measured to determine the difference between each compression rod. The difference in film density and the maximum percentage error was determined to demonstrate the accuracy of the system. Determination of Displacement Repeatability To further investigate the accuracy of the system, the gap between the bottom of the compression rods and the bottom of the wells was measured while the system was assembled into the bioreactor. This was done by using the actuator to compress commercially available Fibre-Strand body filler (6371, The Matin Senour Comp., Cleveland, OH, USA) until the gap between the bottom of the compression rod and well bottom was filled. The actuator remained at this position until the body filler hardened completely. After plunger removal, a micrometer (2.54 μm resolution) was then used to measure the thickness of the body filler. This process was repeated three times with the same plunger and dish orientation. Compliance of the System The compliance of the system was determined by placing a flat stainless steel plate over the wells and running a load-deformation test in the absence of menisci. The test was repeated three times and the load-deformation data recorded. Application of the System Pigs used in the experiment were 18 weeks old, sacrificed 24 h after death (received from Mayo Clinic, Rochester, MN, USA). The explants were collected from both the lateral and medial meniscus using sterile technique. The explants were 6 mm in diameter and cut using a biopsy punch, perpendicular to the superior surface to maximize the amount of superior tissue preserved. To ensure two parallel flat faces on the cylindrical explants, a specialized cutting device was utilized. Explants were held such that a fixture containing two razorblades, 5 mm apart, cut the tissue perpendicular to the long axis of the cylindrical explant. This ensured that each explant was the same height with parallel faces (Fig. 3a). FIGURE 3.a – Meniscus following biopsy, b – Maximum axial strain following different partial meniscectomy simulated by FE analysis. Various percentages (5%, 10%, 30% and 60%) removed from different portion of the medial meniscus (A – anterior, C – central, P – posterior). Knee loaded with 1200 N axial force. Explants were incubated at 37°C (5% CO2) for 48 h in growth media (89% DMEM/F-12, 10% FBS, 1% Penn/Strep) which was changed after 24 h. For mechanical testing, the explants were placed in the six well bioreactor filled with 400 μl of the test media (97% DMEM/F-12, 2% FBS, 1% Penn/Strep). The explants were loaded for 2 h at 1 Hz, at one of the following levels: 5%, 10%, 15% or 20% strain, 0.05 or 0.1 MPa unconfined dynamic compression. Each loading group consisted of six explants taken from six different animals (except 15% test – 4 animals). Control explants were placed in wells, but not exposed to compression. Previous experiments (FE analyses) showed that under two times body weight the intact knee meniscus experiences about 10% maximum compressive strain. Removal of 5–10% of the meniscus minimally affects the maximum strain level, however, removal of 30–60% of the meniscal body increases the maximum strain to 15% or greater. Generally speaking, 0% strain is likely underloading the tissue, 5–10% is approximately physiological and 15–20% is considered overloading18 (Fig. 3b). Following compression the explants were bisected into superior and deep zone, by cutting them in half (Fig. 4), weighted and placed in 24-wells plates, in 1 ml of fresh test media. Samples were post-incubated at 37°C (5% CO2) for 24 h. Post-incubation media was collected and stored at −80°C for future analysis.FIGURE 4.Cross-section of the meniscus with the direction of the cut and explants showing superior and deep zones. The content of sulfated glycosaminoglycan (GAG) released to the conditioned media was assayed using dimethylmethylene blue (DMMB) dye solution.5,6 The standard curve was generated with a known concentration of shark cartilage chondroitin sulfate C. All samples were run in duplicates. The concentration was normalized to the wet weight of the explant. GAG release was then normalized to the no load control samples for each animal. Data Analysis The concentration of the GAG released to the media was calculated using a standard curve (R2 > 0.85). The final value was an average from duplicates. All data is presented as mean ± standard error. A one way ANOVA followed by Fisher’s PLSD post-hoc testing was used to measure statistical differences (p < 0.05 was considered significant). Paired t-tests were used at each stress or strain level to determine differences between superficial and deep zone PG release to the media. RESULTS Accuracy Evaluation of the System The repeatability test showed an average of 0.4773 ± 0.0003 MPa. The pressure film demonstrated equal pressure in each well for each load (Fig. 5). There was less than a 1% error (Table 1). FIGURE 5.Pressure film impressions at 0.477 MPa under the six bioreactor compression rods.TABLE 1.Results of pressure film verification.0.477 MPaPressure (MPa)Test #Rod 1Rod 2Rod 3Rod 4Rod 5Rod 6AverageStd. Dev.10.47730.47680.47710.47710.47680.47680.47700.000220.47680.47690.47680.47680.47690.47720.47690.000130.47730.47680.47690.47700.47690.47710.47700.000240.47730.47680.47690.47680.47690.47680.47690.0002 Determination of Displacement Repeatability The micrometer measurements from the body filler showed that the bioreactor was extremely repeatable. Well 3 had the largest standard deviation in height, 3.4 ± 0.01 mm, whereas well 1 had the lowest standard deviation 3.4 ± 0.0015. Compliance of the System The results of the compliance test indicate a linear load-deformation response (R2 > 0.99). The slope of the load-deformation curve was 3465 ± 200 N/mm. Thus, for the range of strains and stresses seen in this study, the compliance is negligible. Application of the System To determine an appropriate post incubation time, preliminary tests were run. PG concentration in the media was monitored following 2, 4, and 6 h and 1, 2, and 3 days following 10% dynamic compression and for controls samples (Fig. 6). The shortest time showing a strong signal was chosen to minimize the duration of the experiment (1 day). FIGURE 6.PG concentration following different incubation time (n = 2 for controls (C), n = 3 for 10% compression test (10%)). Data presented is mean ± standard error. The highest break down of PG was measured following 20% compressive strain for both the superior and deep zones of the meniscal explants (Fig. 7). There were significant differences between release of GAGs into the media for the overloaded condition (20%) versus physiological loading (10%) for both superior and deep zones. No significant differences were found between superficial and deep zones for any compression level. Explants exposed to pressures of either 0.05 or 0.1 MPa did not show any significance differences in either the superior or deep zone for PG breakdown (Fig. 8).FIGURE 7.PG concentration following different compression tests (n = 6 except 15% test (n = 4)). Data presented is mean ± standard error. *Statistically different than 20% (p < 0.05).FIGURE 8.PG concentration following different load level compression tests (n = 6). Data presented is mean ± standard error. Displacement controlled tests showed a rapid drop in load within the first 1000 cycles with little change in the following 6200 cycles (Fig. 9). For displacement tests at or below 15% strain, loads settled just below 0.05 MPa while the 20% strain test remained above 0.1 MPa throughout the duration of the test. The change in pressure measured from the beginning of the test to the end is shown in Table 2. Load controlled tests (Fig. 10) showed a rapid increase in compressive displacement within the first 2000 cycles. The 0.05 MPa load level reached maximum displacement near 3000 cycles and remained at that level to the end of the test. The 0.1 MPa tests reached 18% strain near 3000 cycles but steadily increased to 20.7% strain by the final cycle. The differences in strain from the start to the end of the tests can be found in Table 3.FIGURE 9.Stress vs. time for four representative displacement control tests. Only peak values during each cycle are plotted. Stress was calculated by dividing the peak load by the initial cross-sectional area.Table 2.Change in pressures over the duration of the stress-relaxation tests.StrainPressure (MPa)StartEnd5%0.166 ± 0.108 *#0.038 ± 0.010 #10%1.141 ± 0.103 #0.046 ± 0.010 #15%2.185 ± 0.8270.035 ± 0.026 #20%3.548 ± 0.4290.128 ± 0.020Data represents mean ± standard deviation. n = 6 for all groups. * Significantly different than 15% (p < 0.05), # significantly different than 20% (p < 0.05).FIGURE 10.Strain vs. time for two representative load control tests. Only peak strain values during each loading cycle are plotted. Strain was calculated by dividing the peak displacement by the original height of the explant.Table 3.Change in strains over the duration of the creep tests.Pressure (MPa)Strain (%)StartEnd0.052.6 ± 0.5311.6 ± 1.360.13.0 ± 0.1220.7 ± 1.45*Data represents mean ± standard deviation. n = 6 for all groups. * Significantly different than 0.05 MPa (p < 0.05). DISCUSSION The explant compression system meets the criteria necessary to obtain a realistic representation of physiological forces present in the knee joint. This system is able to apply known pressures to six explants at once, which is important when trying to gather data for hypothesis testing. It is capable of applying physiological and supraphysiological levels of load and displacement, and has the ability to test in load or displacement control. SMI programming allows for flexibility in frequency, duration, amplitude, and waveform. The system is small enough to fit in a standard incubator and is made of materials that can endure autoclaving and alcohol. An important feature to this system is the ability to keep the explants and media sterile from the culture hood to the incubator. The plunger, dish, and cap form an enclosure that allows easy transport between the culture hood and incubator without allowing open air and bacteria to infect the sample. Since the cap incorporates a linear bearing it does not need to be removed for testing. Utilizing the system features and designing the correct protocol will help maintain a physiological loading sterile environment. This bioreactor is capable of creating higher loads and greater displacements than previous systems used for compressing explants.7,8,15 The Biopress system (Flexcell International, Hillsborough, NC, USA) is capable of loads as high as 69 N, whereas the current systems actuator can thrust to 2225 N. An advantage of our system over the biaxial tissue-loading device presented in Frank et al. (2000), is that it can create displacements over 10 mm with a resolution of 0.4 μm.8 Our actuator also has a bi-directional repeatability of ±7.62 μm compared to the ±25 μm used in Sah et al. (2003).11 In addition, the present system is capable of 1 Hz cyclic compression in a sinusoidal wave using displacement or load control. The bioreactor can be used to compress any tissue that fits under an 8 mm compression rod and in a 10 mm deep well. All surfaces are machined to a smooth, frictionless finish, to ensure the sample is exposed to pure unconfined compression. Although we believe this bioreactor to be an improvement over others, we realize that there are some limitations that need to be compensated for. For example, this system can only perform unconfined compression currently. However, the system could be modified to run confined compression tests as well by outfitting a new plunger/well assembly. Since the stress tests were conducted at 0.05 and 0.1 MPa, and resulted in strains between 2.6% and 20%, the compliance of the system was negligible. Similarly, for the 5% strain control tests, the loads were small enough that the compliance of the system was negligible. However, for the 10%, 15% and 20% strain control tests, stresses were between 3.548 and 0.035 MPa. Thus, at the larger stress levels, the compliance of the system was greater. For instance, during the 10% strain control test, initially stress levels of 1.1 MPa would lead to approximately 57 μm of compliance, and thus, for the first 150 of the 7200 s test, strains were closer to 8.8% instead of 10%. For the 15% strain test, strains were closer to 13% for the first 200 s of the test, and as the material relaxed, the last 1 h and 57 min of the test were at 15% strain. We hypothesized that PG breakdown (as determined by GAG in the media) would be high for both the underloading and overloading condition. PG breakdown was significantly increased at 20% strain. Meniscal tissue is a mechanically sensitive, and mechanical loading has been shown to regulate gene expression.14 Hence, we expected that loading the tissue below normal physiological levels of magnitude (underloading) would result in PG breakdown. Lack of increased PG breakdown for control samples (underloading) is surprising and needs to be further explored. Perhaps the degraded PG is not being released into the media for the control samples, whereas the overloaded samples have mechanical compression to help move the broken down PG into the media. Future studies will measure PG breakdown in the tissue explants following compression to determine the integrity of the PG within the tissue. These data suggest that removing 30–60% of meniscal tissue, and thereby increasing tissue strains over 15% results in an increase in PG breakdown and tissue destruction. Thus, not only does meniscectomy affect the underlying articular cartilage but the remaining meniscal tissue appears to begin to breakdown, possibly leading to a change in meniscal material properties. This data is supported by previous work that showed following an ACL transection, degenerative changes were seen in the meniscus prior to any articular cartilage changes.10 DiMicco et al. showed an increase in GAG release from bovine cartilage that had been exposed to an injurious level single uniaxial, unconfined compression.1 Inhibitors of biosynthesis or degradative enzymes did not affect PG breakdown, suggesting that the breakdown was a mechanical consequence of compression. GAG release 1–7 days post injury was slowed by metalloprotease inhibitors. Shin et al.12 showed that dynamic compression (0.1 MPa for 24 h at 0.5 Hz) increased both GAG synthesis as well as release to the media compared to unloaded controls. Our loading scheme was designed to simulate approximate physiological walking conditions (2 h, 1 Hz). It is difficult to compare our GAG data to a study of a single injurious insult, or dynamic compression for 24 h. We created an approximate daily physiological loading environment that might occur during walking, and simulated loads from unloaded to overloaded due to meniscectomy. In the future, we will investigate the mechanisms of GAG release, whether it is mechanical damage or activation of enzymatic activity. It was surprising that explants tested at 0.05 MPa showed greater PG breakdown than explants tested at 0.1 MPa for the superficial zone. One possible reason for this result might be related to the cell viability. A compression of 0.1 MPa may induce more cell death than 0.05 MPa of compression and hence fewer cells may be available for production of metalloproteases that may contribute to the breakdown of PG. Current studies are underway to document the degree of cell death in the explants. It is interesting to compare the load control results to the displacement control results. Based on Fig. 9, the 20% strain test applied an initially very high load (∼3.5 MPa), but equilibrated at 0.1 MPa. Similarly, the 0.1 MPa test reached a steady state of 20% strain (Fig. 10). Hence, we would expect to see similar levels of PG breakdown in the 20% strain test as well as the 0.1 MPa test. This is also true for the 10% strain test and the 0.05 MPa stress tests. For example, in the deep zone the 20% strain control test resulted in 3.7 ± 1.4 ug/ml per gram of tissue, whereas the 0.1 MPa test resulted in 1.9 ± 0.5 ug/ml per gram of tissue. This large difference could be due to either “lift-off” (separation between loading platen and sample during a 1 Hz test)4 or the initially high stress that was reached initially in the 20% displacement control tests. In contrast to the differences seen above, the 10% strain test and 0.05 MPa stress test resulted in 1.3 ± 0.2 and 1.8 ± 0.5 ug/ml per gram of tissue, respectively. Previous 3-D computational studies of an entire human knee joint have shown that mean contact pressures changed from 1.57 MPa for an intact healthy meniscus to 3.09 MPa following 60% meniscectomy.18 Maximum contact pressures on the superior surface of the meniscus changed from 4.7 to 7 MPa when 60% of the meniscus was removed.18 The differences between stress and strain control studies noted above make the data presented difficult to interpret due to the non-physiological nature of in vitro unconfined compression studies. Clearly, this indicates that meniscal tissue may respond differentially to stress versus strain, the duration of strain levels, or most likely the loading history. Before definitive clinical implications can be made regarding the effects of meniscectomy on meniscal tissue, we must first determine if the levels of PG breakdown noted in this study correspond to changes in the load-bearing capacity of the tissue and its function in the knee joint. One of the limitations of this study is lack of investigation of differences between the medial and lateral menisci as well as specific location within the meniscus from which explants were harvested. This likely resulted in higher standard deviations. Future studies with a larger set of animals are proposed in which lateral versus medial, anterior versus posterior and inner radial versus outer radial differences could be studied. Previous researchers have shown a difference between inner and outer radial explants.14 Bisection of explants has previously been shown to release many growth factors.17 Since all samples in this study were bisected, the relative differences found are still significant. It should be noted that the measured response may be due not only to mechanical stimuli but also growth factors released due to cutting the samples. In addition to bisecting the samples, other factors, such as FBS concentration and post-incubation time may effect the absolute concentrations of GAG analyzed in this study. Therefore, only relative comparisons between treatment groups should be considered. In summary, this simple and practical experimental system allows for reproducible application and quantification of homogeneous stresses or strains to explants tissues, thereby providing a systemic and quantitative method for correlating external mechanical stimuli to cellular and molecular mechanisms of mechanotransduction.

Graefes_Arch_Clin_Exp_Ophthalmol-4-1-2292470

Do dissociated or associated phoria predict the comfortable prism?

Background Dissociated and associated phoria are measures of latent strabismus under artificial viewing conditions. We examined to what extent dissociated and associated phoria predict the “comfortable prism”, i.e. the prism that appears most comfortable under natural viewing conditions. Introduction Since Ogle’s work [14] dissociated phoria has been distinguished from associated phoria. Dissociated phoria is defined as a deviation from the orthovergence position that occurs when no fusionable contours are provided. Associated phoria is a deviation of the eyes that appears under prism correction of fixation disparity: associated phoria equals the “aligning prism” [1] that nullifies fixation disparity. To avoid the technical requirements for measuring the eye position objectively, fixation disparity is commonly determined according to the observer’s directional perception of monocular Nonius lines, embedded in binocular contours. Both dissociated [11, 18] and associated phoria [9, 19] have been recommended as indicators for prismatic correction in the case of asthenopia, but recently associated phoria has been preferred, because the stimulus for associated phoria appears to be more natural, in that both eyes are presented with a nearly identical configuration [1, 4, 5, 16, 20]. Moreover, Yekta et al. [20] found the correlation of asthenopic complaints to be significant only with associated, not with dissociated phoria. Kromeier et al. [8], however, suggested that dissociated and associated phoria should be similar, provided the accommodative demand in the two procedures is equal. Their argument was based on the consideration that the feedback loop for fusion is opened under both conditions. Concerning dissociated phoria, it is obvious that the feedback loop cannot operate; fusionable contours are absent, hence there is no error signal. Concerning associated phoria, the argument is more complicated because of the fundamental difference between fixation disparity and associated phoria. When the observer looks at a test for fixation disparity, any tendency to deviate from orthoposition is largely kept in check, because the disparity of the binocular contours provides an error signal: the fusional feedback loop is functioning. In the test for associated phoria, however, prisms are added. Prompt and repeated adjustment of the prismatic power nullifies any disparity of fusionable contours. This continuous and artificial resetting of the error signal to zero means that the fusional feedback loop cannot fulfill its purpose, namely stabilisation of a certain vergence angle. The experimental results of Kromeier et al. supported their theoretical consideration. These authors suggested that the discrepancies between dissociated and associated phoria encountered in previous studies [15, 20] might be due to different accommodative stimuli. To clarify this issue, we compared dissociated and associated phoria using the same object for fixation, thus keeping the accommodative demand similar. We further investigated to what extent dissociated and/or associated phoria predict the “comfortable prism”. This term defines the prismatic power reached by the observers when they looked at a fully fusionable display, and tuned a pair of counterrotating prisms, called Herschel’s or Risley’s rotary prisms [17], such that the viewing of a fully fusionable object appeared most comfortable. Hence, the comfortable prism anticipated the real-life situation for which prismatic spectacles might be considered. Methods Apparatus Stimuli were generated by a PowerMacintosh G4 and presented at a distance of 4.0 m on a 19” Philips GD403 CRT monitor with a resolution of 800 × 600 pixels. By means of liquid crystal shutter goggles (ELSA 3D Revelator), separate images were presented to the two eyes. The goggles were synchronized to the CRT refresh such that the frames were alternately presented to the right and left eyes. The refresh rate was 120 Hz, i.e. 60 Hz for each eye, just above the flicker fusion frequency. The phosphor persistence of the monitor, measured with a photoelectric cell, was down to 10% after 4.0 ms, i.e. within a shorter time than the frame time of 8 ms. An alternate cover test assured us that this technique provided a complete separation between the images of the two eyes. Herschel’s counterrotating prisms, manufactured by Zeiss, were mounted in front of their right eye (Fig. 1). Fig. 1Herschel’s prisms mounted in front of the right eye. The observer, looking through shutter goggles, adjusts the prisms by means of a lever arm. A cogwheel belt transmits the prismatic power to a potentiometer These prisms, equipped with a lever arm, allowed a variation of the prismatic power between 30 cm/m base in and 30 cm/m base out. A calibration performed with a laser beam showed that the readings on the scale of the prisms varied only about 2% around the true inflection. A cogwheel belt connected the prisms with a potentiometer that provided a signal linearly proportional to the prism power. The electric signal was directly transmitted to a computer. During the trials the room was dark; in the intervals the room was moderately illuminated so that the test person was able to see a coloured print (Henri Matisse “Fleurs et céramique”, 35 × 40 cm, mean luminance 13.8 cd/m2) mounted directly above the monitor. Test figures Dissociated phoria As depicted in Fig. 2a, the left eye was presented with a fixation cross, flanked by two O letters (“O X O”). Each of the three letters had a diameter of 10 arcmin and a line thickness of 1.5 arcmin. Above and below the cross were two Nonius lines, 20 arcmin high and 5 arcmin wide. The interval between the lines and the centre of the “O X O” was 10 arcmin. A bright circular background of 40 arcmin radius, embedded in a random dot pattern, surrounded these dark features. The right eye was presented with a bright dot of 4.3 arcmin diameter on a homogenous dark field. The luminance of the dark features was 0.2 cd/m2, and that of the bright features 22.6 cd/m2 (measured through the liquid crystal shutter goggles). The luminance outside the monitor was 0.1 cd/m2. Fig. 2Test figures. a Test for dissociated phoria. All the elements contained in the Mallett figure, including the two Nonius lines, are displayed to the left eye. The right eye is presented with a white dot on a homogenous dark field. b Test for associated phoria, according to Mallett [9]. c Test for the comfortable prism. An identical figure is displayed to both eyes Associated phoria As depicted in Fig. 2b, the configuration was similar to Mallett’s test for fixation disparity [9]. The “O X O” served as a fusional stimulus. Vertical Nonius lines were used as monocular markers, in that the line above the fixation cross was shown to the right eye, and the line below the fixation cross to the left eye. Comfortable prism Both eyes were presented with all the Mallett elements including both Nonius lines (Fig. 2c). Procedure and instructions The observers were seated in a comfortable chair. During the trials, the observers had to rest their chin on a support, to lean their forehead against a bar, and to adjust Herschel’s prisms with the lever arm. Before each trial, the experimenter set the prisms to ±0. To limit prism adaptation [12], the time for the adjustment was restricted to 30 seconds. In the interval between the single trials (about 60 to 120 s), the observers were encouraged to move their head freely and to look around in the room or to behold the Matisse painting. Each of the three conditions (dissociated phoria, associated phoria, comfortable prism) was tested in a block of eight trials. The conditions were presented in the following order: Associated phoria. If the observers saw an offset between the Nonius lines, they had to align them with the least possible excursion of the lever arm so that they appeared as stable to each other as possible. This instruction resembles the suggestion of Karania and Evens [6] to observe whether one or both of the Nonius lines ever move. If an observer did not see an offset at the start of the trial, or the Nonius lines oscillated around zero, he or she was encouraged to “play” a little with the lever arm and set Herschel’s prisms such that the Nonius lines became aligned as stably as possible. The observers were repeatedly reminded to look at the centre of the “O X O”, not at the Nonius lines. Further, the observers were instructed to ignore if one Nonius line, or even if both of them seemed to disappear every now and then (fading). Most observers had such an experience. They had to adjust Herschel’s prisms according to the percept when they saw both Nonius lines.Dissociated phoria. The observers were asked to align the white dot with the midline of the “O X O”, neglecting any vertical offset of the white dot. While doing so, they were urged to strictly look at the centre of the “O X O”, not at the white dot.Comfortable prism. The observers were asked to look at the centre of the “O X O” and to set the prisms such that viewing appeared most relaxing. They were encouraged to start with rather bold searching excursions and to refine the tuning gradually during the available 30 seconds. To get used to the manoeuvring of Herschel’s prisms, the experiment was preceded by a few trials in which the observers practised bringing the white dot to the midline of the “O X O” (test for dissociated phoria) and aligning the Nonius lines to each other (test for associated phoria). To examine the reproducibility, we repeated the whole experiment for all observers in a second session after an interval of 24 to 130 days. Observers Twenty observers participated in the study (members of our department or recruits via a public advertisement, aged between 20 and 71 years, median 26.5 years). The observers were selected according to the following 4 criteria: (1) Visual acuity of each eye (with full spherical and cylindrical correction) at least 1.0decimal (= 6/6Snellen), (2) absence of strabismus, ascertained with the unilateral cover test, (3) absence of a prism in the spectacles, and (4) presence of random dot stereopsis (Lang Test 1). The observers were refracted without dilating their pupils, using streak retinoscopy and crossed cylinders, taking particular care to avoid any uncorrected hyperopia. During the experiment, the observers wore full spherical and cylindrical spectacle corrections. Care was taken to ensure that the glasses remained centred throughout the study. We did not inquire whether the observers had asthenopic symptoms or not. We explained to the observers that the study intended to optimise the comfort of seeing. Otherwise, the observers were kept naive as to the purpose of the study. Each observer provided informed written consent to participate in the experiments. The study followed the tenets of the Declaration of Helsinki and was approved by the institutional human review board. Data acquisition and analysis The prismatic power derived from the potentiometer attached to Herschel’s prisms was recorded by PowerLab with a sampling rate of 100 Hz. To enable analysis of the dynamic behaviour of the adjustments we recorded the full 30 seconds of each trial. Offline examination was accomplished with Igor Pro® (Wavemetrics, Inc., Lake Oswego, OR, USA) and Statview® (Abacus Concepts, Inc., Berkeley, California, USA). Statistical analysis was performed with Statview® and SPSS® (SPSS Inc., Chicago, Illinois, USA). We used multifactorial ANOVA (including condition, session and trial) and paired comparisons. For post-hoc tests, Bonferroni adjustment was performed. Results Inspection of curves The curves obtained from the 20 observers varied considerably between the 8 trials. For example see Fig. 3. There was no trend, e.g. with the early trials (thin lines) yielding smaller values than the late ones (progressively thicker lines), or vice versa (p = 0.34). Most observers adjusted Herschel’s prisms stepwise, reaching the final value after about 20 seconds. Therefore, we limited the numerical evaluation to the remaining 10 seconds. Fig. 3Adjustment of Herschel’s prisms by observer #10 for dissociated phoria, associated phoria, and the comfortable prism. First session. Positive values indicate base out prisms (eso deviation), negative values base in prisms (exo deviation). The upper panels show the original recordings of the eight single trials. The lower panels show the mean values (solid lines) with the standard error of the mean (SEM, dashed lines). Note that the scale of the ordinate for the comfortable prism differs from those in the two phoria conditions For dissociated phoria, most observers saw an offset right at the beginning of each trial and promptly moved the lever arm accordingly. For associated phoria, some observers initially did not see an offset between the Nonius lines. In this case, the observers followed the instruction to “play” a little with the lever arm and find a position in which the lines appeared as stable as possible. For the comfortable prism, most observers started with rather bold searching excursions and refined the tuning gradually (Fig. 3, right panels). Only exceptionally, an observer ended up with a rather large range for the comfortable prism. Statistical analysis We averaged the 1000 values recorded during the last 10 seconds in each of the eight trials, as most observers reached the final value after about 20 seconds. As stated above, there was no trend between the eight trials, e.g. with the early trials yielding smaller values than the late ones, or vice versa (p = 0.34). This allowed us to treat the 8 trials together and calculate an intertrial mean±SEM (standard error of the mean) for each observer (Fig. 4). Among the 20 observers, the range of the intertrial mean was for dissociated phoria from +9.3 eso to −5.9 cm/m exo deviation, for associated phoria from +11.2 eso to −3.3 cm/m exo deviation, and for the comfortable prism from +4.8 eso to −4.1 cm/m exo deviation. Fig. 4Mean of the last 10 seconds of eight trials ±SEM for each of the 20 observers. The values obtained in the first session are represented in the left columns, those obtained in the second session in the right columns. DP = dissociated phoria, AP = associated phoria, CP = comfortable prism. The observers are identified with #1 to #20. Ordinate: positive values = eso deviation, negative values = exo deviation. Note that the ordinates in #7 and #19 are scaled down To assess the intertrial variability over all 20 observers, we averaged the 95% confidence interval (±1.96 SD) per session. The values were as follows. Dissociated phoria: first session ±1.3 cm/m, second session ±1.0 cm/m; associated phoria: first session ±1.6 cm/m, second session ±1.6 cm/m; comfortable prism: first session ±2.0 cm/m, second session ±2.2 cm/m. Hence, the variability was similar in the two sessions. Combining the two sessions resulted in a greater overall scatter (95% confidence interval = ±1.96 SD): ±5.8 cm/m for dissociated phoria, ±5.4 cm/m for associated phoria, and ±4.3 cm/m for the comfortable prism. The marked increase in the overall scatter indicates that the values changed between the two sessions. In some cases the mean values of the second session were even outside the 95% confidence interval of the first session: 5/20 for dissociated phoria, 9/20 for associated phoria, and 6/20 for the comfortable prism. The change was not always concordant in the three conditions. For example, in observer #13, the associated phoria changed from +0.1 ± SEM 0.4 cm/m to +4.6 ± SEM 0.5 cm/m, i.e. in the eso direction, and the comfortable prism changed from −0.1 ± SEM 0.4 cm/m to −1.8 ± SEM 0.2 cm/m, i.e. in the exo direction. Are these changes clinically relevant? As several practitioners consider prescribing prisms from a minimum of about 1.0 cm/m onwards [13], we identified the observers in whom the change was greater than 1.0 cm/m: for dissociated phoria 6/20 (up to 3.2 cm/m), for associated phoria 11/20 (up to 4.5 cm/m), and for the comfortable prism 12/20 (up to 3.3 cm/m). To find out whether the observers who changed their values by more than 1.0 cm/m could be recognized at the first session, we compared the intertrial variability at the first session in two groups: the observers who changed their values more than 1.0 cm/m with those who changed them less than 1.0 cm/m. There was no statistically significant difference between the two groups (p = 0.67 for dissociated phoria, 0.11 for associated phoria, 0.52 for the comfortable prism). Comparison between dissociated phoria, associated phoria, and the comfortable prism The ANOVA over all trials of all 20 observers (n = 8 × 2 × 20 = 320) revealed that dissociated and associated phoria differed significantly (p < 0.01). In 11 observers, the dissociated phoria was lower, in four higher than the associated phoria; in five observers, dissociated and associated phoria were similar. Figure 5 demonstrates that the majority of observers had an eso deviation for dissociated and associated phoria (13/20 and 17/20, respectively), but an exo deviation for the comfortable prism (17/20). The tendency towards eso deviation in the two phoria conditions was also obvious in the average ±SD over all trials of all 20 observers (n = 8 × 2 × 20 = 320): +1.0 ± 2.9 cm/m eso deviation for dissociated phoria, and +2.0 ± 2.7 cm/m eso deviation for associated phoria, versus −0.6 ± 2.2 cm/m exo deviation for the comfortable prism. The difference between dissociated phoria and comfortable prism, on the one hand, and the difference between associated phoria and comfortable prism on the other were both significant (p < 0.05 and p < 0.001 respectively). Fig. 5Mean of the two sessions ±SEM for associated phoria, dissociated phoria, and comfortable prism. The first column in each graph represents observer #1, followed by #2 etc To investigate whether the tendency towards eso deviation in the two phoria conditions, as compared with the comfortable prism, was brought about by the sequence of tests, we made a spot check. We chose observer #10, because she had a marked eso deviation in the two phoria conditions, but an exo deviation in the comfortable prism. In an extra session, this observer determined the comfortable prism before the tests for dissociated and associated phoria. The values obtained were of the same magnitude as those in the two previous sessions (ANOVA p = 0.084 for tested difference). Discussion We compared three vergence parameters: dissociated phoria, associated phoria, and the comfortable prism. To avoid any (possibly prejudiced) influence of the experimenter, we recorded the prismatic power set by the observer with a potentiometer. The comfortable prism anticipates the real-life situation in which prismatic spectacles are worn. Therefore, the comfortable prism is very close to the endpoint of therapeutical considerations, i.e. prescribing prismatic spectacles. In contrast, dissociated and associated phoria, determined under artificial viewing conditions, are surrogate measures of the comfortable prism. We studied whether these surrogate measures predict the comfortable prism, assuming that dissociated phoria, associated phoria, and the comfortable prism might be similar, because all three parameters are determined by opening the feedback loop for fusional vergence. When measuring dissociated phoria, fusionable contours are absent; hence, there is no disparity error signal. When associated phoria is measured, any disparity of the (defective) fusional pattern, which initially constitutes an error signal, is promptly nullified by repeated prism adjustment. Concerning the comfortable prism, one has to realize that the fusional feedback loop, which uses disparity as its error signal, can stabilize the vergence angle only if the stimuli are presented to the two eyes in a fixed angle, e.g. through a certain unchanging prism. If, however, the observer is asked to adjust a variable prism according to his or her comfort, the fusional feedback loop is open and cannot stabilize the vergence angle. Instead, the fusional feedback loop is replaced by another feedback loop whose error signal is discomfort. Accordingly, the vergence position reached with the comfortable prism may also be described as the “vergence position of rest when both eyes are exposed to identical pictures”. To our surprise, we found marked differences between the three vergence parameters: as indicated in Fig. 5, the majority of observers had an eso deviation under the two phoria conditions (13/20 for dissociated and 17/20 for associated phoria), but preferred base-in for the comfortable prism, corresponding to an exo deviation (17/20). The maximal difference occurred in observer #10 who showed in his second session an eso deviation for associated phoria of +4.9 cm/m, and an exo deviation for the comfortable prism of −3.0 cm/m. The tendency towards eso deviation in the two phoria conditions was also evident in the significant difference between the averages ±SD over all 20 observers: +1.0 ± 2.9 cm/m eso deviation for dissociated phoria, +2.0 ± 2.7 cm/m eso deviation for associated phoria, but −0.6 ± 2.2 cm/m exo deviation for the comfortable prism. Although the test condition for associated phoria is commonly regarded as being more natural than that for dissociated phoria, the associated phoria was not closer to the comfortable prism than the dissociated phoria. This means that the difference in the overall luminance for the two eyes, present in our test for dissociated phoria, did not push the vergence away from the value obtained with the comfortable prism. This finding supports the finding of Kromeier et al. [7, 8] that unequal luminance of the images for the two eyes does not necessarily influence the vergence position. What is the reason for the tendency towards eso deviation in the two phoria conditions? A different accommodative demand can be excluded, because we used a similar target in all three conditions. Rather, we suggest that the shift towards eso deviation is brought about by the dissimilarity between the images of the two eyes. This suggestion pertains even to the associated phoria condition, in which ample fusionable contours are available, and the dissimilarity is limited to the monocular Nonius lines. Nevertheless, this dissimilarity was conspicuous for the observers: most of them reported that one or both Nonius lines seemed to disappear every now and then. Correspondingly, the observers reported occasional disappearance of the white dot when they determined their dissociated phoria, although this was less pronounced than the disappearance of the Nonius lines in the associated phoria condition. These observations raise the possibility that the irritating perception of binocular rivalry drives the eso shift in the two phoria conditions. Compatible with this idea is that the eso shift was stronger in the associated than in the dissociated phoria condition, i.e. in the condition with the more pronounced fading. Considering the open feedback loop for fusional vergence in all three conditions, it is not surprising that the values obtained with all three methods (dissociated phoria, associated phoria, and comfortable prism) were rather variable. This was true in each of the two sessions. The values changed in many observers from the first to the second session beyond the 95% confidence interval of the eight trials in the first session. Comparing the three conditions, they did not always change in the same direction (eso or exo). We do not see a specific reason for these changes, as there was no hint that the general health or attentiveness had changed between the two sessions. Rather, the variability might be understandable as there is no need for the ocular motor system to keep the vergence angle stable under open-loop conditions. Surely, adaptation maintains the open-loop vergence in a certain range [10, 12], but this range appears to be rather wide. What are the practical inferences of our study? As mentioned in the introduction, both dissociated [11, 18] and associated phoria [9, 19] have been recommended as indicators for prismatic correction in the case of asthenopia. We purposely did not select our observers according to the presence or absence of asthenopic complaints. Rather, we avoided using the relief from asthenopic complaints as a criterion, because this approach would have required prolonged wearing of various test prisms, including placebo, in a controlled design, to face the following four problems. First, a causal relationship between heterophoria and asthenopia is hard to establish, because the prevalence of both conditions is high in the general population, so that an unrelated coincidence can easily occur. Second, wearing prisms for hours or days leads to adaptation, so that the patient may later value even prisms that he or she initially disliked. Third, most asthenopic symptoms are waxing and waning. It is highly likely that patients seek remedy at a time when their symptoms are relatively intense. Hence, a subsequent improvement may well be due to a regression to the mean, rather than to a prism [21]. Fourth, a relief from symptoms may be brought about by the supportive behaviour of the therapist. Although we did not use the relief from asthenopic complaints as a criterion, we think that our findings relate to the prescription of therapeutic prisms for such patients. The discrepancy between dissociated and associated phoria on the one hand and the comfortable prism on the other hand would probably also occur in patients with asthenopia, and it is plausible to assume that these patients would benefit more from the comfortable prism, chosen by themselves under natural viewing conditions, than from a prescription based on one of the phoria parameters, which carry artefacts due to the artificial test conditions. Another practical aspect of our study is the finding that both phoria parameters and the comfortable prism can change from one session to the next. Are these changes clinically relevant? Several practitioners consider prescribing prisms from 1.0 cm/m onwards [13]. On this background, our finding that 12 of the 20 observers changed their comfortable prism by more than 1.0 cm/m (up to 3.3 cm/m) within an interval of a few weeks is important. We examined whether patients tending to a long-term change of their open-loop vergence might be identifiable in the first session by a large scatter. Unfortunately, our results did not support this notion: the intertrial variability at the first session was not larger in the observers who changed their values more than 1.0 cm/m, as compared with observers who changed their values less than 1.0 cm/m. Hence, repeated determinations of the comfortable prism on different days appear advisable to identify patients in whom the values remain relatively stable. Only in these patients would a prescription of prisms be reasonable. What are the limitations of our study? Towards the end of our experiments, we considered that it might have been preferable to randomize the sequence of tests between the 20 observers. However, it is unlikely that such a randomization would have produced significantly different results, for the following two reasons: (1) an influence of training or fatigue is unlikely since there was no trend between the eight trials of each test, e.g. with the early trials yielding smaller values than the late ones, or vice versa (p = 0.34), and (2) a spot check in observer #10 with the reverse order of tests confirmed this observer’s marked discrepancy between an eso deviation in the two phoria conditions, and an exo deviation in the comfortable prism (ANOVA p = 0.084 for tested difference). The angles encountered in our 20 observers were relatively small: the dissociated phoria ranged from +9.3 cm/m eso to −5.9 cm/m exo deviation, the associated phoria from +11.2 cm/m eso to −3.3 cm/m exo deviation, and the comfortable prism from +4.8 cm/m eso to −4.1 cm/m exo deviation. An extrapolation of our conclusions to observers with larger angles may not be justified. Our tests corresponded to just one of many “natural” viewing conditions: with respect to luminance, visual angle, and distance (4 meters), the conditions were similar to watching television. According to previous work [2, 3], it is likely that, at closer viewing distances, the values would be shifted in the exo direction. Hence, we suggest that the practitioner should determine the comfortable prism in the viewing distance, for which therapeutic prisms are being considered. This recommendation corresponds with the concept of Jaschinski [4], who suggested that observers should choose their comfortable viewing distance for computer screens in the real-life situation.

Histochem_Cell_Biol-4-1-2413113

Connexons and cell adhesion: a romantic phase

Recent evidence indicates, that gap junction forming proteins do not only contribute to intercellular communication (Kanno and Saffitz in Cardiovasc Pathol 10:169–177, 2001; Saez et al. in Physiol Rev 83:1359–1400, 2003), ion homeostasis and volume control (Goldberg et al. in J Biol Chem 277:36725–36730, 2002; Saez et al. in Physiol Rev 83:1359–1400, 2003). They also serve biological functions in a mechanical sense, supporting adherent connections between neighbouring cells of epithelial and non-epithelial tissues (Clair et al. in Exp Cell Res 314:1250–1265, 2008; Shaw et al. in Cell 128:547–560, 2007), where they stabilize migratory pathways in the developing central nervous system (Elias et al. in Nature 448:901–907, 2007; Malatesta et al. in Development 127:5253–5263, 2000; Noctor et al. in Nature 409:714–720, 2001; Rakic in Brain Res 33:471–476, 1971; J Comp Neurol 145:61–83 1972; Science 241:170–176, 1988), or mediate polarized movements and directionality of neural crest cells during organogenesis (Kirby and Waldo in Circ Res 77:211–215, 1995; Xu et al. in Development 133:3629–3639, 2006). Since, most data describing adhesive properties of gap junctions delt with connexin 43 (Cx43) (Beardslee et al. in Circ Res 83:629–635, 1998), we will focus our brief review on this isoform. Introduction For those readers who are not familiar with the molecular composition of gap junctions, we will briefly recollect the substructure of this cell contact and refer to recent detailed reviews (Duffy et al. 2002; Gaietta et al. 2002; Goodenough and Paul 2003; Meier and Dermietzel 2006; Willecke et al. 2002). Gap junctions are formed by hemichannels (connexons), which consist of an oligomer of six proteins (connexins). At present, at least 20 genes seem to be present in the human and rodent genome (Willecke et al. 2002), which may oligomerize in monomeric or heteromeric patterns to form a hemichannel. A complete gap junction channel is formed by two hemichannels in mirror symmetry (Fig. 1). Heterotypic configurations of different isoforms are allowed for some connexins while others occur exclusively in homotypic configuration. In the history of gap junctions, the junctional plaque has always been considered to occur in a “naked” form without cytoplasmic adjuncts like scaffolding proteins or cytoskeletal elements (Hirokawa and Heuser 1982). However, accruing evidence indicates that gap junctions are associated with a complex system of scaffolding and cytoskeletal proteins, which seem to assemble in cell specific patterns (Duffy et al. 2002 for recent review).Fig. 1General structure of a gap junction plaque. Gap junctions are formed by paired hemichannels (connexons) of two adjacent cells. A single connexon is made by a polymer of six connexins. Only apposed connexons allow intercellular transfer of ions (ionic coupling) and small metabolites (metabolic coupling). Unapposed connexons seem to perform per se functions In the following, we will put main emphasis on heart and brain tissues for which most of the data on gap junctions and cell adhesion have been reviewed. Cardiomyocytes: directed trafficking of connexin43 involves the cytoskeleton and adhesion plaques In the terminal intercalated discs between cardiomyocytes, Cx43 celebrates an example for integrating its hemichannel and cell coupling functions (Gros and Jongsma 1996; Shaw and Rudy 1997; van Veen et al. 2001). Asides the terminal intercalated discs, gap junctions are also localised in the lateral sarcolemma of the heart (Fig. 2), and thus described to form site-to-site and end-to-end connections (Yao et al. 2003). Cardiomyocytes also represent unopposed connexons with hemichannel function in the lateral sarcolemma (Saez et al. 2003; Schulz and Heusch 2006; Yao et al. 2003).Fig. 2a shows cultured cardiomyocytes immunolabelled with an anti-Cx43 antibody (red). Immunolabelling is prevalent in apposed cell membranes, but also in some unapposed domains. Nuclei are counterstained with Hoechst dye. b Immunolabelling of heart tissue with Cx43 antibody. Intercalated discs (red) are intensivley stained. Bar indicates 25 μm To ensure its’ trafficking and functional integration into a gap junction plaque or in form of unpaired connexons (hemichannels) into the plasma membrane, Cx43 has to interact with other proteins. In the intercalated discs of coupled cardiomyocytes for instance, gap junction plaques are embedded into adherens junctions, which are primarily formed by cadherins (Matsuda et al. 2006; Niessen 2007; Zuppinger et al. 2000). Whilst the adherens junction supports the mechanical coupling (Gutstein et al. 2003; Niessen 2007), the gap junction ensures the propagation of action potentials along the cardiomyocytes (Gros and Jongsma 1996; Shaw and Rudy 1997). Multiple models describe the pathway from connexon assembly to the initial gap junction formation and interaction with cadherins in the adherens junction. The most common examples are based on half-life time determined trafficking and junctional protein (cadherins) mediated activation of Cx43. The half-life time of Cx43 is restricted between 1 and 3 h, and implicates a dynamic process of assembly, insertion and replacement of connexons, and pairing of connexons to form gap junctions (Beardslee et al. 1998; Hofer and Dermietzel 1998; Laird et al. 1991). Cx43 synthesis was shown to be located on membrane bound ribosomes, where connexin proteins are rapidly oligomerized into homo or heteromeric connexons (Evans et al. 1999; Martin and Evans 2004). Final packing into hemichannel loaded vesicles occurs in the trans-Golgi network as shown by Musil and Goudenough (1993) followed by directed transport along microtubules to multiple insertion sites in the membrane (Akhmanova and Hoogenraad 2005; Jordan et al. 1999; Lauf et al. 2002; Mimori-Kiyosue et al. 2005; Shaw et al. 2007). Once arrived at the gap junction borders at the membrane, connexons are assumed to be inserted via flipping events into the membrane and to diffuse into the centre of the plaque, whilst elderly connexin proteins are shifted to the plaque periphery for subsequent disposal (Gaietta et al. 2002; Laird 2005; Segretain and Falk 2004). The finding that mislocated Cx43 gap junction plaques in the ischemic myocardium are associated with similarly misplaced adherens junctions (Matsushita et al. 1999), underlines an interdependence between Cx43 and the adherens junction related cadherins (Angst et al. 1997; Li et al. 2005; Luo and Radice 2003; Matsushita et al. 1999). E-cadherin transfections into gap junction incompetent cells, allowed the transfectants to build out functional gap junctions (Matsushita et al. 1999). Furthermore, N-cadherin knockout mice (Luo and Radice 2003) and conditional knockdown of N-cadherin in the heart caused mislocalisation and compromized expression of gap junctions. Conditional knockdown of N-cadherin in the heart was additionally shown to lead to arrhythmogenic death (Li et al. 2005), which may involve aberrant regulation of gap junction function (for reviews see: Duffy et al. 2007) Recently, Shaw et al. (2007) described microtubule-mediated target-delivered transport of Cx43 via microtubule plus-end-tracking proteins (+TIPs) and interaction partners such as p150(GLUED) (Berrueta et al. 1999), a component of the dynein/dynactin complex, which in turn is potent to tether microtubules at the adherens junctions (Chausovsky et al. 2000; Ligon et al. 2001). Studies implicating fluorescence recovery after photobleaching (FRAP) on Cx43-YFP transfected HeLa cells that do not endogenously express Cx43, revealed a rapid Cx43 delivery to gap junction plaques. Deconvolution clarified that microtubules extend directly to the gap junction plaques at the cell’s border and total internal reflection fluorescence (TIRF) microscopy and time lapse imaging revealed the appearance of a preferential and prolonged association of microtubule plus ends with the plaques. Most strikingly, Shaw et al. (2007) were able to show that gap junction plaque formation was disrupted by siRNA knockdown of the dimeric +TIP EB1. EB1 associates directly with the plus ends of microtubules and provides, in turn, dual binding sites for adherens junction related proteins like p150(GLUED) and β-catenin. Furthermore, in this setting gap junction plaques could also be disrupted via Nocodazol and Taxol treatment, peptides, which compromise the homophilic cadherin-cadherin interaction in adherens junctions. This setting according to the recent paper by Shaw et al. (2007) is sketched in the cartoon (Fig. 3). While Nocodazol interrupts formation of microtubules by depolimerization, Taxol lets microtubules remain stable, but interferes with their EB1 interaction partner (Nakata and Hirokawa 2003).Fig. 3Model for microtubulus mediated delivery of vesicle-bound connexons to adherens junctions (adapted from Shaw et al. 2007). Microtubules bind via their +end to EB1. EB1 in turn binds to P150 (GLUED), a component of the dynein/dynactin complex, which interacts with β-catenin through P120-catenin with the adherens junction. This interaction is understood to tether the microtubule to the junction and to serve as a gateway for connexon delivery In this context, actin is discussed to act as an initial sensor of cell-cell interaction, driving the localisation of adherens junctions with assistance from Rho-GTPases (Noren et al. 2001, 2003). Spread of Shigella flexneri requires connexin43 hemichannels How close cytoskeletal (re-)organization and intracellular connexin distribution are related is demonstrated for polarized intestinal cells during Shigella flexneri invasion (Clair et al. 2008). For invading the colonic mucosa (Labrec et al. 1964), the gram negative enteric bacillus requires RhoGTPases, Src and Abl/Arg tyrosine kinases for actin polymerization and formation of cytoplasmic extensions of surrounding cells (Burton et al. 2003; Tran Van Nhieu et al. 2000). The invasion and dissemination of the bacteria causes intense inflammatory responses, and especially ATP-dependent paracrine signalling induced by Cx hemichannel opening (Tran Van Nhieu et al. 2000). E-cadherins were indispensable for the intercellular spreading of S. flexneri (Sansonetti et al. 1994). It is assumed that a cytoskeletal reorganization toward the formation zone of gap junctions is induced in this process to allow the spread of the bacteria (Clair et al. 2008; Tran Van Nhieu et al. 2000). On the epithelial level there is mounting evidence that connexin hemichannels regulate intercellular signalling (Stout et al. 2004), which might be of importance for incoming phagocytic cells during bacterial infection (Ferrari et al. 1997; Griffiths et al. 1995; John et al. 2001; Korcok et al. 2004). The developing brain and hemichannel adhesion The necessity of gap junction adhesion via regulation by its cytoskeleton interaction partners p120 catenin, integrin and actin has become well identified in the developing brain (Xu et al. 2001, 2006). Here, stem cells of the developing neocortex give rise to neurons (Malatesta et al. 2000; Noctor et al. 2001) and provide guidance of the developing neurons to the target zones of the cortical plate, where they are meant to become pyramidal cells of the adult cortex (Rakic 1971, 1972, 1988). Electron microscopy showed, that during the process of migration of neuronal precursors, gap junctions occur between radial fibres and migrating neurons and nestin− and nestin+ cells (Huang et al. 1998a). The most important gap junction protein isoforms are Cx43 and Cx26 in developing brain tissue (Dermietzel et al. 1989). Until now there was evidence that gap junctions between radial glia and migrating neurons served for chemical and electrical communication. Elias et al. (2007) recently found that Cx26 and Cx43 are expressed in β-III tubulin positive migrating neurons in the contacting regions close to vimentin positive radial glial fibres. Using a RNA knockdown of Cx43 and Cx26 by short hairpin RNA (shRNA) constructs in rat, the authors were able to demonstrate a reduced fractioning of neurons in the cortical plate. In addition, transplantation of Cx26 and Cx43 shRNA knocked down donor cells into E17 wildtype mice revealed an intact engrafting of the donor cells into the host brain, but no migration. Immunocytochemistry of the shRNA knocked down transplanted neurons in the recipient brains showed no compromised cell cycle exit and no alterations of differentiation. Furthermore, the expression of the adherens related proteins ZO-1, N-cadherin and β-integrin was not altered, indicating that gap junctions mediate glial-guided radial migration of developing neurons in the cortex. This migration was additionally demonstrated to rely on the adhesive and not on the channel properties of Cx26 or Cx43 (Elias et al. 2007). Dominant negative connexin mutants lacking channel properties were still able to form adhesive contacts. In reverse experiments, the authors demonstrated that channel, but no adhesion forming mutants, were unable to rescue the Cx43 shRNA induced migration defect. Finally, time lapse imaging of Cx43/Cx26 shRNA expressing neurons affirmed their inability to stabilize their processes and to continue to extend along the radial glia. Neural crest cells and colonization Gap junction regulated polarized cell movements and directional migration are not restricted to developmental processes within the central nervous system (Elias et al. 2007; Schaar and McConnell 2005). Studies focusing on Cx43 expression of cardiac neural crest cells indicated a clear relationship between their migratory properties and Cx43 expression (Huang et al. 1998a, b; Li et al. 2002; Lo et al. 1999; Reaume et al. 1995; Sullivan and Lo 1995; Xu et al. 2001, 2006). Neural crest cells are ectomesenchymal cells emerging from epithelial mesenchymal cell transformation in the dorsal neural tube from where they disperse throughout the embryo to generate a variety of tissues (Kirby and Waldo 1995; Xu et al. 2006). Neural crest cells from different axial levels of the neural tube use multiple migratory pathways to reach their terminal destinations. Cardiac neural crest cells (CNCs) have been shown to migrate along a circumpharyngeal pathway to reach the aortic arches and the heart (Kirby et al. 1983; Lumsden et al. 1991). This deployment has been shown to be modulated by Cx43 and cytoskeletal interaction partners with the extracellular matrix (Xu et al. 2006). The finding, that dynamic di- and reassembly of focal contacts is essential for polarized cell movements and directional cell migration moved the heterodimeric receptor group of integrins into the centre of related studies. Integrins cluster to form focal domains within the cell membrane, linking the extracellular matrix to the actin cytoskeleton. Since, it could be shown that neural crest cells express multiple integrins (Delannet et al. 1994; Monier-Gavelle and Duband 1997) and perturbation studies provide evidence that integrins modulate the migratory behaviour of neural crest cells, (Strachan and Condic 2003, 2004, 2008) the question arose whether integrin signalling might be affected in Cx43 expressing versus Cx43 knock out cells (Xu et al. 2006). In neural tube explants of the post-otic hindbrain folds from E8.5 mice, underlying either an Cx43 knockout or Cx43 overexpression, neural crest cells were generated that emerge from the same axial level as CNCs, which migrate to the heart. In contrast to overexpressing CNCs, the Cx43 deficient CNC type was characterized by a severe loss of directionality and reduced adhesion whilst being cultured on a fibronectin matrix. Furthermore, an increase in the fibronectin matrix density leads to reductions in the migratory speed of Cx43 deficient CNCs, as shown via time lapse videomicroscopy (Xu et al. 2001). In fact double-immunostaining against β1-integrin and vinculin as markers for focal adhesion, was significantly reduced in Cx43 deficient CNCs, indicating a reduction in the actin-cytoskeletal linkage for matrix adhesion. A modulatory influence of Cx43 on the actin cytoskeleton became evident in rhodamine-phalloidine stainings, where Cx43 knockout CNCs represented shorter stress fibre bundles (Xu et al. 2001). Additionally, these bundles exhibit no anchoring via vinculin to focal adhesions, as being observed for Cx43 overexpressing CNCs. Furthermore, adhesion and migration of Cx43 deficient CNCs on a fibronectin matrix could be inhibited by semaphorin application, which is described to act as a potent blocker of integrin activation (Brown et al. 2001). This approach also confirmed that Cx43 modulates the retraction of cellular processes. Immunoprecipitation, Western blot and immunocytochemistry pointed out that Cx43 in CNCs does not co-localise with β1-integrin, but with vinculin and actin-filaments (Osborne et al. 2005; Pasterkamp and Kolodkin 2003; Serini et al. 2003; Xu et al. 2006). This finding is supported by the co-localisation of Cx43 with several actin binding proteins, such as ezrin, IQGAP, α-actinin and drebrin (Butkevich et al. 2004). No correlation between gap junctional coupling properties and the density of fibronectin matrix was found for Cx43 knockouts and Cx43 overexpressing CNCs in dye coupling experiments, although an upregulation of gap junction communication with altered integrin-matrix interactions has previously been described for other cell types (Czyz et al. 2005; Lampe et al. 1998; Shanker et al. 2005). In summary, asides their function as gap junction forming elements, unpaired connexons have been shown to modulate the cells’ migratory and adhesive functions whilst being in permanent crosstalk with an elaborate complex of cytoskeletal interaction partners. Perspectives on pathology How important the connexin-cytoskeleton interaction is, becomes elucidated in case of pathology. For instance, cadherin-cadherin interactions might be critically affected during tumour formation. The interaction normally becomes active in cell sorting mechanisms during development (Wheelock and Johnson 2003). Since, Shaw et al. (2007) have shown that Cx43 can reach adherens junctions via microtubule directed delivery, it may be suggested that gap junctions are formed preferentially with cells, expressing the same type of cadherin (Wheelock and Johnson 2003). Additional studies provide evidence, that loss of E-cadherin or upregulation of N-cadherin can increase tumour invasiveness and Cx43 downregulation in malignant cells (Mesnil 2002). It is thought that gap junction channels and their interactions with molecules such as p120 catenin, integrin and the actin cytoskeleton are important for neural crest cell migration (Xu et al. 2001, 2006), and that glioblastoma invasion of the brain parenchyma requires functional gap junctions between tumour cells and astrocytes (Lin et al. 2002). Furthermore, the migration of lung and skin cancer cells has also been associated with gap junction expression, although no clear mechanism has been proposed so far (Ito et al. 2000; Lois et al. 2002). Developmental defects are also related to mutations of the Cx43 gene. As already indicated, Cx43 knockout mice reveal comprised conotruncal heart development, which is associated with a reduction in the number of cardiac neural crest cells targeted to the heart (Xu et al. 2006). Mutations in Cx43 which seem to influence cytoskeletal organization in a strong manner range from disease patterns like deafness, cataracts, germ cell developmental defects, ocludentodigital dysplasia to cardial outflow abnormalities (Polontchouk et al. 2002) and left ventricular remodelling (Kanno et al. 2003). At its final extent, gap junction channels, in particular in form of hemichannels, constitute a new player in the complex interaction of cell adhesion and cytoskeletal activation, which underlies directed migration during development and in mature tissue. It is a romantic phase where anything may go, but time has to approve what will remain forever.

J_Neurooncol-3-1-1915654

Hemangioblastomatosis in a patient with von Hippel-Lindau disease

Case presentation A woman developed a cerebellar hemangioblastoma in 1991 at the age of 17 and von Hippel-Lindau disease was diagnosed. She underwent five craniotomies for recurrent hemangioblastoma. In 2001 she had a laminectomy for an intramedullar hemangioblastoma at C2-C3 level. In 2004 a renal cell carcinoma was found and a partial nephrectomy was performed. A few weeks after her last craniotomy in July 2005 she developed progressive right hemisensory disturbances and sensory ataxia. Spinal MRI showed leptomeningeal contrast enhancement around the entire spinal cord (see Figure). The sensory ataxia progressed and she got tetraparesis, most severe proximally in her arms, especially myotome C4. Although hemangioblastoma are not very radiosensitive, radiotherapy was started because of the rapid clinical deterioration. High-dose total neuraxis radiotherapy was considered to toxic. It was judged that most of her clinical complaints could be attributed to the cervical spine. Also the MRI showed the most severe lesions at the cervical region. Therefore she was treated from C1 to Th1 (30 × 1.8 Gy) combined with dexamethasone (2 × 8 mg). During radiotherapy she developed also symptoms outside the radiotherapy field (thoracal radiculopathy). As the main problem in VHL is uncontrolled angiogenesis by overexpressing VEGF and VEGF receptors, Thalidomide (400 mg, later 600 mg) was added. Thalidomide inhibits angiogenesis induced by VEGF [1]. Despite the treatment she developed a total tetraplegia and died 3 months after diagnosis as a result of respiratory failure. This syndrome appears to result from delayed growth of subarachnoid tumor cells disseminated by surgery. Hemangioblastomatosis is also seen in hemangioblastoma patients without von Hippel-Lindau disease and it is thought that additional genes are likely to be the source of this malignant behaviour [2].

Breast_Cancer_Res_Treat-3-1-2001216

The discovery and mechanism of action of letrozole

Because estrogen contributes to the promotion and progression of breast cancer, a greater understanding of the role of estrogen in breast cancer has led to therapeutic strategies targeting estrogen synthesis, the estrogen receptor, and intracellular signaling pathways. The enzyme aromatase catalyses the final step in estrogen biosynthesis and was identified as an attractive target for selective inhibition. Modern third-generation aromatase inhibitors (AIs) effectively block the production of estrogen without exerting effects on other steroidogenic pathways. The discovery of letrozole (Femara®) achieved the goal of discovering a highly potent and totally selective AI. Letrozole has greater potency than other AIs, including anastrozole, exemestane, formestane, and aminoglutethimide. Moreover, letrozole produces near complete inhibition of aromatase in peripheral tissues and is associated with greater suppression of estrogen than is achieved with other AIs. The potent anti-tumor effects of letrozole were demonstrated in several animal models. Studies with MCF-7Ca xenografts successfully predicted that letrozole would be clinically superior to the previous gold standard tamoxifen and also indicated that it may be more effective than other AIs. An extensive program of randomized clinical trials has demonstrated the clinical benefits of letrozole across the spectrum of hormone-responsive breast cancer in postmenopausal women. Introduction Studies have consistently shown that lifetime exposure to estrogens increases the risk of breast cancer [1]. The degree of risk is increased by persistently elevated blood concentrations of estrogen [2]; clinical indicators of persistently elevated blood estrogen concentrations, for example, age at menarche, first live birth, menopause, alcohol consumption, and obesity [3–5]; and, although still controversial, exposure to exogenous estrogen, for example, some forms of hormone replacement therapy and oral contraceptives [6–12]. The presence of some of these factors also increases the risk of breast cancer being estrogen receptor (ER)-positive [13]. Studies have shown that higher levels of endogenous estrogen and testosterone (which is converted to estrogen by aromatase) increases breast cancer risk, regardless of predicted breast cancer risk [14–16]. These data indicate that estrogen is an important risk factor even in women considered at high risk of developing the disease, for example, those with a family history of breast cancer. Estrogen is thought to contribute to the initiation and contributes to the promotion and progression of breast cancer via two complementary mechanisms [1], the carcinogenic effects of estrogen metabolites, notably hydroxyl metabolites [3, 17, 18], and stimulation of ER signaling pathways, including those initiated by activation of epidermal growth factors, notably the mitogen-activated phosphoinositide 3 kinase pathway [19–30]. Greater understanding of the role of estrogen in breast cancer has led to therapeutic strategies targeting estrogen synthesis (aromatase inhibitors [AIs]) [31], the ER (selective ER modulators [SERMs], pure antagonists) [32], and intracellular signaling pathways (signal transduction inhibitors) [33]. Hormone receptor (HR)-positive tumors are defined as those with ER or progesterone receptor (PgR) expression detectable above a pre-set limit [34]. Patients whose ER or PgR expression is below this pre-set limit are considered HR–. Approximately two thirds of breast cancer patients have HR+ tumors [13] and are candidates for treatment strategies designed to counteract the growth effects of estrogen. This review describes the rational development of the potent AI letrozole, which has therapeutic utility in HR+ tumors across the breast cancer continuum. Mechanism of action of aromatase inhibitors Aromatase Aromatase (cytochrome P-450 [CYP] 19) catalyzes the rate-limiting step (conversion of steroidal C-19 androgens to C-18 estrogens) in estrogen biosynthesis [35–37]. Aromatization is the final step in steroid biosynthesis (Fig. 1) [38]; and, therefore, aromatase is an attractive target for selective inhibition [39, 40]. Aromatase is expressed primarily in the ovary and also in central and peripheral tissues, fat, muscle, liver, and breast [41, 42]. With increasing age, as ovarian estrogen production declines [43], the contribution of peripheral production of estrogens increases [44], and in postmenopausal women, peripheral aromatization of androstenedione produced by the adrenal gland (Fig. 1) [38] becomes the main source of endogenous estrogens [45–49]. Of note, normal and malignant breast tissue contributes to the peripheral synthesis of estrogens [14, 50–53]. Thus, expression of aromatase in breast tumors may contribute significantly to the degree of cellular exposure to estrogens [14]; therefore, it is important to target both intra-tumoral and peripheral aromatase [31]. Fig. 1Aromatization of androgens to estrogens in postmenopausal women. A androstenedione, E1 estrone, E1S estrone sulfate, E2 estradiol, T testosterone. Reprinted from [38] with permission from the Society of Endocrinology The presence of intracellular aromatase activity could explain why estrogen concentrations are 10–20 times higher in peripheral tissue than blood in postmenopausal but not pre-menopausal women [41, 54–58]. Moreover, estrogen concentrations are higher in tumors than in surrounding non-malignant tissue [41, 54–58]. Recent research has increased understanding of how aromatase is regulated by tissue-specific promoters [59] and how genetic variation may affect the pathophysiology of estrogen-dependent disease [60]. Pharmacogenomics may become an increasingly important tool for individualizing hormonal therapy for patients with breast cancer. Aromatase inhibitors Modern third-generation AIs effectively block the production of estrogen without exerting effects on other steroidogenic pathways and have been heralded as a “triumph of translational oncology” [61]. The search for potent and selective inhibitors of aromatase started with the first-generation inhibitor aminoglutethimide [62]. However, aminoglutethimide lacked selectivity for aromatase [63] and inhibited biosynthesis of cortisol, aldosterone, and thyroid hormone [64] as well as aromatase; moreover, aminoglutethimide was also found to induce hepatic enzymes (Fig. 2) [65, 66]. Second-generation AIs included the nonsteroidal inhibitor fadrozole and the steroidal inhibitor formestane (4-hydroxyandrostenedione). Fadrozole was superior to aminoglutethimide in terms of potency, selectivity, and safety [67], but its selectivity was not complete and clinical trials suggested that it was no more effective than tamoxifen [68, 69]. Fig. 2The development of aromatase inhibitors (AIs) has culminated in agents with high specificity and potency for aromatase. Spectrum of action of first- through third-generation AIs: The third-generation AIs act exclusively on the aromatase enzyme and do not appear to exert additional effects. Potency of AIs determined by degree of inhibition of total body aromatase: 4-OHA 4-hydroxyandrostenedione. Reprinted from [66] with permission from the Society of Endocrinology To improve on fadrozole, Novartis synthesized a series of new compounds. Structure-activity relationship studies were then performed to identify the most potent AI from a series of benzyl-azole derivatives of fadrozole [70]. The third-generation AI letrozole (Femara®) was the result of this structure-activity approach to drug design and achieved the research goal of creating a highly potent and totally selective AI [71]. These compounds were also used to design pioneering molecular modeling techniques used to map the active site of aromatase [70, 72]. Other third-generation AIs developed during this period were the nonsteroidal agents vorozole (since discontinued) and anastrozole [73] (Fig. 2) [66] and the steroidal agent exemestane [74]. AIs have been classified as steroidal (type I; for example, exemestane) or nonsteroidal (type II; for example, letrozole and anastrozole) [75]. A comprehensive review of AIs focuses on the pharmacology and clinical development of letrozole [76]. Letrozole pharmacodynamics and pharmacokinetics Potency The chemical structure of letrozole (4,4′-[(1H-1,2,4-triazol-1-yl) methylene] bis-benzonitrile) is compared with other AIs in Fig. 3 [77]. The nitrogen-containing structures like the imidazoles and the triazoles bind to the iron in the heme moiety of CYP-450, whereas the cyanobenzyl moiety present in the nonsteroidal AIs such as letrozole partially mimics the steroid backbone of the enzyme’s natural substrate androstenedione. Furthermore, the triazole compound letrozole was found to be superior to other derivatives of fadrozole in terms of in vivo inhibition of aromatase [70]. Fig. 3Comparison of the molecular structures of aromatase inhibitors. Reprinted from [77] with permission from Elsevier Letrozole is a highly potent inhibitor of aromatase in vitro, in vivo in animals, and in humans. The relative potencies of letrozole, anastrozole, and fadrozole were determined in a variety of model cellular endocrine and tumor systems containing aromatase (hamster ovarian tissue fragments, adipose tissue fibroblasts from normal human breast, the MCF-7Ca human breast cancer cell line transfected with the human aromatase gene, and the JEG-3 human choriocarcinoma cell line) [31]. These studies showed that although letrozole and anastrozole are approximately equipotent in a cell-free aromatase system (human placental microsomes), letrozole is 10–30 times more potent than anastrozole in inhibiting intracellular aromatase in intact rodent cells, normal human adipose fibroblasts, and human cancer cell lines (Fig. 4) [31]. In several other studies, letrozole has consistently demonstrated greater potency compared with anastrozole, exemestane, formestane, and aminoglutethimide (Table 1) [31, 71, 75, 78–82]. Fig. 4Relative potencies with which letrozole, anastrozole, and fadrozole inhibit aromatase from non-cellular and intracellular sources. Reprinted from [31] with permission from ElsevierTable 1Inhibitory concentrations of letrozole, anastrozole, exemestane, fadrozole, 4-hydroxyandrostenedione and aminoglutethimide against the aromatase enzyme derived from various cellular and non-cellular sources. Reprinted from [77] with permission from ElsevierAromatase inhibitorIC50 values (nM), (relative potency; letrozole = 1) Human placental microsomesParticulate fractions of human breast cancerRat ovarian microsomesMCF-7Ca cancer cellsJEG-3 cancer cellsCHO cellsHamster ovarian tissueHuman breastLetrozole2 (1)0.8 (1)Anastrozole8 (0.25)15 (0.053)Exemestane15 (0.13)5 (0.16)4-OHA30 (0.07)30 (0.027)AG20,000 (0.0001)10,000 (0.0008)Letrozole11 (1)0.07 (1)0.07 (1)20 (1)0.8 (1)Anastrozole23 (0.48)0.82 (0.085)0.99 (0.071)600 (0.033)14 (0.057)Fadrozole5 (2.2)0.05 (1.4)0.07 (1.0)30 (0.67)1 (0.80)4-OHA62 (0.18)AG1900 (0.0058)Letrozole1.02 (1)0.35 (1.0)0.45 (1)0.14 (1)Anastrozole5.35 (0.19)3.62 (0.097)5.66 (0.080)17.17 (0.0082)4-OHA0.59 (0.59)1.6 (0.28)0.72 (0.19)Letrozole7 (1)Anastrozole25 (0.28)Fadrozole7 (1)Letrozole1.4 (0)Anastrozole27 (0.052)4-OHA60 (0.023)AG5500 (0.00025)4-OHA 4-hydroxyandrostenedione, AG aminoglutethimideValues quoted are IC50 values representing the concentration needed to achieve 50% inhibition of aromatase activity. The relative potency of each inhibitor compared with letrozole is shown in parentheses The degree of aromatase inhibition can be determined in vivo by measuring uterine weight after treatment with a standard dose of androstenedione in immature female rats [71]. Using this assay, it was found that the in vivo potency of letrozole is more than four orders of magnitude greater than aminoglutethimide (50% effective dose [ED50], 1–3 μg/kg vs. 30 mg/kg, respectively) [71]. It has also been shown that neoadjuvant letrozole profoundly inhibits in situ aromatase activity and reduces endogenous estrogens within the breast in postmenopausal women with large primary breast cancers [75]. In postmenopausal women, letrozole achieves significantly greater plasma estrogen suppression of estrogens and greater inhibition of in vivo aromatization than anastrozole [83]. In the study, levels of aromatase were detectable in 11 of 12 patients during treatment with anastrozole (mean percentage inhibition in the whole group, 97.3%) but in none of the 12 patients during treatment with letrozole (>99.1% suppression in all patients; Wilcoxon, P = 0.0022, comparing the two drug regimens). Suppression of estrone and estrone sulfate was found to be significantly greater during treatment with letrozole compared with anastrozole (P = 0.019 and 0.0037, respectively). Another study conducted in 54 postmenopausal women with invasive breast cancer showed that more complete inhibition of aromatase was achieved with 2.5 mg of letrozole than 1 mg of anastrozole, resulting in significantly greater suppression of estradiol (P < 0.0001), the most bioactive estrogen [84]. This recent study confirms previous observations showing that letrozole produces near complete inhibition of aromatase in peripheral tissues, associated with greater suppression of estrogen than achieved with other AIs [78, 85–90]. Selectivity Letrozole is highly selective for aromatase and unlike first- and second-generation AIs does not significantly affect cortisol, aldosterone, or thyroxine [77]. In vitro studies showed that letrozole was more than three orders of magnitude more selective than aminoglutethimide in its effects on progesterone and corticosterone production, and more than 300-fold more selective against aldosterone than fadrozole [71, 78]. In vivo adrenocorticotrophic hormone (ACTH) stimulation tests in rats showed that letrozole had no significant effect on either aldosterone or corticosterone levels, even at a dose 1,000 times greater than that required for inhibition of aromatase [71]. The selectivity of letrozole has been demonstrated in clinical studies in postmenopausal women. These studies showed that letrozole has no effect on the plasma levels of 17α-OH progesterone, thyroid-stimulating hormone (TSH), luteinizing hormone (LH), follicle-stimulating hormone (FSH), or androstenedione and does not affect normal urine electrolyte excretion or thyroid function [86, 91–93]. Of note, the vast majority of patients treated with letrozole have a normal response to synthetic ACTH [86]. Anti-tumor activity in vivo The potent anti-tumor effects of letrozole have been demonstrated in several animal models [77, 78, 94]. Letrozole induced complete regression of estrogen-dependent, 9,10-dimethylbenz-a-anthracene (DMBA)-induced mammary tumors in adult female rats [95]. The ED50 for letrozole was determined to be 10–30 μg/kg/day. The use of MCF-7 cells transfected with human aromatase gene (MCF-7Ca) and implanted into athymic nude mice has proved to be an effective in vivo model for predicting clinical results with AIs [61, 96, 97]. Using this model, it has been shown that letrozole produces dose-dependent inhibition of tumor growth, resulting in complete inhibition at a daily dose of 10 μg/animal/day [94, 98]. Comparative studies using the MCF-7Ca model have shown that letrozole is more effective at suppressing tumor growth than the pure anti-estrogen fulvestrant and the SERM tamoxifen [99]. While anastrozole was also better than fulvestrant and tamoxifen in suppressing tumor growth, only letrozole was shown to induce tumor regression [99]. Another study, also using the MCF-7Ca model, demonstrated that letrozole potently inhibits mammary tumor growth but does not have the estrogenic effects of tamoxifen, as measured by its uterotrophic effects [100]. The observation that tamoxifen has an agonist effect even when estrogen synthesis is inhibited by letrozole suggests that there may be a degree of antagonism between these compounds [100]. Interestingly, studies in the MCF-7Ca model showed that letrozole is more effective as monotherapy than when combined with tamoxifen [80, 101]. In the study reported by Long et al. [101] tumor volume doubling times were 3–4 weeks in controls, 16 weeks with tamoxifen alone, 18 weeks with tamoxifen plus letrozole, and 34 weeks with letrozole alone. First-line treatment with letrozole was shown to be significantly superior to treatment with tamoxifen alone or with the two drugs combined (at week 16, both P < 0.001). Tumors that progressed during treatment with tamoxifen remained sensitive to second-line letrozole therapy, whereas tumors that progressed on letrozole did not respond to second-line treatment with tamoxifen or fulvestrant. In another series of experiments conducted by the same group using the MCF-7Ca model, letrozole was even effective as third-line therapy for a limited period when administered after treatment with tamoxifen and exemestane [102]. The studies showed that although exemestane was more effective than tamoxifen in controlling tumor growth, letrozole as first-line therapy was the most effective treatment overall, both in terms of the degree of tumor suppression and the length of effectiveness of treatment [102]. The potential of letrozole as a chemopreventive agent was investigated in an in vivo model using aromatase-transgenic female mice [103]. The model provided evidence to show that aromatase overexpression is sufficient to induce and maintain early preneoplastic and neoplastic changes that can be completely abrogated by treatment with letrozole. Carcinogenicity studies have also found that letrozole decreases the incidence of spontaneous mammary tumors and granular cell tumors in rats [104]. Pharmacokinetics of letrozole Clinical pharmacokinetic studies of letrozole have been conducted in healthy volunteers [105–107] and in patients with breast cancer [108, 109]. Following oral administration, letrozole is rapidly and completely absorbed (mean absolute bioavailability of 99.9%) and extensively distributed to tissues. It has a large apparent volume of distribution at steady state (1.87 l/kg [range, 1.47–3.24]), and approximately 60% is bound to plasma proteins, mainly to albumin (55%). The terminal half-life (T1/2) of letrozole is 42 h. The terminal T1/2 was observed to be longer and area under the curve (AUC) greater in patients with breast cancer than in healthy volunteers, possibly due to reduction in metabolic clearance [109]. The major route of elimination of letrozole is metabolism by CYP-450 isoenzymes (CYP 3A4 and CYP 2A6) into an inactive carbinol metabolite. Systemic exposure to metabolites is, therefore, low. Steady-state concentrations of letrozole are reached after 2–6 weeks and maintained for long periods with no evidence of drug accumulation. In marked contrast to the first-generation AI aminoglutethimide, no significant drug interactions have been reported for letrozole; however, when combined with tamoxifen, letrozole plasma concentrations are reduced by between 35% and 40% [110]. Age does not have an effect on the pharmacokinetics of letrozole. Exposure to letrozole, measured by AUC, is increased in renally impaired subjects but remains in the range seen in subjects without impaired function. However, hepatic impairment can markedly increase the T1/2 of letrozole, and caution is required in such patients. Differences in pharmacokinetics, including uptake rates, elimination T1/2, and metabolism and clearance exist between AIs and have been reviewed by Lønning et al. [111]. The clinical significance of such differences is not known. Clinical development of letrozole Letrozole entered clinical trials on the basis of its high potency and selectivity for aromatase, the demonstration of unsurpassed anti-tumor effects in models of human breast cancer, and the development of a convenient oral formulation. Daily doses of 0.1–5 mg have been shown to suppress estradiol, estrone, and estrone sulfate plasma concentrations by 75–95% from baseline, while doses >0.5 mg suppress estrogens to below limit of detection [92, 112, 113]. Based on pharmacokinetic and pharmacodynamic studies, the recommended dose of letrozole is one 2.5 mg tablet once daily. Preclinical models [97, 101] successfully predicted that letrozole would be superior to tamoxifen, the previous gold standard in the treatment of breast cancer. An extensive program of clinical trials has been conducted with letrozole across the spectrum of hormone-responsive breast cancer in postmenopausal women. The first randomized controlled trials demonstrated consistent superiority for letrozole compared with megestrol acetate, aminoglutethimide, and tamoxifen in patients with advanced breast cancer [114–118]. The clinical efficacy of letrozole in advanced breast cancer is described in a review by Dr. Mouridsen in this supplement. Preclinical MCF-7Ca models have also predicted that letrozole should be clinically more effective than other less potent third-generation AIs [99, 102]. Letrozole (2.5 mg/day) and anastrozole (1 mg/day) were directly compared in a randomized, open-label phase IIIb/IV study involving 713 postmenopausal women with advanced breast cancer previously treated with an anti-estrogen [119]. While there was no difference between the treatment arms in the time to progression, letrozole produced a significantly higher overall response rate than anastrozole (19.1 vs. 12.3%, P = 0.013). Letrozole and anastrozole are currently being compared in a large randomized head-to-head trial in early breast cancer (ClinicalTrials.gov identifier NCT00248170) [120]. A review by O’Shaughnessy in this supplement provides the rationale for this trial and a description of its design. The clinical benefits of letrozole in early breast cancer have already been demonstrated in landmark randomized clinical trials. MA.17 was the first trial to show improved clinical outcomes with extended adjuvant hormone therapy [121]. In this trial, letrozole given after initial adjuvant therapy with tamoxifen significantly improved disease-free survival compared with placebo [121, 122]. Full details of this trial are provided in a review by Dr. Goss in this supplement. Subsequently, the Breast International Group 1-98 trial provided high-level evidence for the superiority of letrozole over tamoxifen as initial adjuvant therapy [123]. A detailed description of this ongoing trial, which will also help to define the optimal sequence for hormone therapies in hormone-responsive early breast cancer, is provided in a review by Dr. Thürlimann in this supplement. Letrozole has also demonstrated superior efficacy compared with tamoxifen when used as neoadjuvant therapy [124]. This treatment setting is particularly interesting in terms of drug development because the effects of hormone therapy on breast tumors can be detected early and may be predictive of long-term outcome [125]. Conclusions Letrozole is a highly potent and selective AI that inhibits the enzyme activity of intracellular aromatase at the major sites where it is found, resulting in almost complete suppression of whole body aromatization. By effectively blocking estrogen synthesis, letrozole inhibits the growth or induces the regression of hormone-responsive breast tumors in vivo. Estrogen is implicated as a major risk factor in the majority of breast cancers; therefore, use of the most potent AI is a logical treatment strategy. Studies conducted using in vitro and in vivo models have demonstrated that letrozole is the most potent of the third-generation AIs. Preclinical data obtained from MCF-7Ca xenograft models suggest that the greater potency of letrozole compared with anastrozole and exemestane may translate into clinically meaningful differences in postmenopausal women with hormone-responsive breast cancer. These models accurately predicted that letrozole would be more effective than tamoxifen in the clinical setting. The superiority of letrozole over tamoxifen has been consistently demonstrated in advanced and early breast cancer [118, 123]. Outstanding clinical questions, including what is the most effective AI and what is the optimal sequence for adjuvant hormonal therapy, will be answered by the results of ongoing trials involving letrozole. In conclusion, experimental data indicating that letrozole efficiently inhibits aromatase activity have been confirmed clinically, leading to approved indications across the spectrum of breast cancer. The broad range of indications for letrozole in unique clinical settings is reshaping the management of hormone-sensitive breast cancer.

Breast_Cancer_Res_Treat-3-1-2092410

BRCA1 and BRCA2 germline mutation analysis in the Indonesian population

Specific mutations in BRCA1 and BRCA2 genes have been identified in specific populations and ethnic groups. However, little is known about the contribution of BRCA1 and BRCA2 mutations to breast cancers in the Indonesian population. One hundred-twenty moderate to high risk breast cancer patients were tested using PCR-DGGE, and any aberrant band was sequenced. Multiplex ligation-dependent probe amplification (MLPA) was performed on all samples to detect large deletions in the two genes. Twenty-three different mutations were detected in 30 individuals, ten were deleterious mutations and 20 were “unclassified variants” with uncertain clinical consequences. Three of seven (c.2784_2875insT, p.Leu1415X and del exon 13–15) and two of four (p.Glu2183X and p.Gln2894X) deleterious mutations that were found in BRCA1 and BRCA2 respectively, are novel. Several novel, pathogenic BRCA1 and BRCA2 germline mutations are found in early onset Indonesian breast cancer patients, these may therefore be specific for the Indonesian population. Introduction Breast cancer is the most common cancer in women. In 5% to 10% of breast cancer cases, the disease results from a hereditary predisposition [1, 2], which can to a large extent be attributed to mutations in either of two tumour suppressor genes, BRCA1 (MIM# 113705) and BRCA2 (MIM# 600185) [3–5]. These genes are involved in DNA repair as well as transcriptional regulation [6, 7]. Women carrying pathogenic germline mutations in either of these genes tend to develop breast cancer at an early age [8, 9]. The BRCA1 and BRCA2 genes encode large proteins of 1,863 and 3,418 amino acids, respectively. Over 300 distinct mutations in BRCA1 and BRCA2 have been described [10, 11]. These mutations are widely scattered across both genes and most affect the structure and function of the gene. Nevertheless, a significant proportion (34% of BRCA1 and 38% of BRCA2 mutations) (http://www.nhgri.nih.gov/Intramural_research/Lab_transfer/Bic) are missense mutations that alter one amino acid, but do not truncate the protein and are rare sequence variants of unknown functional consequence. Moreover, a number of base substitutions do not alter the amino acid sequence or result in amino acid changes not associated with disease (polymorphisms) [12]. Hence the biggest challenge in interpreting the mutation analysis of BRCA1 and BRCA2 genes is to distinguish between harmless polymorphisms and deleterious mutations associated with increased cancer risk. In addition, mutations specific for certain populations and ethnic groups have been identified in both genes. For example, specific BRCA1 and BRCA2 mutations were reported for Ashkenazi Jews [13]. Other common BRCA1 mutations were especially found in Italian, Canadian, Belgian or Dutch breast cancer families [14–16]. In Indonesia, the contribution of the BRCA1/BRCA2 mutations to the population incidence of early-onset breast cancer is largely unknown. In one pilot study, however, a new BRCA2 mutation was identified [17] indicating that it was worthwhile to more extensively study the Indonesian population, which was the aim of this study. The accumulating knowledge about the prevalence and nature of BRCA1 and BRCA2 mutations in specific populations may facilitate the interpretation of genetic analysis with regard to breast cancer risk of individual patients. Materials and methods Patients A total of 120 unrelated breast cancer patients and 16 of their family members from three Indonesian cities (Jakarta and Jogjakarta on the Java island, Denpasar on the Bali island) were analyzed. Breast cancer patients at moderate to high-risk of a hereditary predisposition were selected according to the following criteria: (A) Breast cancer before the age of 41 (n = 102); (B) Two cases of breast cancer in the same family before the age of 60 (n = 9); (C) Three or more cases of breast cancer in the same family (n = 2); (D) Bilateral breast cancer (n = 7). Subjects were asked to fill out questionnaires to evaluate their personal and family histories, and blood specimens were collected for determination of BRCA mutations. Informed consent was obtained from all the subjects in this study. DNA extraction and PCR amplification Genomic DNA was isolated by the saturated salt extraction procedure as described in [18]. All 22 coding exons of BRCA1 and 26 coding exons of BRCA2 were amplified using primer sequences developed by the University of Groningen, The Netherlands [19]. Primers for DGGE were obtained from Ingeny (Goes, The Netherlands). Genomic DNA was amplified using 100–200 ng of template DNA, 10 pmol of the mixture of 40-mer primers, 30 mM of MgCl, 3 mM dNTPs (Invitrogen) and 0.7 unit of Platinum Taq (Invitrogen) in 9 μl PCR reactions. In order to speed up the test, the PCR reaction was placed in 384 well plates using a pipetting robot (TECAN Miniprep 75). PCR conditions were performed as previously described [17]. Denaturing Gradient Gel Electrophoresis and DNA sequencing A 4–6 μl aliquot of each PCR product with relatively large melting temperature differences were pooled as previously described [17] with some modifications for optimal results. The fragment pool was designed based on melting profiles and sequence. Electrophoresis was performed in 0.5 TAE buffer at 58°C, 120 V for 16 h for BRCA1 gene, and 55°C, 100 V for 18 h for BRCA2. Gels were stained with ethidium bromide and photographed under a UV transilluminator. The aberrantly migrating samples were re-amplified using sequencing primers and sequencing was performed using Big Dye Cycle-sequencing kit according to the manufacturer’s instructions. The reaction products were analyzed using an ABI 3100 DNA Sequencer (Applied Biosystems, Torrence, CA, USA) and sequence files were edited using the Bio Edit program. The classification of gene alterations was performed in accordance with the entries in the Breast Cancer Information Core (BIC, Bethesda, MD). Multiplex ligation-dependent probe amplification (MLPA) The principle of the MLPA technique has been described elsewhere [20]. The MLPA test for BRCA1 (P02) and BRCA2 (P45) mutations were obtained from MRC-Holland, Amsterdam, The Netherlands. The fragments were analyzed on an ABI model 310 capillary sequencer (Applied Biosystems, Torrence, CA, USA) using Genescan-TAMRA 500 size standards (Applied Biosystems). Fragment analysis was performed with Genescan software. Results and discussions We identified 120 incident Indonesian breast cancer cases diagnosed before the age of 41 years, or having family history of breast cancer, or harboring bilateral breast cancer during September 1999–April 2005 (Jogjakarta) and during July 2004–April 2005 (Jakarta and Denpasar). In addition, 16 of their family members were analyzed. The entire coding regions and exon-intron junctions of BRCA1 and BRCA2 were screened in these 136 persons of breast cancer patients and their families using PCR-DGGE (Fig. 1) followed by sequencing (Fig. 2) for samples with aberrant migrating bands. To optimize the screening, MLPA, a relatively new technique, was also performed in all samples (Fig. 3). Here, we report on 116/120 women (96.7%) for whom BRCA1/2 analysis were completed. The remaining four patients (all from group A) had to be excluded due to the small amount of extracted DNA that did not allow complete screening of the BRCA1 and BRCA2 genes. Fig. 1DGGE analysis of fragments 11.15 g, 11.4 and 11.10 of the BRCA2 gene in ten unrelated breast cancer patients. The arrows show altered band mobility compare to other patientsFig. 2Sequence electropherogram of a normal individual showing (A) wild-type BRCA2 exon 11 sequence and (B) of breast cancer patient (B-3-5) showing c.2699_2704delTAAATG mutationFig. 3MLPA analysis of BRCA1 gene of patient sample (blue) compare to the normal control (red). X and Y axis represent peak size and peak height respectively. There are reduced peaks in the patient sample compared to the normal control in exons 13, 14 and 15 (arrows) indicating deletions BRCA1 and BRCA2 pathogenic mutations The analysis of 116 unrelated breast cancer patients with breast cancer revealed that nine patients (7.8%) carried pathogenic germline mutations especially the early onset patients: 3 within BRCA1 (2.6%) and 6 within BRCA2 (5.2%) which is comparable to previous studies [21]. We only found BRCA1 and BRCA2 mutations in groups A (“early onset”, n = 7 out of 98, 7.1%) and B (two cases of breast cancer in the same family before the age of 60, n = 2 out of 9 (22.2%)) (Table 1). There were twice as many BRCA2 mutations as BRCA1 mutations. Although the absolute numbers are low and no firm conclusions can therefore be drawn, this is comparable to other Asian regions [22–24] but seems to discern the Indonesian population from non-Asian ethnic groups where the reverse trend is seen. Table 1BRCA1 or BRCA2 germline mutations in Indonesian women with early onset breast cancerPatientAgeageneExonMutationbmutation typePathogenic mutationBICcAE25BRCA111c.2784_2785insTframeshift+noB1031BRCA113p.Leu1415Xnonsense+noAA40BRCA113–15−dlarge rearrangement+noAB34BRCA211c.3040_3043del4frameshift+1B566BRCA211p.Glu2183Xnonsense+noB665BRCA211p.Glu2183Xnonsense+noB-III-530BRCA211p.Leu824Xnonsense+noAZ40BRCA211p.Leu824Xnonsense+noW-II37BRCA221p.Gln2894Xnonsense+noQ-II40BRCA12c.101–10T>CIVS±6P-III-1919BRCA19p.Val191IleMissense±6J2232BRCA111p.Leu1209ValMissense?noAZ40BRCA116p.Met1652IleMissense±35B1 24BRCA120c.5313–31A>GIVS?noB731BRCA124p.Arg1835GlnMissense?no21633BRCA124p.Thr1852IleMissense?noP-III-1919BRCA25p.Gln147ArgMissense±6B324BRCA210p.Gln609GluMissense?noC-II-739BRCA211p.Met1149ValMissense±5AO28BRCA211p.Met1149ValMissense±5AQ44BRCA211p.Met1149ValMissense±5BH38BRCA211p.Met1149ValMissense±517236BRCA211p.Gln699LeuMissense?noJ3229BRCA211p.Arg2108CysMissense±16J633BRCA211p.Val950IleMissense?no20637BRCA225c.9485–16T>CIVS±4BC35BRCA227p.Ile3412ValMissense±10916633BRCA227p.Ile3412ValMissense±109J2435BRCA227p.Ile3412ValMissense±10920637BRCA227p.Lys3326Xnonsense ±289aAge at time of diagnosisbGen Bank Accession number, BRCA1: U14680, BRCA2: U43746cnumber of times reported in BICdnot determined, detected by MLPA Seven pathogenic mutations were found in nine probands: three in BRCA1 (c.2784_2785insT, pL1415X (c.4361_4362insT), del exon 13–15) and four in BRCA2 (c.3040_3043delGCAA, p.Glu2183X (c.6775G>T), p.Leu824X (c.2699_2704delTAAATG), p.Gln2894X (c.9008C>T)). All these mutations were classified as pathogenic as they are predicted to result in protein truncation. The three pathogenic mutations found in BRCA1 were not previously reported in the BIC database as well as two novel nonsense mutations (p.Glu2183X and p.Gln2894X) identified in BRCA2. The p.Glu2183X mutation was found in 2 related patients that had breast cancer above the age of 60. One of seven pathogenic mutations found in BRCA1 and BRCA2 showed a significant clinical impact on the patient (Table 2). Patient AE with a one nucleotide insertion (Thymine) between nucleotide 2784 and 2785 (c.2784_2785insT) in exon 11 of BRCA1 suffered from bilateral breast cancer at a relatively early age (25 years). The insertion leads to frameshift and creates a premature stop codon in exon 11. The mutation takes place in the sequence within BRCA1 encoding for aminoacids 758–1064 which interact with RAD51 protein that is required for homologous recombination (HR) repair of double strand breaks (DSBs) [25], which is one of the most important functions of the BRCA1 protein. This patient presented in a late stage (stage III for both breasts) and only survived for 9 weeks after treatment. Her mother did not carry this mutation. Although her father may be carrier, the mutation is probably de novo as there was no family history of breast or other cancers. Table 2Clinicopathological features of Indonesian breast cancer patients with deleterious BRCA1 or BRCA2 germline mutationsPatientAgeaGene with germline mutationMutationbstageDiagnosisMenopausal statusfamily history of cancerSurvival statusAE25BRCA1c.2784_2785insTIIIB/IIIAIDC, bilateralpreNoDOD 9 wB1031BRCA1p.Leu1415XIIDCpreNoDOD 57 wAA40BRCA1-cIIIBIDC N+preNoAWDAB34BRCA2c.3040_3043del4IIIBIDC N+preSister, IntDOD 17 wB563BRCA2p.Glu2183XIVTubularpostSister,BrAWDB665BRCA2p.Glu2183XIIIIDCpostBrother, BrAWDB-III-530BRCA2p.Leu824XIIDCpreNoAWDAZ40BRCA2p.Leu824XIVIDCpreSister, CvDOD 46 wW-II37BRCA2p.Gln2894XIIIAIDCpreNoDOD 107 waAge at time of diagnosisbGen Bank Accession number, BRCA1: U14680, BRCA2: U43746cnot determined, detected by MLPAIDC: invasive ductal carcinoma; DOD: dead of disease; bil: bilateral breast cancer; N+: with metastatic to lymph node; Int: intestinum cancer, Br: breast cancer; Cv: cervical cancer The second pathogenic mutation with a significant clinical manifestation was a cytosine for thymine substitution on nucleotide 9008 of BRCA2 leading to a premature stop codon in position 2894, c9008C>T (p.Gln2894X). Patient W presented at age 37 in a late stage and survived for only 107 weeks after initial treatment. She had no family history of breast or other cancers. This mutation lies within exon 21 of BRCA2 which is the proposed site for interaction with the DSS1 protein that seems to have a fundamental role in enabling the BRCA2-RAD51 complex to associate with sites of DNA damage [26]. The c.2699_2704delTAAATG (p.Leu824X) in BRCA2 that has been reported previously by us in the Indonesian population [17], was found in one other patient in the present study (Table 1). This mutation lies in exon 11 BRCA2, within the BRC repeats domain. The truncating mutation causes loss of three quarters of the protein leading to lack of interaction with the RAD51 protein. Different from BRCA1, the repair of DSBs by HR is the most important function of the BRCA2 protein [27]. Patient B-III-5 was diagnosed with early stage breast cancer at age 30 with no family history of breast or other cancers. Her sister carried the same mutation, but with no present clinical manifestation as yet. Patient AZ who was diagnosed at 40 years of age, presented in late stage, only survived 46 weeks after initial treatment. This patient also harbored a mutation in exon 16 of BRCA1, a G to A substitution in nucleotide 5075 (c.5075G>A), which leads to amino acid change from Methionine to Isoleucine, (p.Met16521Ile) which has to date been reported 35 times in BIC as a UV mutation. As the c.2699_2704delTAAATG mutation was found in two unrelated patients, this mutation could be a good candidate as a founder mutation. None of the families with more than 3 cases of breast cancer and families with bilateral breast cancer showed pathogenic mutations in the BRCA1 and BRCA2 genes. Family U had four first-degree relatives that were affected by breast cancer. Two of four members had bilateral breast cancer. In spite of this high familial breast cancer incidence, no BRCA1/2 mutations were found. BRCA1 and BRCA2 unclassified variants Sixteen (7 BRCA1 and 9 BRCA2) rare mutations of so far unknown significance (“unclassified variants”, UVs) were detected in 18 patients: 13 missense changes and 3 intronic variants. Of these 16 UVs, 7 were novel, whereas the other UVs have been previously reported in the BIC database (Table 1). From the 18 patients which carried UV mutations, two patients were detected in families from group D; one patient in a group B family and the other fifteen patients in families from group A. Seven UV were found in the BRCA1 gene, two mutations occurring in the intronic region between exons 1 and 2 (c.101–10T>C) and between exons 19 and 20 (c.5313–31A>G), and five missense mutations identified: p.Val191Ile (c.690G>A), p.Leu1209Val (c.3744T>G), p.Met1652Ile (c.5075G>A), p.Arg1835Gln (c.5623G>A) and p.Thr1852Ile (c.5674C>T). Four out of seven BRCA1 missense mutations; p.Leu1209Val (c.3744T>G), c.5313–31A>G, p.Arg1835Gln (c.5623G>A) and p.Thr1852Ile (c.5674C>T) were have not been described previously in the BIC. The p.Leu1209Val may not be a significant change as both Leucine and Valine belong to the same group of non polar amino acids. However, p.Arg1835Gln is possibly an important alteration since a positively charged Arginine is replaced by an uncharged Glutamine, which may have an effect on the structure and/or function of the protein. Another potentially important alteration concerns p.Thr1852Ile, where the hydrophilic amino acid Threonine is replaced by a hydrophobic Isoleucine. The sites of mutation of both p.Arg1835Gln and p.Thr1852Ile also have to be considered as they lie within the site for the activation domain of the BRCA1 protein [28]. The intronic UV c.5313–31A>G also deserves further investigation as it may theoretically have an effect on splicing. However, according to splice site finder (http://www.genet.sickkids.on.ca/∼ali/splicesitefinder.html), the splicing sites in the wild type and mutant alleles are similar, so therefore we can suggest that the c.5313–31A>G has no effect on splicing. Nine different UVs of the BRCA2 gene were found in fourteen patients (Table1), and three of them were novel; p.Gln609Glu 9c.2053C>G), p.Gln699Leu (2324A>T) and p.Val950Ile (3076G>A). One truncating mutation near the C-terminal end of BRCA2, p.Lys3326X (c.10204A>T) is probably not pathogenic. Since the truncating mutation is at the very end of the protein, it is possible that protein functions are not affected. Most of the few entries in databanks describing nonsense mutations near the C terminus of BRCA2 between codon 3308 and 3408 are described as UVs. Thus, the effect of this truncating mutation on cancer predisposition remains unclear. The p.Val950Ile may not be a significant change as both Valine and Isoleucine belong to the same group of non polar, hydrophobic amino acids. However, p Gln609Glu and p.Gln699Leu are potentially important alterations as for p.Gln609Glu, a non acidic, polar, hydrophilic Glutamine is replaced by a negatively charged Glutamic acid, whereas for p.Gln699Leu, an uncharged hydrophilic Glutamine is replaced by a hydrophobic Leucine. As it takes place within the BRC repeats of the BRCA2 protein, the p Gln699Leu alteration might affect protein structure and function. To know more about the importance of amino acid substitutions for protein function, we compared the amino acid sequence of interest in seven other species, i.e. Mus musculus, Rattus rattus, Bos taurus, Gallus gallus, Canis familiaris, Macaca mullata and Monodelphis domestica. The missense mutation p.Leu1209Val lies in the conserved region of exon 11 of the BRCA1 gene as the sequence is maintained in seven other species, whereas p.Arg1835Gln and p.Thr1852Ile are only conserved in four and three other species (comparison of p.Arg1835Gln and p.Thr1852Ile with Bos taurus sequence is not possible because the BRCA1 gene is shorter). Therefore, even tough the Leucine to Valine changes may not give any effect on amino acid charge, its conservation in evolution is suggestive of a functional role. Interestingly, p.Gln609Glu and p.Gln699Leu of BRCA2 that result in a quite dramatic amino acid subtitution that might lead to protein structure changes, are only conserved in four and five species respectively. As for the p.Val950Ile, the conservation in evolution is quite low. Although p.Gln609Glu is less conserved, we still believe that Glutamine to Glutamic acid substitution may have an effect on protein conformation as two adjacent acidic amino acids will be formed as the result of the substitution. Glycosylation moiety of an amino acid also plays a role in protein function. Amino acid substitutions involving Serine, Threonine and Asparagine, should also be checked for their O-GlcNac potential and threshold. Here we have a Threonine to Isoleucine substitution (p.Thr1852Ile) that after checking with YinOYang (http://www.cbs.dtu.dk/services/YinOYang) showed no significant threshold changes between the wildtype and the mutant allele. The possible effect of amino acid changes in proteins can also be assessed using similarity scores (based on Grantham table [29]), in which a value above 100 for an amino acid substitution indicates a higher chance of impact on protein function. Among seven novel UVs in the BRCA1 and BRCA2 genes found in the present study, only p.Gln699Leu in BRCA2 has a similarity score above 100, whereas p.Gln609Glu and p.Val905Ile in BRCA2 have the lowest score (Table 3). Table 3The amino acid properties of novel unclassified mutations in BRCA1 and BRCA2 within an Indonesian breast cancer populationGeneAmino acid changeChange of chargeChange of amino acid groupSimilarity scorea# species with conserved sequenceBRCA1Leu to ValNoneNo327a,b,c,d,e,f,gBRCA1Arg to GlnPos to no chargeYes434a,c,f,gBRCA1Thr to Ilepolar to non polarYes893a,c,gBRCA2Gln to GluNo charge to negYes294a,b,c,gBRCA2Gln to LeuPolar to non polarYes1135a,b,d,e,fBRCA2Val to IleNoneNo292f,gabased on Grantham table [Grantham et al. [29], a score above 100 indicates significance changesa = Macaca mullata, b = Bos taurus, c = Canis familiaris, d = Rattus rattus, e = Mus musculus, f = Gallus gallus, h = Monodelphis domestica Overall, we propose that among the seven novel UVs, there are three mutations that are possibly pathogenic: p.Leu1209Val for its location in a conserved region, and p.Gln609Glu and the p.Gln699Leu because of two adjacent acidic amino acid being formed and a high similarity score, respectively. When comparing the three different Indonesian regions, the percentages of breast cancer patients with pathogenic BRCA1/2 mutations was significantly higher in Denpasar (Bali island) than in Jogjakarta and Jakarta (Java island) ((25% (3/12), 7.2% (6/83) and 0% (0/25) respectively (P = 0.0255, chi-square test)). The percentages of breast cancer patients with UV mutations in Jakarta, Jogjakarta, and Denpasar were 16% (4/25), 12% (10/83), and 25% (3/12), respectively (n.s.). Although the number of patients is too small to draw firm conclusions, these data may point to geographic differences within Indonesia. It was initially suggested that the BRCA1 and BRCA2 genes would be responsible for most cases of inherited breast cancer, but more recent studies suggest that they would account for a far smaller proportion, with considerable variation among different populations [30]. We found that the incidence of mutations in these genes varies, depending on the diagnostic group. In this sense, mutations were present in (22/102) 21.6% of early onset patients (group A), 28.7% (2/7) in patients with bilateral breast cancer (group D) and (2/9) 22.2% of patients with two cases of breast cancer before the age of 60 (group B). The proportion of families affected by BRCA1/2 mutations depends on the population analyzed and on the criteria used to select the patients. Family history of breast cancer was, however, absent or not suggestive of a hereditary predisposition in three-fourth of the deleterious mutations carriers and in more than 90% of UV carriers. This suggests that BRCA screening policies based on family history only would miss a considerable proportion of mutation carriers. In conclusion, a relatively high percentage of early onset Indonesian breast cancer patients carry a germline mutation in either BRCA1 or BRCA2. Several novel, pathogenic BRCA1 and BRCA2 germline mutations have been found, as well as a variety of novel “unclassified variant” mutations that may therefore be specific for the Indonesian population. It is likely that some of the “unclassified variant” mutations may have a functional role in breast cancer development, which deserves to be explored further.

Purinergic_Signal-4-1-2246001

Remyelination after chronic spinal cord injury is associated with proliferation of endogenous adult progenitor cells after systemic administration of guanosine

Axonal demyelination is a consistent pathological sequel to chronic brain and spinal cord injuries and disorders that slows or disrupts impulse conduction, causing further functional loss. Since oligodendroglial progenitors are present in the demyelinated areas, failure of remyelination may be due to lack of sufficient proliferation and differentiation of oligodendroglial progenitors. Guanosine stimulates proliferation and differentiation of many types of cells in vitro and exerts neuroprotective effects in the central nervous system (CNS). Five weeks after chronic traumatic spinal cord injury (SCI), when there is no ongoing recovery of function, intraperitoneal administration of guanosine daily for 2 weeks enhanced functional improvement correlated with the increase in myelination in the injured cord. Emphasis was placed on analysis of oligodendrocytes and NG2-positive (NG2+) cells, an endogenous cell population that may be involved in oligodendrocyte replacement. There was an increase in cell proliferation (measured by bromodeoxyuridine staining) that was attributable to an intensification in progenitor cells (NG2+ cells) associated with an increase in mature oligodendrocytes (determined by Rip+ staining). The numbers of astroglia increased at all test times after administration of guanosine whereas microglia only increased in the later stages (14 days). Injected guanosine and its breakdown product guanine accumulated in the spinal cords; there was more guanine than guanosine detected. We conclude that functional improvement and remyelination after systemic administration of guanosine is due to the effect of guanosine/guanine on the proliferation of adult progenitor cells and their maturation into myelin-forming cells. This raises the possibility that administration of guanosine may be useful in the treatment of spinal cord injury or demyelinating diseases such as multiple sclerosis where quiescent oligodendroglial progenitors exist in demyelinated plaques. Introduction Demyelination contributes to functional deficits in central nervous system (CNS) disorders [1–3]. Traumatic spinal cord injury (SCI) induces local inflammation and demyelination in the white matter around the lesion resulting in disrupted axonal conduction [4–7]. Remyelination is critical for recovery, but does not occur spontaneously [8–11]. Spinal cords contain endogenous progenitors that can proliferate and differentiate into mature oligodendroglia and remyelinate axons under certain circumstances [12–14]. After acute SCI in rats, oligodendrocyte precursors proliferate in the first 2 weeks but are “silent” thereafter [15], possibly because the oligodendroglial precursors cannot differentiate into mature oligodendroglia capable of remyelinating axons. Enhancement of proliferation of oligodendrocytes or progenitors may lead to repair and restoration of function after SCI. Neurotrophic factors restore physiological function by promoting oligodendrocyte proliferation and axonal sprouting [16–19], regulating the proliferation of oligodendrocyte progenitors [3] and modulating the differentiation [20–22] and the maturation of oligodendrocyte precursors into myelin-forming cells [23, 24]. Guanosine has trophic effects on many cell types [25–28], and we have some evidence that guanine acts similarly. For example, it stimulates proliferation of a variety of cells in culture [25, 28] and promotes synthesis and release of several potentially neuroprotective trophic factors from a variety of cells, including neuron growth factor (NGF) from astrocytes as well as basic fibroblast growth factor (bFGF) and transforming growth factor beta (TGF beta) [25, 27]. Extracellular guanosine itself also enhances outgrowth of nerve processes from PC12 cells and from primary cultures of embryonic rat brain neurons [25]. As well, it acts synergistically with NGF to promote outgrowth of neurites [29, 30]. Our preliminary data [31] showed that 35 days after moderate traumatic injury of the spinal cord of rats, when the initial spontaneous incomplete recovery of function had plateaued, systemic administration of guanosine (intraperitoneally 8 mg/kg body weight) for 7 days enhanced locomotor functional recovery that continued to improve throughout the 7-day duration of the treatment and correlated with the increase in myelination. In the present study, we treated rats for up to 14 days to determine whether guanosine-induced functional improvement was due to triggering the endogenous adult progenitors. Meanwhile, we also examined the uptake and tissue distribution of systemically administered guanosine. Materials and methods All experiments were performed in compliance with the requirements of the Animals for Research Act of Ontario, Canada and the Guidelines of the Canadian Council on Animal Care, and had been approved by the Animal Research Ethics Board of McMaster University. Spinal cord injury Adult female Wistar rats (280–300 g weight, Charles River) were anaesthetized with isoflurane (3–5%): O2 (1 l/min). Buprenorphine (0.03 mg/kg body weight, subcutaneously) was administered prior to surgery for pain relief. Moderate spinal cord injury was induced by crushing exposed cords with modified coverslip forceps [31–33]. The forceps were closed slowly over 2 s and the compression sustained for 15 s. The muscles and fat pad over the lesion were sutured and the skin was closed with stainless steel clips. Post-operatively, the rats were kept quiet and warm [31, 34, 35]. Behavioural assessment and drug administration Rats were handled daily for 2 weeks pre-operatively to acclimatize them to the experimental procedures and behavioural test. After the spinal injury, locomotor recovery was assessed weekly for 35 days in an open field walking task (OFWT) before treatment. Cagemates (two animals) were placed in the centre of the open field, formed by a child’s circular plastic swimming pool (1.3 m in diameter). They were observed for 5-min periods and scored for general locomotor ability using the Basso-Beattie-Bresnahan (BBB) locomotor rating scale [36, 37]. Rats were rated on a scale of 0 to 21, with 0 being no function and 21 being normal. If the animal stopped moving for a minute, it was placed again in the centre of the open field; otherwise it was left alone for the duration of the 5-min test period.On day 35 following the injury, rats were randomly assigned to three groups. One group of rats was used for baseline behavioural and histological analysis. The other rats received either daily intraperitoneal (i.p.) injections of 8 mg/kg guanosine or the same volume of saline containing 0.001 N NaOH [31] for 2 weeks. The open field walking testing was performed once a day from days 35 to 48 post-surgery (0 to 14 days post-treatment). Groups of 9–12 rats were killed by perfusion/fixation for immunohistochemical analysis at 1, 3, 7 or 14 days after treatment. BrdU administration Bromodeoxyuridine (BrdU) was used to label proliferating cells after treatment. BrdU (50 mg/kg in sterile saline; Sigma, B-5002) was injected intraperitoneally daily for 1, 3, 7 or 14 consecutive days, beginning the day treatment was initiated, thereby reflecting cell proliferation over the entire treatment period. Tissue processing and immunohistochemistry On post-operative days 35, 36, 38, 42 or 49 (representing baseline, 1, 3, 7 or 14 days treatment), rats were deeply anaesthetized with sodium pentobarbital (50–60 mg/kg body weight, i.p.) immediately following the final locomotor testing and perfused transcardially with 100 ml 0.1 M phosphate-buffered saline (PBS) containing 0.1% heparin, followed by 300–500 ml of 4% paraformaldehyde (PFA). T9 to L1 segments of the spinal cords were removed and incubated in the same fixative for 2 h at room temperature and then cryoprotected in 30% sucrose PBS solution. A segment of each cord, extending from 5 mm rostral to 5 mm caudal to the lesion site was embedded in medium (Tissue-Tek® O.C.T. compound, Sakura Finetek USA, Inc., Torrance, CA, USA). Serial transverse sections were cut at 15- to 20-μm intervals on a cryostat and mounted onto slides (ColorFrost/Plus; Fisher, Pittsburgh, PA, USA). Every fifth section was stained with the lipophilic dye Luxol fast blue to stain myelin [38]. Adjacent sections were stained immunohistochemically with the various antibodies described below.For BrdU immunohistochemistry, 6% hydrogen peroxide was used to reduce endogenous peroxidase activity, followed by three 10-min PBS rinses and incubation in 1 N HCl at 37°C for 30 min. After rinsing, a blocking solution [5% bovine serum albumin (BSA) in 0.1% Triton-100] was applied for 1 h; sections were subsequently incubated in rat anti-BrdU (1:200 in 1% BSA in 0.1% Triton-100; Serotec, Raleigh, NC, USA) at 4°C overnight in a humid chamber. Sections were then rinsed and incubated with biotinylated horse anti-rat IgG secondary antibodies. Sections were then incubated with Elite ABC (Vector Laboratories, Burlingame, CA, USA) for 1 h, followed by 3,3′-diaminobenzidine (DAB) substrate kit for peroxidase (Vector Laboratories, Burlingame, CA, USA) for 5–10 min. Sections were dehydrated, coverslipped and examined under a microscope.To determine the fate of newly born cells after SCI, double-fluorescent immunolabelling was performed, combining BrdU with one of the cell-specific phenotypic markers listed below. For double immunofluorescence, sections were pretreated with HCl as described above and incubated with BrdU antibody mixed with one of the following antibodies in PBS at 4°C overnight in a humid chamber: rabbit anti-MBP (myelin basic protein) polyclonal antibodies (1:50; Chemicon Int., Temecula, CA, USA), rabbit anti-NG2 polyclonal antibodies (chondroitin sulfate proteoglycan for oligodendroglial progenitors; 1:200; Chemicon Int. Temecula, CA, USA), mouse anti-Rip monoclonal antibodies (mature oligodendrocyte-specific marker: Reactive Immunology Protein; 1:200; Chemicon Int., Temecula, CA, USA; [39]), rabbit anti-glial fibrillary acidic protein (GFAP) polyclonal antibodies (1:600; Zymed® Lab-SA System Kit, San Francisco, CA, USA) and mouse anti-microglial-specific marker OX-42 monoclonal antibodies (1:200; CD11b, Research Diagnostics Inc., Flanders, NJ, USA). For Rip and OX-42 monoclonal antibodies single staining, sections were developed with fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse IgG. For BrdU and polyclonal antibodies (NG2 or MBP or GFAP) double immunolabelling, sections were developed using a mixture of FITC-conjugated goat anti-rat IgG and rhodamine-conjugated goat anti-rabbit IgG in 1% normal goat serum and 0.25% Triton X-100; 1:200 (Invitrogen, Carlsbad, CA, USA) for 2 h. To examine the contribution of Schwann cells from the peripheral nervous system (PNS) in the remyelination process, sections were immunolabelled with specific markers for Schwann cell myelin P0 (a rabbit polyclonal anti-P0 antibody was kindly donated by Prof. Marie T. Filbin, Department of Biology, CUNY, NY, USA; 1:200) and secondary antibody using rhodamine-conjugated goat anti-rabbit IgG. For detection of apoptotic cells, a terminal deoxynucleotidyltransferase (TdT)-mediated dUTP nick end labelling (TUNEL) stain was performed using the ‘In situ Cell Death Detection Kit, Fluorescein’ (Roche Molecular Biochemicals, Mannheim, Germany), according to the manufacturer’s instructions. Quantification To quantify the amount of myelin present at the injury site before and after treatment, the Luxol fast blue-stained section from each cord (n = 7–10 for each group) that contained the greatest lesion area was selected together with the adjacent two caudal and two rostral Luxol blue-stained sections. Luxol fast blue staining was also done on cords of five additional unoperated normal animals for comparative purposes. Digital photographs were taken of the sections. The total area of each section and the area of the fast blue-stained portion were measured using a computerised Bioquant BQ-TCW98 image analysis program by an investigator who was blind to group assignment.For quantification of cell proliferation, five to seven sections taken from the penumbra of the lesion and spaced about 100 μm apart were analysed for each animal (n = 7–9 animals per group). Cellular proliferation was determined by counting the total number of BrdU+ nuclei in spinal cord cross sections. Low power sections were digitized and manually outlined using an image analysis system. Any cavities present in the sections were excluded from analysis. Positively labelled nuclei were counted automatically. The threshold was set such that only positively labelled profiles of the appropriate size were counted. Data are expressed as cells per section.The total number of NG2+ progenitor cells and the number of those that proliferated (double-labelled for NG2 and BrdU: NG2+/BrdU+) was determined by manual counting. Cell counting was done conservatively, and a positive profile was counted as a cell only when clearly defined borders could be detected completely around and adjacent to the entire nucleus. Data are expressed as the number of immunostained cells per section.Mature oligodendrocytes in injured cords were identified with Rip, a specific marker for mature oligodendrocytes that recognises an unknown epitope [39] and has been used successfully in the CNS to label both oligodendrocyte processes and myelin sheaths. The number of mature oligodendrocyte (Rip+ cells) in the tissue sections was determined by manual counting. Evaluation of exogenous guanosine distribution A guanosine solution (8 mg/kg) was administered intraperitoneally to rats with spinal cord injury induced 5 weeks previously. Samples of spinal cord were taken from the site of the lesion and immediately above or below the lesion just before the intraperitoneal injection of guanosine and at 7.5, 30 and 60 min after the guanosine injection. There were between three and six animals at each time point. Samples of spinal cord were placed in 0.5 ml of cold 0.4 M perchloric acid solution and immediately homogenized. Samples were then centrifuged (4°C, 10,000 g, 10 min) to remove proteins precipitated by perchloric acid and the supernates containing the soluble molecules were neutralized with KOH solution and centrifuged (4°C, 10,000 g, 5 min). Samples were filtered with 0.2-μm filters (Millipore, Billerica, MA, USA) and stored at −70°C before high-performance liquid chromatography (HPLC) [40] or capillary electrophoresis (CE) analysis. Tissue concentrations of guanosine and guanine To evaluate the concentration of guanosine and its principal degradation product, guanine, in spinal cords before and after the intraperitoneal injection of guanosine, either HPLC or CE was used. There was no significant difference in the values obtained by these two methods so the data represent the mean of values obtained by both analyses. HPLC analysis was carried out using an Agilent 1100 series HPLC (Agilent Technologies, Waldbronn, Germany) and an ion-pair technique. Separation was carried out with a reverse-phase analytical column (LiChroCART 125-4 LiChrospher 100 RP-18 5 μm, Merck, Darmstadt, Germany) and the flow rate was 1.5 ml/min. A 15-min linear gradient was applied from 100% buffer A (60 mM KH2PO4 and 5 mM tetrabutylammonium phosphate, pH 6) to 100% buffer B (30% methanol plus 70% buffer A). CE separations were carried out using an HP3DCE capillary electrophoresis system (Agilent Technologies, Waldbronn, Germany) equipped with a diode array detection system. Analysis was performed using a 80.5 cm × 50 μm ID × 375 μm OD uncoated fused silica capillary from Agilent Technologies (Waldbronn, Germany) with an effective length of 72 cm to the detector window. The capillary temperature was kept constant at 37°C and the running buffer was 10 mM borate, pH 9.3. The samples were injected by pressure (50 mb for 40 s) and the separation was conducted at 22 kV. Detection took place at 254 nm and purines were identified by their migration times and UV spectra using the diode array detector (Agilent Technologies, Waldbronn, Germany). Statistical analysis Data are expressed as mean ± SEM. Behavioural scores were analysed by the Kruskal-Wallis non-parametric analysis of variance (ANOVA). Post hoc comparisons were made using the Dunnett’s test. Correlation between behavioural and histological outcomes was analysed using regression analysis. Significance was set at p < 0.05. Results and discussion After SCI, motor function in the legs partially recovered over 3–4 weeks, after which there was no further improvement (Fig. 1a). Daily guanosine for 14 days (8 mg/kg intraperitoneally beginning 35 days after injury) produced improvement in locomotor performance over the next 2 weeks; control rats showed no improvement (Fig. 1b). This treatment regimen was associated with significantly increased amounts of myelin in the penumbra of the cord lesions shown by Luxol fast blue staining (Fig. 1c–e). The extent of Luxol fast blue staining in the injured cords correlated with each animal’s open field walking scores (r = 0.72; p < 0.05; Fig. 1f), indicating that guanosine-induced functional recovery may be related to the increase in myelin. Immunostaining for CNS myelin basic protein (MBP; Fig. 1g,h) and a specific marker for Schwann cell myelin, P0 (Fig. 1i,j) showed that Schwann cells from the peripheral nervous system did not enter spinal cord and, therefore, did not contribute to the remyelination. Fig. 1Systemic treatment with guanosine improves locomotor function and correlates with increased myelination in rats with stable chronic spinal cord injury. a After spinal cord crush, locomotor function was assessed in an open field walking task (OFWT) and scored using a standard scale (“Materials and methods”). No further improvement in locomotor performance occurred after 28 days. b On days 35–48 after injury (days 1 to 14 of treatment in the figure; the first test done immediately prior to the first treatment), animals were randomly divided into two groups. One group received guanosine (▴), the other group received vehicle (♦). The locomotor function in the control group did not change significantly throughout the 14 days of vehicle administration. In contrast, rats that received guanosine showed a marked improvement in locomotor behaviour, which was significantly (p < 0.05) better than vehicle-treated control animals from the 5th day of treatment onward. c, d Luxol fast blue staining of cross sections of spinal cords from guanosine- (d) and vehicle- (c) treated animals at the lesion site 48 days after the initial lesion (after 14 days treatment) shows increased Luxol fast blue-stained material in d compared to the vehicle-treated control group in c. DC dorsal column at the lesion. Scale bar = 50 μm. e The total area of each section and of the Luxol fast blue-stained portion were measured following the protocol in “Materials and methods”. The quantitative results show that the cords from guanosine-treated animals had significantly greater myelinated area (*p < 0.05), as estimated from the Luxol fast blue staining, than cords from vehicle-treated control animals, which in turn had significantly less myelinated area (**p < 0.001) than unoperated animals. f Correlation between behavioural and histological outcomes was analysed using regression analysis. The data show that the Luxol fast blue-stained area in the injured cords from animals treated with guanosine correlated with their behavioural recovery in open field walking scores, correlation coefficient of 0.72, indicating that the recovery of function may be related to the increase in myelin. Myelin immunofluorescent staining of cross sections of spinal cords from vehicle- (g) and guanosine- (h) treated animals at the lesion site 14 days after treatment shows that guanosine-treated cords (h) contained more myelin basic protein (MBP)-positive profiles from central nervous system at the injured site compared to the vehicle-treated control group (g). IS injured site. Whereas, in i (vehicle-treated cord) and j (guanosine-treated cord) immunolabelling with specific markers for Schwann cell myelin P0 indicated that Schwann cells from the peripheral nervous system did not contribute to guanosine-induced remyelination. Scale bar = 50 μm for all To determine whether existing cells elaborated the myelin or whether it was a product of other cells that proliferated and subsequently elaborated myelin, the number of proliferating cells was determined by counting all BrdU+ nuclei in cross sections of the lesion penumbra at different times after treatment. In the first 3 days after treatment, cell proliferation at the injury site in both guanosine-treated and control animals was not different from baseline (immediately before treatment; Fig. 2c), although at 3 days there was a trend toward more BrdU+ cells in guanosine-treated animals than controls (112 ± 25 vs 79 ± 16). At 7 and 14 days after guanosine treatment, the number of BrdU+ cells increased significantly; there were 382 ± 42 BrdU+ cells per section at 7 days and 477 ± 49 at 14 days compared with only 88 ± 20 at 7 days and 95 ± 22 at 14 days in controls (p < 0.05; Fig. 2a–c). There were few dividing cells beyond the lesion margins. Fig. 2Quantitative assessment of immunostaining shows that systemic treatment with guanosine stimulates endogenous oligodendroglial progenitors (NG2+ cells) to proliferate and mature in rats with stable chronic spinal cord injury. a, b Bromodeoxyuridine (BrdU) was used to label proliferating cells after treatment. BrdU immunostaining in cross sections from vehicle- (a) and guanosine- (b) treated animals at the lesion 14 days after treatment shows increased BrdU+ nuclei in b compared to the vehicle-treated control group in a. Scale bar = 50 μm. c Quantitative analysis shows that 7 or 14 days after guanosine treatment there was a significant increase in the number of BrdU+ cells compared with vehicle-treated controls. d, e Demonstrate examples of double-fluorescent immunostaining using antibodies against BrdU (in green) and a marker (NG2) for oligodendroglial progenitors (in red) in cross sections from vehicle- (d) and guanosine- (e) treated animals at the lesion 14 days after treatment. Scale bar = 50 μm. f–h Show quantification of proliferating progenitors (NG2+/BrdU+ double-labelled cells; f) and total number of NG2 cells (g). Data indicate that after 7 or 14 days administration guanosine significantly stimulates proliferation of oligodendroglial progenitors (f, g). Concurrently, guanosine also significantly increases the number of mature oligodendroglia shown in i and j compared to vehicle-treated controls (h, j) using fluorescent immunostaining with a specific marker (Rip) to label the mature oligodendroglia. Scale bar = 50 μm Oligodendrocytes that survive the demyelinating insult are not thought to be able to contribute to remyelination [41, 42]. Instead, endogenous oligodendrocyte progenitor cells local to the lesion are believed to be the source of new myelinating cells [43]. Previous studies have illustrated that NG2+ cells are the major cycling cell within the resting adult CNS [14, 44] and express many characteristics of oligodendroglial lineage cells, not least their ability to differentiate into oligodendrocytes when isolated into culture. Their role in demyelination and remyelination has been extensively studied and there is evidence that they are the cells responsible for myelin repair [43, 45–48]. To determine whether the proliferating cells were related to NG2+ oligodendroglial progenitor cells, we counted the total number of NG2+ progenitor cells and the number that had proliferated (NG2+/BrdU+) from 2 mm rostral to 2 mm caudal to the lesion site (Fig. 2d,e). There was minimal turnover in the NG2+ cells in the spinal cord during the first 3 days in both guanosine-treated and control animals (13 ± 5 in the guanosine group vs 9 ± 4 cells per section in controls; Fig. 2f). By 7 days, guanosine significantly stimulated progenitor proliferation (85 ± 18 vs 13 ± 4 NG2+/BrdU+ cells/section; Fig. 2f); these cells were observed throughout the penumbra with the majority localised in the dorsal columns. Enhanced proliferation of the NG2+ progenitors after guanosine persisted, and by 14 days, double-labelled progenitors in the guanosine-treated group were eightfold more than in controls (108 ± 12 vs 13 ± 5 NG2+/BrdU+ cells/section; Fig. 2f). The total number of NG2+ cells was also determined. During the first 3 days, there was no significant difference in the numbers of NG2+ cells in cords at baseline after vehicle treatment or guanosine treatment (Fig. 2g). By 7 days, guanosine treatment increased the total NG2+ cells from 96 ± 22 cells/section at baseline to 266 ± 38 cells/section—more than 2 times higher than in the control group (110 ± 24 /section; Fig. 2g). By 14 days, the numbers of double-labelled progenitors in the cords of guanosine-treated animals were threefold higher than controls (347 ± 36 NG2+ cells/section for guanosine-treated animals; 115 ± 24 cells/section for controls; Fig. 2g). Mature oligodendrocytes were identified with Rip+ staining (Fig. 2h,i). Guanosine had no effect on mature oligodendroglia 3 days after treatment; at 7 and 14 days, there was a significant increase in Rip+ cells in the penumbra of the cords of animals that had received guanosine compared to controls (7 days: 468 ± 40 vs 205 ± 25; and at 14 days: 492 ± 43 vs 219 ± 23 per section, p < 0.01; Fig. 2j). Guanosine treatment stimulated astrocyte proliferation in the penumbra as early as 3 days and throughout the 14 days of treatment (Fig. 3a,b) and was detected by antibodies to glial fibrillary acidic protein (GFAP). Astrocytes may stimulate oligodendroglial progenitor proliferation or induce their migration [49]. Guanosine did not alter the numbers of OX-42+ microglia/macrophages until 14 days after treatment. At day 14, guanosine significantly increased OX-42+ cells compared to vehicle treatment (Fig. 3c–e; p < 0.05). Fig. 3Cross sections of the cords at the injury site staining with GFAP and OX-42 from guanosine-treated and vehicle-treated rats 14 days after treatment following chronic spinal cord injury. a, b Demonstrate examples of GFAP-positive astrocytes (in red) in cross sections from vehicle- (a) and guanosine- (b) treated animals at the injury site 14 days after treatment. c, d Show OX-42-immunofluorescent staining of cross sections of spinal cords from vehicle- (c) and guanosine- (d) treated animals at the lesion site 14 days after treatment. Quantification showed more OX-42-immunolabelled cells in cords of guanosine-treated rats (d, e) compared to cords of vehicle-treated rats (c, e; p < 0.05). Scale bar = 50 μm for all Guanosine inhibits apoptosis due to a number of stimuli [50–52]. After stroke and spinal cord injury, intrinsic stem cells and progenitors proliferate and differentiate but many die through apoptosis [15, 53–55]. In these experiments, guanosine did not prevent cell death (measured by TUNEL) (Fig. 4), indicating that the protective effects of guanosine do not result from a reduction in precursor cell apoptosis. Fig. 4Guanosine treatment did not affect apoptosis in the cords with chronic, stable spinal cord injury. For detection of apoptotic cells, a terminal deoxynucleotidyltransferase (TdT)-mediated dUTP nick end labelling (TUNEL) stain was performed using the ‘In situ Cell Death Detection Kit, Fluorescein’. No obvious difference in number of TUNEL-positive cells was observed between vehicle- (a) or guanosine- (b) treated spinal cord sections. Scale bar = 50 μm for all Before the intraperitoneal injection of guanosine, there were detectable amounts of guanosine and its metabolic product guanine in the spinal cords of injured rats (time 0, Fig. 5a,b). There were no significant differences in guanosine or guanine in the samples of the spinal cords above, at or below the lesion. After guanosine injection, the amount of both guanosine and guanine increased, reaching a maximum by 30 min; the amounts of guanosine and guanine were about twofold and fivefold higher than those detected before the injection (Fig. 5a,b). As with the basal levels (time 0), there was no difference in the amounts of guanosine above, at or below the lesion. Interestingly, guanine levels were always higher than those of guanosine at each of the times tested after injection (Fig. 5a,b). Fig. 5Guanosine (a) and guanine (b) in spinal cord samples of rats with chronic spinal cord injury before and after intraperitoneal injection of guanosine. Guanosine (8 mg/kg) was administered intraperitoneally to rats 5 weeks after the spinal cord injury. Before or at the indicated time points after the injection, samples of spinal cord from 3–6 animals were taken from the site of the lesion and immediately above and below the lesion. The tissue content of guanosine and its metabolic product guanine was measured by HPLC and CE analysis. Data are the mean ± SEM New oligodendrocytes in injured spinal cords come from a population of endogenous progenitor cells present in the CNS that can differentiate into mature cells capable of myelinating axons [3, 12, 13, 56]. The data from the present study showed that the increase in the number of NG2+ cells is associated with an increase in Rip+ cells. Our data indicated that guanosine treatment induced an increase of mature oligodendroglia (Rip+ cells), likely due to proliferation and differentiation of their progenitor (NG2+) cells. Therefore, one would expect that Rip+ cells increase after NG2+ cells. However, this was not observed. But guanosine may well stimulate the proliferation of oligodendroglial progenitor cells at several stages of development. This would account for the observed results. Additionally, we cannot exclude that other possible mechanisms are involved in the process; for example, mature oligodendroglia could migrate from the adjacent normal tissue to the injury site soon after guanosine treatment. Whether the effects on remyelination are due to guanosine, guanine or both is not clear. However, since a putative receptor for guanosine has been identified in the CNS, and since PNP (purine nucleoside phosphorylase) in CNS permits the interconversion of guanine and guanosine enabling guanine to act as a reservoir for guanosine, it is possible that guanosine may be responsible for causing the proliferation of the endogenous progenitor cells and the subsequent remyelination process, although this is by no means certain. But, since guanine itself is difficult to administer because of its poor solubility, administration of guanosine is the most effective delivery mechanism. The mechanism by which remyelination is facilitated after chronic, traumatic SCI is unknown. Guanosine has been shown to stimulate the production and/or release of a variety of trophic factors [25]; for example bFGF, NGF and NT-3 have been demonstrated to increase the proliferation and survival of oligodendrocyte progenitors [3, 16, 57], induce oligodendroglial genesis [58–61], stimulate the proliferation of oligodendrocyte precursors and enhance remyelination after spinal cord traumatic injury [18, 19] and after toxin-induced demyelination [62]. There are no published reports of the effects of guanine in the vivo systems. However, in cultured cells, guanosine stimulates cell division apparently through a pertussis toxin-sensitive process involving MAP kinase [63]. Moreover, there is evidence of a G protein-coupled receptor in brain membranes that might be a candidate for a guanosine receptor [64]. Further studies will permit us to distinguish amongst these possibilities. The present findings are the first to show that systemic treatment with guanosine after chronic SCI induces an improvement in function that is accompanied by the formation of mature oligodendrocytes. This effect of guanosine/guanine may be attributable to direct or indirect stimulation of endogenous oligodendrocyte lineage progenitors in the spinal cord and remyelination of axons at the injury site. These findings raise the possibility that the administration of guanosine may be effective in treating spinal cord injuries and in the treatment of demyelinating diseases such as multiple sclerosis where quiescent progenitors exist in demyelinated plaques [65–70].

Naunyn_Schmiedebergs_Arch_Pharmacol-3-1-1915621

Pharmacological characterisation of capsaicin-induced relaxations in human and porcine isolated arteries

Capsaicin, a pungent constituent from red chilli peppers, activates sensory nerve fibres via transient receptor potential vanilloid receptors type 1 (TRPV1) to release neuropeptides like calcitonin gene-related peptide (CGRP) and substance P. Capsaicin-sensitive nerves are widely distributed in human and porcine vasculature. In this study, we examined the mechanism of capsaicin-induced relaxations, with special emphasis on the role of CGRP, using various pharmacological tools. Segments of human and porcine proximal and distal coronary arteries, as well as cranial arteries, were mounted in organ baths. Concentration response curves to capsaicin were constructed in the absence or presence of the CGRP receptor antagonist olcegepant (BIBN4096BS, 1 μM), the neurokinin NK1 receptor antagonist L-733060 (0.5 μM), the voltage-sensitive calcium channel blocker ruthenium red (100 μM), the TRPV1 receptor antagonist capsazepine (5 μM), the nitric oxide synthetase inhibitor Nω-nitro-l-arginine methyl ester HCl (l-NAME; 100 μM), the gap junction blocker 18α-glycyrrhetinic acid (10 μM), as well as the RhoA kinase inhibitor Y-27632 (1 μM). Further, we also used the K+ channel inhibitors 4-aminopyridine (1 mM), charybdotoxin (0.5 μM) + apamin (0.1 μM) and iberiotoxin (0.5 μM) + apamin (0.1 μM). The role of the endothelium was assessed by endothelial denudation in distal coronary artery segments. In distal coronary artery segments, we also measured levels of cyclic adenosine monophosphate (cAMP) after exposure to capsaicin, and in human segments, we also assessed the amount of CGRP released in the organ bath fluid after exposure to capsaicin. Capsaicin evoked concentration-dependent relaxant responses in precontracted arteries, but none of the above-mentioned inhibitors did affect these relaxations. There was no increase in the cAMP levels after exposure to capsaicin, unlike after (exogenously administered) α-CGRP. Interestingly, there were significant increases in CGRP levels after exposure to vehicle (ethanol) as well as capsaicin, although this did not induce relaxant responses. In conclusion, the capsaicin-induced relaxations of the human and porcine distal coronary arteries are not mediated by CGRP, NK1, NO, vanilloid receptors, voltage-sensitive calcium channels, K+ channels or cAMP-mediated mechanisms. Therefore, these relaxant responses to capsaicin are likely to be attributed to a non-specific, CGRP-independent mechanism. Introduction Capsaicin, a pungent constituent of red pepper, is known to activate sensory C-fibres via transient receptor potential vanilloid receptors type 1 (TRPV1; Caterina et al. 1997; Gunthorpe et al. 2002; Szallasi 2006), which are nonselective cation channels. Activation of these channels increases influx of mono- and divalent cations, which leads to an increase in the intracellular Ca2+ concentrations. Consequently, an array of neuropeptides like calcitonin gene-related peptide (CGRP), substance P and neurokinin A are released (Franco-Cereceda et al. 1988; Mitchell et al. 1995). These neuropeptides play a role in the regulation of normal vascular smooth muscle tone and are also implicated in several pathological conditions like ischemic preconditioning (Chai et al. 2006), preeclampsia (Dong et al. 2005) and migraine (Arulmani et al. 2004b). In the pathophysiology of migraine, vasodilatation of cranial blood vessels, especially extracerebral intracranial blood vessels, which are richly innervated by nerves containing a number of peptides such as CGRP, seems pivotal (Arulmani et al. 2006). Vasodilatation of these intracranial arteries leads to activation of nociceptors, which stimulate the pain centres in the brain (Goadsby et al. 2002). Involvement of CGRP in the pathophysiology of migraine is further strengthened by the observation that olcegepant (BIBN4096BS, Doods et al. 2000), a CGRP receptor antagonist, is effective in the acute treatment of migraine attacks (Olesen et al. 2004). Capsaicin has been widely used in various in vivo models of migraine to induce cranial vasodilatation, which is attributed to endogenous release of neuropeptides, especially CGRP (Akerman et al. 2003; Arulmani et al. 2004a; Gupta et al. 2006a). The involvement of CGRP is further substantiated in all these in vivo studies by the fact that the capsaicin-induced vascular responses are all amenable to blockade with CGRP receptor antagonists. Repeated capsaicin challenges are reported to deplete a number of neurotransmitters, including CGRP in various animal models (Tang et al. 1997; Tang et al. 1999; Zhou et al. 2002). In addition to the in vivo models, also in vitro vascular models involving meningeal (Gupta et al. 2006b) and coronary (Edvinsson et al. 2002; Gupta et al. 2006c) arteries have been used in migraine research, but vasorelaxation in these models was induced by exogenously administered CGRP. Several major vascular beds, including the meningeal and coronary (Franco-Cereceda 1988), are richly innervated with capsaicin-sensitive sensory fibres containing CGRP (Edvinsson et al. 1987). Meningeal blood vessels have been shown to be involved in the plasma protein extravasation in rat dura mater (Delepine and Aubineau 1997; Seabrook et al. 1996), a model of neurogenic inflammation, one of the putative mechanisms in migraine pathophysiology. In this model, capsaicin pre-treatment has been shown to block plasma protein extravasation, again by depleting neuropeptides (Delepine and Aubineau 1997). Therefore, it is of interest to investigate the release of endogenous CGRP in various blood vessels and capsaicin seems most appropriate for such studies. Interestingly, very few studies have investigated capsaicin-induced relaxations in human isolated blood vessels (Franco-Cereceda 1991a; Franco-Cereceda et al. 1987). Based on CGRP-like immunoreactivity, it was claimed that capsaicin-induced relaxations are mediated by CGRP (Franco-Cereceda 1991b), but no CGRP receptor antagonists were used to unequivocally demonstrate the role of CGRP in these responses. Therefore, we were interested in studying capsaicin-mediated relaxations in human and porcine isolated arteries as a model to study endogenous release of CGRP in view of its relevance in the pathophysiology of migraine. Remarkably, some recent studies indicate that capsaicin-induced relaxations are mediated by non-CGRP-mediated mechanisms in guinea pig ileum, rabbit coronary arteries and equine tracheal smooth muscle preparations (Fujimoto et al. 2006; Yeon et al. 2001; Zhu et al. 1997), which is contrary to the observation that capsaicin-induced relaxations are mediated principally by CGRP. In these studies, relaxant responses to capsaicin were mainly attributed to different Ca2+-activated K+ channels. Therefore, in the present study, we tried to characterise capsaicin-induced relaxant response in human and porcine vessels, with special emphasis on the role of CGRP, and also investigated other mechanisms, using different pharmacological tools. Materials and methods Tissue preparation The study protocol was approved by the ethical committee of Erasmus MC. Human coronary artery was obtained from ‘heart-beating’ organ donors (13 men, 19 women; 13–65 years) who died due to non-cardiac disorders. The hearts were provided by the Heart Valve Bank, Rotterdam, The Netherlands, after donor mediation by Bio Implant Services Foundation/Eurotransplant Foundation (Leiden, The Netherlands). The meningeal arteries were obtained from patients (four men, six women; 41–75 years) undergoing craniotomy at the neurosurgical unit (n = 6) or from the Department of Pathology (postmortem, n = 4) within 24 h of death, at Erasmus MC, Rotterdam. During the surgical procedure or autopsy, the dura mater together with a small piece of the meningeal artery was cut and placed in a plastic container filled with ice-cold (0–4°C), physiological salt solution. The artery segment was immediately transported to the laboratory and placed in cold oxygenated Krebs bicarbonate solution of the following composition (mM): NaCl 119, KCl 4.7, CaCl2 1.25, MgSO4 1.2, KH2PO4 1.2, NaHCO3 25 and glucose 11.1; pH 7.4. Excess tissue was removed. The meningeal artery (internal diameter of 350–800 μm) was used the same day or stored overnight in cold oxygenated modified Krebs solution and used the following day. Porcine hearts and heads (pigs of either sex; 6–12 months of age) were collected from a local slaughterhouse. Porcine basilar (internal diameter of 150–200 μm) as well as meningeal (internal diameter of 100–250 μm) arteries were dissected out from the skull and were placed in a cold oxygenated Krebs solution as described earlier for human meningeal arteries. In the laboratory, both human and porcine proximal (internal diameter, 2–3.5 mm) and distal (internal diameter, 200–600 μm) coronary arteries were dissected out of the right ventricle; distal coronary arteries were dissected with the aid of a microscope and stored in bicarbonate solution of following composition (mM): NaCl 118, KCl 4.7, CaCl2 2.5, MgSO4 1.2, KH2PO4 1.2, NaHCO3 25 and glucose 11.1; pH 7.4. Functional studies Both human and porcine proximal coronary artery segments (3–4 mm length) were suspended with the help of stainless steel hooks in 15-ml organ baths with a pretension of 15 mN (optimal tension shown in earlier experiments). Distal coronary and cranial artery segments were cut into ring segments of 1–2 mm length and were mounted in Mulvany myographs between two parallel titanium wires with a tension normalised to 90% of l100 (distance when transmural pressure equals 100 mmHg), thus achieving optimal conditions for active force development (Nyborg et al. 1987). Vessels were allowed to equilibrate for 30 min in Krebs solution at 37°C, a similar equilibration period was repeated after each physical or pharmacological challenge. In case of human arteries, the cyclo-oxygenase inhibitor indomethacin (0.1 μM) was added to prevent prostaglandin synthesis during the whole experimental protocol. Two successive challenges to KCl (30 mM) were performed to test the reproducibility of responses. Endothelial integrity was assessed by observing relaxation to substance P (10 nM) after precontraction with U46619 (9,11-dideoxy-11α, 9α-epoxymethano-prostaglandin F2α; 10 nM–1 μM), followed by washout of the agonists. KCl (18–30 mM, except where higher concentrations are indicated) or U46619 (10 nM–1 μM), in case of the K+ channel blockers, was used to obtain a stable contraction plateau of around 60–75% of the maximal contraction reached with KCl (100 mM). Subsequently, capsaicin was added in a cumulative manner in log steps. The different antagonists were used to antagonise capsaicin responses, and the concentration of each of these antagonists was decided on the basis of highest concentration in the literature unless mentioned otherwise. Concentration response curves were constructed in the absence or presence of the CGRP receptor antagonists olcegepant (1 μM; Gupta et al. 2006b,c) and CGRP8–37 (10 μM; Gupta et al. 2006b,c), the neurokinin NK1 receptor antagonist L-733060 (0.5 μM; Seabrook et al. 1996), the voltage-sensitive calcium channel blocker ruthenium red (100 μM, 45–90 mM KCl was required to get a similar precontraction as in the corresponding control segments; Franco-Cereceda and Rudehill 1989), the nitric oxide synthetase inhibitor Nω-nitro-l-arginine methyl ester HCl (l-NAME; 100 μM; Batenburg et al. 2004a,b), as well as the vanilloid receptor antagonist capsazepine (5 μM; Fujimoto et al. 2006; Gazzieri et al. 2006). At still higher concentrations of capsazepine (>5 μM) and L-733006 (>0.5 μM), even higher concentrations of preconstricting agents could not produce a workable precontraction. We also investigated the role of different K+ channels by using various inhibitors, 4-aminopyridine (1 mM; Fujimoto et al. 2006; Yeon et al. 2001), a voltage-dependent K+ channel (Kv) blocker and charybdotoxin (0.5 μM), a blocker of Ca2+-dependent K+ channels for large conductance (BKCa) and intermediate conductance (IKCa), in combination with apamin (0.1 μM), a blocker of small-conductance Ca2+-dependent K+ channels (SKCa; Batenburg et al. 2004b; Fujimoto et al. 2006; Zhu et al. 1997). The effect of the RhoA kinase inhibitor, Y-27632 (1 μM; Fujimoto et al. 2006) was investigated both in the absence or presence of 4-aminopyridine (1 mM). We also investigated the effect of the gap junction inhibitor 18-α-glycyrrhetinic acid (10 μM) on capsaicin-induced relaxations. Where indicated, the endothelium was removed with a human hair, and removal was confirmed by observing a relaxation of less than 10% of precontraction of U46619 after addition of substance P. Further, in porcine distal coronary artery, we also studied the effect of repeated administration of capsaicin (50 μM, four times) to verify the reproducibility of the responses in view of possible depletion of endogenous peptide pools or other agents. Measurements of cAMP Human proximal and distal, as well as porcine distal, coronary artery segments were incubated in a medium containing isobutylmethylxanthine (0.5 mM) for 30 min in the absence or presence of olcegepant (1 μM). The arterial segments were exposed to KCl (30 mM), challenged with ethanol (vehicle of capsaicin, final concentration in the baths 0.56%), capsaicin (10 μM) or h-αCGRP (100 nM, serving as a positive control) for 5 min and then snap frozen. The samples were stored at −80°C until cyclic adenosine monophosphate (cAMP) assay. To determine cAMP, tissues were homogenised in 0.5 ml 0.1 M HCl using a stainless steel ultraturrax (Polytron, Staufen, Germany). Homogenates were centrifuged at 3,300×g, and cAMP was measured in 300 μl supernatant using the ELISA kit according to the instructions of the manufacturer (R&D Systems Europe, Abingdom, UK). Measurements of CGRP in organ bath fluid Human distal coronary artery segments were subjected to a similar protocol as during the functional studies, while bath fluids were collected after construction of the concentration response curve. The bath fluids were collected from the segments treated with vehicle and capsaicin (100 μM), and Krebs solution was used as a control. Bath fluids were stored in tubes containing aprotinin (0.6 TIU/ml) and stored at −80°C. A competitive radioimmunoassay (Peninsula Lab INC., San Carlos, CA, USA) was used according to the instructions of the manufacturer to measure the CGRP concentrations in the bath fluid. Data presentation and statistical analysis The relaxant responses elicited by the agonists are expressed as percentage relaxation of the tone induced by 30 mM KCl or U46619 (in case of the K+ channel blockers). All data are presented as means±SEM, and n represents the number of blood vessel segments, all obtained from different subjects. The effect of all potential inhibitors of the relaxations to capsaicin was investigated in a paired parallel setup; that is, relaxations in segments with inhibitors were always compared to relaxations obtained in control segments from the same subject. The concentration response curves for all agonists were analysed using nonlinear regression analysis, and the potency of agonists was expressed as pEC50 using Graph Pad Prism 3.01 (Graph Pad Software, San Diego, CA, USA), setting the Emax of capsaicin in the presence of potential inhibitors to that in the respective control segment in case it was lower than the control Emax. The blocking potency of the antagonists was estimated by calculating median effective concentration (EC50) ratios, and apparent pKB values were calculated for the antagonists at each given concentration, with the slope set to unity. Statistical differences between concentration response curves to capsaicin in the absence and presence of potential inhibitors were determined using Student’s paired t-test with α set to 1.00–0.95(1/n) (Motulsky 2003) to correct for multiple comparisons. For the measurements of CGRP in the organ bath fluids, we could not exclude a non-Gaussian distribution due to the large degree of variability in the data. Therefore, the levels of CGRP in bath fluids were analysed by the non-parametric Kruskal–Wallis test, followed by Dunn’s post hoc multiple comparison test. Significance was assumed at P ≤ 0.05. Compounds Human α-CGRP and α-CGRP8–37 were obtained from Polypeptide, (Wolfenbüttel, Germany), olcegepant (BIBN4096BS, 1-piperidinecarboxamide, N-[2-[[5-amino-1-[[4-(4-pyridinyl)-1-piperazinyl]carbonyl] pentyl] amino]-1-[(3,5-dibromo-4-hydroxyphenyl) methyl]-2-oxoethyl]-4-(1,4-dihydro-2-oxo-3(2H)-quinazolinyl)-, [R-(R*,S*)]-) was a gift from Boehringer Ingelheim Pharma (Biberach/Riss, Germany); 4-aminopyridine was purchased from ICN Biomedicals (Aurora, OH, USA); L-733060 was purchased from Tocris (Bristol, UK); apamin, capsaicin, capsazepine, 3-isobutyl-1-methyl-xanthine, l-NAME, charybdotoxin, ruthenium red, substance P, U46619 and Y-27632 were purchased from Sigma-Aldrich (Zwijndrecht, The Netherlands), and KCl was obtained from Merck (Darmstadt, Germany). Capsaicin was dissolved in 70% ethanol, and the dilution series was also prepared in ethanol 70%. Capsazepine was dissolved in methanol; olcegepant was dissolved in a small amount of 1 N HCl and then diluted with distilled water. The other compounds were dissolved in distilled water, and all compounds were stored in aliquots at −80°C. Results Functional studies Human arteries Substance P relaxed artery segments precontracted with U46619 (10 nM–1 μM); responses were equi-efficacious in distal coronary (80 ± 5% of contraction to U46619, n = 28) and meningeal (75 ± 8%, n = 8) artery and significantly less in proximal coronary artery (27 ± 15%, n = 4). Both in the meningeal and distal coronary arteries, capsaicin induced concentration-dependent relaxations. In human proximal coronary artery, relaxant responses were only observed at the highest concentration of 100 μM, and the maximum relaxant response (34 ± 14% of contraction to 30 mM KCl) was significantly less than that observed in the distal arteries (94 ± 1% of contraction to 18–30 mM KCl). In human meningeal arteries, there was no difference in capsaicin-induced responses between arteries obtained perioperatively or postmortem; therefore, these data were pooled for further analysis. Capsaicin was equipotent and equi-efficacious in human distal coronary and human meningeal artery. In human distal coronary artery segments, the lower concentrations of capsaicin (0.1 nM–1 μM) in some cases induced contractions, but in all cases, we uniformly only measured the relaxant responses. The relaxations to capsaicin in proximal and distal coronary as well as meningeal arterial segments were insensitive to blockade by the CGRP antagonist olcegepant (1 μM; Fig. 1, Table 1). Fig. 1Effect of capsaicin or its vehicle in the absence or presence of various pharmacological agents or interventions in precontracted human and porcine distal coronary arteriesTable 1Effect of various antagonists/interventions on capsaicin-induced relaxations in human isolated artery segmentsAntagonist or other intervention (n)Emax (%)Δ EmaxpEC50Δ pEC50Human distal coronary artery (Control) (32)94 ± 15.27 ± 0.12Olcegepant (1 μM) (10)89 ± 48 ± 44.84 ± 0.090.33 ± 0.16CGRP8–37 (10 μM) (5)96 ± 33 ± 44.79 ± 0.060.16 ± 0.08Capsazepine (5 μM) (13)91 ± 31 ± 25.10 ± 0.130.07 ± 0.16Ruthenium red (0.1 mM) (9)92 ± 35 ± 35.01 ± 0.130.21 ± 0.22L-733060 (5 μM) (7)94 ± 22 ± 26.03 ± 0.78−0.54 ± 0.34Denuded endothelium (5)90 ± 69 ± 55.34 ± 0.48−0.21 ± 0.58l-NAME (0.1 mM) (7)94 ± 23 ± 25.23 ± 0.48−0.24 ± 0.4518-α-Glycyrrhetinic acid (10 μM) (3)96 ± 20 ± 25.08 ± 0.28−0.15 ± 0.38Olcegepant (1 μM) + L-733060 (5 μM) (3)90 ± 61 ± 36.04 ± 0.230.06 ± 0.21 (Control) (10)97 ± 15.91 ± 0.324-Aminopyridine (1 mM) (6)95 ± 22 ± 35.64 ± 0.380.94 ± 0.46Charybdotoxin (0.5 μM) + apamin (0.1 μM) (8)97 ± 10 ± 15.80 ± 0.370.23 ± 0.37Iberiotoxin (0.5 μM) + apamin (0.1 μM) (5)96 ± 20 ± 25.94 ± 0.650.15 ± 0.38Y-276323 (1 μM) (3)99 ± 1−1 ± 25.39 ± 0.21−0.58 ± 0.14Y-276323 (1 μM) + 4-Aminopyridine (1 mM) (3)100 ± 0−3 ± 45.12 ± 0.180.33 ± 0.28Human proximal coronary artery (Control) (4)34 ± 144.30 ± 0.14Olcegepant (1 μM) (4)36 ± 164.40 ± 0.17Human meningeal artery (Control) (10)91 ± 5 5.04 ± 0.09Olcegepant (1 μM) (10)96 ± 1−5 ± 45.03 ± 0.07−0.07 ± 0.08Capsazepine (5 μM) (4)81 ± 98 ± 74.90 ± 0.310.11 ± 0.31Ruthenium red (0.1 mM) (3)74 ± 155 ± 145.13 ± 0.420.02 ± 0.62L-733060 (5 μM) (4)79 ± 128 ± 154.80 ± 0.31−0.09 ± 0.31Emax is the maximum relaxant response, expressed as percentage of the respective precontraction; pEC50 is the −logEC50, where EC50 is the concentration of agonist required to produce half the maximal response. The arteries were precontracted with KCl (18–30 mM) except where KCl (45–90 mM; bold) or U46619 (ital). As the relaxant responses to capsaicin were small in human proximal coronary arteries and the availability of human meningeal arteries was very limited, further experiments were only carried out in human distal coronary artery. In this preparation, the CGRP receptor antagonist CGRP8–37 (10 μM), the TRPV1 receptor antagonist capsazepine (5 μM) and the NK1 receptor antagonist L-733006 (0.5 μM) also did not attenuate capsaicin-induced relaxations (Table 1). Similarly, there was no significant difference in relaxant responses in endothelium-intact or endothelium-denuded segments (Table 1). Also, in the absence or presence of the NO synthase inhibitor l-NAME (100 μM) or in the presence of the gap junction blocker 18-α-glycyrrhetinic acid, capsaicin caused equipotent relaxations compared to the respective control segments (Table 1). Various K+ channel blockers, namely, 4-aminopyridine (1 mM), charybdotoxin (0.5 μM) + apamin (0.1 μM) and iberiotoxin (0.1 μM) + apamin (0.1 μM) were also unable to block the relaxant responses to capsaicin. The RhoA kinase inhibitor Y-27632 (1 μM) alone or in combination with 4-aminopyridine was also unable to block the relaxant responses to capsaicin. Ruthenium red (100 μM), a nonselective blocker of Ca2+ transport through membrane channels, also did not significantly block these responses (Table 1). It is noteworthy that at higher concentrations, capsazepine and L-733060 significantly attenuated the responses to their respective preconstriction agents, and also in case of preincubation with Y-27632 and ruthenium red, 2- to 20-fold higher concentrations of precontracting agents were required to induce 60–75% of the maximal contraction reached with KCl (100 mM). At still higher concentrations of capsazepine (>5 μM) and L-733060 (>0.5 μM), even higher concentrations of preconstricting agents could not produce a workable precontraction. Additionally, the combination of olcegepant (1 μM) + L-733060 (0.5 μM) in human distal coronary artery or 4-aminopyridine (1 mM) + Y-27632 (1 μM) both in human and porcine distal coronary artery were also unable to block the responses to capsaicin. The vehicle of capsaicin (0.56% ethanol) did not induce any significant relaxations. It should be noted that the potency of capsaicin was higher on a precontraction with U46619 (10 nM–1 μM; pEC50 of capsaicin, 5.91 ± 0.32) than on a precontraction with KCl (18–30 mM; pEC50 of capsaicin, 5.27 ± 0.12; P = 0.029). However, irrespective of the precontracting agent used, the antagonists behaved in a similar fashion towards capsaicin. For example, the effects of 4-aminopyridine on capsaicin-induced relaxations were similar after precontraction with KCl (pEC50, 5.01 ± 0.05 and 5.12 ± 0.41; n = 7, in the absence and presence of 4-aminopyridine, respectively) and with U46619 (pEC50, 6.58 ± 0.47 and 5.64 ± 0.38; n = 6, in the absence and presence of 4-aminopyridine, respectively). In human meningeal artery, capsazepine, L-733006 and also ruthenium red did not block the responses to capsaicin (Table 1). Porcine arteries In porcine arteries, substance P induced relaxations in precontracted arteries to a varying degree in different vessels. The relaxations in distal coronary (64 ± 6% of contraction to U46619, 10 nM–1 μM, n = 26) basilar (44 ± 3%, n = 3) and meningeal (57 ± 15%, n = 3) artery were similar, whereas that in proximal coronary (8 ± 8%, n = 4) artery was significantly smaller. In porcine proximal and distal coronary arteries, capsaicin induced concentration-dependent relaxations. Unlike in human coronary arteries, the maximum response was not significantly different between porcine proximal and distal coronary arteries (Table 2). In both the arteries, olcegepant (1 μM) did not block the responses to capsaicin. Further experiments were carried out in distal coronary arteries and, similar as in the human distal coronary artery, capsazepine (1 μM), ruthenium red (0.1 mM), L-73360 (0.5 μM), 4-aminopyridine (1 mM), l-NAME (0.1 mM), charybdotoxin (0.5 μM) + apamin (0.1 μM), iberiotoxin (0.5 μM) + apamin (0.1 μM), Y-27632 (1 μM), 18-α-glycyrrhetinic acid (10 μM) and endothelium denudation did not affect capsaicin-induced relaxations (Fig. 1, Table 2). Additionally, Y-27632 (1 μM), alone or combined with 4-aminopyridine (1 mM), was also unable to block the responses to capsaicin. Similar as in the other arteries, the vehicle did not induce a relaxation (Fig. 1). Interestingly, in contrast to what we observed in the human distal coronary artery, precontracting the arteries with either KCl or U46619 did not change the relaxant responses to capsaicin. Four consecutive challenges to capsaicin (50 μM) did not significantly affect the magnitude of the responses (Fig. 2). Table 2Effect of various antagonists/interventions on capsaicin-induced relaxations in porcine isolated artery segmentsAntagonist or other intervention (n)Emax (%)Δ EmaxpEC50Δ pEC50Porcine distal coronary artery (Control) (56)96 ± 15.27 ± 0.09Olcegepant (1 μM) (10)92 ± 23 ± 25.26 ± 0.11−0.02 ± 0.11CGRP8–37 (10 μM) (5)97 ± 3−2 ± 54.78 ± 0.060.00 ± 0.03Capsazepine (5 μM) (7) 99 ± 1−3 ± 25.15 ± 0.260.26 ± 0.16Ruthenium red (0.1 mm) (12)88 ± 37 ± 44.89 ± 0.090.47 ± 0.23L-733060 (5 μM) (7)88 ± 34 ± 44.89 ± 0.09−0.04 ± 0.15Denuded endothelium (9)91 ± 42 ± 25.62 ± 0.490.21 ± 0.16l-NAME (0.1 mM) (6)90 ± 88 ± 84.84 ± 0.150.28 ± 0.31 (Control) (15)99 ± 05.08 ± 0.1518-α-Glycyrrhetinic acid (10 μM) (3)99 ± 10 ± 15.73 ± 0.75−0.59 ± 0.534-Aminopyridine (1 mM) (7)96 ± 23 ± 24.84 ± 0.15−0.46 ± 0.35Charybdotoxin (0.5 μM) + apamin (0.1 μM) (11)99 ± 10 ± 15.27 ± 0.15−0.04 ± 0.20Y-276323 (1 μM) (9)96 ± 23 ± 25.17 ± 0.190.04 ± 0.21Y-276323 (1 μM) + 4-Aminopyridine (1 mM) (8)96 ± 34 ± 35.22 ± 0.200.00 ± 0.11Porcine proximal coronary artery (Control) (4)100 ± 05.33 ± 0.42Olcegepant (1 μM) (4)90 ± 65.79 ± 0.16Porcine basilar artery (Control) (3)97 ± 14.70 ± 0.05Olcegepant (1 μM) (3)100 ± 04.97 ± 0.24Capsazepine (5 μM) (3) 100 ± 04.80 ± 0.01Porcine meningeal artery (Control) (3)99 ± 14.82 ± 0.02Olcegepant 1 μM (3)99 ± 14.88 ± 0.04Emax is the maximum relaxant response, expressed as percentage of the respective precontraction; pEC50 is the −logEC50, where EC50 is the concentration of agonist required to produce half the maximal response. The arteries were precontracted with KCl (18–30 mM) except where KCl (45–90 mM; bold) or U46619 (ital).Fig. 2Effect of four consecutive challenges to capsaicin (50 μM) in porcine distal coronary arteries precontracted with KCl (30 mM) There were no significant differences in efficacy or potency of capsaicin in porcine basilar (Emax, 97 ± 1%; pEC50, 4.70 ± 0.05; n = 3) and meningeal (Emax, 99 ± 1%; pEC50, 4.82 ± 0.02; n = 3) arteries as compared to porcine distal coronary arteries. In both these arteries, responses to capsaicin were not affected by olcegepant (1 μM). Measurements of cAMP Capsaicin (10 μM) did not affect cAMP levels in comparison to its vehicle or the control, which was only exposed to 30 mM KCl. In contrast, αCGRP (100 nM), which was used as a positive control, significantly increased cAMP levels in both human and porcine distal coronary arteries, which was blocked after incubation with olcegepant (1 μM; Fig. 3). Fig. 3Changes in cAMP levels in human (n = 4–7) and porcine (n = 4–9) distal coronary artery segments after exposure to various pharmacological agents. *Significantly different (P < 0.05) from KCl (30 mM)-treated segments Measurements of CGRP levels in organ bath fluid Capsaicin and its vehicle induced a significant CGRP release from human distal coronary artery segments into the organ bath fluid (Fig. 4) as compared to levels observed in Krebs solution. There was no significant difference between CGRP levels obtained after incubation with capsaicin or its vehicle. Fig. 4CGRP levels measured in bath fluids (Krebs buffer, control) after capsaicin or vehicle challenge in precontracted human distal coronary artery segments. *Significantly different (P < 0.05) from Krebs buffer Discussion In the present study, we investigated the role of CGRP in capsaicin-induced relaxations in human and porcine isolated arteries. In all arteries investigated, there does not seem to be any relevant role of CGRP in the relaxant responses to capsaicin. Further, the effects of capsaicin appear to be mediated by non-specific mechanisms. In all arteries investigated, capsaicin induced concentration-dependent relaxations, although these had a limited potency and efficacy in human proximal coronary artery. The responses to capsaicin were resistant to blockade with olcegepant in all tissues studied, which suggests no involvement of CGRP receptors. The potency of capsaicin is in line with what has earlier been observed in human coronary arteries (Franco-Cereceda 1991a) and guinea pig ileum (Fujimoto et al. 2006). As expected in view of its resistance to olcegepant, CGRP8–37 (10 μM) also did not block capsaicin-induced relaxations in human distal coronary artery segments. This is contrary to the observations by Franco-Cereceda and Rudehill (1989), where the authors claim the relaxations to capsaicin are mediated by CGRP on the basis of the observation that the sustained relaxations in the human coronary induced by capsaicin are similar to those induced by exogenous CGRP, unlike the transient relaxations to substance P, which are followed by rapid tachyphylaxis. In another publication, these authors claim involvement of CGRP in responses to capsaicin in the human isolated coronary artery on the basis of increased CGRP-like immunoreactivity after exposure of the artery segments to capsaicin (Franco-Cereceda 1991b). It should be noted that in the above study, the authors show an increase in CGRP-like immunoreactivity in large arteries, where no functional studies were performed. Further, the classical prerequisite for demonstrating the involvement of a particular pharmacological agent, by using the corresponding antagonist to block functional responses, was lacking in this study. Admittedly, the same group did demonstrate that CGRP8–37 inhibited the responses to capsaicin in porcine coronary arteries (Franco-Cereceda 1991b), although we could not block these responses even at 10-μM concentration. As CGRP-induced responses are mediated by increases in cAMP (Gupta et al. 2006c), we measured the levels of this second messenger after exposure to capsaicin and CGRP. We observed no increase in cAMP after addition of capsaicin, in contrast to the increased levels of cAMP after exposure of the vessel segments to (exogenous) α-CGRP. These increased levels were, as expected, blocked by the CGRP receptor antagonist olcegepant. Interestingly, after exposure to capsaicin, we observed CGRP release in the organ bath fluid, where the human distal coronary arteries were mounted in accordance with earlier observations (Franco-Cereceda 1991a). However, a similar increase was observed after administration of vehicle, while the control was not studied in the study of Franco-Cereceda et al. (1991a). Although capsaicin is known to activate TRPV1, there are reports that ethanol, also via activation of TRPV1, induces the release of CGRP as well (Trevisani et al. 2002, 2004). Moreover, as obvious from Fig. 1, the vehicle did not induce any relaxations in the precontracted arteries, and hence, the released CGRP cannot account for relaxations induced by capsaicin. Interestingly, the concentration of CGRP detected in the bath fluid of about 3 pM should have been about 10,000 times higher in the vessel segments (∼0.5 mg tissue in 5-ml organ bath fluid). Thus, the concentration of CGRP in the vessel segments should have been in the nanomolar range, which is equal to or even higher than the pEC50 in human distal coronary artery under similar experimental conditions (Gupta et al. 2006c) and should, thus, have induced a detectable relaxation. Therefore, it is most likely that the radioimmunoassay displayed cross-reactivity to another ligand, not CGRP. Admittedly, our observation that ruthenium red, even at a very high concentration (100 μM), did not block the responses to capsaicin, is in contrast with the observations described by Franco-Cereceda (1991a), where ruthenium red completely blocked the responses to capsaicin. Additional evidence for the fact that CGRP and TRPV1 are not involved in relaxant responses to capsaicin is provided by the fact that capsazepine, a competitive antagonist of TRPV1 (Caterina et al. 1997), did not block responses to capsaicin in our study. In view of reports of involvement of various K+ channels in the relaxant responses to capsaicin in various smooth muscle preparations (Yeon et al. 2001; Zhu et al. 1997), we also investigated various K+ channel inhibitors in similar or even higher concentrations, but none of these blocked the responses to capsaicin. 4-Aminopyridine, a blocker of delayed rectifier K+ channels, which is reported to block the responses to capsaicin in rabbit coronary artery (Yeon et al. 2001) and guinea pig ileum (Fujimoto et al. 2006), was ineffective in human as well as porcine distal coronary artery. Similarly, charybdotoxin, a BKCa and IKCa blocker, which blocked capsaicin responses in equine tracheal smooth muscle (Zhu et al. 1997), did not block these responses at similar concentrations and even in combination with apamin, a SKCa blocker. We also used the RhoA kinase inhibitor Y-27632, but unlike in guinea pig ileum, (Fujimoto et al. 2006) it also did not antagonise responses to capsaicin in the isolated preparations used in the present study. The capsaicin-mediated responses appear to be endothelium-independent, as denudation of the endothelium did not significantly change the capsaicin-induced responses. Further, the vehicle of capsaicin was without effect and, hence, cannot account for the relaxant responses to capsaicin. Moreover, repeated administration of capsaicin in porcine coronary artery did not significantly decrease the responses, suggesting that stored neuropeptides are not responsible for the relaxations, as these would most likely be depleted after repeated challenges to capsaicin (Sams-Nielsen et al. 2001). Unlike in the present study, relaxations in precontracted guinea pig isolated pulmonary artery could not be repeated with capsaicin indicating depletion of neurotransmitters (Maggi et al. 1990). Although we did not include a positive control for the various K+ blockers in the current study, in the same set up at our laboratory, relaxant responses to l-S-nitrosocysteine were blocked by the combination of charybdotoxin (100 nM) and apamin (100 nM) in porcine distal arteries (Batenburg et al. 2004b). Further, the concentrations of the various inhibitors that we used in our study were equal to (Yeon et al. 2001; Zhu et al. 1997) or even higher (Yeon et al. 2001) than those employed by others. Taken together, our observations in human and porcine distal coronary artery suggest that capsaicin-induced responses are not mediated by CGRP, substance P or TRPV1 receptors and also do not involve various Ca2+-activated K+ channels. The relaxations to capsaicin are mediated by a cAMP independent pathway. The major component of capsaicin-induced relaxations, therefore, appears to be mediated by non-specific actions of capsaicin, rather than from the release of neuropeptides like CGRP.

Environ_Manage-3-1-2094719

Using Soil and Water Conservation Contests for Extension: Experiences from the Bolivian Mountain Valleys

Soil and water conservation (SWC) contests among farmer groups were organized in five rural villages in the Bolivian mountain valleys. The contests were aimed at quickly achieving widespread sustainable results. This article analyzes the effectiveness of these contests as an extension tool. Mixed results were obtained. In three villages, participation rates in the SWC activities introduced in the contests were still high even 2 years after project withdrawal. These were all villages where a solid foundation for sustainable development had been laid before the contests were held. Two years later, most families were still involved in maintenance of the SWC practices introduced in the contests, and many farmers had started to experiment with different soil management practices. However, replications of these SWC practices were not widespread, Conservation Leaders did not continue with their training activities, and the quality of maintenance of the practices was often not satisfactory. In order to become a more effective extension tool and achieve widespread impact, SWC contests must receive continued support by a catalyst agency. Moreover, other SWC contests should also be organized in which practices are not predefined. Given that SWC contests are a low-budget extension tool, local municipalities could become more actively involved. Introduction Agricultural development programs and extension services, both in developed and developing countries, struggle with frustratingly low adoption rates of soil and water conservation (SWC) practices. In response, Savenije and Huijsman (1991) called for a “making haste slowly” approach that focuses more emphasis on developing and implementing solutions together with rural people. Numerous experiences, principally from Non-Governmental Organizations (NGOs), have proven that these more participatory (slow) approaches often work and that sustainable livelihoods can be established. Successes to date, however, consist of scattered small-scale projects; wide-scale impact (haste) has yet to be achieved. The major dilemma remains combining widespread and short term impact (Farrington 1998). In SWC extension in particular, participatory approaches are too limited in scope, while existing governmental extension services are often not participatory enough and not sustainable. Hence, strategies and accompanying methodologies and tools are needed that achieve sustainable results in a short time and on a widespread scale. Despite urgent calls to modify attitudes and develop strategies for the extension of SWC (Sombatpanit and others 1996), the agricultural extension service in Bolivia has completely disappeared, and a governmental strategy for tackling soil erosion does not exist. NGOs have partially filled this gap, and several small-scale successes were obtained with the participatory introduction of SWC practices. However, these successes have not been reproduced at a large enough scale to have an impact at national (or even departmental) level. Government officials are often not even aware of these experiences, and NGOs themselves generally do not have the capacity for wide-scale dissemination of technology options (Farrington 1998). As a consequence, land degradation during the past decades has increased (Kessler and Stroosnijder 2006), and rural poverty remains a serious threat to sustainable development in Bolivia. In response, a project executed by the Japan Green Resources Corporation (JGRC) developed and validated a strategy to motivate poor Bolivian mountain farmers to engage in adequate natural resources management, the logical strategy for SWC (Kessler 2007a). Unlike other small-scale participatory experiences, the logical strategy not only aims at slowness but also at making haste (i.e., in finding participatory methodologies and tools that achieve sustainable widespread success in SWC). Scaling-up and integrating SWC in a future extension system at a national level is the ultimate objective. One of the logical strategy’s most innovative tools is conducting SWC contests among organized groups of farmers. The objective of this article is to analyze the effectiveness of these SWC contests as an extension tool. This is principally done by monitoring and evaluating farmers’ participation in maintaining and replicating SWC practices after execution of the contests. The results will be used to discuss possibilities of using SWC contests in an extension strategy at a national level. Study Area The study was conducted in the five research villages of the JGRC project. They are situated in the north-Chuquisaca region of the inter-Andean valleys of Bolivia (Fig. 1), which is a semi-arid region located at an altitude of 2500 to 3100 m above sea level. The majority of the families in this region still manage a mixed farming system, with mainly subsistence agriculture and a flock of goats and sheep. Potatoes are still the main crop and are the only crop that receives small amounts of manure. Other important crops are maize, wheat, and barley. Animal traction is mostly used for land preparation. Chuquisaca is one of the least developed areas in Bolivia, with the second lowest Human Development Index (HDI) in Bolivia, equaling only 0.49 (United Nations Development Programme 1997). Similarly, a study by the JGRC project revealed that the Human Poverty Index (HPI) for the research villages is among the highest in the world (on average 45%), with lack of access to drinking water and health facilities being the principal constraints. Hence, the region is very poor and faces enormous socioeconomic and physical constraints. Fig. 1Location of the study area in the inter-Andean valleys of Bolivia In three of the research villages, the “experimental villages” (Tomoroco, Kaynakas, and Sirichaca), the logical strategy was developed and tried out over a period of 4 years. The other two villages were “validation villages” (Talahuanca and Patallajta); in these villages the strategy was validated for 2 years. The villages were carefully selected, in order to obtain a representative sample of villages for the north-Chuquisaca region. Therefore, geographical, climatic, cultural, and socioeconomic characteristics are different among the villages. Rainfall varies from 350 mm/year (in Tomoroco) to 750 mm/year (in Kaynakas). Sirichaca has more flat lands for potato production, whereas the other villages have steeper and more erosion-prone slopes. SWC is a barely developed activity in this region. Since the colonial era, when farmers started using a combination of Spanish and traditional agricultural practices to maintain adequate production levels, practices have not changed much. As a consequence, once cultivable land began to be used more intensively, traditional conservation practices were no longer able to control erosion and maintain soil fertility. Most of the SWC practices currently found in the research villages are still based on traditional knowledge, but given their ineffectiveness in conserving soil and water, farmers have become increasingly sceptical about prospects for sustainable agriculture in their villages. Therefore, migration rates are high in most of the villages. Attempts of development agencies to promote improved SWC practices have not been successful, mainly because of failing intervention strategies and the lack of adequate extension tools. Research Methodology The SWC contests were conducted in the “experimental villages” in 2001, and in the “validation villages” in 2003. In each ‘‘experimental village’’ only one SWC contest, which focused on all physical SWC practices, was conducted; in the “validation villages” two SWC contests were conducted, focusing on two SWC practices each. During the SWC contests, the number of labor days invested per family was monitored for each SWC practice. Moreover, the rate of participation in the contests (i.e., percentage of families actively involved in SWC activities) was evaluated in each village. The most important data for this study were obtained in the ex-post evaluation in 2005, 2 years after the JGRC project’s withdrawal from the villages. The ex-post evaluation was conducted in all villages, except Sirichaca. In this village an evaluation was not considered necessary, due to the disappointing results and low participation rates during the contests. In the other villages, a random sample of 30 families was taken for the ex-post evaluation. Although the diversity of farmers within each village is high, and farmers who are progress-driven and economically better-off tend to invest more in SWC practices than others (Kessler 2006), this was not considered a variable. Hence, large families, as well as, for example, widows were included in the sample. Similarly, farmers from the higher economic stratum (with more land) as well as very poor small farmers were considered. In the ex-post evaluation, the percentage of families actively involved in both maintenance and replications of SWC practices was assessed by means of a field survey. If at least one type of SWC practice was properly maintained, the family was considered to be actively involved in maintenance. If at least one SWC practice was replicated on other fields after the project’s withdrawal, the family was considered to be actively involved in replications. Similarly, the active usage of other (nonphysical) SWC practices was evaluated during the ex-post evaluation. Finally, in order to obtain data concerning the popularity of the executed SWC practices, quality of maintenance of these practices was evaluated within the same sample of families. For each family an average score was given for the quality of maintenance of each type of SWC practice, ranging from very bad (or abandoned) to very good. Based on these scores, and in order to be able to compare the villages, average scores for each village were calculated. The Need for More Effective Extension Tools Although Bolivia has tried out different extension service models (Bojanic 2001), these were always technology-centered and top-down, with weak research-extension linkages. Most widely used was the Training & Visit extension system, in which knowledge trickled down from the research institutes to the extension worker and finally to the farmers. Farmer participation in this model was mainly functional. In 1991 a World Bank project strengthened the country’s research capacity, but extension remained extremely weak, resulting in poor adoption rates of improved practices (Bojanic 2001). In contrast to other developing countries, where low adoption rates forced extension agencies to apply more people-centered extension approaches, the governmental extension service in Bolivia simply disappeared. The research and extension component remained an independent foundation only for the study of potatoes. Currently, a demand-driven research and extension system (SIBTA, the Bolivian Agricultural Technology System) is in place that promotes innovative technologies in support of productive chains. Given the system’s focus on promoting regional cash-crops, (poor) subsistence farmers are largely excluded. In Bolivia, as in many other Latin America countries, NGOs fulfilled the more intensive extension tasks. In contrast to the functional type of participation of public sector organizations, most NGOs aim at an empowering type of participation (Farrington 1998). Several NGO-initiated experiments with Farmer Participatory Research (FPR), were based on the farmer-first approach (Chambers and others 1989). Their pioneering work with a large set of participatory techniques and innovative extension approaches has led to the realization that community participation and integrated (multi-sector) approaches are essential elements for success. In Latin America, farmer-to-farmer extension has had a notable impact and has led to adequate natural resources management in Mexico (Ramos 1998), Honduras (Sherwood and Larrea 2001), and Nicaragua (Braun and Hocdé 2000), among other countries. Given that NGOs can spend considerable resources in a few villages, and often invest in time-demanding and costly face-to-face participation, they have achieved higher participation rates than governmental extension services. However, the principal constraint of these extension approaches is their limited scope: they are difficult to replicate on a wide scale in the absence of local support organizations (Farrington and others 2002). With the prime objective of agricultural extension being to reach all farmers, this is a crucial constraint. The JGRC project’s logical strategy for SWC is not the first attempt to develop effective strategies that combine farmer participation at the grassroots level with extension and scaling-up. Killough (2005), for example, proposes “participatory extension through [the] accompaniment model.” This extension approach is a middle road between the traditional extensionist-centered approaches and the more recent farmer-led approaches. In this model, professional extension workers help farmer-promoters conduct on-farm experiments and provide farmer-to-farmer training. Local empowerment (through the development of farmer leaders) and strengthening local institutions are primary goals of this holistic approach. It is therefore important that such extension approaches promote integrated human development. Evidence from Honduras shows that farmers would have likely abandoned SWC practices long ago, if the extension approach had not addressed essential aspects of the interaction between human development and agriculture (Sherwood and Larrea 2001). Farmers conceptualized this interaction and holistic approach as “the human farm.” Implementing such an integrated and farmer-based extension approach is urgently needed in Bolivia; only then can a wide-scale impact be achieved. Although municipalities have considerable budgets for rural development, they do not have the human capacity to provide this service, and they regard extension as a state responsibility. Hence, there is a need for (1) profound changes at institutional level to give priority to extension, and (2) effective extension tools that can be used in a farmer-based extension approach. SWC contests are an example of such a tool. Using SWC Contests Within a Farmer-Based Extension Approach Organizations in several countries have experimented with farmer contests or competitions. The State Farm Bureaus in the United States regularly hold such contests, in particular for young farmers. Management, growth and progress in farming operations are major factors in judging the contests. The outstanding farmers receive financial awards, which are usually made available by sponsors. Farmer competitions in New Zealand and Australia are more focused on developing technical skills and knowledge, with conservation and land management practices being one of the several competition modules. In developing countries all kinds of farmer contests are held, but rarely focused on SWC. Chuma and Murwira (1999) report of farmer competitions in Zimbabwe organized to stimulate the process of experimentation and revival of farmer knowledge regarding best farming practices. The best results with respect to SWC have been achieved in two rural development projects in Bolivia and Peru, where SWC contests among farmers were used to speed up execution of SWC practices. Both projects make use of the “learning from the best” principle (i.e., learning from the best families and villages) (Van Immerzeel and De Zutter 2005). Knowledge management (i.e., combining knowledge and capacities from different farmers to find sustainable and fast solutions) is essential in this approach. The contests are used as a catalyst tool to disseminate this knowledge and motivate farmers to learn from the best, to experiment and innovate, and to win prizes by improving on what they have learned (Van Immerzeel and De Zutter 2005). In Peru, the rural development project, MARENASS (Management of Natural Resources in the Southern Highlands), uses farmer contests to promote new technological practices for improving natural resources management, agricultural production and living conditions. An important characteristic of the project is the transfer of decision-making and responsibility for planning and financial resources to the villages. Each participating village receives financial support to hire direct technical assistance (Posthumus 2005). These external service providers can be farmers with much experience (“the best farmers”), consultants or technical staff members. When a contest is organized, the villages themselves select and contract these privatized services to provide training to a number of selected farmers. By means of farmer-to-farmer training, the trained farmers in turn teach the other villagers the new techniques they have learned. Contests are held both at village level, with farmers competing against each other, and on district level between villages (Posthumus 2005). Jury members are selected by the participants; the families, or villages that best apply the recommendations provided by technical staff earn a cash prize. The innovations of MARENASS have been successful, and the project is still in progress (International Fund for Agricultural Development 2005); in 360 villages about 60% of all households have been reached (De Zutter 2004). Nevertheless, since internal problems in some villages limit their participation in the project, the impact of activities on watershed level is quite low (Posthumus 2005). In Bolivia, the SID (Strategies for International Development) project, Pachamama Urupa, approaches the dual problems of soil erosion and rural poverty with the understanding that neither can be resolved without simultaneously addressing both. Competitions among villages are organized to encourage participation in natural resources management, and to recognize the most successful farmers (Strategies for International Development 2005). SID hires farmers who are skilled in one or more of the land conservation and reclamation practices as part-time staff. These para-professional extension agents train selected farmers in about five villages, who in turn train all the other participants of the contest in their respective villages (Borda 2002). The competitions are flexible enough to permit farmers to find their own solutions to their problems (i.e., to experiment and innovate). All farmers participate in the judging, which stimulates the sharing of knowledge and the adoption and improvement of the practices. The winning villages and families receive farm tools, seeds, and animals as prizes. Participation rates in the competitions are about 80% to 90%. The project has recently won the World Bank’s “Development Marketplace Competition” for innovative ideas in international development. The MARENASS and SID projects are quite similar; they aim at wide-scale adequate natural resources management through a farmer-based extension approach that builds on local knowledge and capacities. Massive participation in farmer contests and farmer-to-farmer training are crucial aspects of this approach. Moreover, progressive learning and improvement is stimulated through constant interchanges, the participatory judging process and the closing ceremonies. Two important differences between the projects are that:in Bolivia the practices being judged in a contest are specifically defined, while in Peru the contests have a more general character (e.g., soil conservation) and practices are not specified;in Bolivia money is not involved (and prizes are in the form of goods), while in Peru farmers are responsible for contracting the trainers and cash prizes can be won.Based on these experiences, SWC contests are undeniably a very promising tool for farmer-based extension strategies, and for achieving fast and widespread sustainable impact in natural resources management. In the following section we will focus on the SWC contests of the JGRC project. SWC Contests in the JGRC Project In the logical strategy of the JGRC project, SWC practices are executed within a framework of integrated rural development. They are always accompanied by – and often integrated in – other activities that aim at improving local livelihoods. An important feature of this strategy is the laying of a solid foundation for sustainable development before starting a SWC contest, with activities focusing, for example, on better village organization, responsible participation and effective collaboration (Kessler 2007b). The project’s strategy stresses the human dimension of sustainable development: genuine participation of stakeholders is essential (Kessler 2007a). The objective of the SWC contests is to train farmers in basic SWC practices. Moreover, the contests encourage farmers to experiment and to decide which practices best fit their specific conditions. Reflection and dialog – two key features of Participatory Research & Extension (Percy 2005) – are constantly used. Participatory research with a selected group of farmers, as well as farmer-to-farmer training and knowledge transfer, are crucial before, during and after the SWC contests. Lessons learned through the farmer-to-farmer movement that began in the Guatemalan highlands (Bunch 1982) are, therefore, taken into account (e.g., to use small-scale experimentation) to start slowly and small, to achieve early recognizable success and to limit the introduction of technology. The essential multiplier effect is provided by the SWC contests and farmer-to-farmer training. In the next section we explain the SWC contests in more detail. In all the activities the project’s extension worker plays a crucial role. At the end of the section we also present the differences between the approach of the aforementioned two projects and the JGRC project. Activities Preceding the SWC Contests Preceding the SWC contests (i.e., during the laying of a solid foundation for sustainable development in a village), SWC activities start with a group of about 10 Conservation Leaders (CLs). CLs are chosen by the assembly, taking into account personal characteristics such as responsibility, honesty and willingness to innovate. They receive intensive training from the project’s extension worker, aiming at the generation of a progress-driven attitude and at conducting experiments on their fields. Experimentation focuses both on physical SWC practices (stone lines, diversion ditches, bench terraces, etc.) and agronomical soil management practices (more efficient manure use, composting, green manure, etc.). CLs are also stimulated to establish some test-sites for comparing with and without cases, with the objective of obtaining more visible results and making on-site comparisons. This might convince visiting farmers that the positive effects are indeed a result of the practices, and not of different physical conditions between their farm and the CLs’ farms. Apart from experimentation and providing demonstrations on their own farms, CLs also have many less tangible tasks such as mobilizing the villagers to become involved in development activities. They are thus both promoters and technicians. Finally, training of CLs in techniques for knowledge transfer is given before and during the farm visits. Once a solid foundation is laid in the village and CLs are sufficiently trained, the group decides when the first SWC contest will be held and which practices will be executed. In the dry season, practices like stone lines and gully control measures are considered and in the wet season practices that require digging such as diversion ditches and bench terraces. The village is informed in the assembly and by distributing information leaflets. During a period of at least a few weeks, families have the opportunity to decide whether to participate in the contest. CLs have an active role in motivating their neighbors, and in starting to organize groups based on vicinity. Eventually, each CL should lead a group of five to eight families. During these weeks of group formation, possible conflicts must be resolved, especially between neighbors. Given that effective group collaboration is crucial for a successful SWC contest, group formation should be given the required time. Execution and Evaluation of the SWC Contests A contest generally deals with two SWC practices. In a later stage a certain contest can be repeated and/or more integrated contests can be considered, combining different practices. The first contest is the most important one, because it serves as a selection tool for distinguishing the interested families (those with a progress-driven attitude) from the others. In this first contest a subsidy is given – as an incentive to all participants – for the purchase of a set of tools that are essential to conduct SWC practices. Each family pays 20% of the original cost. Families that start participating in a later stage have no access to the subsidized tools, and should themselves catch up with the already executed practices. Each SWC contest takes about one or two months. Twice a year a contest can be held; one in the dry season and the other one in the wet season. The criteria for evaluating the contests are clearly indicated before starting: (1) executed quantity, (2) quality of the work, (3) knowledge, and (4) group collaboration. Major emphasis is given to training and learning during the contests. An essential technique for conducting most of the SWC practices is the adequate handling of the A-frame, which is used for establishing the contour lines in a field. This is generally taught during the first contest. Once the contest starts, each group – under the guidance of their CL – decides how the work will be done. Reciprocal group work (or “ayni” in Quechua) is mostly used. In this system, farmers work on each others’ farms on a rotation basis. It is especially useful for labor-intensive work, and hastens the pace of execution. The host family provides food; money is never used. Reciprocity in the Andes region is based on mutual trust. Torrico and others (1994) argue that it still contains many religious aspects, and that, therefore, quality of the work is never discussed. Similar to the “alayon”, a traditional form of village cooperation in the Philippines (Moneva and others 2000), the ayni serves as a venue for group learning, problem solving and the promotion of equitability among farmers. SWC practices are mainly conducted on fields situated near the farmer’s homestead; later they can be replicated on other fields. The project’s extension worker regularly monitors group work, and assists the CLs in their training. In the beginning, CLs often encounter problems and feel uncertain; regular meetings help to solve this. Once a contest has finished, each CL measures the quantity of practices executed by his or her group. The verification of this quantity, as well as the evaluation of quality and knowledge, is done by means of cross-visits with other CLs. Quality criteria are harmonized between CLs before starting the evaluation. Knowledge is evaluated by asking some practical questions. For these three criteria (quantity, quality and knowledge) scores of 1 (bad) to 3 (good) can be obtained. This is written down on evaluation sheets. Based on observations, the extension worker evaluates the fourth criterion: group collaboration and cohesion. All criteria are given the same weight-factor in the final calculation. During the final closing ceremony the groups receive a reward for their efforts. Recognition of efforts is important because it engenders a sense of pride and it increases self-confidence (Cinnéide and Conghaile 1990). The most recommendable rewards are seeds for green manure and vegetables, which contribute to more sustainable agriculture. Additionally, the winning groups also receive, for instance, potato, maize, or barley seeds, all in small quantities. These prizes are useful products and their value is small enough to avoid participating in the SWC contests for the wrong reason. In this respect, money or food must never be used as prizes. Activities Succeeding the SWC Contests Maintenance of the newly constructed practices is the first priority, and complementary vegetative and soil management practices are essential in order to achieve impact on soil productivity. Vegetative conservation practices (grasses, bushes, or trees) also strengthen most SWC practices and make them more sustainable. They require, however, controlled grazing and strict rules at village level that are respected by all villagers. Hence, only when such regulations are collectively agreed upon will vegetative practices be viable and will SWC practices work. Given the importance of vegetation, establishing tree nurseries in each village – preferably at family level – is part of the holistic approach of the logical strategy. By means of an Integrated Project (Kessler 2007a), a group of farmers interested in agroforestry is trained to become trainers in this topic. Through farmer-to-farmer training, technical knowledge regarding agroforestry practices can be spread to a large number of farmers. Similarly, other groups of farmers specialize in, for example, manure management (improved stables, manure storage and collection methods), composting or green manure practices. The spreading and replication of SWC practices to all the other fields that need to be conserved is the responsibility of each family. Although, ideally, groups that participated in the SWC contest will continue to work in ayni, most families will have to do it on their own. The role of the CLs is to provide support whenever it is requested. Differences with the MARENASS and SID Projects Although similar in many aspects, especially in considering SWC as being part of an integrated approach to sustainable rural development, the JGRC project conducted its SWC contests in a slightly different way than the MARENASS and SID projects. This concerned four crucial aspects:The emphasis on training in specific SWC practices during each contest instead of leaving more space for experimentation. The reason: the SWC contests are foremost an extension tool, i.e., they aim at providing farmers with basic practical knowledge of some simple SWC alternatives. After the contests farmers experiment with innovations and will adapt and improve the practices.The emphasis on laying a solid foundation for sustainable development before starting the SWC contests. The reason: only farmers with a progress-driven attitude will continue to experiment and innovate after the project’s withdrawal.The use of contests between groups instead of families or villages. The reason: to stimulate group formation and collaboration within a village, i.e., to maintain or re-establish the traditional work in ayni, and to contribute to better internal relations and knowledge exchange.The subsidizing of tools to stimulate participation in the contests, but no (or insignificant) prizes for winning groups. The reason: not having the tools is often a major limitation for participating, but once farmers participate, they must become convinced by the result of their work and not by the prizes they can win. Results and Discussion Table 1 shows that similar numbers of family labor days were invested during the SWC contests in all villages. Groups generally worked two aynis (or two complete labor days) on each group member’s fields. Stone lines were the most popular practice executed during the contests, with labor accounting for 40% to 70% of total labor days invested. In Tomoroco and Kaynakas – with more steeply sloping land – considerable investments were made in bench terraces, while in Sirichaca and Patallajta gully control works (especially the smaller ones) and earth bunds (due to the absence of stones) were given more attention. Table 1Average number of labour days invested per family for each SWC practice, during the SWC contestsSWC practicesExperimental villagesValidation villagesTomorocoKaynakasSirichacaTalahuancaPatallajtaStone lines 8.15.64.55.25.4Diversion ditches 0.80.81.22.82.8Gully control0.30.61.80.52.5Bench terraces 2.12.7–––Earth bunds –0.11.60.21.1Infiltration ditches 0.20.3–––Individual terraces–0.51.4––Total 11.510.710.58.911.8Source: M&E data in 2001 (experimental villages) and 2003 (validation villages) Table 2 shows that the percentage of families participating in the construction of practices during the SWC contests was lowest in Sirichaca (48%) and Patallajta (66%); these are both villages in which a solid foundation for sustainable development was never laid. In Tomoroco the participation rate was highest with 86%. These data show that in the villages where activities concerning organization, collaboration and environmental awareness raising had already been successfully executed, the SWC contests were able to mobilize more people. Table 2Percentage of families actively involved in SWC activitiesSWC activityExperimental villagesValidation villagesTomorocoKaynakasSirichacaTalahuancaPatallajtaConstruction of SWC practices during the SWC contests86754884 66Maintenance of SWC practices two years after project withdrawal9169–8453Replications of SWC practices two years after project withdrawal78252030Source: M&E data in 2001 and 2003 (during the contests) and ex-post evaluation data in 2005 However, initial motivation is easy; continued motivation is what really matters (Savenije and Huijsman 1991). The effectiveness of the SWC contests can only be properly measured by evaluating the continued use of SWC practices after the project’s withdrawal. Table 2 shows the results of the ex-post evaluation, in which the percentage of families that have maintained and replicated SWC practices was assessed. Concerning maintenance, in Tomoroco this percentage is highest and has even increased; presently, 91% of all families perform adequate maintenance of one or more SWC practices. Quality of maintenance was moderate to good in Tomoroco (Table 3); stone lines and bench terraces were especially well maintained. In Kaynakas 6% fewer families are currently involved in SWC activities; some families have not maintained their practices or have even removed stone lines and gully control measures. In some cases stones from stone lines were used for fruit tree terraces; such farmers experimented with and adapted practices. Farmers often refine their practices under environmental pressure (Veihe 2000). They may, for example, consider stones more effective for terrace building than for stone bunds. This is supported by Table 3: in Kaynakas the quality of maintenance is better for bench terraces than for stone lines. Bench terraces are popular and productive for vegetables. Table 3Quality of maintenance of SWC practices two years after project withdrawalSWC practicesExperimental villagesValidation villagesTomorocoKaynakasTalahuancaPatallajtaStone lines++/−+++/−Diversion ditches+/−+/−+-Gully control+/−-+/−+/−Bench terraces++n.a.n.a.Earth bundsn.a.n.a.––++ very good; + good; +/− moderate; - bad; – very bad (abandoned); n.a. not applicable (not executed)Source: Ex-post evaluation data in 2005 Abandonment of SWC practices is highest in Patallajta, where two consecutive severe drought years and lack of impact of SWC practices caused general disillusion among the villagers. Migration increased and fields were left unattended. Table 3 shows that quality of maintenance is bad to moderate in this village; earth bunds, which were heavily damaged during a high-intensity rain storm, were all abandoned. In neighboring Talahuanca, however, despite severe drought and similar damage to earth bunds, all participating families in the SWC contests are still actively involved in maintenance, except for earth bunds. The presence in Talahuanca of a solid foundation for sustainable development before starting the SWC contests explains the differences between both ‘validation’ villages. Maintenance of gully control measures was generally given little attention in all of the villages, although most people are of the opinion that these practices work very well. Concerning replications, a field survey in 2003 revealed that in Tomoroco and Kaynakas wide-scale replications were conducted during the two years in which the JGRC project was in effect. All participating families in the SWC contests constructed new SWC practices. This was mainly attributed to the inclusion of SWC activities in other activities of the JGRC project (namely Integrated Projects, see Kessler 2007a). Stone lines and diversion ditches were mostly replicated and covered large areas of the agricultural land in both villages. The average investment in these villages in maintenance and replications of SWC practices was estimated to be 20 labor days per family during these two years (2001 to 2003). However, more important from a sustainability viewpoint is what happened after the project’s withdrawal (starting in 2003). Table 2 shows that in Tomoroco 78% of all families constructed replications of SWC practices in the subsequent two years; in the other villages this was much lower (about 25%). Stone lines, especially, were replicated, given their relatively low labor requirements. The differences between the villages are explained by the fact that Tomoroco has more potential land for constructing stone lines, while in Talahuanca and Kaynakas the most important fields had already been protected. Moreover, on the steeper slopes of Kaynakas, stone lines sometimes disturb land preparation. In this village bench terraces were found more useful, but their replication requires higher investments. Replications of stone lines were found in only one village outside the project area, near Tomoroco.. Hence, the SWC contests have achieved mixed results in the five villages. They were effective in three villages: Tomoroco, Kaynakas and Talahuanca. The most positive outcome is that in these villages (on average) more than 80% of the families are currently involved in one way or another in SWC activities, and this was achieved without using incentive schemes or cash prizes. Most villagers also consider the executed SWC practices useful, and more than half of the farmers plan to replicate more measures in the near future. The three villages have in common that they all have solid foundations for sustainable development, which were laid before the SWC contests were conducted (in the first Phase of the logical strategy). This has triggered a renewed interest in alternatives to improve living conditions, including better soil management. The most negative outcomes are that in two of these villages (Kaynakas and Talahuanca) replication rates are currently very low, and in all villages CLs are no longer active as trainers. It seems that the dynamics of the process came to a halt after the project’s withdrawal. Some farmers cautiously experiment and replicate SWC practices, but most of them only maintain existing practices and wait for tangible results before investing in new ones. Similarly, the CLs find themselves in a vacuum; they are rarely asked for advice, and there is no common objective to keep the CLs’ groups active. The catalyst of the process, the project, is no longer there. The CLs were expected to fulfill this motivating role after the project’s withdrawal, but this has not happened. On the other hand, in Tomoroco and Kaynakas the landscape has visibly changed due to the installed SWC practices, and internal regulations concerning controlled grazing are being complied with (Kessler 2007a). According to the respondents in the ex-post evaluation, SWC contests work; 80% of respondents are positive about the actual impact of the contests. Most importantly, for many farmers the experiences acquired during the contests with alternative techniques and practices have served as a basis for experimentation. Under the marginal conditions of poor farmers, adapting innovations is more important than adopting innovations (Van de Fliert and Braun 2002). The message that alternatives are available to improve productivity has come through; people are interested and have started to experiment with new techniques. In this respect, it is interesting to observe the usage of other SWC practices that were not included in the contests, but are now practiced by interested farmer groups and CLs. Table 4 shows that several of these practices are currently in use by a considerable number of families, especially improved traditional practices like crop rotations (including leguminous crops) and mixed cropping systems. Hence, knowledge transfer from farmer to farmer does also occur spontaneously. If these SWC practices would have been included in the SWC contests, their adoption rate could have been much higher. Therefore, a first recommendation of this paper is to also conduct SWC contests in which practices are not previously defined. The contests described in this paper filled an important knowledge gap and provided farmers with basic information. However, the failure of earth bunds and the removal of other practices by several families prove that only executing contests with some predefined practices is not enough. Stimulating people’s creativity and having each individual family decide which management practices best fit their conditions is considered crucial in the MARENASS and SID projects (Van Immerzeel and De Zutter 2005). Local adaptations of existing practices will become available sooner if more space is left for farmers’ initiatives during the contests. Hence, the JGRC-type of contests can be maintained as a training tool, but other (more general) SWC contests should be organized as a follow-up activity. Table 4Percentage of families using other SWC practices two years after project withdrawalSWC practicesExperimental villagesValidation villagesObservationsTomorocoKaynakasTalahuancaPatallajtaManure use361650Improved traditional practiceGreen manure200100New practiceCrop rotations60283015Improved traditional practiceStrip cropping2412150New practiceAgroforestry28642545New practiceMixed cropping68487065Improved traditional practiceComposting4020–10New practiceSource: Ex-post evaluation data in 2005 A second recommendation is to strengthen the role of local organizations in providing follow-up support to the SWC contests. The major weakness of the JGRC project was its incapacity to institutionalize the process: SWC activities continued at a high rate when the project fulfilled its catalyst function, but drastically decreased after the project’s withdrawal. Of course, municipalities were always involved in the activities, but their genuine participation was not achieved. Longer-lasting institutional support in organizing and facilitating more SWC contests would have strengthened the CLs’ role, and would have kept the dynamic process going. To some extent, farmers can respond to land degradation without external support, but they need continued provision of technical assistance and information in order to make progress (Paudel and Thapa 2001). Farmers have often lost the self-confidence and capacity to adapt and innovate (Reijntjes and others 1998); without external support they will continue to farm in the way they have in the past (Percy 2005). Given that SWC contests are a low-budget extension tool, they can be easily organized by municipalities and local NGOs, for example. However, despite the involvement of local leaders during the SWC contests, the active involvement of many extension workers is needed, especially when larger areas are to be covered in order to achieve a wide-scale impact. This can be a major limitation for local institutions; policies at the macro-level that enable the implementation of a farmer-based extension approach are therefore required. Earlier in this paper we already mentioned that in Bolivia profound changes at institutional level are required that give priority to extension. Only when such changes are made can SWC contests become an effective extension tool “to make haste slowly.” In Bolivia, to date, governments have never committed themselves to extension; strategies were not clearly defined and the extension service constituted a large burden on the state budget (Bojanic 2001). Presently, many participatory approaches and tools are available (Chambers and others 1989) that have proven their effectiveness over the last decades. Based on case studies in Thailand and Laos, Connell (2000) concludes that there are several opportunities for the institutionalization of participatory approaches in mainstream extension, but that they all require significant political commitment. Particularly regarding environmental problems, different interest groups often pull in complementary and opposing directions (Röling and Pretty 1997). The challenge ahead for effective extension is to combine efforts. This has been done in Chile, where the government has contracted private technology companies to cater for the larger commercial farmers, and NGOs for small subsistence-oriented farmers. Rivera and Qamar (2003) for example propose a mixture of funding and service delivery modalities; governments could provide funding and NGOs could deliver the extension services. If the political willingness is present, this could also present an interesting opportunity in Bolivia. Conclusions The challenge we face in SWC is to quickly achieve widespread sustainable results, i.e., “to make haste slowly” (Savenije and Huijsman 1991). Based on the success of SWC contests elsewhere in the Andes (Van Immerzeel and De Zutter 2005), the JGRC project used this innovative tool in five rural Bolivian villages to put “making haste slowly” into practice. Mixed results were achieved with SWC contests between farmer groups. On the one hand, in villages where a solid foundation for sustainable development had already been laid, participation rates in the SWC contests were high. Most farmers were still involved in SWC activities even two years after project withdrawal, without receiving any incentive. In these villages large areas are currently protected with physical SWC practices and farmers have also started to experiment with other soil management practices. On the other hand, in the same villages the renewed system of collaboration focused on SWC was lost when the project withdrew, and Conservation Leaders did not continue with their training activities. Despite the visible widespread impact of the contests, sustainability is thus not yet assured. Farmers easily become disillusioned and unmotivated in the absence of a catalyst to keep the process of SWC contests and farmer-to-farmer training going. Moreover, tangible results such as higher soil productivity take a long time to appear. Farmers are opportunistic; in poor regions like the Bolivian mountain valleys, especially, they will grasp any opportunity to increase income. Two recommendations were given in this paper to make SWC contests more effective in an extension strategy. First, in addition to the contests described here, other SWC contests should be organized in which practices are not predefined; this will stimulate peoples’ creativity in developing adaptations of existing practices. Second, commitment is required from local institutions to support SWC contests as an extension tool; only then can Conservation Leaders continue their activities, and can a widespread impact be achieved. Given the responsibility of Bolivian municipalities for rural development, they must be the first to become actively involved in extension. However, steering and support with adequate strategies from departmental and state institutions is indispensable; this will motivate municipalities to take natural resources conservation and rural development more seriously. Without such support, any attempt to spread SWC practices via participatory extension methodologies – such as farmer contests – will likely fail; no matter how logical and well-designed the strategy may be.

Purinergic_Signal-3-3-2096642

Ecto-5′-nucleotidase and intestinal ion secretion by enteropathogenic Escherichia coli

Enteropathogenic Escherichia coli (EPEC) triggers a large release of adenosine triphosphate (ATP) from host intestinal cells and the extracellular ATP is broken down to adenosine diphosphate (ADP), AMP, and adenosine. Adenosine is a potent secretagogue in the small and large intestine. We suspected that ecto-5′-nucleotidase (CD73, an intestinal enzyme) was a critical enzyme involved in the conversion of AMP to adenosine and in the pathogenesis of EPEC diarrhea. We developed a nonradioactive method for measuring ecto-5′-nucleotidase in cultured T84 cell monolayers based on the detection of phosphate release from 5′-AMP. EPEC infection triggered a release of ecto-5′-nucleotidase from the cell surface into the supernatant medium. EPEC-induced 5′-nucleotidase release was not correlated with host cell death but instead with activation of phosphatidylinositol-specific phospholipase C (PI-PLC). Ecto-5′-nucleotidase was susceptible to inhibition by zinc acetate and by α,β-methylene-adenosine diphosphate (α,β-methylene-ADP). In the Ussing chamber, these inhibitors could reverse the chloride secretory responses triggered by 5′-AMP. In addition, α,β-methylene-ADP and zinc blocked the ability of 5′-AMP to stimulate EPEC growth under nutrient-limited conditions in vitro. Ecto-5′-nucleotidase appears to be the major enzyme responsible for generation of adenosine from adenine nucleotides in the T84 cell line, and inhibitors of ecto-5′-nucleotidase, such as α,β-methylene-ADP and zinc, might be useful for treatment of the watery diarrhea produced by EPEC infection. Introduction Ecto-5′-nucleotidase (CD73, EC 3.1.3.5) is a key extracellular enzyme which catalyzes the hydrolysis of extracellular nucleoside monophosphates to their corresponding nucleosides, with 5′-AMP being the preferred substrate. Ecto-5′-nucleotidase is highly expressed in endothelium, liver, and intestinal mucosa and at varying levels in other tissues including lymphocytes, kidney, and certain cancers. Ecto-5′-nucleotidase is tethered to the extracellular surface of mammalian cells by a glycosylphosphatidylinositol (GPI) lipid anchor. Most of the research on this enzyme has been in the cardiovascular system because of its role in the production of adenosine, which protects cells against ischemia by multiple mechanisms including vasodilation [1, 2]. Enteropathogenic Escherichia coli (EPEC) is a common cause of watery diarrhea in children in developing countries. While some aspects of EPEC infection, such as adherence, are well studied, the way that EPEC triggers watery diarrhea has been obscure, since EPEC produces no toxins. We recently proposed a theory that release of adenine nucleotides from host intestinal cells, followed by breakdown to adenosine, could trigger watery diarrhea by activation of adenosine receptors on intestinal cells [3]. In the gastrointestinal tract, ecto-5′-nucleotidase was found to be a key enzyme necessary for enterocyte responses to 5′-AMP, the neutrophil-derived secretagogue [4–6]. At that time the significance of ecto-5′-nucleotidase was felt to be in the setting of invasive, inflammatory pathogens because polymorphonuclear neutrophils (PMNs) release 5′-AMP during the process of chemotaxis into areas of infection or inflammation. Since that time, however, we have discovered that a noninvasive pathogen, EPEC, triggers a large release of ATP directly from host intestinal cells without any requirement for PMNs to be present [3]. Other diarrheal pathogens not considered classically invasive, such as Aeromonas hydrophila, also cause ATP release from enterocytes [7]. Extracellular ATP is broken down to ADP, AMP, and then adenosine, where it triggers a vigorous fluid secretory response in intestinal epithelium. We suspected that ecto-5′-nucleotidase also played a role in the diarrheal fluid secretion induced by EPEC and initiated this study to determine if it is the case, and, if so, whether inhibitors of ecto-5′-nucleotidase could block fluid secretion triggered by adenine nucleotides and EPEC infection. We used the T84 colon carcinoma cell line as a model because we have used these cells to study EPEC infection, they express ecto-5′-nucleotidase, and they can be studied in the Ussing chamber to measure chloride secretion. In order to facilitate our study we also developed a nonradioactive, non-high-pressure liquid chromatography (non-HPLC) method for measuring ecto-5′-nucleotidase in living T84 cell monolayers. Materials and methods Bacterial strains used E. coli strains used included laboratory E. coli strain HB101 (O: rough), commensal strain HS (O9: H4), and classic human EPEC strains E2348/69 (serotype O127: H6), B171-8 (O111: NM), and JCP88 (O119: B14) as described in several publications [8–12]. EPEC mutants included JPN15, an E2348 derivative which has lost the EPEC adherence factor (EAF) plasmid [13], UMD874, the espF mutant derived from E2348, which is deficient in host cell killing [3, 14], and SE1010, with a mutation in sepZ (also called espZ), which is defective in type III secretion [15]. Bacteria were added to yield a multiplicity of infection (MOI) of 100:1. Materials The following reagents were obtained from Sigma-Aldrich Chemicals: α,β-methylene-ADP, adenosine, adenosine 5′-monophosphate (AMP), tetramisole (also called levamisole), polymyxin B, neomycin, purified phosphatidylinositol-specific phospholipase C (PI-PLC, from Bacillus cereus), and zinc acetate. BIOMOL (Plymouth Meeting, PA, USA) was the source of the BIOMOL GREEN reagent used in the phosphate release assay for nucleotidase activity and of U73122, a PI-PLC inhibitor. U73122 is 1-(6-[17 beta-3-methoxyestra-1,3,5- (10) triene-17-yl] amino/hexyl) 1H-pyrroledione. A cell permeant PI-PLC activator, 3M3-FBS, was from the Calbiochem Division of EMD Biosciences (La Jolla, CA, USA). 3M3-FBS is 2,4,6-trimethyl-N-(m-3-trifluoromethylphenyl)benzenesulfonamide. Phosphate-free DMEM medium was purchased from MP Biomedicals (formerly ICN Biomedicals, Aurora, OH, USA). UNIFILTER plates were from Whatman (Clifton, NJ, USA). Bacterial culture E. coli strains were grown overnight in Luria-Bertani (LB) broth at 37°C with 300 rpm shaking, then subcultured for 2 h in serum-free DMEM/F12 medium supplemented with 18 mM NaHCO3, 25 mM hydroxyethylpiperazine ethanesulfonic acid (HEPES) buffer, pH 7.4, and 1% D-mannose as previously described [3]. For experiments with bacteria in minimal medium, bacteria were subcultured at a dilution of 1:2,000 into minimal medium (M9 salts plus casamino acids supplemented with 2 mM glucose). For convenience we used M9-CA liquid broth packets (E. coli Fast Media, MBI-Fermentas, Hanover, MD, USA) and added 2 mM glucose before use. Cell culture T84 colon cancer cells were grown in DMEM/F12 medium supplemented with 7.5% fetal bovine serum (Gibco/Invitrogen, Grand Island, NY, USA), 18 mM NaHCO3, 20 μg/ml vancomycin, and 15 μg/ml gentamicin as previously described [16]. Ussing chamber studies of secretion were performed on T84 cell monolayers grown in Snapwell inserts (Corning Costar, Corning, NY, USA). The Snapwell inserts, which had a 0.4 μm pore size, were coated with 32 μg collagen per well by applying 0.16 ml of 0.2 mg/ml type III collagen (Sigma; dissolved in warm 0.2 M acetic acid) to the Snapwell and allowing it to dry in the tissue culture hood under UV light. T84 cells were seeded onto the Snapwell inserts at ~1.2 × 106 cells per well and allowed to grow to confluency for 7–9 days. At this time the monolayers had transepithelial electrical resistances (TER) of 400–1,000 Ω · cm2. Assay for ecto-5′-nucleotidase by phosphate release An assay for ecto-5′-nucleotidase activity in living cells was developed based on the ability to detect inorganic phosphate (Pi) released from 5′-AMP. This method has been used to detect activity of protein phosphatases such as PTEN [17] and lipid phosphatases [18] and is based on sensitive detection of low levels of Pi using the BIOMOL GREEN reagent, an enhanced and stabilized formulation of malachite green. To carry out the assay a phosphate-free buffer was used consisting of (in mM): NaCl, 154; KCl, 2: MgCl2, 4; NaHCO3, 18: HEPES, pH 7.4, 25; and glucose, 10. This buffer is referred to as nucleotidase buffer. To measure ecto-5′-nucleotidase activity, the cell monolayer was rinsed once with sterile normal saline, then the medium was replaced with warm nucleotidase buffer. For cells in a 48-well plate, 0.25 ml of nucleotidase buffer were added per well, and the cells were allowed to rewarm to 37°C in the CO2 incubator. During pipetting the multiwell plate was kept warm using a metal heating block set at 37°, and a stopwatch was used to time the AMP addition and to terminate the assay. The procedure used for measuring monolayer activity was slightly different from that used to measure 5′-nucleotidase activity released into the supernatant, as described below. Cell-bound or monolayer activity To measure cell-bound ecto-5′-nucleotidase activity in cell monolayers, 5′-AMP was added to yield a final concentration of 0.2 mM to quadruplicate wells. Two other wells were left without addition of AMP (the “no AMP blank”). After a 10-min incubation at 37° an aliquot (usually 50 μl) was removed and quickly transferred to a well of a 96-well plate to terminate the reaction. E. coli-induced release of nucleotidase activity into the supernatant medium For nucleotidase release experiments, the cell monolayer was changed to warm, phosphate-free DMEM, then infected with an E. coli strain for 35 min to allow adherence, then the medium was changed to nucleotidase buffer and the infection was allowed to continue for 2 or 3 h. Note that in this procedure any nucleotidase activity that is released in the first 35 min is discarded and not detected by our method. However, this two-stage procedure with the medium change was necessary because EPEC bacteria did not adhere normally if they first encountered the host cell in nucleotidase buffer. After a period of infection, supernatant medium was collected with a multichannel pipettor and transferred to the wells of a Whatman UNIFILTER plate (a 96-well with 0.45-μm membrane for sterile filtration). Sterile filtrates were prepared by centrifugation with collection of the filtered medium into another 96-well plate placed beneath the UNIFILTER as previously described [19]. Once again, experimental conditions were usually done in groups of six, with two wells not receiving any AMP (no AMP blanks) and four wells receiving 0.2 mM AMP. Again, the usual assay condition was 10 min at 37° before the reaction was stopped by addition of 10 μl of 1 M HCl (“stop solution”). BIOMOL GREEN detection of phosphate released from AMP Stopped samples in a 96-well plate were brought to 100 or 110 μl volume with water if necessary, then treated with 100 μl of BIOMOL GREEN reagent. A standard curve of inorganic phosphate was prepared and run with every experiment; standards and unknown samples were incubated at room temperature for 20 min to allow a green color to develop, then the 96-well plate was read on a multiwell plate spectrophotometer at 620 nm. Unknown values were calculated from the standard curve using a hyperbolic curve fit using GraphPad Prism software, version 4.0. Results of monolayer activity were expressed as nmol of Pi produced/min per 106 cells. For experiments showing nucleotidase release, the results were often expressed as nmol Pi released/ min per well since the assay was done on a cell-free filtrate and because we often noted some detachment of cells during the longer incubations of 2–3 h needed to observe release.Although we believed we were developing a new method for assay of ecto-5′-nucleotidase by phosphate release, during the course of this work another group reported using a virtually identical method, also based on detection of phosphate released from 5′-AMP [1]. Detection of CD73 by Western immunoblot To prove that the released 5′-nucleotidase activity we measured was of host cell rather than bacterial origin, we performed immunoblots on the supernatants of infected T84 cells with antibodies against CD73. Initial attempts at immunoblotting using a commercially available monoclonal anti-CD73 antibody (Abnova Corp., Taipei, Taiwan) were unsuccessful. Dr. Linda F. Thompson, Oklahoma Medical Research Foundation, kindly sent us mouse monoclonal antibodies against human CD73 which had been generated by Dr. Wolf Gutensohn several years earlier. Of these, the two antibodies that gave the best results were designated CD73.4 and CD73.6 by Dr. Gutensohn; both were of isotype IgG2b and were used at a concentration of 1 μg/ml. After washings, the secondary antibody was goat anti-mouse IgG2b conjugated to peroxidase at a dilution of 1:3,000 (Roche Molecular Biochemicals, Indianapolis, IN, USA). Blots were developed by chemiluminescence as previously described [20]. Ussing chamber studies A Snapwell insert containing a monolayer of T84 cells was placed in the plexiglass “slider” and inserted into the Ussing chamber (Physiologic Instruments, San Diego, CA, USA) at 37°C and continuously short-circuited by a four electrode, automatic voltage-clamp apparatus which measured short-circuit current (Isc) and transepithelial resistance (TER); chamber fluid resistance was automatically subtracted. Transepithelial resistance was determined by passing 10-s 10-mV current pulses through the tissues. Short-circuit current was measured by passing sufficient current through the tissues via Ag/AgCl electrodes to reduce the spontaneous transepithelial potential to zero. The composition of the tissue bathing solution was (in mM): 140 Na+, 124 Cl−, 21 HCO3-, 5.4 K+, , 1.2 Mg2+, 1.2 Ca2+, and 10 glucose. Raw short-circuit current (Isc) values were converted to μA per cm2 by dividing by the area of the Snapwell monolayer (1.13 cm2). Other details of the Ussing chamber methods were exactly as described [16]. Protein assay Protein assay was by the Coomassie blue dye binding assay method of Bradford, using a Bio-Rad kit [21]. Expression of ecto-5′-nucleotidase RNA by reverse transcription and real-time polymerase chain reaction (PCR) T84 cells grown in 24-well plates were infected with EPEC for 35 min, then the medium was changed to remove unbound bacteria. Three hours after the medium change, ciprofloxacin was added to 25 μg/ml to kill EPEC and the incubation was continued 1 more hour. Old medium was removed, and the cell monolayer was lysed in extraction buffer with 10% β-mercaptoethanol (RNeasy Kits, Qiagen, Valencia, CA, USA). RNA was subjected to reverse transcription using Invitrogen Superscript III reverse transcriptase; 5 μl of purified RNA was used per 50 μl reaction, and gene-specific primers at 0.2 μM were used. Reverse transcription reaction was at 55° for 1 h. Copy DNA from reverse transcription was diluted 100-fold, then analyzed by quantitative real-time PCR using the same oligonucleotide primers. For ecto-5′-nucleotidase the primers used were 5′-TTC CAC CCT GAA GAA GGC CTT TGA-3′ (forward) and 5′-ATA ACT GGG CAC TCG ACA CTT GGT-3′ (reverse). As a normalizing gene we used glyceraldehyde phosphate dehydrogenase (GAPDH) as described by Khan et al. [22] except that we redesigned longer primers which were 5′-TCG ACA GTC AGC CGC ATC TTC TTT-3′ and 5′-ACC AAA TCC GTT GAC TCC GACC CTT-3′. PCR was carried out using a MyiQ Single-Color qRT-PCR machine from Bio-Rad (Hercules, CA, USA) using SYBR Green as the dye to monitor the amplification. Relative expression was calculated by the ΔΔCt (“Livak”) method as described [23], where Ct is the number of cycles to threshold. SYBR Green PCR reagents were from Bio-Rad and to reduce the cost, the PCR reaction volume was reduced to 25 μl. PCR was performed using a two-step protocol with an annealing temperature of 58.7° and denaturation at 95° for 30 s each (i.e, no extension step) for 35 cycles. Thermal melt curve analysis was performed at the end of the PCR amplification and showed a single sharp peak for the genes analyzed. Data analysis and presentation All error bars shown in graphs and error values reported in the text are standard deviations. Significance was tested by one-way analysis of variance (ANOVA) with the Tukey-Kramer post-test for multiple comparisons, using InStat software for the Macintosh from GraphPad software (San Diego, CA, USA). Graphs were prepared using Prism 4.0 software, also from GraphPad. Asterisks shown on graphs indicate a p value of < 0.05. Results Ecto-5′-nucleotidase activity was readily measurable in T84 cell monolayers using the phosphate release assay method developed in this study. Figure 1 shows the characteristics of the ectoenzyme in this system. Figure 1a shows a curve of activity vs substrate concentration, demonstrating that the enzyme shows Michaelis-Menten kinetics. In four experiments similar to that shown in Fig. 1a, the KM was 229 ± 30 μM and the Vmax was 2.7 ± 1.8 nmol/min per 106 cells (mean ± SD of 4 experiments). While the KM was very consistent over time and from experiment to experiment, the Vmax was more variable as we compared experiments separated by months or years. We noted that highly T84 passaged cells (passage number of 70 or greater) showed lower levels of ecto-5′-nucleotidase activity than less passaged cells from the same source (experiments not shown). The KM we observed in intact T84 cells was higher than that reported for the purified enzyme derived from a colon carcinoma cell line, BCS-TC2 cells [24]. Figure 1b shows the raw data of Pi generated over time in the presence and absence of added AMP substrate and demonstrates that the assay was linear for 10 min; therefore, 10 min was chosen as the standard assay duration. Fig. 1Characterization of ecto-5′-nucleotidase activity in T84 cell monolayers. 5′-Nucleotidase activity was measured by the release of inorganic phosphate (Pi) from AMP in phosphate-free buffer using a colorimetric method as described in “Materials and methods.” For a, c, and d, the spectrophotometric reading in the absence of AMP (“no AMP blank”) was subtracted from the total reading. In b the “no AMP blank” was not subtracted and is shown in b as labeled. a Michaelis-Menten curve of enzyme velocity vs concentration of AMP substrate. b Time course of Pi release vs time. c Lack of inhibition by tetramisole, an inhibitor of alkaline phosphatase. d pH dependence of 5′-nucleotidase activity. e, f HPLC traces showing conversion of AMP (e) to adenosine (Ado, f) in T84 cell monolayers. g Sensitivity of the AMP-hydrolyzing activity to the inhibitor α,β-methylene-ADP, confirming it as ecto-5′-nucleotidase Alkaline phosphatase is another abundant intestinal ectoenzyme which can hydrolyze 5′-AMP [25]. Alkaline phosphatase can be distinguished from 5′-nucleotidase by its alkaline pH optimum and sensitivity to inhibition by tetramisole. Figure 1c shows that the ectonucleotidase we measured on T84 cells was not inhibited by tetramisole (also known as levamisole). Furthermore, its activity did not increase with increasing pH (Fig. 1d) providing evidence that the activity being measured was not alkaline phosphatase. In addition, we used HPLC to confirm that 5′-AMP was being converted to adenosine in the supernatant medium of T84 cells (Fig. 1e and f). Morover, the 5′-nucleotidase activity we measured by phosphate release was sensitive to inhibition by α,β-methylene-ADP, a known inhibitor of ecto-5′-nucleotidase (Fig. 1g and see also Fig. 4). Lastly, after we initiated this study, other researchers also reported using phosphate release from 5′-AMP as their assay for ecto-5′-nucleotidase [1]. We have found that the increased throughput of the phosphate release assay increases its usefulness compared to the much slower HPLC methods for ecto-5′-nucleotidase. Figure 2 shows the effect of EPEC infection on ecto-5′-nucleotidase activity released into the supernatant medium during an experimental EPEC infection of cultured T84 cells. Wild-type EPEC strains E2348/69 and JCP88 triggered a release of 5′-nucleotidase into the supernatant medium while nonpathogenic E. coli strains HB101 and HS did not. The supernatant medium was subjected to sterile filtration prior to assay to remove bacterial cells as well as any detached host cells, so that the activity reflects a soluble and not a cell-bound form. In experiments where we determined both the supernatant release and the monolayer activity in the same experiment, supernatant release by wild-type EPEC was 30–40% of monolayer activity 3 h after the medium change (Fig. 2a and data not shown). To determine if the 5′-nucleotidase could possibly be derived from the E. coli bacteria rather than from the host we also assayed sterile filtrates of E. coli for 5′-nucleotidase activity and it was virtually absent from the culture filtrates of all E. coli strains tested, including EPEC (Fig. 2b, light gray bars). Even after treatment with a concentration of polymyxin B sufficient to cause > 99% bacterial cell lysis the amount of bacterial nucleotidase-like activity was far less than the amounts we observed in the supernatant in Fig. 2a. In addition, our protocol included a medium change step in order to remove unbound bacteria. Therefore we concluded that the 5′-nucleotidase activity being released in Fig. 2a was of host cell origin. Fig. 2Effect of E. coli infection on 5′-nucleotidase activity in the supernatant medium. a Nonpathogenic E. coli strains (HB101 and HS) or EPEC strains E2348/69 and JCP88 were subcultured in DMEM medium for 2 h, then used to infect T84 cells at a multiplicity of infection of 100:1 in phosphate-free DMEM. After 35 min to allow adherence, the medium was changed to nucleotidase buffer and aliquots were collected at various times after the medium change, filtered through a 0.45-μm filter to remove bacterial cells, then assayed for 5′-nucleotidase activity. Since the activity was measured in the cell-free sterile filtrates the activity is expressed as nmol/min per well. bE. coli suspensions were subjected to sterile filtration without treatment (light gray bars) or following treatment with 50 μg/ml polymyxin B, a lytic antibiotic (dark gray bars). Then the nucleotidase activity of a 100-μl aliquot was measured. The 100 μl volume was chosen because this was the volume of inoculum needed to achieve an MOI of 100:1 for the slowest growing strain (E2348/69) in a typical 48-well plate of T84 cells To determine if this hypothesis was correct and to try to determine the mechanism of EPEC-induced 5′-nucleotidase release, we compared the 5′-nucleotidase release triggered by wild-type EPEC strain E2348/69 with the plasmid-cured derivative, JPN15, and its espF mutant, UMD874 (Fig. 3a). The espF mutant is defective in host cell killing and in inducing damage to monolayer integrity [14, 26]. Interestingly, the espF mutant was not attenuated in its ability to trigger 5′-nucleotidase release, but instead it consistently outperformed the wild-type strain in this regard. In six experiments 5′-nucleotidase release induced by the espF mutant exceeded that of the wild-type E2348/69 by an average of 2.3 ± 0.14 fold (p = 0.03 by paired t-test), suggesting that host cell death or damage is not the mechanism of EPEC-induced nucleotidase release. Fig. 3a Effect of EPEC mutants and PI-PLC inhibitors and activators on induced 5′-nucleotidase release into supernatant. Release of 5′-nucleotidase activity into supernatant medium was measured as described in “Materials and methods” and in the legend to Fig. 2. Once again, activity in the cell-free sterile filtrates is expressed as nmol/min per well. However, for purposes of comparison 48-well plates contain ~0.25 × 106 T84 cells per well at confluency. b U73122, an inhibitor of PI-PLC, was used at a concentration of 2 μm and was re-added after the medium change; *significantly decreased compared to the EPEC strain alone, p < 0.05. cm-3M3-FBS, a cell-permeant sulfonamide activator of PI-PLC, was added at the concentrations and for the times indicated. d Inhibitory effect of neomycin on PI-PLC-induced 5′-nucleotidase release; this experiment was performed on cells grown in a 24-well plate, with ~0.8 × 106 cells per well; *significantly decreased compared to PI-PLC alone. e PI-PLC was again added to a final concentration of 0.1 U/ml for the times shown; *significantly different from the corresponding control. This figure is a composite of experiments that were separated in time and with cells of different passage number; therefore the absolute amount of activity varies among the figure parts. f Effect of EPEC infection on expression of RNA encoding ecto-5′-nucleotidase in T84 cells, by reverse transcription and real-time PCR. T84 cells were infected for 35 min, then the supernatant medium was changed to remove unattached bacteria, then the T84 cell monolayer harvested for RNA extraction 4 h after the medium change. Reverse transcription and PCR conditions were as described in “Materials and methods.” Expression of ecto-5′-nucleotidase was normalized to that of glyceraldehyde 3-phosphate dehydrogenase (GAPDH). f Normalized expression of a single representative experiment (mean ± SD of 6 PCR wells). g Immunoblot analysis of the proteins released into the supernatant after 2 h of infection with EPEC or an EPEC mutant (SE1010, sepZ/espZ), using monoclonal antibody against CD73. Lane 1, supernatant from uninfected, control T84 cells. Lanes 2 and 3, supernatant from cells infected with wild-type E2348/69. Lanes 4 and 5, supernatants from cells infected with the sepZ mutant. GPI-linked CD73 has an apparent molecular size of 72 kDa (not seen in this blot) and the GPI-cleaved soluble portion runs at 51 kDa (heavy band in lanes 2 and 3) Ecto-5′-nucleotidase is tethered to the extracellular surface of the cell by a glycosylphosphatidylinositol (GPI) lipid anchor [24, 27]. GPI lipid anchors are cleaved by phosphatidylinositol-specific phospholipases C (PI-PLC), and EPEC has been shown to activate at least one type of PI-PLC in the host cell, PI-PLC-γ [28]. Therefore we considered whether PI-PLC activation by EPEC could be a nonlethal mechanism by which EPEC triggered 5′-nucleotidase release from host intestinal cells. Figure 3b shows an experiment in which we tested whether a PI-PLC inhibitor could block EPEC-induced nucleotidase release. Indeed, 2 μM U73122 did reverse the nucleotidase release triggered by wild-type EPEC strains. Conversely, a newly described cell-permeable activator of PI-PLC, called m-3M3-FBS [29], triggered nucleotidase release from uninfected cells in a time- and concentration-dependent manner (Fig. 3c). Experiments to test if m-3M3-FBS would enhance EPEC-induced release were inconsistent, with some experiments showing a mild enhancement or additive effect and others showing no enhancement or mild antagonism (results not shown). In addition we also tested whether purified PI-PLC added to the apical surface of the T84 cell monolayer induced 5′-nucleotidase release and it did (Fig. 3d and e). Figure 3d also shows that 50 μg/ml neomycin could partially block 5′-nucleotidase release by purified PI-PLC. In addition to its antimicrobial effects, neomycin is an inhibitor of several types of PLC enzymes including PI-PLC. Figure 3e shows that after a 2-h treatment with PI-PLC not only was 5′-nucleotidase activity increased in the supernatant, but that monolayer activity was depleted as well (right portion of Fig. 3e). In experiments similar to Fig. 3e done with EPEC infection, we could not detect any decrease in monolayer activity 2–3 h after infection, despite the liberation of 5′-nucleotidase activity into the supernatant. We considered the possibility that increased host cell synthesis of ecto-5′-nucleotidase might be occurring in response to the cleavage of the enzyme from the cell surface. We analyzed the abundance of RNA encoding ecto-5′-nucleotidase by reverse transcription and real-time quantitative PCR in response to EPEC infection. Figure 3f and Table 1 show that infection with wild-type EPEC and the EPEC espF mutant (UMD874) increased the abundance of ecto-5′-nucleotidase RNA compared to uninfected control cells. The increase in expression of ecto-5′-nucleotidase RNA by wild-type EPEC was 33% above control. While this is not a large increase, it does explain the maintenance of normal levels of ecto-5′-nucleotidase on the surface of EPEC-infected cells, because under the most optimum conditions the release of enzyme activity into the supernatant is about 30–40% of monolayer activity (e.g., Fig. 3e). Table 1Effect of EPEC infection on expression of RNA encoding ecto-5′-nucleotidase in T84 cells, by reverse transcription and real-time PCRConditionNormalized expression ratioaUninfected control1.0, by definition+1 U/ml PI-PLC1.23 ± 0.26HS-infectedb0.66 ± 0.22E2348/69-infectedb1.33 ± 0.21cUMD874 (espF)-infectedb1.40 ± 0.19caCells were infected for 35 min, then the supernatant medium was changed to remove unattached bacteria, then the T84 cell monolayer harvested for RNA extraction 4 h after the medium change. Expression of ecto-5′-nucleotidase was normalized to that of glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Data shown are the means ± SD of 5 similar experimentsbBacteria were added with the intent of achieving a multiplicity of infection (MOI) of 100:1; actual MOIs in the 5 experiments ranged from 71 to 144cSignificantly increased compared to uninfected control by paired t-test (p = 0.04) The 5′-nucleotidase released into the supernatant appeared to be of host cell origin, because it could be triggered by chemical stimuli in the absence of EPEC infection (Fig. 3c–e). The host origin was also proven using a mouse monoclonal antibody against human CD73. Wild-type EPEC infection of T84 cells triggered the release into the supernatant of a 51 kDa protein recognized by a mAb against CD73 (Fig. 3g, lanes 2 and 3). In contrast, the EPEC sepZ/espZ mutant, which is defective in type III secretion, triggered the appearance of a fainter band which was of ~55 kDa in size. The origin of this faint band is unknown, but it is possible it is generated by proteolysis rather than PI-PLC activation. For example, E2348/69 expresses EspC, an autotransporter protease whose export is independent of type III secretion. Figure 4 shows that 5′-nucleotidase activity in living T84 cells was susceptible to the same inhibitors as the purified enzyme studied by others. Figure 4a shows that α,β-methylene-ADP inhibited ecto-5′-nucleotidase activity and that its inhibition has the characteristics of a competitive inhibitor. In the presence of α,β-methylene-ADP the enzyme shows an apparent increase in KM but no increase in the Vmax, and the inhibition can be overcome at high concentrations of AMP substrate. In contrast in Fig. 4b one observes that zinc acetate behaves as a noncompetitive inhibitor of the enzyme, with a decrease in the Vmax but no increase in the KM (in fact there is a small, paradoxical decrease in KM in the presence of zinc). Figures 4c and d show the inhibitory dose-response curves at a single 0.2 mM concentration of AMP substrate. For both inhibitors, when concentration is expressed on a logarithmic scale, the curves are well fit by a sigmoidal, single-site model, allowing determination of the inhibitory concentration 50% (IC50) as shown in Figs. 4c and d. In addition to α,β-methylene-ADP and zinc acetate, ecto-5′-nucleotidase activity was also inhibited by 3–10 mM sodium fluoride as reported for the purified enzyme (data not shown). 5′-Nucleotidase activity released into the supernatant was also sensitive to inhibition by α,β-methylene-ADP and zinc acetate with similar potencies (data not shown). Fig. 4Effect of the 5′-nucleotidase inhibitors on enzyme activity in T84 cell monolayers. a, b Michaelis-Menten curves of enzyme activity vs substrate concentration in the absence (open symbols) and presence (closed symbols) of the inhibitor shown. Enzyme parameters were calculated by nonlinear curve fitting using GraphPad Prism and the fitted parameters of Vmax and KM are shown below the graphs. c, d Dose-response curves for inhibition of ecto-5′-nucleotidase activity by α,β-methylene-ADP (c) and zinc acetate (d) using a 0.2 mM concentration of AMP substrate. In all parts of Fig. 4 the inhibitor was added before the AMP, and in c and d the curves were fitted to a sigmoidal inhibitory dose-response curve and the inhibitory concentration 50% (IC50) value derived from the curve fit is shown In Fig. 4 the inhibitor was always added to the cell monolayer prior to the AMP substrate. Using the Ussing chamber apparatus to measure chloride secretion, we realized we could monitor the effects of inhibition of 5′-nucleotidase activity dynamically and assess the ability of inhibitors to reverse a secretory response already underway. In preliminary experiments we found that to reverse an ongoing secretory stimulus we had to add inhibitors at concentrations considerably above their IC50 values. The requirement for addition of inhibitors well above the IC50 is due to the fact that adenosine is a very potent secretagogue in intestinal tissues and a very high degree of inhibition of ecto-5′-nucleotidase is needed to reduce the production of adenosine below the secretory threshold. Figure 5a shows that secretion triggered by 10 μM AMP is reversed by theophylline, an adenosine receptor antagonist. Figure 5a emphasizes the fact that 5′-AMP has no secretory ability unless it is converted into adenosine. Adenosine then acts via apical adenosine A2b receptors to trigger chloride secretion toward the apical side of the monolayer. Figure 5b shows that chloride secretion triggered by 1 μM AMP is partially antagonized by 2 mM zinc acetate even when the inhibitor is added after secretion has started. Addition of 1 μM adenosine restored the short-circuit current (Isc) to its previous trend, showing that zinc does not block adenosine receptors or “poison” the secretory machinery. Under the same experimental conditions, 200 μM α,β-methylene-ADP produced a much larger degree of reversal of Isc than did zinc (Fig. 5c). As with zinc, α,β-methylene-ADP did not prevent a subsequent, normal Isc response to addition of adenosine (Fig. 5c, right-hand portion), showing that it is acting as a 5′-nucleotidase inhibitor and not an adenosine antagonist. The efficacy of inhibition by α,β-methylene-ADP in this paradigm, however, was strongly dependent on the concentration of AMP used to trigger secretion. At higher concentrations of AMP (e.g., 20 μM), the degree of inhibition by α,β-methylene-ADP was markedly less (Fig. 5d) than when 1 μM AMP was used as the stimulus. In contrast, zinc acetate produced the same, modest inhibition of Isc regardless of the concentration of AMP used. Figure 5e summarizes the effectiveness of the two inhibitors, zinc and α,β-methylene-ADP, when tested against various different concentrations of AMP as the initial stimulus. Figure 5e shows that the effectiveness of α,β-methylene-ADP declines significantly with increasing concentrations of AMP used, whereas zinc’s inhibitory effects are fairly constant over a range of AMP concentrations. The results of Fig. 5 (b–e) vividly illustrate, at the level of of an electrophysiologic response, the difference between a competitive and a noncompetitive mode of action of an enzyme inhibitor (compare with Fig. 4a and b). Fig. 5Effect of 5′-nucleotidase inhibitors on the chloride secretory response to AMP in T84 cells in the Ussing chamber. Short-circuit current (Isc) was measured in T84 cells grown in Snapwell inserts as described in “Materials and methods”; in this cell line and electrode configuration, a positive Isc (upward deflection on the graphs) represents chloride secretion in the basolateral to apical direction. All of the additions of agonists and inhibitors in these parts of the figure was on the apical (mucosal) side of the monolayers only. a Reversal of AMP-induced secretory response by theophylline, an adenosine receptor antagonist. b Partial inhibition of AMP-induced secretion by 2 mM zinc acetate, with restoration of the tracing to its previous trend by adenosine, demonstrating that adenosine receptors are not blocked by zinc. c Substantial reversal of Isc by α,β-methylene-ADP when a low concentration of AMP (1 μM) is used as the stimulus. Note that adenosine fully restores secretion to its previous level. d Lesser degree of reversal of Isc by α,β-methylene-ADP when a higher concentration of AMP (20 μM) is used as stimulus. e Summary of many experiments in which fixed concentrations of zinc and α,β-methylene-ADP were tested for inhibitory efficacy against varying concentrations of AMP as test stimulus. In Fig. 5e the slope of the α,β-methylene-ADP curve was −1.73 ± 0.5 by linear regression (95% confidence limits of the slope value −3.27 to −0.2, significantly different from zero, p < 0.04). In contrast the slope of the zinc curve was −0.04 ± 0.23 and did not differ significantly from zero Since the discovery of EPEC-induced ATP release we have hypothesized that EPEC bacteria may gain a nutritional advantage by this inflicting kind of damage on the host cell [3]. The lumen of the gastrointestinal tract is a purine-limited environment, as demonstrated by in vivo gene expression screens such as in vivo expression technology (IVET) [30] and signature-tagged mutagenesis (STM) [31]. Nucleotides and nucleosides enhance growth of EPEC strains in vitro, and these growth-promoting effects are most obvious when adenosine is added to minimal medium (Fig. 6). Compared to minimal medium, addition of 30 μM adenosine markedly accelerated growth of EPEC strains such as E2348/69 (Fig. 6). Addition of 5′-AMP also increased EPEC growth, although not as much as adenosine. Alpha-β-methylene-ADP alone has no effect, stimulatory or inhibitory, on EPEC growth, but it did completely reverse the growth-promoting effects of 5′-AMP (Fig. 6a and b, right-hand bars). Alpha-β-methylene-ADP did not block the enhanced growth seen with adenosine (not shown). Similarly, 100 μM zinc acetate did not affect growth in minimal medium (Fig. 6c, left-hand bars) or the enhanced growth seen with adenosine (Fig. 6c, two middle bars), but zinc did abolish the enhancement of growth by AMP. E. coli 5′-nucleotidase is susceptible to inhibition by α,β-methylene-ADP and zinc; in fact, the crystal structure of the E. coli enzyme was solved with α,β-methylene-ADP bound in the catalytic pocket [32]. Bacterial utilization of extracellular nucleotides is usually by dephosphorylation followed by uptake of the nucleoside (in this case adenosine) and inorganic phosphate [33]. Although we began studying α,β-methylene-ADP and zinc as inhibitors of the host cell enzyme, CD73, a side effect (or side benefit) of these compounds may be inhibition of bacterial 5′-nucleotidase activity as well, an effect that is able to block the nutritional benefits that E. coli bacteria derive from extracellular AMP (Fig. 6). Fig. 6Effect of 5′-nucleotidase inhibitors on growth of EPEC strain E2348/69 in minimal medium supplemented with 30 μM adenosine or AMP. Minimal medium here refers to M9 salts, 2% casamino acids, and 2 mM glucose. Growth was at 37°C with 300 rpm shaking after a 1: 2,000 dilution from an overnight culture in LB medium. a, c Culture turbidity by spectrophotometric absorbance at 600 nm. b Bacterial protein content. a, b Growth-enhancing effects of AMP are blocked by α,β-methylene-ADP. c Growth-enhancing effects of AMP are blocked by zinc acetate; *significantly decreased compared to AMP alone (p < 0.05) Discussion We became interested in ecto-5′-nucleotidase (CD73) because of its likely role in the generation of adenosine from adenine nucleotides released during EPEC infection. This host enzyme has not received a great deal of scrutiny in the context of microbial pathogenesis, and when it was studied its role was felt to be in the setting of a polymorphonuclear neutrophil (PMN) response, since activated PMNs release 5′-AMP during migration into an infected area [4, 34]. EPEC behaves clinically as a noninvasive pathogen and does not trigger an influx of fecal leukocytes, but nevertheless triggers ATP release from the host. Therefore our results suggest that ecto-5′-nucleotidase plays a role in the response to a broader range of pathogens, including noninvasive enteric pathogens, than was previously thought. One important finding of our study is that EPEC infection triggers a release of ecto-5′-nucleotidase from a cell-bound to a soluble, free form (Figs. 2 and 3). The significance of this finding is not clear at present but there are several possibilities to be considered. First, studies by other investigators have shown that cleavage of ecto-5′-nucleotidase from its GPI lipid anchor increases its activity significantly via an increase in the Vmax [27]. This could result in an increase in the activity of 5′-nucleotidase present in an EPEC-infected portion of the intestine, which could be an advantage to EPEC (see Fig. 6 and additional discussion below). The observed increase in transcription of ecto-5′-nucleotidase in response to EPEC infection (Fig. 3f and Table 1), if accompanied by a similar increase in protein synthesis and surface expression of the enzyme, would replace ecto-5′-nucleotidase lost due to cleavage of the lipid anchor and also contribute to an increase in total activity. Third, release of nucleotidase from its lipid anchor would change its distribution from a mucosal location to a soluble form in the intestinal lumen, where it could possibly be more accessible to substrate. This could be important in the human or animal EPEC infection because expression of ecto-5′-nucleotidase is highest in the intestinal crypts [35], whereas EPEC adheres to villus tips as well as crypts. Last, after many days of infection the mucosa could be stripped of ecto-5′-nucleotidase activity, leading to malabsorption of nucleotides in the diet. Indeed, malabsorption of electrolytes has been demonstrated after 7 days of infection in a rabbit model of EPEC infection [36]. In addition to ecto-5′-nucleotidase, several other important digestive enzymes have GPI lipid anchors, including alkaline phosphatase and aminopeptidase N (CD13) and these also could be liberated by EPEC-induced activation of PI-PLC. Another GPI-linked protein of interest in the intestinal tract is decay-accelerating factor (CD55), the cellular receptor for adhesion by diffuse adherent E. coli [37, 38] and which is also involved in internalization of Salmonella into vacuoles. Several virulent pathogens have been found to secrete their own microbial ATPases or nucleotidases into the extracellular medium, including Entamoeba histolytica [39], Trichinella spiralis [40], and Vibrio cholerae [41]. Researchers have speculated that secreted microbial nucleotidases may serve in protection against killing by host immune cells, acquisition of nutrients from the host, or dampening of host immune responses. EPEC does not appear to secrete nucleotidase activity under any conditions we have encountered, but may achieve the same goal by triggering the release of 5′-nucleotidase from the host cell. Ecto-5′-nucleotidase acts on nucleoside monophosphates, with 5′-AMP as the preferred substrate, and is inactive toward ATP. Therefore, the generation of AMP from ATP requires another enzyme or enzymes. CD39, or ectonucleoside triphosphate diphosphohydrolase, is the major enzyme responsible for this conversion in the cardiovascular system and probably in the gastrointestinal tract as well [42, 43]. CD39 is also deserving of further study in the context of microbial pathogenesis. One surprising finding of our study was that the espF EPEC mutant was not attenuated but in fact better than the wild-type EPEC in release of ecto-5′-nucleotidase (Fig. 3a). Although this mutant is defective in host cell killing, its phenotype has been instructive in many ways. For example, the ability of the espF mutant to trigger a robust ATP release led to us the discovery of a new pathway for ATP release from the host cell; this “second pathway” is not dependent on cell death but is dependent on the cystic fibrosis transmembrane regulator (CFTR) for ATP efflux [19]. The competence of the espF mutant in release of ATP and of ecto-5′-nucleotidase from the host cell may help explain why this mutant is only modestly attenuated compared to wild-type in some animal models of EPEC infection, such as Citrobacter rodentium in mice [44]. On the other hand, the sepZ/espZ mutant is attenuated in its ability to trigger ecto-5′-nucleotidase release (Fig. 3f), showing that type III secretion is necessary for PI-PLC activation and release of the enzyme from its lipid anchor. The role of ecto-5′-nucleotidase in triggering EPEC-induced fluid secretion is highlighted in the Ussing chamber studies shown in Fig. 5. Even after AMP has induced a chloride secretory response, addition of the nucleotidase inhibitor α,β-methylene-ADP is able to reverse the short-circuit current in a manner similar to that of an adenosine receptor antagonist, theophylline (compare Fig. 5a and c). Tracings such as those in Fig. 5 seem to indicate that to sustain a secretory response there must be on ongoing production of adenosine from AMP. In other words, there seems to be a “sink” for depletion of adenosine from the extracellular medium, either by cellular reuptake or by adenosine deaminase. The rapid turnover of adenosine implied in these experiments suggested that inhibitors of ecto-5′-nucleotidase might be able to block EPEC-induced fluid secretion in animal models of infection. In discussions of zinc on ecto-5′-nucleotidase it is worthwhile to note that zinc supplements have been shown to reduce the duration and severity of watery diarrhea in children in developing countries around the world [45–48]. In most of these trials the microbial etiology of the diarrhea was not carefully investigated, so it is unknown if zinc is particularly efficacious in diarrhea due to certain pathogens, such as EPEC, or if it beneficial “across the board” for multiple agents. Likewise, the mechanism by which zinc exerts its beneficial effects has not been carefully studied in the field trials. In earlier reports it was assumed that zinc was acting by correcting a zinc deficiency, but this assumption has been called into question by the finding that zinc supplements have a beneficial effect on diarrhea even in children with normal serum zinc levels at the outset [49]. Kelleher et al. showed that zinc provided some added benefit beyond a probiotic (Lactobacillus rhamnosus) in rhesus monkeys infected with the human EPEC strain E2348/69; the monkeys studied were not zinc deficient [50]. Inhibition of ecto-5′-nucleotidase or K+ channels [51] by zinc could slow diarrhea but this should be considered a pharmacological effect of the mineral rather than repletion of a deficiency. Inhibition of 5′-nucleotidase also blocks the growth-enhancing effects of AMP on EPEC bacteria (Fig. 6); since this is a direct effect on bacterial cells this action of zinc would not depend on the nutritional status of the host. The effects of zinc on EPEC infection and on diarrheal diseases in non-zinc-deficient children deserve further study.

J_Mol_Biol-1-5-2291451

Plasticity Within the Obligatory Folding Nucleus of an Immunoglobulin-like Domain

A number of β-sandwich immunoglobulin-like domains have been shown to fold using a set of structurally equivalent residues that form a folding nucleus deep within the core of the protein. Formation of this nucleus is sufficient to establish the complex Greek key topology of the native state. These nucleating residues are highly conserved within the immunoglobulin superfamily, but are less well conserved in the fibronectin type III (fnIII) superfamily, where the requirement is simply to have four interacting hydrophobic residues. However, there are rare examples where this nucleation pattern is absent. In this study, we have investigated the folding of a novel member of the fnIII superfamily whose nucleus appears to lack one of the four buried hydrophobic residues. We show that the folding mechanism is unaltered, but the folding nucleus has moved within the hydrophobic core. Introduction Studies of structurally related proteins have clearly indicated that for many proteins the folding mechanism is determined primarily by the native state topology.1,2 This is evident from comparative Φ-value analyses of proteins that share similar folds but are very different in sequence. Such studies include representatives from all protein classes; all-α proteins,3–9 all-β proteins,10–22 and mixed α/β proteins.23–28 The folding mechanisms of these proteins range from purely hierarchical, where secondary structural elements form before any tertiary structure, to pure nucleation-condensation, where secondary and tertiary structure form concomitantly.29 A study of representative members of the homeodomain superfamily family has suggested that their folding mechanisms are dependent on inherent secondary structural propensity, and the authors propose that all folding mechanisms are in fact variations of the same theme:3 as the propensity for forming secondary structures decreases, the folding mechanism shifts from pure hierarchical to polarized transition states, and ultimately to the classical nucleation-condensation mechanism first shown for CI2.30 Separate investigations have suggested that a folding nucleus consists of obligatory and critical components: the specific interactions necessary to establish the correct topology form first, followed by a subset of surrounding residues that provide the critical stabilising interactions.31–33 As part of the “fold approach” we have studied the folding of a number of proteins with an Ig-like fold,16–18,34–36 all of which are composed of two anti-parallel β-sheets packed against each other. The deep hydrophobic core is always formed from the packing of the four central B, C, E and F β-strands,37 but the number and position of the edge strands varies between the superfamilies.38 Although proteins in different superfamilies share the same fold, they are apparently unrelated in sequence and are found in proteins with a wide variety of functions. The stabilities of the Ig-like proteins studied to date range from about 1 kcal mol− 1 to 9 kcal mol− 1, and the folding rate constants vary by six orders of magnitude. However, there is a correlation between folding rate and thermodynamic stability, which suggests that the interactions that are critical in stabilising the fold also govern the folding process.34 All members of the fold studied to date fold via a nucleation-condensation mechanism where the obligatory folding nucleus comprises a set of structurally equivalent buried hydrophobic residues in the B, C, E and F-strands that form a “ring” of interactions in the core (Figure 1). Early packing of the residues, which are distant in sequence, ensures formation of the correct native state topology.35,36 The critical nucleus surrounds this obligatory nucleus, but the degree of structure formation varies between different proteins. The residues that form the obligatory folding nucleus are highly conserved within immunoglobulin domains but are conserved only in terms of residue type in fibronectin type III (fnIII) domains. However, there are rare examples of proteins that appear to have a disparate nucleation pattern. Here, we have identified a fnIII domain in which one of the hydrophobic residues in the conserved folding nucleus has been replaced by a surface polar residue, and we ask how the folding mechanism has been affected. An extensive protein engineering Φ-value analysis reveals that the folding mechanism is unaltered, but that a spatially different set of core residues is used to form the obligatory folding nucleus, where interactions within each sheet establish the correct hydrogen bond registry between the core β-strands. Subsequent interactions between two such pairs are able to bring the β-sheets together and set up the complex Greek key topology. Results Residue conservation within the folding nucleus of fnIII domains A non-redundant multiple sequence alignment from Pfam was used to analyse the residue conservation in the putative folding nucleus of fnIII domains.39 These nucleation positions were identified through comparison with the third fnIII domain from human tenascin (TNfn3), which has been studied extensively in our laboratory. In most fnIII domains (73%), all four residues in the proposed folding nucleus positions are hydrophobic. Further analysis reveals that where there is a polar residue in one of these folding positions, it is almost invariably in the C or E strand, and these hydrophilic residues are usually arginine or lysine. These can act as hydrophobic residues, since the long aliphatic side-chains can traverse the core and allow the charged terminus to reside on the surface of the protein.40 Small hydrophilic residues, such as asparagine or aspartate, are found very rarely (only in ∼ 3% of all cases). Most fnIII domains (65%) have a single aromatic residue in the proposed folding nucleus. Analysis of the distribution of aromatic residues in the four obligatory folding nucleus positions shows clearly that aromatic amino acids are located preferably within the C–F sheet (Figure 2). Furthermore, the type of aromatic residue present is affected by the solvent accessibility of the β-strand: the C-strand position is partly solvent-accessible, and hence the majority of aromatic residues occurring within this strand are tyrosine (thereby allowing hydrogen bonding of the hydroxyl group with solvent molecules). In contrast, the F-strand position is deep within the core and phenylalanine is almost always the aromatic residue of choice. Approximately 20% of the sequences have more than one aromatic residue in the folding nucleus, and again these residues are almost exclusively located in the C–F sheet (86% of all such sequences). This asymmetry is likely caused by the presence of an adjacent conserved tryptophan in the B-strand that is essential for stability but not involved in the folding nucleus.16,18 Interestingly, about 15% of the fnIII domains appear to fold without any aromatic residues at the supposed folding positions, suggesting that a large side-chain is not crucial for the formation of the obligatory nucleus. Selection of CAfn2 as a candidate All known fnIII structures were surveyed to find a candidate protein that was missing a hydrophobic residue at one of the four putative folding positions. Only one candidate protein was identified, the second fnIII domain in chitin A1 from Bacillus circulans (CAfn2), which has a surface-exposed asparagine in the putative nucleus position in the C-strand. It has an aromatic residue in the F-strand folding position (Phe66).41 In this work we intended to compare the folding of CAfn2 with TNfn3, as this is the most extensively studied “typical” fnIII domain. The structure of CAfn2 was superimposed on the structure of TNfn3 to reveal a RMSD of only 1.7 Å over all structurally equivalent positions (69 residues), even though the two proteins have just 13% sequence identity (Supplementary Data Figure 1). The major differences are restricted to the loops and turns, which are different in length in the two proteins. CAfn2 has the same Greek key topology as all other fnIII domains, (Figure 1), and possesses both the highly conserved Trp residue in the B-strand and the conserved tyrosine corner motif (Supplementary Figure 1).42 A comparison of the structures of CAfn2 and TNfn3 reveals that the lengths of the highly conserved EF-loop and the AB-turn are identical in these proteins, whereas the rest of the loops show some variation. Most notably the C and C′-strands are shorter in CAfn2 and are joined by a short tight turn (Figures 1 and 3). The packing interactions are almost identical in the two proteins, with the majority of the contacts being made within the same sheet, and the interactions between sheets occurring in “layers”.18 To help visualize the interactions that occur within the proteins, all residue positions are described according to both the β-strand and the core layer in which they reside (Figure 3). Importantly, however, inspection of the supposed obligatory folding nucleus of CAfn2 clearly shows that the C-strand residue, N40, does not pack against the other putative nucleus residues (Figure 1). How, therefore, does this domain fold? Characterization of wild-type CAfn2 The equilibrium stabilities of wild-type CAfn2 and all its mutant proteins were determined through the use of standard denaturation curves fit to a two-state equation.43 From a number of repeated measurements of wild-type CAfn2 the free energy of unfolding (ΔGD–N) was estimated to be 6.7( ± 0.3) kcal mol− 1 at pH 5.0 and 25 °C. Kinetic studies reveal a single unfolding phase, but two refolding phases. Since the slower phase accounts for less than 15% of the total amplitude, and is apparently independent of the concentration of denaturant, it was attributed to proline isomerisation. Both arms of the chevron plot are linear (Figure 4), and the βT for wild-type CAfn2 is estimated to be 0.56. This is similar to that of TNfn3,44 indicating that the two transition state structures are of a similar compactness. Effect of mutations Using the Φ-value analysis of TNfn3 as a basis, a total of 23 non-disruptive mutations were made throughout the CAfn2 protein. Where possible, all mutations were conservative deletions.45 Contacts lost upon mutation are given in Table 1. The difference in free energy between wild-type and mutant (ΔΔGD–N) was calculated using an average m-value, <m>, of 1.15(± 0.02) kcal mol− 1 M− 1. It has been shown that m-values are hard to determine with accuracy (the range of m-values observed is typical for that observed in most large-scale protein engineering studies) so that use of an average m-value reduces the error in ΔΔG.46 Most mutations are destabilizing but ΔΔGD–N values range between − 1.5 kcal mol− 1 and + 6.0 kcal mol− 1. Chevron plots for all mutants are shown in Figure 4, with the kinetic data given in Table 2. Using these data, Φ-values were calculated from refolding data at 0 M denaturant. Note that for three highly destabilized proteins (L22A, Y36L and L58A) there were too few data points in the refolding arm to determine the gradient (mkf) accurately. In these cases an average mkf (1.06 M− 1) was used to fit the data. Note that this has no effect on the final Φ-value determined at 0 M denaturant. Only one mutant, V38A, has a folding m-value that is significantly different from this mean value. Several mutant proteins exhibit roll-over in the unfolding arm, and some mutants have increased mku values. Such behavior in unfolding has several possible explanations. It has been ascribed to “Hammond” behaviour, where there is a broad transition state barrier,47 or to population of a high-energy intermediate.48 We do not have sufficient data to distinguish these two possibilities and, since there is no roll-over in the wild-type protein, it is not possible to determine Φ-values for the “late” transition state using unfolding data. However, the model used has no effect on analysis of the “early” transition state, at 0 M denaturant.49 Structure of the transition state A number of positions in each β-strand of CAfn2 were probed using Φ-value analysis. Φ is a measure of the extent of structure at a given residue in the transition state (‡). A Φ-value of 1 indicates that the interactions are fully formed in ‡, whereas a Φ-value of 0 indicates that the structure is as unfolded in ‡ as in the denatured state. The precise interpretation of fractional Φ-values is ambiguous but is usually taken to mean that the residue is partly structured in ‡.50 However, it is generally accepted that, particularly when comparing homologous proteins, the best approach is to look at patterns of Φ-values rather than considering the absolute values of individual residues.2 The CAfn2 Φ-values range from 0 to 0.5, indicating that none of the positions analyzed is completely structured at the transition state (Table 2). In general, the Φ-values in the A and G-strands are close to 0, while those in the central B, C, C′, E and F β-strands are higher, and the Φ-values in these central strands are higher in the central layers of the core than at the extremes (Figure 3), as observed in TNfn3,18 the tenth fnIII domain of fibronectin (FNfn10)16 and the titin immunoglobulin domain TI I27.17 The Φ-values were classified into low (Φ ≤ 0.2), medium (0.2 < Φ < 0.4) and high (Φ ≥ 0.4) classes. These Φ-values are mapped onto the CAfn2 structure in Figure 5. A and G-strands All mutations in the A and G-strands gave low Φ-values, indicating very little structure formation in these strands in the transition state (Table 2; Figures 3 and 4). Both mutated sites in the A-strand, L10A(A3) and S12A(A2), pack onto the neighbouring B-strand, whilst inter-sheet interactions are formed with residues from the F and G-strands (Table 1). The ΔΔGD–N for the S12A(A2) mutation is too low for a reliable Φ-value to be determined. The three residues probed in the G-strand, S81(G4), V84(G3) and V86(G2), make interactions mainly with residues from the A and F-strands (Table 1). Strands B, C, E and F All core positions in CAfn2 were mutated, with the exception of W24 in the B-strand. The distribution of Φ-values reveals that the hydrophobic core is only partially formed in the transition state (Table 2; Figures 3 and 4). The highest values occur at positions V38(C4) and I55(E3), with moderate Φ-values at four other positions: I20(B2), L22(B3), V68(F4) and A70(F5). C′-strand The C′-strand is connected by two short loops to the central C and E-strands. Three of the four mutations within this strand, L44A(C′3), T46A(C′4), and V48A(C′5), have high Φ-values (Table 2; Figures 3 and 4). T46 and V48 interact mainly with the buried residues from the C and E-strands, whereas the L44 contacts are limited to residues within the C and C′-strands (Table 1). Discussion CAfn2 folds by a nucleation-condensation mechanism There has been much discussion of folding mechanisms in recent years. The two “extremes” are represented by the framework model, where local secondary structure forms before tertiary structure, and nucleation-condensation, where secondary structure and tertiary structure form concomitantly. Such extremes have distinct patterns of Φ-values. In the framework model, Φ-values will fall into two groups, one set close to 1 and the other close to 0. This has been termed a polarised transition state. In a nucleation-condensation mechanism, the transition state structure will be more diffuse, involving most of the protein, and Φ-values will all be generally between 1 and zero. Furthermore, in the nucleation condensation mechanism, the pattern of Φ-values is generally distinctive, with a subset of residues having slightly higher Φ-values, with Φ-values gradually becoming lower as structure condenses around the early “nucleus”. The pattern of Φ-values shows CAfn2 to have a diffuse nucleus, with two-thirds of the residues having Φ-values between 0.10 and 0.45. Moreover, these are arranged in the structure as one would predict from a nucleation-condensation pattern, with higher Φ-values at the centre of the core, becoming lower towards the edges of the molecule. This suggests strongly that CAfn2 folds, as do other Ig-like proteins, via a nucleation-condensation folding mechanism. However, again like other Ig-like proteins, there is a significant number of residues, in the peripheral A and G-strands, and in loops that have Φ-values close to 0, suggesting that in the final stage of folding these peripheral strands and loops pack onto the central region of the protein. Identification of the obligate (embryonic) folding nucleus Oliveberg and co-workers have suggested that residues that constitute the transition state for folding in a nucleation-condensation mechanism might be divided into two sets.31,32 The first set of residues make up the “embryonic” or obligate folding nucleus, defined as the set of primary contacts that are obliged to form to establish the topology of the protein. The second set is the residues that pack onto this embryonic nucleus forming the “critical contact layer”, providing sufficient interactions to drive the folding process downhill. Note that residues that form the embryonic nucleus have to form a network of contacts that establish the topology of the protein, but that residues in the critical contact layer may contribute significantly towards stabilising the transition state for folding. Identifying the most likely obligate folding nucleus from a pattern of Φ-values is non-trivial, especially for complex Greek key structures. This has been discussed in detail.18 In summary, it is not possible simply to “pick” residues with the highest Φ-values as being those that form the folding nucleus: one also has to consider the packing of the residues. In TNfn3, as is observed here for CAfn2, residues in the B-strand have generally low Φ-values compared to the Φ-values in the other central β-strands. This does not necessarily mean that the residues in the B-strand are less important for folding. Residue L22(B3) in CAfn2, for example, forms about half of its inter-strand contacts with residues in the A and G-strands, which have Φ-values of ∼ 0. Thus, the contacts with the C (V38), E (A53 and I55) and F (F66 and V68) strands must be more formed than the moderate Φ-value would indicate. Similarly, the high Φ-values in the C′-strand probably reflect the fact that the residues in this strand make the vast majority of their tertiary contacts with residues in the C and E-strands, which are themselves partially formed. In TNfn3 it was suggested that this C′-strand is “obliged” to fold when the adjoining C and E-strands pack together. Thus, we would now, following the nomenclature suggested by the Oliveberg model, assign residues in the C′-strand to the critical contact layer and not to the obligatory embryonic nucleus. For TNfn3, the residues with the highest Φ-value in the B, C, E and F-strands were initially chosen as putative nucleus residues. Examination of the structure showed that these residues are all found in the same core layer, and that they pack to form a “ring” of interactions in the core of the protein. It was suggested that this “obligatory” nucleus alone was sufficient to establish the topology of the native protein. This picture of the folding transition state was confirmed by subsequent restrained molecular dynamics simulations.36 A similar method has been used to identify the folding nucleus in a structurally related immunoglobulin domain.17,35 Using the same strategy, the Φ-value pattern of CAfn2 was investigated to identify the putative obligate folding nucleus, a set of residues with significant Φ-values that interact such that these interactions are sufficient to establish the topology of the protein. The layer that contains residues with consistently high Φ-values in the B and E-strands is layer 3 (I55 and L22), as in TNfn3. However, for the C and F-strands, the layer with the highest Φ-values is layer 4 (V38 and V68). Examination of the structure of CAfn2 shows that although residues L22 and I55, and V38 and V68 sit in different core layers, these four residues are still able to pack together in the centre of the core to form a ring of contacts (Figure 6). We suggest that the residues surrounding the obligate nucleus, residues in the C′-strand and more peripheral residues in the B, C E and F-strands, pack onto these obligate nucleus residues and, together, form the critical contact layer required to stabilise the transition state structure sufficiently to drive folding. There is a caveat we should make. We note that in CAfn2 it is more difficult to select which residues are likely to form part of the obligate nucleus than it was in TNfn3. Consider the B-strand. I20(B2) exhibits a Φ-value that is slightly lower, but is within error of L22(B3). However, I20 forms no contacts with the nucleating residues from the opposite sheet, (V38(C4) and V68(E4)), suggesting that it does not form part of the obligatory nucleus that establishes the topology of the molecule. Furthermore, as was the case in TNfn3, we were unable to determine a Φ-value for the highly conserved Trp in the position B4. Simulations confirmed for TNfn3 that this Trp residue had a low Φ-value (as we had inferred from the pattern of Φ-values surrounding the Trp residue). Trp 24 makes 150 side-chain–side-chain contacts in CAfn2, and 65% of these contacts are with residues that have Φ-values that are (or are predicted to be) low (Φ ∼ 0.15, 52 contacts) or zero (46 contacts). Less than one-third of the contacts made by Trp24 are with residues in the putative obligatory nucleus (with L22, V38 and V68) and no contact is made with I55. Thus, we tentatively propose that if Trp24 does have a role in the folding nucleus, it is more likely to be in the critical layer than in the topology-defining obligate nucleus. Also consider residue A70 in the F-strand in position F5. The Φ-value for this residue is very slightly higher than that for V68 in layer F4. However, Ala to Gly mutation must be considered to be non-conservative and, furthermore, Ala70 makes no contact within the proposed obligate nucleus i.e. it cannot have a role in establishing the Greek key topology. Comparison of the transition states of CAfn2 and TNfn3 The Φ-values for TNfn3 are generally higher than those in CAfn2, ranging from 0 to 0.6;18 however, the pattern of Φ-values is similar. For both proteins, the mutational results can be separated into two classes. The first group consists of residues in the central β-strands, which show significant formation of structure in the transition state. The second group consists of mutations probing the terminal A and G-strands, and residues from the extremities of the central strands. These two populations are clearly observed in a Brønsted plot (Figure 7). Nevertheless, there are important differences between the two domains, which are apparent in the pattern of Φ-values (Figure 8). In the C–F-sheet, the highest Φ-values in TNfn3 are found in core layer 3 (V70, 0.54; Y36, 0.53), with the Φ-values in layer 4 being significantly lower (L72, 0.29; L34, 0.35). However, in CAfn2 the Φ-values in layer 3 are very low (F66, 0.07; N40, 0.01), whereas the Φ-values in layer 4 (V68, 0.25; V38, 0.40) are significantly higher (Figures 3 and 8). This indicates that the absence of a buried hydrophobic residue in position C3 has forced the obligate folding nucleus of CAfn2 to “migrate down” one layer within the core (Figure 6). Perhaps unexpectedly, a corresponding “downwards migration” has not occurred in the B–E-sheet; (even if Trp24 was important, the Φ-value for A53(E4) is unambiguously low (0.14)). Such migration is not necessary; analysis of the CAfn2 structure shows clearly that residues L22(B3) and I55(E3) form significant interactions with V68(F4) and V38(C4) in the opposite sheet. Thus, these inter-sheet interactions would be sufficient to establish the Greek key topology. Further support for this migration hypothesis comes from Φ-values in the EF-loop. TNfn3 exhibits moderate Φ-values in this loop, (Y68(F2), 0.42; L62(E2), 0.33), which indicates that it is significantly structured in the transition state. It was argued that this loop is “obliged” to be structured in the transition state of TNfn3 to allow for formation of the adjacent folding nucleus: the more distant BC-loop exhibits lower values. However, in CAfn2 the folding nucleus has shifted away from the EF-loop (Figure 6) and consequently it is less restrained within the transition state (Φ-values for Y64(F2) and L58(E2) are 0.04 and 0.16, respectively). Both TNfn3 and CAfn2 display high Φ-values in the C′-strand. We suggest that these residues are not involved in the obligatory folding nucleus, but result from short CE-loops that force the C′-strand to pack as the nucleus forms;18 thus, these residues form part of the critical contact layer. In CD2d1, an immunoglobulin Ig variable domain, the nucleating C and E-strands are joined by a much longer loop comprising three β-strands, C′, Cʺ and D. In this case these strands do not pack until late in folding.18,19 In summary, for both proteins we observe the formation of a specific nucleus in the core of the protein involving formation of long-range tertiary contacts between a single residue from each of the B, C, E and F-strands. Formation of this “obligate” nucleus establishes the topology of the protein. Other residues pack around this obligate nucleus to form the critical contact layer until sufficient contacts have formed to surmount the free-energy barrier. This is typical of a nucleation-condensation folding mechanism. The peripheral strands and the loops pack late, mainly after the rate-limiting step for folding. Conclusion: plasticity within the obligatory folding nucleus in Ig-like domains Unlike other classes of proteins, such as the homeodomain proteins, all Ig-like proteins appear to fold by the same, nucleation condensation mechanism. The obligate nucleus is defined by the interactions that are necessary to establish the complex Greek key β-sheet topology of the native state. Previous biophysical studies of members of the Ig-like fold have shown that this folding nucleus always comprises a ring of interacting residues within the hydrophobic core: one residue from each of the B, C, E and F-strands. Whereas the obligatory nucleus in the immunoglobulin superfamily proteins is highly conserved and is based around the invariant tryptophan located within the C-strand, members of the fnIII superfamily show more variability. Instead of restricting a particular structural position to a specific amino acid, each position simply needs to possess a hydrophobic residue. Here, we have shown that the fnIII nucleus is more flexible still, and that when this pattern of residue conservation is lost upon mutation, fnIII proteins can “migrate” the folding nucleus, thereby revealing plasticity in the early stages of the folding process, while retaining the same folding mechanism. Such plasticity in the folding of Ig-like proteins has been observed previously; the Ig domain TI I27 has been shown to fold by alternative, parallel pathways.51 Although the wild-type protein folds only through one pathway under physiological conditions, extremes of temperature and denaturant, or mutations within its obligate folding nucleus also result in a switch of folding pathway. Lindberg and Oliveberg have suggested recently that a “malleable” protein folding energy landscape will allow proteins to retain efficient folding during the course of evolution, even though the finer details of the folding pathway are dependent on individual sequence.52 It is possible that the ability of these Ig-like domains to alter their folding pathway on mutation has contributed to the success of this fold, and has contributed to its abundance in the proteome (over 40,000 Ig-like domains are listed in the current Pfam database). Materials and Methods Protein expression and purification The fibronectin type III domain used in this work consists of 88 residues (SwissProt P20533, residues 559–646, PDB 1K85) of the Bacillus circulans chitinase A1. The synthetic gene was produced using overlapping primers and standard PCR techniques, and was inserted into a modified version of pRSETA vector (Invitrogen) containing an N-terminal His-tag followed by a thrombin cleavage site. Site-directed mutagenesis was performed using the QuikChange Kit (Stratagene). The identity of wild-type and mutants was confirmed by DNA sequencing. Protein expression was carried out in Escherichia coli C41 cells.53 Transformed cells were grown to an absorbance at 600 nm of 0.6 at 37 °C before induction with IPTG and growth overnight at 28 °C. The cells were harvested and lysed by sonication. The soluble fraction was bound to Ni2+-agarose resin, washed several times to remove weakly bound proteins, and eluted from the Ni2+-agarose resin in a high concentration of imidazole. After dialysis to remove the imidazole, the proteins were cleaved overnight with thrombin. Uncleaved protein and remaining His-tag were removed by using small amounts of Ni2+-resin before further purification by gel-filtration chromatography using a Pharmacia Biotech Superdex 75 column. When not used immediately, proteins were flash-frozen and stored at − 80 °C. Equilibrium measurements The stability of the CAfn2 wild-type and mutant proteins was determined by equilibrium urea denaturation in 50 mM sodium acetate buffer, pH 5.0 (15 mM HOAc, 35 mM NaOAc) at 25 °C. The solutions were left to equilibrate at 25 °C for at least 2 h before measurements were recorded. All experiments were carried out in thermostatted cuvettes at 25 °C. The experiments used an excitation wavelength of 280 nm, and an emission wavelength of 360 nm. Data were fit to an equation describing a two-state transition.43 Change of free energy on mutation The change of free energy on mutation, ΔΔGD–N, was determined using equation (1):54Where [urea]50% is the concentration of urea at which 50% of the protein is unfolded for wild-type (wt) and mutant (mut) proteins, and <m> is the mean m-value determined from all measurements on wild-type and mutant proteins. Kinetic measurements All kinetic experiments were done using an Applied Photophysics stopped-flow fluorimeter. The excitation wavelength was 280 nm and the emission was monitored at wavelengths > 320 nm. All experiments were carried out in 50 mM sodium acetate buffer (pH 5.0) at 25 °C. The final concentration of all proteins was 1 μM. Refolding rates at 0 M denaturant were determined using CAfn2 unfolded at pH 12.4 as described.18 Between three and five traces were averaged for each concentration of denaturant. The refolding data were fit to an equation using a single-exponential term. An average refolding m-value of 1.06 M− 1 was used for mutations L22A, Y36L and L58A. Fitting data to an equation with two exponentials did not improve the residuals. The unfolding data were fit to an equation describing a single-exponential process with curvature. Φ-Value analysis The Φ-value for folding was determined using equation (2):54.where ΔΔGD–‡ is the change in the difference in free energy between D and the transition state (‡) upon mutation and calculated from refolding data as follows:where kf and kf′ are refolding rate constants for wild-type and mutant proteins (at 0 M urea), respectively.

Eur_Spine_J-2-2-1602190

Thoracic cord compression caused by disk herniation in Scheuermann’s disease

We present the case of a 14-year-old male with Scheuermann’s disease and significant neurological deficit due to thoracic disk herniation at the apex of kyphosis. He was treated with an anterior decompression, anterior and posterior fusion in the same setting using plate, cage and a segmental instrumentation system. The patient had an excellent outcome with complete neurological recovery. Introduction Scheuermann’s disease (juvenile kyphosis dorsalis) is a structural kyphosis of the thoracic spine initially described by Scheuermann in 1921 [19]. It occurs commonly in adolescents (0.4–8.3% of the general population) and in most cases is characterized by minimal deformity and few clinical symptoms. Scheuermann’s disease is rarely associated with neurological complications [5, 20]. The purpose of the present paper is to present the case of a patient with Scheuermann’s disease and severe neurological deficit due to a thoracic disc hernia at the level of kyphos, to review the literature and compare with previously published similar cases. Case report A 14-year-old male was admitted with a severe spastic paraparesis. He had gradually developed a gait disorder and bilateral numbness of lower limbs in the past 5 months. During that period his symptoms deteriorated rapidly. He was unable to walk and a week before his admission he developed spastic paraplegia. There was no history of trauma. Clinical examination revealed spastic paraparesis below T9. The superficial sensation was diminished below the T9 dermatome. He had a muscle strength of grade II according to the Medical Research Council (MRC) scale. A barely detectable contraction could be seen below the knees by palpation over the muscles (M1 on the MRC scale). Plantar responses (Babinski sign) were present on both sides. Knee and ankle reflexes were brisk bilaterally, clonus was present in both ankles as well. Patient had hypalgesia and decreased temperature sensation. Light touch and pin prick sensation were diminished over both lower limbs and absent from the knees downward. Rectal tone and sensation while urinating were decreased. All laboratory findings were normal including an investigation for systemic diseases. Plain radiographs of the spine in standing position revealed anterior wedging of more than 5° of several adjacent vertebrae (T7:12°, T8:14°, T9:17°, T10:15°) at the apex of the kyphosis and vertebral endplate irregularities. The thoracic kyphosis from T5 to T12 measured 66° (Fig. 1). These findings are in agreement with the criteria for diagnosis of Scheuermann’s disease. A magnetic resonance imaging scan revealed thoracic disc heniation with spinal cord compression at T8–T9 level (apex of the kyphosis). The intervertebral spaces from T6 to T10 were also very narrow and the vertebral bodies had an anterior wedging (Figs. 2, 3). Fig. 1a, b Anteroposterior and lateral radiographs of the thoracic spine in standing position, at the time of admission. The thoracic kyphosis from T5 to T12 is 66°Fig. 2a, b Magnetic resonance imaging scan revealed thoracic disc herniation with spinal cord compression at T8–T9 level (apex of the kyphosis)Fig. 3Three-dimensional reconstruction computed tomography of the thoracic spine preoperatively Surgical treatment was decided based on the severity of the neurological deficit. A right seventh rib transthoracic approach to the spinal column was performed followed by decompression at T8–T9 level. A disc fragment was found and removed behind the thoracic body of T9. We performed an anterior fusion using plate, screws, an interbody titanium cage and also a posterior fusion from T6 to T12 using a double-rod multihook and transpedicular screws segmental instrumentation system. Bone grafts were applied between T8 and T9 and intertransversaly from T7 to T10 (Fig. 4).Fig. 4a, b Postoperative anteroposterior and lateral radiographs of the thoracic spine. Note the anterior and posterior spinal fusion Postoperatively the patient had a surprisingly rapid improvement of the neurological deficit. Four weeks after the operation he could walk steadily and 2 months later his neurological examination was entirely normal. At 2 years follow-up the patient had a normal gait, he was symptoms free and there was no increase of the spinal deformity. Discussion Scheuermann in 1921 [19] described a condition, which he called juvenile dorsal kyphosis, distinguishing it from the more common postural kyphosis. The etiology of Scheuermann’s disease is unknown. Scheuermann proposed that the kyphosis resulted from avascular necrosis of the ring apophysis of the vertebral body [19]. Schmorl suggested that the vertebral wedging was caused by herniation of disc material into the vertebral body (Schmorl’s nodes). According to Ippolito and Ponseti [10] a biochemical abnormality of the collagen and matrix of the vertebral endplate cartilage may be another contributing factor. Bradford et al. [6] claimed that osteoporosis may be responsible for the development of Scheuermann’s disease. Mechanical factors and repetitive trauma have been also considered to play a significant role in the appearance of the disease [20]. Case reports in monozygotic twins support the theory that there is also a genetic contribution [9]. In conclusion, the etiology of Scheuermann’s disease remains unclear and probably is multifactorial. The criteria for the diagnosis of Scheuermann’s disease are: (1) more than 5° of wedging of at least three adjacent vertebrae at the apex of the kyphosis, (2) endplates irregularities and (3) a thoracic kyphosis of more than 45° [5]. Our patient had typical roentgenografic features of Scheuermann’s disease and his kyphosis was 66° from T5 to T12. Neurological complications in Scheuermann’s disease are rare [2–4, 7, 11, 15, 18, 21, 24]. Three different types of neural compression have been reported: (1) extradural spinal cyst, (2) compression of the cord at the apex of the kyphos and (3) disk hernia at the apex of the kyphos [4]. To the best of our knowledge only 20 cases fall in the last category [2–4, 8, 12, 14, 17, 21–24]. Data of all these cases are presented in Table 1. Table 1Cases of Scheuermann’s disease causing spinal cord compression as the result of thoracic disc herniationReferencesLevelAge/sexFindingsAppro achProcedureResults (follow up)Muller [14]T1040 MParaplegiaNot reportedNot complete recoveryVan Landingham [23]T7–T8–T917 MSpastic paraparesis, no sphincter dysfunctionPT7–T9 laminectomyComplete recovery (4 months)Roth et al. [17]T9–T1061 MSpastic paraparesisPT9–T10 laminectomyParaplegiaTurinese and Raven [22]T5–T7–T816 MSpastic paraparesis, sphincter dysfunctionBed rest, Minerva jacketComplete recovery Bradford and Garcia [4]T7–T816 MSpastic paraparesisPT7–T9 laminectomyResidual hypereflexia(120 months) Ryan and Taylor [18]T8–T918 MSpastic paraparesisTT8–T10 partial vertebrectomy, anterior spinal fusionSpastic paraparesis(36 months)Yablon et al. [24]T7–T829 MSpastic paraparesis, sphincter dysfunctionCT/TT7–T8 discectomy and fusionComplete recovery (14 months)Lesoin et al. [12] (six cases) T9–T1049 FMonoparesis (left)PT8–T10 laminectomyNot complete recovery (12 months)T8–T955 FSpastic paraparesisTT8–T9 discectomySmall motor deficit (8 months)T11–T1261 MMonoparesis (right)PPosterior spinal fusionNot complete recovery (10 months)T8–T927 MLower back pain, Babinski’s sign (right)PLT8–T9 discectomy, T7–T10 posterior spinal fusionComplete recovery (3 months)T8–T930 MLower back pain, slight motor deficit (left)PLT8–T9 discectomy, T8–T9 posterior spinal fusionComplete recovery (3 months)T8–T935 MSpastic paraparesisPLT8–T9 discectomy, T8–T9 posterior spinal fusionSmall motor deficit (3 months)Bohlman and Zdeblick [3] ( two cases)T11–T12T12–L138 FMonoparesis. Back and leg pain (for 16 years)CTT11–L1 discectomy and fusionComplete recovery (36 months)T8–T925 FBack and leg pain (for 7 years)TT8–T9 discectomy and fusionSignificant pain relief (26 months)Stambough et al. [21]T6–T7–T821 MSpastic paraparesisTT6–T8 discectomy, anterior spinal fusion Complete recovery (24 months)Bhojraj and Dandawate [2] (three cases)T11–T1216 MSpastic paraparesis, urinary hesitancyPT10–T12 laminectomy, posterior interbody fusion Complete recovery (36 months)T12–L124 MSpastic paraparesis, urinary hesitancyPT12–L1 laminectomy, posterior interbody fusionComplete recovery (24 months)T11–T1216 FSpastic paraparesisPT11–T12 laminectomy, posterior interbody fusionTransient worsening postoperatively, subsequently complete recovery (15 months)Chiu and Luk [8]T11–L235 FSpastic paraparesisTT11–L2 discectomy, spinal fusion Small motor deficit (24 months)This report (2005)T8–T914 MRapidly progressive spastic paraparesis, finally spastic paraplegiaTT8–T9 discectomy, anterior fusion, T6–T12 posterior fusionComplete recovery (25 months)M Male, F female, P posterior, T transthoracic, CT costotransversectomy, PL posterolateral Despite the fact that this condition has a uniform sex distribution [15, 20], only 6 of all 20 cases were female. Males with Scheuermann’s disease are at greater risk for the development of spinal cord compression in the second decade of life [1, 15]. This is in agreement with our case (male, 14 years old). This difference between the sexes is probably due to the fact that males have more longitudinal spinal growth than females and their maximum growth in trunk length, as well as the maximum deformity of the kyphosis, occurs about 2 years later. In all cases in which the degree of deformity was reported, the average angle was only 56.3°. The severity of the neurological complications had no obvious correlation with the magnitude of the deformity [11, 18]. In our case the deformity was 66°. Lonstein et al. [13] found that in 43 cases with neurological deficits secondary to spinal deformity (what ever the etiology) the average angle of kyphosis was 95°. He suggested that the greater the angle of kyphosis, the greater the risk of neurological impairment. Neurological complications are less likely to occur when the deformity is present over a large number of segments, instead of sharply angular deformities. Disc ruptures tend to occur at the apex of the thoracic kyphosis [13]. The majority of them have occurred at T7–T8 or at T8–T9. This is also confirmed by our case (at T8–T9). It is interesting that a relatively small disc herniation such as in our case produces such a major neurological deficit. Factors that may influence the onset of cord compromise are the tenuous vascularization and the relatively small canal diameter of the thoracic cord [16]. Additionally, in young patients the disc material is not degenerated and can act as a solid mass. Bradford and Garcia [4] suggested that the high incidence of Schmorl’s nodes in Scheuermann’s disease can prevent disc herniation by decompressing the disc space. In our case there were no Schmorl’s nodes either at this level or at any other level. Surgical treatment is indicated in the cases of Scheuermann’s disease with acute cord compression and generally combines anterior release with posterior instrumentation and fusion. A posterior decompression is ineffective because the compressive force is acting from the front [11, 13, 15, 18]. Although some authors have advocated posterior laminectomy and decompression, Patterson and Arbit [16] showed that 45% of patients undergoing laminectomies for thoracic disc herniations either had no relief or deteriorated. For the 20 reported cases with thoracic disc herniation and Scheuermann’s disease six patients were treated with anterior spinal fusion, four patients with laminectomy, four with posterior spinal fusion, three patients with interbody fusion and pedicular screw plate fixation and one patient was treated with discectomy without spinal fusion. Only one patient was treated conservatively with bed rest and Minerva jacket without any decompressive surgery [22]. In one case [14] the treatment was not reported. If a significant or progressive kyphosis, i.e. 55° or more exists, a combined posterior spinal fusion and instrumentation should be considered [18, 21]. Due to the age of the described patient and to the degree of kyphosis we performed discectomy, decompression, anterior and also posterior fusion using a double-rod multihook and transpedicular screws segmental instrumentation system. The majority of the reported cases had good results. In 10 out of 20 cases the patients had complete recovery while only in one case the final outcome was paraplegia [17]. Our patient had an excellent result with complete recovery in 3 months. At the final follow-up 2 years postoperatively, he is neurologically entirely normal and the spinal fusion is solid. This case is interesting because of the severity of the symptoms. The patient had a progressive bilateral spastic paraparesis and finally a spastic paraplegia. Spastic paraparesis was the most common symptom in all cases but neurological symptoms were not so severe to result in paraplegia. Only one case had paraplegia, as the described patient [14]. It is impressive that despite the significant neurological deficit there was a complete recovery. It seems that when symptomatic compression of the spinal cord occurs, surgery is the best option. Conclusion Thoracic disc herniation is an extremely rare complication of Scheuermann’s disease, resulting in cord compression and marked neurological deficit, most commonly in young patients. This rare form of spinal cord compression once demonstrated, should be treated as soon as possible with surgical decompression and stabilization of the spine anteroposteriorly.

Eur_Arch_Otorhinolaryngol-3-1-1914267

Donor site morbidity of the fasciocutaneous radial forearm flap: what does the patient really bother?

The objective of this study was the evaluation of donor site morbidity in head and neck cancer patients after reconstruction using a free vascularized radial forearm flap with emphasis on subjective complaints. Fifty patients who underwent at least 6 months before a reconstruction using a free vascularized radial forearm flap were asked to fill out two questionnaires regarding cosmetics and sensibility and forearm disabilities. Furthermore, a function test including movement extensions (flexion–extension, ulnar–radial deviation and pronation–supination), strength (pinch and grip) and temperature (digiti I and V) of the donor and non-donor site were measured and compared. Thirty-five percent of the patients reported no complaints regarding cosmetics and sensibility and 75% mentioned no forearm disabilities. There was no difference in movement extensions, temperature and grip strength between donor and non-donor sites. The difference in pinch strength appeared to be significant (p < 0.001). The total score of the questionnaire on forearm disabilities correlated significantly with extension, pronation and grip strength of the donor arm. Donor site morbidity of the radial forearm flap measured by objective functional tests was limited but subjective self-ratings revealed complaints regarding cosmestics and sensibility and to a lesser extent regarding forearm disability. The present data may be used for solid patient counselling. Introduction The free radial forearm flap (FRFF), introduced by Yang et al. in 1981 [1], has established itself as a versatile, reliable and widely used method for reconstruction of defects in the head and neck region [2, 3]. The pliability and thinness of FRFF allow its use in complex reconstructions. The vascular pedicle of the FRFF provides adequate vessel diameter and length for microvascular anastomosis. A split skin graft is most frequently used for the forearm donor site reconstruction. Donor site morbidity includes functional and aesthetical outcome. Hand and wrist function is the most important issue in the assessment of morbidity of the FRFF. In a previous study we found that donor site morbidity of the radial forearm flap measured by objective functional tests is negligible but in the patient’s perception this is substantial [4]. Some retrospective studies describe a reduced forearm and/or wrist mobility and a reduced strength [5], stiffness of the wrist with movement disability, dysaesthesia, swelling of hand and/or wrist and pain [6], reduced sensibility of the radial nerve area and a less aesthetic result [7, 8], while others describe rare or no movement disabilities after using the FRFF [9, 10]. The subjective morbidity after harvest of the FRFF is one of the reasons that some reconstructive surgeons shift their attention to other fasciocutaneous flaps for reconstruction of defects in the head and neck [11, 12]. The goal of the present study is to investigate these subjective complaints in more detail and in relation to the objective function tests in a larger group of patients. Methods Patients All patients who visited the outpatient clinic of the department of otolaryngology/head and neck surgery of the VU Medical Center during the period of March till July 2004 and had a reconstruction of the surgical defect by using a FRFF at least 6 months before were asked to participate in the study. Exclusion criteria were arthritis of the upper extremities or a recent trauma of the wrist and/or lower arm. During this period 69 patients, who underwent such a operation, visited the outpatient clinic. Fifty patients (28 men and 22 women, aged between 26 and 77 years, mean 59.2 ± 11.1 years) met the inclusion criteria. Eight of these 50 patients had had a forearm or wrist injury before surgery (five patients had a wrist fracture, two had a Dupyutren’s contracture and one had surgery because of a ganglion on dorsal side of the wrist), of which four patients at the donor side. None of these patients had pre-existing complaints and/or range of motion disability of the hand or wrist. In 45 patients the left arm was used for the harvest of the FRFF and in five patients the right arm. In three patients the donor side was the side of the dominant hand. Harvest of the flap was done simultaneously with the ablative procedure whenever possible. All patients underwent a preoperative and intra-operative Allen’s occlusion test to rule out inadequate blood supply from the ulnar artery. Fasciocutaneous flaps were raised under a tourniquet in a conventional subfascial or superfascial manner about 2 cm proximal to the wrist skin fold. The superficial radial nerve and branches of the lateral antecubital nerve were preserved. The cephalic vein was used as donor vein. The radial artery was not reconstructed in any patient. The donor defect was closed with a split skin graft (0.6 mm) taken from the upper thigh at the same site. In order to aid healing the arm a pressure dressing of a paraffin gauze and foam was placed over the skin and the arm was immobilized for 7 days in a dorsal hand-to-upper-arm splint. Donor site morbidity questionnaires Subjectively the donor site morbidity was measured by means of a questionnaire regarding cosmetics and sensibility (Table 1) and a questionnaire regarding forearm disabilities of the operated arm (Table 2). Response possibilities were yes, no, or not applicable. All themes relate to the last 24 h. A total score on each questionnaire was calculated by dividing all yes-scores by the number of questions answered yes or no. The total score ranged from 0 to 100 with 0 indicating no problems (no disability) and 100 indicating severe problems (severe forearm and/or wrist disabilities). Table 1Questionnaire regarding sensibility and cosmetics of the donor arm (n = 50)QuestionsNumber of positive answer (%)No complaints17 (35)Complaints 1. Can you wear a wristwatch or bracelet?12 (24) 2. Does the hand feel numb?11 (22) 3. Does the scar itch?11 (22) 4. Do you experience problems in the cold?4 (8) 5. Does the appearance bother you?13 (27)In case of complaints, multiple answers were possibleTable 2Questionnaire regarding forearm disabilities in the last 24 h (n = 50)Questions Number of positive answer (%)No forearm disabilities38 (75)Disabilities 1. I wake up at night because of my forearm1 (2.0) 2. I have complaints lying on my forearm 4 (8.0) 3. I have complaints during daily life activities2 (4.0) 4. I have complaints during movements of my wrist3 (6.0) 5. I have complaints during leaning on my elbows or hands3 (6.0) 6. I have complaints with writing (or typing)1 (2.0) 7. I have complaints holding my wheel of my car or bike2 (4.0) 8. I have complaints during lifting an object5 (10.0) 9. I have complaints opening or closing a door2 (4.0) 10. I rub my wrist or forearm more than once a day9 (18.0) 11. I am irritable for people in my environment due to my forearm1 (2.0) Donor site morbidity tests Three tests were used for objective measurement of donor site morbidity. All measurements were performed at the operated and non-operated side. The first test covers movement extensions of the wrist. The angles of the maximal flexion and extension of the wrist, the ulnar and radial abduction and the lower arm pronation and supination were measured by the Mediclino Inclinometer (Bodybow-Holland, Nieuwegein, The Netherlands) with an accuracy of 2° and a range of 0–180° (Fig. 1). Fig. 1Inclinometer used for measuring of movements The second test measures the strength of the hand. The grip strength (strength measured in the hand) was measured in kg/m2 with a hydraulic dynamometer (Smith & Nephew Roylan, Germantown, WI, USA) with an accuracy of 2.0 kg and a range of 0–90.0 kg (Fig. 2). The pinch-strength (strength of digiti I and V) was measured with a pinch gauge (B&L Engineering, Santa Fe Springs, CA, USA) with an accuracy of 0.5 kg and a range of 0–12.5 kg (Fig. 3). Fig. 2Hydraulic dynamometer used for measuring grip strengthFig. 3Pinch gauge used for measuring pinch strength The third test measures the skin temperature of digiti I and V with a Tempcontrol MT 100 KC and probe (Tempcontrol Industrial Electronic Products, Voorburg, The Netherlands) on the skin surface with an accuracy of 0.1°C and a range of −200 to 1350°C. Statistical analyses Descriptive statistics were generated for the range of outcome variables in the study. Student’s t-tests were used to determine the differences in objective tests between the donor and non-donor site. Spearman correlation-coefficients were calculated to assess correlations between subjective and objective forearm disability. For all tests, a two-sided p-value less than .05 was considered statistically significant. Results Regarding subjective evaluation of cosmetics and sensibility, 35% of the patients had no complaints. The other patients scored positively (thus indicating complaints) on several of the questions (Table 1). Regarding forearm disabilities, 75% of the patients had no complaints (Table 2). The other patients showed a variety of complaints, especially on rubbing and lifting an object. The mean total score on the forearm disability questionnaire was 4.6 (SD 12.6) with a range from 0 to 82. The results of the objective tests revealed that no differences were measured between the donor and non-donor arm regarding movement, grip and temperature. Pinch values appeared to be significantly different (Table 3). Table 3Movement extensions, strength and temperature of donor and non-donor wrist and hand Donor sideNon-donor sideMeanSDMeanSDExtension57.311.759.811.0Flexion71.18.070.29.2Ulnar abduction57.78.258.98.0Radial abduction15.38.414.85.1Pronation85.410.885.311.0Supination79.612.280.412.9Grip28.710.429.99.7Pinch7.42.48.02.4*Temperature digit I29.33.329.43.6Temperature digit V29.03.929.14.1*Significant difference (p-value < .01)) The total score on the forearm disabilities questionnaire appeared to be significantly related to extension (r = 0.29, p < 0.05), pronation (r = 0.30, p < .05) and grip strength (r = 0.30, p < .05). Discussion The most important goal of reconstructive surgery in head and neck cancer patients is the optimal restoration of function. The FRFF is a very reliable flap which can restore function in the head and neck function very well [13, 14, 15]. However, the donor site morbidity after harvesting the flap is another important issue. While the main postoperative concern was the reconstruction site, during follow-up the donor site become more important to patients [16]. Hand and wrist functions are important in the assessment of morbidity of the FRFF. The only objectively observed difference in our study was pinch strength between operated and non-operated sides. It must be emphasized that in 94% of the patients the FRFF was harvested from non-dominant side. Therefore, the pinch strength on the operated arm may be already less than on the non-operated side before surgery. All other functional tests revealed no statistically significant differences between operated and non-operated sides. Also Ho et al. [17] did not find any significant difference between the operated and the non-operated arm for strength, range of motion and dexterity. The questionnaires on the other hand revealed donor site complaints related to the FRFF. This difference between objective and subjective findings was also reported by others [4, 7, 8, 9]. The questionnaire on forearm disabilities showed that the vast majority (75%) of the patients had no functional complaints at all. The most frequent positive answer was about rubbing the forearm, which may be more a sensibility than a functional problem. In a study of Toschka et al. [18] postoperative hand function received a subjective rating of 80–100% of the preoperative function by 89% of their patients. Other studies report no subjective impairment of function in 63–100% of patients [7, 10]. Ho et al [17] found using a questionnaire consisting of a list of activities of daily living (ADL) that required use of wrist and forearm, that postoperative function appears to be quite satisfactory. While the forearm disability questionnaire showed that functional problems are limited, the cosmetics and sensibility questionnaire revealed that only 39% of the patients had no cosmetic or sensible complaints. The most frequent (27%) complaint was on the appearance of the donor site. Other studies found complaints of poor aesthetic results in up to 28% of patients, particularly female patients [7, 9, 10, 16]. Bardsley et al. [10] examined the cosmetic result by a subjective assessment on a scale of 0–10. The cosmetic result was acceptable in men (mean score 1.5) but was less so in women (mean score 4). Ito et al. [16] scored the dissatisfaction about the FRFF donor site of 23 patients in five items: color (4%), scar width (0%), depression (30%), wrist mobility (0%) and sensation (4%). The majority of patients (61%) had no complaints at all. The mean number of items of which patients were dissatisfied was 0.39. On the other hand in the study of Toschka et al. [18] 94% of the patients rated the aesthetic outcome as fair or good. Lutz et al. [8] reported 98% of patients rating the aesthetic outcome as satisfactory. Only extension, pronation and grip strength had a clear correlation with the score on the forearm disabilities questionnaire. For all other objective test no significant correlation with this score was found. Therefore, most objective test results may not be indicative for donor site complaints by the patient. This finding has been confirmed earlier by others [4, 7, 8, 9]. To diminish donor site morbidity several harvesting and donor site closure technique modifications have been proposed. The technique of harvesting the flap may have some impact on donor site function. In the early days of this flap when radial bone was included for mandibular reconstruction, morbidity was substantial, even leading to wrist fractures [7]. Suprafascial dissection is claimed by some to lead to superior results from a standpoint of wound healing, but there are no comparative studies that confirm this [8, 18]. Wolff et al. [19] reported on a small series of prefabricated fascial-split-thickness skin flap: after a split-thickness skin graft is transplanted to the forearm fascia the flap can be raised with complete preservation of the forearm skin and microsurgically transplanted like a conventional radial flap. The prelaminated fasciomucosal radial forearm flap as described by Nehrer-Tairych et al. [20] claims to provide better cosmetic and functional donor site effects, although the series of patients they describe is rather small. Avery et al. [5] reported good aesthetic results of repair of the FRFF donor site by full-thickness skin graft from the inner upper arm. Ho et al. [17] compared the functional and aesthetic outcomes of FRFF donor sites reconstructed with full-thickness skin graft, split skin graft alone and split skin graft overlying an acellular dermal matrix and found that all three methods of reconstruction have comparable low morbidity, postoperative satisfactory aesthetic and functional outcomes. Negative pressure wound dressing has been used for rapid healing and decreased donor site complications [21]. Bardsley et al. [10] used a ulnar artery-based transposition flap for primary closure to reduce wound healing and improve cosmetic results. Hsieh et al. [22] reported good results of primary closure of the FRFF donor site with a bilobed flap based on the fasciocutaneous perforator of the ulnar artery. Several types of tissue expansion have been reported to allow the use of local tissues for primary closure [23]. The preferred technique of closure is difficult to assess, because of limited comparative studies, limited number of patients and lack of significant differences. The presented data can be used for solid counselling patients who are scheduled for FRFF reconstruction. Elaborate pre-surgical counselling may reduce the impact of functional and cosmetic impairment at the FRFF donor site [24]. Moreover, these data may serve as benchmark for future studies that use other free fasciocutaneous flaps. Conclusion Donor site morbidity of the radial forearm flap measured by objective functional tests is negligible but in the patient’s perception this is substantial. Subjective measurements show especially problems in cosmesis and sensibility. The present data may be used for solid patient counselling.

J_Antimicrob_Chemother-1-1-2386079

Cellular pharmacokinetics of telavancin, a novel lipoglycopeptide antibiotic, and analysis of lysosomal changes in cultured eukaryotic cells (J774 mouse macrophages and rat embryonic fibroblasts)

Background Telavancin is a lipoglycopeptide with multiple mechanisms of action that include membrane-destabilizing effects towards bacterial cells. It shows bactericidal activity against forms of Staphylococcus aureus (phagolysosomal infection) with different resistance phenotypes [methicillin-resistant S. aureus, vancomycin-intermediate S. aureus or vancomycin-resistant S. aureus]. We examine here the uptake, efflux and intracellular distribution of telavancin in eukaryotic cells as well as its potential to induce lysosomal changes (in comparison with vancomycin and oritavancin). Introduction Telavancin is a novel lipoglycopeptide derivative of vancomycin1 with marked bactericidal activity against vancomycin-susceptible and vancomycin-resistant organisms2 due to a multifunctional mechanism of action that combines inhibition of cell wall synthesis and disruption of bacterial cell membrane permability.3 It shows a high penetration in tissues,4 including human alveolar macrophages.5 In a recent study, we showed that telavancin exerts time- and concentration-dependent bactericidal activity against intraphagocytic Staphylococcus aureus, disregarding their resistance phenotypes (methicillin-resistant S. aureus, vancomycin-intermediate S. aureus or vancomycin-resistant S. aureus).6 The present investigation was initiated to further document and rationalize this observation by examining the uptake and subcellular disposition of telavancin in eukaryotic cells. Because previous studies had disclosed marked lysosomal alterations in eukaryotic cells exposed to another lipoglycopeptide, oritavancin,7 we also undertook to assess the impact of telavancin in this context, using vancomycin as a comparator. The study was performed with both phagocytic (J774 macrophages) and non-phagocytic (fibroblasts) cells because this allowed us, in the past, to obtain a comprehensive picture of the behaviour of other antibiotics accumulating in cells and causing specific lysosomal alterations.8–11 Materials and methods Cells, cell cultures and assessment of membrane integrity J774 mouse macrophages and rat embryo fibroblasts were cultivated, as previously described,7 in RPMI 1640 or in Dulbecco's modified Eagle's medium, respectively, both supplemented with 10% fetal calf serum (unless stated otherwise). The integrity of the pericellular membrane upon exposure to the antibiotics was assessed by measuring the release of lactate dehydrogenase in the culture medium, as described previously.6 Determination of cellular antibiotic concentration Cells incubated with 14C-labelled telavancin were washed three times in ice-cold 0.9% NaCl, collected by scraping in distilled water and used for radioactivity determination (liquid scintillation counting) and protein assay (a detailed description of the methodology has been published previously12). In pilot experiments, cell-associated antibiotic was also measured by a microbiological technique (with Micrococcus luteus ATCC 9341 as test organism and antibiotic medium 11).12 A correlation coefficient (R2) of 0.95 was found between the two methods for cells incubated 24 h at concentrations spanning from 10 to 100 mg/L (n = 12). Cell fractionation studies Cells were collected by gentle scraping in 0.25 M sucrose/1 mM EGTA/3 mM imidazole pH 7.4, homogenized and fractionated, as described previously.12 In brief, homogenates were separated into an ‘unbroken cells/nuclei’ fraction by a low-speed centrifugation. The resulting cytoplasmic extract was separated into a ‘granules/membranes’ fraction and a final supernatant by high-speed centrifugation. The granule/membrane fraction was then further analysed by isopycnic centrifugation in a sucrose gradient. Protein and [14C]telavancin contents were determined in the fractions in parallel with the activity of marker enzymes of the main organelles, namely, inosine 5′-diphosphatase (E.C. 3.6.1.6.; plasma and endoplasmic reticulum membranes), cytochrome c oxidase (E.C. 1.9.3.1.; mitochondria) and N-acetyl-β-glucosaminidase (E.C. 3.2.1.30.; lysosomes). Results are expressed as the percentage of enzyme activity, protein or drug recovered in each fraction. For isopycnic centrifugation, distributions were standardized for equal density increments ranging from 1.08 to 1.21, as described previously.13 Biochemical studies Cell sheets were washed three times in ice-cold 0.9% NaCl, collected by scraping in distilled water and lysed by sonication. Total phospholipids and cholesterol were extracted and assayed, as described previously.7 Proteins were assayed by the Folin–Ciocalteau/biuret method.14 Electron microscopy Sample preparation was performed as described previously.15,16 Observations were made in a Zeiss electron microscope operated at 80 kV. Morphometric analysis was performed on pictures taken at random, using the Image J software available from the NIH web site (http://rsb.info.nih.gov/ij/) and examining a total cell surface of cell profiles of 1000–2000 µm2. For each cell profile, we manually selected the zones occupied by electron-dense material (for which a limiting membrane could not always be clearly recognized) or corresponding to large vesicles filled with heterogeneous material (see Figure 3 for examples). Results were expressed as percentage of the whole cell surface profile, which is numerically identical to the percentage of cell volume.17 Figure 3 Selected pictures illustrating typical changes observed in macrophages (a) and fibroblasts (b) after incubation with 90 mg/L telavancin for 24 or 72 h, respectively. Both cells show markedly enlarged and ill-shaped lysosome-like bodies with a pleiotropic material that is highly osmiophilic (often organized in a distorted concentric fashion). (c) Example of loose material and (d) example of highly osmiophilic material organized in multiple separate bodies. Bars are 1 µm. The dotted lines show examples of profiles selected for morphometric analysis: i, a large profile showing mainly a moderately osmiophilic and pleiomorphic material; ii, a small profile showing mainly osmiophilic material [possibly a polar section of a lysosome with mixed content (as in iii), but cut as shown by the arrow]. Materials Telavancin hydrochloride (powder for microbiological evaluation and >90% purity) and [14C]telavancin trifluoroacetate (33.8 mCi/mmol and radiochemical purity >91%) were supplied by Theravance, Inc. (South San Francisco, CA, USA). The labelled drug was mixed with unlabelled telavancin to obtain a specific activity of 5 mCi/mmol. Stock solutions were prepared at a final concentration of 1–2 mg/L by vigorous vortexing in distilled water (the use of acidified DMSO recommended by the manufacturer was avoided because preliminary experiments disclosed a decrease in cell viability). Oritavancin (supplied as diphosphate salt fully hydrated) was obtained from Intermune (Brisbane, CA, USA). Vancomycin was procured as VANCOCIN® from GlaxoSmithKline, Belgium. Cell culture or microbiology media were from Invitrogen (Paisley, Scotland, UK) and Difco (Sparks, MD, USA). Other reagents were of analytic grade and purchased from E. Merck AG (Darmstadt, Germany) or Sigma-Aldrich-Fluka (St Louis, MO, USA). Results Influence of telavancin on pericellular membrane integrity In preliminary experiments, we measured the release of lactate dehydrogenase, a cytosolic enzyme, from cells exposed to telavancin (90 mg/L, corresponding to the human Cmax18), in comparison with vancomycin (50 mg/L, corresponding to the human Cmax19) and oritavancin (25 mg/L, corresponding to the human Cmax20). No difference in controls (5.0 ± 0.2%) was seen with telavancin (4.9 ± 1.0%) or vancomycin (5.0 ± 1.5%) in J774 macrophages after 24 h of incubation, whereas oritavancin induced a small but significant increase in enzyme release (15.5 ± 1.5%; P < 0.001; n = 3 for all conditions). No significant effect was seen for fibroblasts after 72 h of incubation between controls (13.4 ± 3.7%) and cells incubated with telavancin (16.9 ± 2.5%), vancomycin (17.1 ± 1.3%) or oritavancin (18.2 ± 3.0%; n = 3 for all testing conditions). Kinetics of uptake and release of telavancin Figure 1(a and b) shows the kinetics of uptake of telavancin in J774 macrophages and fibroblasts incubated with an extracellular concentration of 90 mg/L. The uptake proceeded linearly over time at a rate that was similar in both cell types (see figure captions for details). Figure 1(c and d) examines the uptake of telavancin at increasing extracellular concentrations and at fixed time points (24 h for macrophages and 72 h for fibroblasts). The uptake was linearly related to the extracellular concentration in both cell types, allowing us to calculate a clearance rate for all conditions shown in Figure 1 of ~10 µL/mg of cell protein per h. For macrophages, we also examined the drug efflux (after an initial uptake of 12 h in the presence of 90 mg/L telavancin), which occurred at an apparent rate ∼5.7-fold lower than that of influx (Figure 1a). Figure 1 (a and b) Kinetics of uptake of telavancin (filled symbols and continuous line) in J774 macrophages (a) or embryo fibroblasts (b) incubated for the indicated times with an extracellular concentration of 90 mg/L at 37°C in medium supplemented with 10% fetal calf serum. (a) Kinetics of efflux of the drug from J774 macrophages exposed to telavancin (90 mg/L) for 12 h and re-incubated in a drug-free medium for an additional 24 h (open symbols and broken line) are also shown. (c and d) Cellular concentration of telavancin in J774 macrophages (c) or embryo fibroblasts (d) incubated at 37°C for 24 h or 72 h, respectively, in the presence of telavancin at the extracellular concentrations indicated on the abscissa. Results are given as arithmetic means ± SD (n = 3) and analysed by linear regression to calculate the corresponding clearances (µL/mg of protein/h): J774 macrophages, 9.6 ± 0.6 (R2 = 0.98; a) and 10.0 ± 0.4 (R2 = 0.99; c); fibroblasts, 8.2 ± 0.4 (R2 = 0.97; b) and 9.0 ± 0.7 (R2 = 0.98; d). The influence of serum on telavancin uptake was examined in J774 macrophages by performing experiments in culture medium not supplemented with calf serum. The incubation was limited to 5 h to ensure maintenance of cell viability. Telavancin uptake remained linear over the 10–90 mg/L range of extracellular concentrations, but proceeded at a rate ~1.7-fold faster than in the presence of serum (data not shown). Subcellular distribution of telavancin The subcellular distribution of cell-associated telavancin was examined in homogenates obtained from J774 macrophages incubated for 24 h with the drug at an extracellular concentration of 90 mg/L. The distribution of cell-associated telavancin and N-acetyl-β-glucosaminidase (lysosome marker) among the unbroken cells/nuclei, granules/membranes and supernatant fractions was similar (35% and 27%, 55% and 67% and 10% and 6%, respectively). The granule/membrane fraction was therefore subfractionated by isopycnic centrifugation. The density distribution of telavancin, in comparison with that of N-acetyl-β-glucosaminidase (lysosomes), cytochrome c oxidase (mitochondria) and inosine 5′-diphosphatase (plasma/endoplasmic reticulum membranes), is shown in Figure 2. The distribution pattern of telavancin was largely superimposable on that of the N-acetyl-β-glucosaminidase, clearly dissociated from that of cytochrome c oxidase and also distinct from that of inosine 5′-diphosphatase. Figure 2 Density distribution of marker enzymes [N-acetyl-β-glucosaminidase (lysosomes), cytochrome c oxidase (mitochondria) and inosine 5′-diphosphatase (plasma and endoplasmic reticulum membranes)] and of telavancin after isopycnic centrifugation in a linear sucrose gradient of a granule fraction prepared from homogenates of J774 cells that were incubated for 24 h with 90 mg/L telavancin at 37°C in medium supplemented with 10% fetal calf serum. The ordinate shows the percentage of each constituent recovered in each fraction. Ultrastructural alterations Electron microscopy was used to examine whether telavancin induces morphological alterations in the subcellular organelles of cells incubated in its presence. Figure 3 shows selected pictures of alterations that could be observed in J774 macrophages and fibroblasts (90 mg/L; 24 and 72 h, respectively). There was an enlargement of lysosomes that were filled with a pleiotropic material, made of partly highly osmiophilic, concentric structures and partly of a more loose appearance material. To gain quantitative information on these changes, a morphometric analysis was performed on pictures taken at random. Figure 4 shows the relative abundance of these abnormal lysosomal profiles in cells incubated with three different concentrations of telavancin, in comparison with control cells or cells exposed to 50 mg/L vancomycin or 20 mg/L oritavancin. In macrophages (24 h incubation) as well as in fibroblasts (72 h incubation), telavancin induced a concentration-dependent increase in the percentage of cell volume occupied by overloaded lysosomes. The morphometric analysis showed that: (i) changes induced in fibroblasts were more marked than those in macrophages (which may have been a consequence of the longer incubation time in fibroblasts); (ii) incubation of macrophages with 25 mg/L telavancin or of fibroblasts with 90 mg/L caused changes similar to those seen with 50 mg/L vancomycin in either cell types and (iii) alterations induced by telavancin at the highest concentration tested (90 mg/L) were considerably milder than in cells incubated with 20 mg/L oritavancin in both cell types. Figure 4 Morphometric analysis of the material accumulated in macrophages (left-hand panel) or fibroblasts (right-hand panel) after 24 and 72 h of incubation, respectively, in control conditions or in the presence of 5, 25 or 90 mg/L telavancin (TLV), 50 mg/L vancomycin (VAN) or 20 mg/L oritavancin (ORI). Results are expressed as percentage of the cell surface occupied by the electron-dense material and/or large vesicles filled with a material of undetermined nature. Surface analysed was ~2000 µm2 for macrophages and 1000 µm2 for fibroblasts. Influence of telavancin on cell phospholipid and cholesterol contents In view of the ultrastructural changes observed, we looked for changes in phospholipids and cholesterol content of cells exposed to 90 mg/L telavancin in comparison with vancomycin (50 mg/L) and oritavancin (25 mg/L). Figure 5 illustrates the data obtained after 24 h in macrophages and 72 h in fibroblasts. Neither telavancin nor vancomycin caused detectable increase in the phospholipid content. In contrast, oritavancin induced significant increases in the phospholipid content in both cell types. Telavancin and vancomycin caused a slight but not significant increase in the cholesterol content, whereas oritavancin again induced a marked, statistically significant increase. Figure 5 Total phospholipid content (upper panels) or total cholesterol content (lower panels) of J774 mouse macrophages (left-hand panels) or rat embryo fibroblasts (right-hand panels) exposed for 24 and 72 h, respectively, to glycopeptides at their human Cmax (VAN, vancomycin 50 mg/L; TLV, telavancin 90 mg/L and ORI, oritavancin 25 mg/L). Results are expressed as percentages of control values for total phospholipids and cholesterol. Values are arithmetic means ± SD (n = 6–8). Values for control macrophages and fibroblasts were, respectively, 204 ± 10 and 295 ± 8 nmol/mg of protein for total phospholipids and 72 ± 3 and 130 ± 16 nmol/mg of protein for total cholesterol. Statistical analysis (ANOVA): ***P < 0.001, **P < 0.01, *P < 0.05 when compared with the matching control. Other differences were not significant. Discussion The present study provides evidence that telavancin is taken up by cultured macrophages where it becomes associated with lysosomes, as assessed by cell fractionation studies. These results rationalize our previous data, showing that telavancin exerts a marked antibiotic activity against intraphagocytic S. aureus, which is known to develop in the phagolysosomes of infected macrophages.21,22 The accumulation level reached by telavancin in macrophages at 24 h (∼45-fold) is intermediate between that recorded previously in the same conditions for vancomycin (∼8-fold) and oritavancin (∼370-fold).12 Its uptake is not specific to phagocytic cells, as it is also observed with fibroblasts, both cell types taking up the drug at similar rates. The entry of drugs into lysosomes may occur through either diffusion/segregation or pinocytosis,23 as illustrated by the behaviour of macrolides8 and aminoglycosides,24 respectively. Diffusion–segregation seems unlikely in view of the slow efflux of telavancin, as this process is usually considered to be reversible (although binding of the drug to intracellular constituents could explain this slow efflux, as is observed for azithromycin in fibroblasts16). Pinocytosis, therefore, appears more plausible, especially if considering the size of the molecule that would prevent its fast diffusion through membranes. However, the clearance rates recorded here (∼10 µL/mg of protein/h) are ~15- to 30-fold higher than those reported for fluid-phase pinocytosis markers (∼0.7 µL/mg of protein/h in macrophages15 and ∼0.3 µL/mg of protein/h in fibroblasts24). Telavancin uptake, therefore, should involve a process of adsorptive pinocytosis (through binding to cell surface; see discussion and modelling in Silverstein et al.25) However, the lack of saturation of telavancin uptake upon increase of its extracellular concentration is intriguing as this (i) is an hallmark of adsorptive pinocytosis25 and (ii) was observed for oritavancin.12 The initial uptake clearance rate of oritavancin, however, is considerably larger (∼150 µL/mg of protein/h),12 which indicates that binding of telavancin should be much weaker and may not actually show saturation in the concentration range investigated here. A weaker membrane binding of telavancin is actually consistent with its lower lipophilicity when compared with oritavancin (reported logP values of 0.6 for telavancin versus 4.1 for oritavancin) [Advanced Chemistry Development Software Solaris V4.67, Sci Finder Scholar 2006, American Chemical Society, Washington, DC, USA]. Beyond its therapeutic interest, the lysosomal accumulation of telavancin may also be responsible for the morphological alterations observed in cells exposed to the drug. Yet, these changes appear quantitatively minor, as is also the case for vancomycin. Moreover, and in contrast to what is observed with oritavancin,7 telavancin did not significantly affect phospholipid or cholesterol cellular levels. This difference may be related to the lower uptake of telavancin, although we cannot exclude, at this stage, a true difference in the intrinsic capacity of the two molecules to interfere with lipid metabolism. Further studies are required to determine the exact nature of the accumulated material and to decipher the underlying molecular mechanisms. These may be more complex than those evidenced for aminoglycoside antibiotics, which mainly induce an accumulation of phospholipids.26 Interestingly, telavancin was without detectable effect on lactate dehydrogenase release. This suggests that the membrane-destabilizing properties exerted by telavancin towards bacterial membranes,3 which probably contribute to its marked bactericidal effect, involve constituents that are specific to or more abundant in prokaryotic cells. Independent studies have shown that telavancin accumulates in human alveolar macrophages, reaching a cellular concentration of ~50 mg/L within 8–24 h after systemic administration of therapeutic doses.5,27 In our experiments, macrophages exposed for 24 h at 90 mg/L have an apparent cellular concentration of 4 mg/mL [based on an estimated mean cellular volume of 5 µL/mg of protein (see Van Bambeke et al.12 and references cited therein)]. The lower cellular uptake of telavancin reported in vivo may result from (i) its high protein binding (∼90% to 93%) and (ii) the fluctuation of its serum levels related to its once-daily administration and its half-life of 7.5 h.18 As extensively discussed in previous publications,6,7,12,22 our cellular model suffers from the limitation that it does not take into account these two important pharmacokinetic determinants. This may lead to an overestimation of the accumulation of telavancin versus that of vancomycin, for which the free fraction in vivo oscillates between 10% and 55%.19 Thus, assuming that only the free fraction of telavancin is available for the uptake by macrophages in vivo, its effective extracellular concentration will oscillate between ∼10 and 0.5 mg/L, which, in our model, would create a cellular concentration of ∼0.24 µg/mg of protein (i.e. ∼48 mg/L). At this concentration, which is of the same order of magnitude as that observed in vivo, telavancin does not cause any morphological change in our model. Taken together, our data further document the tissue-directed pharmacokinetics of lipoglycopeptides.28 As telavancin is clearly superior to vancomycin with respect to bactericidal activity towards both extracellular and intraphagolysosomal S. aureus6 while showing a similar cellular safety profile (as far as lipid metabolism and subcellular morphology are concerned), the data support its potential interest for the treatment of infections where intracellular foci are present.29 Further studies using in vivo models are, however, required to confirm the improved intracellular efficacy and cellular inocuity of telavancin. Funding F. V. B. is Maître de Recherches of the Belgian Fonds National de la Recherche Scientifique. Funding: Belgian Fonds de la Recherche Scientifique (FRS–FNRS) and Belgian Fonds de la Recherche Scientifique Médicale (FRSM; grant nos 1.5.223.05 F, 3.4.549.00 F and 3.4.639.04 F); Belgian Federal Science Policy Office (Research project P5/33; research action P5); Fonds Spéciaux de Recherches and Actions de Recherches Concertées of the Université catholique de Louvain; Grant-in-Aid from Theravance, Inc. (for the pharmacokinetic studies with telavancin). Transparency declarations F. V. B. and P. M. T. are members of the European Advisory Board of Targanta Inc. (current owner of oritavancin). The remaining authors have none to declare.

Eur_J_Nucl_Med_Mol_Imaging-3-1-1914271

Cardiovascular molecular MR imaging

Introduction Cardiovascular molecular imaging is a rapidly evolving field of research, aiming to image and quantify molecular and cellular targets in vivo. MR imaging has some inherent properties that make it very suitable for cardiovascular molecular imaging. Until now, only a limited number of studies have been published on cardiovascular molecular imaging using MR imaging. Introduction Conventional imaging modalities, including magnetic resonance (MR), are primarily based on anatomical, functional or metabolic properties to study (patho)physiology. Molecular imaging is a rapidly evolving field of research, aiming to image and quantify molecular and cellular targets in vivo. Molecular imaging can be applied to a wide range of scientific and clinical fields of interest. One of the most promising applications of molecular imaging is in the field of cardiovascular imaging. Imaging of cardiac anatomy, dimensions and function has some limitations concerning, for example, prediction of therapy outcome. Addition of specific information on, for instance, plaque composition and total plaque burden can be very helpful in guiding therapy. Imaging of molecular processes is desirable because cardiovascular disease may be detected earlier, risk stratification may be more accurate, monitoring of innovative therapies may be improved, or a more accurate prognosis may be provided [1]. MR imaging has some inherent properties that make it very suitable for cardiovascular molecular imaging. The interaction between inherent tissue properties and specific contrast agents may lead to more specific clinical conclusions and prediction of therapy outcome. Thereby, cardiovascular molecular MR imaging may help in diagnosing cardiovascular disease, and in deciding whether expected beneficial effects of (invasive) therapy counterbalance the risk of complications of therapy. A conventional approach to molecular MR imaging concerns MR spectroscopy. Furthermore, there are two main innovative contrast agents that may be used clinically soon: (1) iron oxide MR contrast agents and (2) fibrin-targeted MR contrast agents. MR spectroscopy MR spectroscopy (MRS) allows non-invasive characterisation of myocardial metabolism. In principle, MRS is a pure form of molecular ‘imaging’ technique. Clinically, several nuclei allow noninvasive MRS of the heart. Initially, human MRS research was focussed on the 31P nucleus to study high-energy phosphate metabolism. An example concerning the effects of diabetes type 2 on myocardial high-energy phosphate metabolism is shown in Fig. 1 [2]. Another interesting new application of 31P-MRS was published by Smith et al. [3], who measured myocardial creatine kinase (CK) metabolite concentrations and adenosine triphosphate synthesis through CK, the primary energy reserve of the heart, to test the hypothesis that ATP flux through CK is impaired in patients with left ventricular hypertrophy (LVH) and chronic heart failure. It turned out that myocardial ATP levels were normal, but creatine phosphate levels were 35% lower in LVH patients than in normal subjects. Furthermore, the myocardial CK rate constant was normal in LVH, but halved in patients with LVH combined with chronic heart failure. Thereby, they could show that it is not the relative or absolute CK metabolite pool size but rather the kinetics of ATP turnover through CK that distinguishes failing from non-failing hypertrophic hearts. These observations support the hypothesis that a deficit in myofibrillar energy delivery contributes to chronic heart failure pathophysiology in human LVH. The same technique was applied by Weiss et al. to study ATP flux through CK in the normal, stressed and failing human heart [4]. The latter studies are elegant examples of the capability of MR to measure non-invasively the concentration of metabolites and even the rate constant of enzyme turnover. Fig. 1Left panel: planscan of the volume of interest on transverse and sagittal spin-echo MR images. Right panel: phosphorus-31 magnetic resonance spectroscopy (31P-MRS) obtained at rest from the anterior left ventricular wall in a patient with diabetes type 2 (left) and a healthy subject (right). Note the decreased signal amplitude of phosphocreatine in the type 2 diabetes patient. Courtesy of Diamant et al. [2] Another interesting application of metabolic imaging is 23Na MR imaging, as show by Jansen et al. They applied this innovative spectroscopic imaging technique as a diagnostic modality for early detection of myocardial ischaemia and viability in a rat model [5]. They tested whether 23Na MR imaging can be used to assess viability after low-flow ischaemia. 23Na MR chemical shift imaging was alternated with 23Na MR spectroscopy. Na image intensity increased with increasing severity of ischaemia. 23Na image intensity at end low-flow ischaemia was well correlated with CK release during reperfusion, as well as with infarct size. Therefore, their study indicates that 23Na MR imaging is a promising tool for the assessment of myocardial viability. Ouwerkerk et al. applied 23Na MR imaging to measure cardiac tissue sodium concentrations in the human myocardium [6]. They used a surface coil at 1.5 T MR to non-invasively quantify regional myocardial sodium concentrations in the left ventricular free wall, septum and adipose tissue. Their 23Na MR imaging results were consistent with prior invasive measurements on biopsy and autopsy specimens. In the past, 1H-MRS of the myocardium was first applied to non-invasively study creatine depletion in non-viable infarcted myocardium [7]. Total creatine was measured in the posterior and anterior left ventricle and septum, and was significantly lower in regions of infarction than in non-infarcted regions of myocardium in patients or in the myocardium of healthy controls. Therefore, they showed for the first time that spatially localised 1H-MRS can be used to measure total creatine non-invasively throughout the human heart. The detection of regional creatine depletion may provide a metabolic means to distinguish healthy from infarcted non-viable myocardium. Szczepaniak et al. used 1H-MRS to measure myocardial lipid content [8]. Studies in rat tissue ex vivo and in healthy humans in vivo provided evidence that 1H-MRS constitutes a reproducible technique for the measurement of myocardial triglyceride. Increased myocardial triglyceride content was accompanied by elevated LV mass and suppressed septal wall thickening as measured by cardiac imaging. More recently, 1H-MRS of the human myocardium was improved by implementing the respiratory navigator technique to monitor diaphragmatic motion, and thereby correct data acquisition prospectively for breathing motion. First (unpublished) results from our institution show improved reproducibility of human cardiac 1H-MRS measurements when using the respiratory navigator technique, as compared with conventional continuous breathing. Respiratory navigator gated 1H-MRS was recently applied in an experimental setting in our institution to evaluate the effects of a very low calorie diet on myocardial triglyceride content. First (unpublished) results show that after a short very low calorie diet, there is an increase in intramyocardial triglyceride content (Fig. 2). Fig. 2The surface coil was positioned just below the mitral valve level of the heart (a, b). Spectroscopic volume localisation in the interventricular septum on four-chamber (c) and short axis (d) views. Special care was taken to avoid contamination from epicardial fat. (e) Typical water-suppressed 1H-MR spectrum of myocardial tissue located in the interventricular septum. Peak heights are in arbitrary units Gadolinium-based contrast agents Gadolinium-based contrast agents can be applied to study regional myocardial perfusion. After a rapid intravenous contrast injection, there is marked signal enhancement first in the RV cavity, then in the LV cavity, and subsequently in the LV myocardium [9]. The peak signal intensity is related to the concentration of the contrast agent in the local tissue and is directly proportional to the coronary blood flow. Perfusion MR at rest and after infusion of pharmacological agents (adenosine and persantine) have been compared with standard methods (angiography or radionuclide scintigraphy) and demonstrated reasonable sensitivity (67–83%) and specificity (75–100%) [9]. Currently, multiple MR imaging techniques are available to assess myocardial viability. Cardiovascular MR imaging can be used to assess end-diastolic wall thickness and contractile function at rest [10]. Segments with an end-diastolic wall thickness <6 mm most likely represent transmural scar formation, and contractile function will not improve after myocardial revascularisation. Dobutamine MR imaging can be used to evaluate contractile reserve, in a similar manner to dobutamine echocardiography. Gadolinium contrast-enhanced MR imaging [11] allows for detection of the extent and transmurality of scar tissue (Fig. 3). Recently reported sensitivity and specificity are in the range of 74% and 82% respectively. Fig. 3Left panel: short-axis MR images at rest and during dobutamine stress. Note the lack of improvement in myocardial wall motion in the anteroseptal region when dobutamine stress is applied. Right panel: delayed gadolinium contrast-enhanced MR images in two-chamber, four-chamber and short-axis views. Note the almost transmural delayed enhancement of the anteroseptal/apical region, corresponding to the region without contractile response due to dobutamine stress in the left panel. The anteroseptal region is considered as ‘non-viable’ myocardial tissue Based on as yet unpublished scientific developments, it is expected that gadolinium-based delayed enhancement of the vessel wall may become reality. This MR imaging technique may allow fast total body screening for total plaque burden, an important predictor for morbidity and mortality. In general, gadolinium-based contrast agents are not perfectly suited for molecular imaging because of the inherent high threshold of detectability. Therefore, new contrast agents are under development to potentiate the effect of distortion of the magnetic field. Iron oxide MR contrast agents Superparamagnetic iron oxide (SPIO) particles can be detected at micromolar concentrations of iron, and offer sufficient sensitivity for MR imaging [12]. SPIO-based cellular imaging has become an established technique to label and detect cells of interest. Imaging of macrophage activity was the initial and is still the most significant application, with several products either approved for or in clinical trials [12]. Another exciting application of SPIOs is labelling of myocardial stem cells. In a swine model for myocardial infarction, magnetically labelled stem cells were injected in the infarcted myocardial region (Fig. 4). Using delayed contrast-enhanced MR imaging, the infarcted area can be identified with high accuracy. New technical developments may even allow specific delivery of magnetically labelled therapeutics to the infarcted myocardial region [12]. Combined with measurements of myocardial function, MR imaging seems an excellent modality for planning, delivery and follow-up of myocardial stem cells as therapy for ischaemic heart disease. Fig. 4Detection of delivery and migration of Feridex-labelled myocardial stem cells in a swine model. Hypointense lesions in spin-echo (SE),gradient-echo (GRE) and delayed-enhanced (Delayed) MR imaging (upper panel) of injection sites (arrows) within 24 h of intramyocardial injection. Cells were injected in the myocardial infarct (MI). Long-axis MR images (lower panel) show hypointense lesions (arrows) caused by labelled myocardial stem cells acquired within 24 h and 1 week. LV left ventricle, RV right ventricle. Courtesy of Bulte andKraitchman [12] Another promising application of SPIO MR imaging is visualisation of vessel wall inflammation. SPIOs are ‘digested’ by macrophages, involved in inflammatory processes. Imaging of the SPIO-induced magnetic inhomogeneities allows for imaging of inflammation. Such an approach is currently only available in a research setting; it is, however, expected that these contrast agents will become available for clinical application soon. Fibrin-targeted MR contrast agents Exciting recent developments allow selective and non-invasive molecular MR imaging of thrombus [13]. The principle of the contrast agent is that it is targeted to fibrin. In an elegant study by Spuentrup et al., a swine model was used to test the innovative fibrin-targeted MR imaging contrast agent, which can be administered intravenously [14]. The imaging protocol consisted of coronary MR angiography to demonstrate the lumen of a coronary artery, combined with molecular thrombus MR imaging. Thereby, anatomical information could be linked to specific information of vessel wall components. In an area of focal coronary artery stenosis, intraluminal thrombus could be detected (Fig. 5). Fig. 5Coronary thrombus visualisation with a fibrin-targeted molecular MR imaging contrast agent. Left panel: double-oblique white blood coronary MR angiography (multiplanar reconstruction) demonstrating the lumen of the left anterior descending artery with bright signal (arrowheads). Right panel: double-oblique MR images after administration of the fibrin-targeted contrast agent (multiplanar reconstruction, same orientation). Note the increased signal (arrow) in the left anterior descending coronary artery, corresponding to thrombus. Courtesy of Spuentrup et al. [14] The same contrast agent can be applied to detect, for example, right atrial thrombus, a potential source of more distal emboli (Fig. 6). An atrial clot could be visualised easily with this molecular MR imaging technique, by intravenous administration of the fibrin-targeted contrast agent. Furthermore, MR clot imaging can be combined with functional imaging of the heart in the same imaging session. An even more exciting application of this fibrin-targeted contrast agent is detection of pulmonary emboli (Fig. 7). Fig. 6Molecular MR imaging of atrial clot in a swine model. The left panel shows pre-contrast (upper) and post-contrast (lower) coronal images. Note the presence of high MR signal in the area of the right atrium, indicating an atrial clot. The right panel shows increased MR signal in the left atrium (LA), corresponding to a left atrial clot (arrow). These clots are potential sources for more distal emboli. LV left ventricle. Courtesy of Spuentrup et al. [14]Fig. 7Examples of molecular MR imaging of pulmonary embolus. Two examples are shown, each consisting of three adjacent coronal slices (horizontal). The MR imaging technique is such that signal from surrounding blood pool and lung parenchyma is suppressed. The upper row shows pulmonary embolus (arrow) in the right lower lobe. The lower panel shows bilateral pulmonary emboli (arrows) after intravenous administration of fibrin-targeted MR imaging contrast agent. Courtesy of Spuentrup et al. [14] The above-described applications of molecular MR imaging may be especially suitable for fast screening for cardiovascular disease in an emergency setting. Patients presenting with chest pain in the emergency room can be studied by MR imaging to confirm or rule out ischaemic heart disease or pulmonary embolus. Molecular MR imaging using fibrin-targeted contrast agents allows selective visualisation of acute coronary, cardiac and pulmonary thrombi. Additional functional cardiac imaging can help determine the functional effects of detected thrombi. Conclusion Molecular MR imaging is an exciting and rapidly evolving new area of cardiovascular imaging. MR imaging seems very suitable for molecular imaging, although many technical difficulties have to be overcome. The main current limitation is the low sensitivity of MR imaging to detect small changes in magnet homogeneity. We expect that in the next decade, currently promising MR molecular imaging agents will be introduced into the clinical arena to guide diagnosis and therapy of cardiovascular disease.

Int_J_Biol_Sci-3-3-1802012

Advances in Swine Transcriptomics

The past five years have seen a tremendous rise in porcine transcriptomic data. Available porcine Expressed Sequence Tags (ESTs) have expanded greatly, with over 623,000 ESTs deposited in Genbank. ESTs have been used to expand the pig-human comparative maps, but such data has also been used in many ways to understand pig gene expression. Several methods have been used to identify genes differentially expressed (DE) in specific tissues or cell types under different treatments. These include open screening methods such as suppression subtractive hybridization, differential display, serial analysis of gene expression, and EST sequence frequency, as well as closed methods that measure expression of a defined set of sequences such as hybridization to membrane arrays and microarrays. The use of microarrays to begin large-scale transcriptome analysis has been recently reported, using either specialized or broad-coverage arrays. This review covers published results using the above techniques in the pig, as well as unpublished data provided by the research community, and reports on unpublished Affymetrix data from our group. Published and unpublished bioinformatics efforts are discussed, including recent work by our group to integrate two broad-coverage microarray platforms. We conclude by predicting experiments that will become possible with new anticipated tools and data, including the porcine genome sequence. We emphasize that the need for bioinformatics infrastructure to efficiently store and analyze the expanding amounts of gene expression data is critical, and that this deficit has emerged as a limiting factor for acceleration of genomic understanding in the pig. 1. Overview Dramatic advances in our understanding of the porcine transcriptome have occurred over the past decade, and especially in the past few years. The pig transcriptome has been analyzed to address biomedical, agricultural and fundamental biological questions, using more and more sensitive and comprehensive tools. Few reviews of porcine transcriptional profiling have been published. Blomberg and Zuelke 1 recently outlined several techniques that have been used for porcine profiling, although their focus was on serial analysis of gene expression (SAGE). This review will specifically discuss results in pigs using the major technologies of EST sequencing; PCR-centric screening, analysis and assay approaches; and array hybridization methods. Due to space limitations, in this review we will not describe the technical aspects of each technique in detail; there are many reviews and specialized resources for cDNA library and EST production; differential display PCR; SAGE; suppression subtractive hybridization (SSH); real-time quantitative PCR; and of course microarray based techniques. Further, many reports have been made using techniques such as differential display or microarrays to identify genes differentially expressed in biological states of interest. The majority of these reports use a porcine tissue or cell line to ask questions of biomedical interest, and the use of porcine-derived biological material is not directly relevant to the study. Space limitations require us to describe in detail only those efforts with broad interest to the pig genomics community. All papers found that differential display screening or the use of microarrays in assaying expression in porcine tissues report are listed in Supplementary Tables 1 and 2, respectively. Reports on EST sequencing will be discussed, but we will not discuss publications that focus on physical and/or genetic mapping of such ESTs. Finally, while still in its infancy, we will discuss published and other publicly reported work on the development and use of bioinformatics and databases to analyze porcine gene expression data. 2. Tools, Techniques and Results for Porcine Transcriptome Analysis Porcine Expressed Sequence Tag Projects describe Significant Portions of the Swine Transcriptome Soon after Adams and co-workers 2 suggested that the sequencing of random clones from cDNA libraries would be an effective means to obtain human gene sequence data rapidly, researchers have been reporting porcine ESTs 3, 4, 5. Many groups have now submitted EST sequence data; Table 1 shows the projects with major contributions to global EST sequencing efforts (totaling over 564,000 entries). The NCBI UniGene website for pig (http://www.ncbi.nlm.nih.gov/UniGene/UGOrg.cgi?TAXID=9823) shows over 100 libraries have more than 250 EST sequences deposited (over 150 additional libraries with less than 250 sequences/library have been reported). These sequences come from thirteen categories of tissues, ranging from a low of 3 libraries found for the adipose, brain, and conceptus categories to 31 libraries in the genito-urinary category. In NCBI Build #25 (as of July 30, 2006), the mRNAs (4,410), high-throughput cDNAs (1,153) and ESTs (463,885) have been clustered into 37,861 UniGene sets. On the other hand, a slightly earlier build of the Porcine Gene Index (SsGI Release 12.0; June 20, 2006 (recently moved from TIGR to Dana Farber Cancer Institute; http://compbio.dfci.harvard.edu/tgi/cgi-bin/tgi/gimain.pl?gudb=pig) shows 575,730 ESTs and 6,854 expressed transcripts from a total of 257 cDNA libraries. This release identifies 64,746 tentative clusters (TCs) and 88,117 singleton ESTs and ETs. Many of these ESTs have been submitted within the past three years, showing the great strides taken recently; since January 2004 (Release 8.0), the number of TCs in the Pig Gene Index has nearly tripled. Additional information on these databases and others is discussed in the section on bioinformatics. In addition to the many uses for such sequencing data that can be made in structural biology and in mapping genes, this large amount of data has also been used to estimate the level of expression of a gene by calculating the number of instances when the transcript has been randomly sequenced in a library. While care must be taken in such comparisons to use non-normalized libraries and EST data of sufficient size, several authors have discussed such estimates. Davoli and co-workers 6 used EST sequence frequencies to compare to equivalent data for human transcripts to characterize their skeletal muscle library; similar estimates were performed by Smith and others 7 to identify highly expressed genes in early embryo. Several groups 8, 9, 10 have used data from non-normalized libraries to estimate such “virtual expression patterns” across multiple tissues or developmental stages. Using sequence frequencies, we were able to identify genes that are expressed highly selectively in one tissue, as well as genes changing expression during embryonic development through calculating frequencies in libraries created from different embryonic stages 8. Jiang and colleagues 9 identified sequence clusters that were over-represented across specific stages of ovarian follicle development and between growing follicles and corpora lutea. Whitworth and others 10 recognized genes showing differential frequency in very early embryogenesis, finding many transcripts with different frequency in comparisons between unfertilized oocytes and 4-cell embryos, and between these embryos and the more differentiated blastocysts. Interestingly they found transcripts with different frequency between in vivo and in vitro produced four cell embryos as well as between in vivo and in vitro blastocysts. Differential frequency has also been used by several groups 8, 11, 12 to determine the efficacy of normalization of cDNA libraries, which is performed to remove the highly expressed (and therefore high frequency) transcripts, allowing for increased efficiency in finding new gene transcripts through random EST sequencing. Finally, both UniGene and SsGI (now the Dana Farber Cancer Institute Pig Gene Index) provide global views of gene expression levels based on EST frequency (see later section). Through comparison of frequencies of ESTs across species, it is possible to identify species-specific differences in expression. Zhao et al. 13 identified a highly expressed transcript in porcine placenta (1 % of ~4,500 placenta and early conceptus sequences available) which had no ortholog in the human EST database, although hundreds of thousands of placenta sequences had been deposited. A highly similar sequence was present in the human genome, with a conserved open reading frame. Further cDNA sequencing and RACE-based transcript mapping analysis of this gene (PLET1) in human, mouse and pig placenta was performed and it was determined that the human gene is expressed, but not spliced due to mutations in splicing signals, thus very few transcripts for human PLET1 appear in cDNA libraries. This study is one of the first examples of the use of porcine EST frequency data to identify a splicing-defective gene in the human genome 13. Serial Analysis of Gene Expression in Porcine Tissues Another sequencing-based method to estimate transcript abundance is serial analysis of gene expression (SAGE). This is a powerful approach whose sole purpose is developing a statistically robust estimate of the relative amounts of expressed sequences in the genome of interest; no cDNA clones or libraries are created. Double-stranded cDNA is digested with specific restriction enzymes, adapters are ligated to cDNA fragments, and the resulting fragments are concatenated and sequenced extensively to obtain sequences for tens to hundreds of thousands of 10-20 bp sequence “tags”. These tags can be mapped back to known cDNA sequence and the resulting data is analyzed for relative abundance. Blomberg and colleagues 14 have applied SAGE to determine the content and expression pattern of developing porcine conceptus at the critical stage of peri-implantation, where the developing conceptus goes through a remarkable morphological elongation stage just prior to implantation. They identified over 400 putative transcripts with differential expression between ovoid (before elongation) and filamentous concepti, and identified several pathways involved in this process, including steroidogenesis and oxidative stress response. In a follow-up paper, they describe the production of SAGE libraries for tubular conceptus and a comparison of non-amplified and amplified SAGE libraries, which showed that amplification, a useful procedure when tissue is limiting, can be used to generate SAGE data accurately reflecting transcript levels similar to non-amplified libraries 15. They then compared expression patterns between this intermediate conceptus form and the ovoid and filamentous expression patterns, and identified over 600 genes with putative differentially expression between at least two stages. Genes differentially expressed at all stages had GO annotations for involvement in cell cycle, cellular organization, cell-cell interaction and general metabolism. Specific differentially expressed or constitutively expressed genes involved in tissue remodeling, glycolysis, cell cycle and tissue/cell type differentiation were further studied; many patterns, but not all, were confirmed by QPCR. In most cases of dissimilar results between SAGE and QPCR, the QPCR results was in agreement with one but not both stage-differences (ovoid versus tubular but not tubular to filamentous, etc.), and the authors remark that these differences may be due to unknown splicing variants which may be resolved once further information on the porcine genome sequence becomes available 15. Screening for Differentially Expressed Genes using Differential Display and Suppression Subtractive Hybridization Technology “Open-ended” techniques such as differential display (DD) and suppression subtractive hybridization (SSH) have been used to profile the transcriptional response to various treatments or to contrast tissues or other variables. Such screening techniques are complementary to “closed” profiling technologies such as microarrays as they allow the discovery of novel, heretofore un-sequenced transcripts. The DD technique uses specific primers to amplify specific subsets (reduced representations) of the total mRNA pool of a cell or tissue, allowing resolution of a limited number of PCR products for each specific primer pair on a sequencing gel. These PCR products can be compared by side-by-side electrophoresis of PCR products amplified from RNAs representing different tissues or treatments to be compared/screened. Research using DD to screen porcine transcripts have been published starting in 1996 and reported from 1-20 differentially expressed products/genes. The majority of these reports (over 20) focused on biomedical research question, using porcine cells or tissues. Due to space limitations, these papers will not be discussed here, although all papers are listed in Supplementary Table 1. On the other hand, papers reporting the characterization of larger sets of genes, from about 50 to 200 or more genes, began appearing in 2001. These reports describe the use of DD to study response to viral infection; liver-specific expression and hepatic response to dietary changes; and gene expression changes in response to selection for improved reproductive traits. These six papers will be summarized here. Wang and colleagues 16 identified 48 total ESTs that increased (35) or decreased (13) their expression in alveolar macrophage cells as a result of porcine reproductive and respiratory syndrome virus (PRRSV) infection. Differential expression for seven genes was tested and all confirmed by Northern blot analysis of infected or mock-infected macrophages. Of interest was the finding that the majority of the annotated genes in this group were mapped to metabolic and physiological pathways; only one EST was clearly annotated as an immune response gene (inflammatory mediator gene chemotactic factor II) and this gene was down-regulated by PRRSV 16. In a recent report, also investigating host response to viral infection, Bratanich and Blanchetot 17 used DD to study genes differentially expressed in lymph nodes between healthy piglets and those suffering from porcine multi-systemic wasting syndrome (PMWS). These authors sequenced and confirmed nine DD products that were found only in the affected animals or showed an increase in PMWS animals as compared to controls. Northern analysis of two genes, hyaluronan-mediated motility receptor (RHAMM) and RNA splicing factor, further confirmed the expression response for these two genes 17. Two groups have successfully applied DD to study liver gene expression. Ponsuksili and co-workers 18 used DD to screen for genes selectively expressed in liver, in order to identify candidate genes for metabolic traits. In a comparison of normal liver and nine other tissues, they found 240 candidate EST fragments that were expressed only in liver or predominately in liver with up to 4 tissues showing expression. These fragments represented 200 known (91) or unknown (109) transcripts. Most of the genes represented by multiple clones were known to encode proteins that are secreted by the liver into the bloodstream. Multi-tissue Northern blot analysis confirmed the expression pattern for complement component 3 (C3), while semi-quantitative PCR confirmed C3 and an additional 14 genes 18. A second group 19 exploited the DD technique to find genes whose hepatic expression was altered by feeding pigs on a restricted protein diet formulated using either casein (CAS) or soy protein isolate (SPI). This group identified and confirmed 86 differentially displayed PCR products, of which 44% has similarity to known genes. Nearly all of the known genes were annotated to pathways expected to be involved in the restricted diet, including protein and amino acid metabolism, oxidative stress, regulation of gene expression, fat and energy metabolism and others. Schwerin and collaborators further showed that six genes involved in stress response were differentially expressed between pigs fed the two diets, with five of six genes showing higher liver expression in the SPI-fed pigs over the CAS-fed pigs. The authors proposed that one reason for this stress response in the SPI-fed animals is the amino acid imbalance in this diet, as compared to the CAS-based diet 19. Finally, two reports looking at gene expression changes due to selection for improved reproductive traits were published recently. Gladney and colleagues 20 used DD of ovarian follicles to identify genes differentially expressed in control animals as compared to animals in a population selected for improved litter size based on ovulation rate and embryo survival. Tissue was collected at 2 or 4 days post prostaglandin F2alpha injection, which was used to synchronize follicular development. Overall, 84 DD products were identified and sequenced; the majority of which were predicted to differentially expressed although several equally expressed transcripts were also characterized. Four genes were confirmed as DE by Northern blot. Of interest was the finding that calpain I light subunit (CAPN4) expression was lower in the select line as compared to controls. As the similar small subunit calpain gene ACG-2 has been linked to apoptosis, the authors speculate that a decrease in CAPN4 expression in the select line could lead to decreased apoptosis and to a larger number of follicles available for ovulation 20. The same group published a companion paper, using DD to look at changes in anterior pituitary expression in the same selection line and controls 21. Anterior pituitary tissue was collected similarly to the follicles above, and DD analysis was performed on tissue pooled across PGF2alpha treatment days. A total of 162 bands were successfully sequenced, representing 125 distinct sequences, a majority of which (58%) had sequence similarity to annotated genes. Three genes were validated by Northern blot, demonstrating that follicle stimulating hormone beta (FSHB) increased expression in the selection line, while both ferritin heavy chain and G-beta-like protein genes decreased expression relative to controls. The former finding is interesting, as FSHB has a direct role in increasing the development of mature ovarian follicles, and this clearly indicates FSHB may have been a direct target of the selection applied to this population 21. Another useful approach to gene discovery is SSH. This approach uses hybridization of an excess of “driver” cDNA (from one state/treatment) to remove sequences in the “tester” cDNA (from the other state/treatment) that are present in both populations in similar amounts. Highly differentially expressed sequences can be enriched for by preferential recovery of the resulting set of un-hybridized “tester” cDNAs. The tester and driver cDNA populations can be reversed, to allow screening for both up- and down-regulated genes. This technique was used by Narayanan and co-workers to 22 identify genes with altered abundance at different stages of the conversion of hepatocyte cells in culture into spheroids, an aggregate of cells that is functionally similar to liver tissue. This group identified 65 genes and 14 novel sequences and deposited 70 ESTs; demonstrating the down-regulation of cytochrome P450 family members as well as genes involved in heme biosynthesis, and the up-regulation of acute phase proteins and genes encoding proteins involved in Ca-dependent vesicle trafficking. Ross and co-workers 23 have used SSH to screen for genes differentially expressed during porcine conceptus elongation, a process important for efficient implantation in the pig. This group reported 142 sequences that changed expression levels during elongation; they deposited 16 distinct sequences to GenBank. Decreased expression during elongation was observed for several ribosomal RNAs, and an increase was seen for IL1B, thymosin beta 4, mitochondrial proteins, HSP70 and S-adenosyl homocysteine hydrolase (SAHH). The latter two genes, as well as an un-annotated transcript, were verified by QPCR. SAHH is part of the pathway to generate folate, an essential metabolite for embryogenesis, which is known to increase in the uterine lumen at the elongation stage of conceptus development 23. Other studies of reproductive biology have used SSH as well. Bonnet and colleagues 24 isolated transcripts responding to follicle stimulating hormone (FSH) treatment of granulose cells in culture. Sixty-four independent sequences predicted to respond to FSH were further analyzed and 25 transcripts were confirmed by Northern or QPCR to be regulated by FSH. Annotation of these 25 genes shows FSH may alter pathways controlling peroxidase activity and chromatin remodeling 24. To study the mechanism of heterosis, Xu and collaborators 25 exploited SSH to identify genes expressed differently in Longissimus dorsi tissue between Yorkshire dams and their Landrace x Yorkshire offspring. The HUMMLC2B gene was identified and confirmed by semi-quantitative PCR methods to be higher in the dam compared to her cross-bred offspring. The HUMMLC2B gene was expressed equally throughout muscle development (fetal to 6 month old samples were tested) and expression was found primarily in skeletal muscle tissue. As HUMMLC2B has been implicated in Ca+2 dependent signaling pathways, and a down-regulation was seen in the F1 hybrid offspring, these results may indicate Ca signaling influences porcine heterosis traits 25. Finally, researchers screening for genes responding to Salmonella enterica serotype Choleraesuis (SC) inoculation have reported the use of SSH to identify 88 sequences that change expression at 24-48 hours post-inoculation in mesenteric lymph nodes 26. Seven SSH-identified and two related genes were selected for further study and all were verified by quantitative real-time PCR. The differential expression of these genes indicate that modulation of cytoskeletal components and heat shock pathways are involved in the host immune response to SC in the porcine mesenteric lymph tissues 26. Quantitative PCR Methods as the “Gold” Standard in Confirming Differential Expression Results There have been many reports of quantitative real-time PCR (QPCR) being used to measure the abundance of porcine transcripts. Many recent publications focus on a small number of genes in the context of verifying differential display results 27, 28, as well as microarray or other profiling data (see specific results in the following microarray section), as QPCR has been the standard technique used to validate such results. This comparison is not as direct as one might wish, as properly run QPCR measures the abundance of a single transcript or transcript segment. Thus microarray data, which may result form hybridization of multiple transcript isoforms to a probe or probes on an array, may not be confirmed by a QPCR assay even though the microarray data is not necessarily incorrect. In fact, some microarray-based technologies, such as the Affymetrix platform, have multiple oligonucleotides that interrogate multiple locations along a gene transcript. Thus data may be available to identify differential abundance of different segments of transcripts, allowing measurement of differential expression of splice variants. On the other hand, QPCR is clearly more sensitive and often shows an increase in the level of difference (fold change) between the class or treatments being compared than seen in microarray data. It is therefore important to recognize both the strengths and weaknesses of each of these techniques in such confirmation studies. QPCR is traditionally not viewed as a high-throughput screening tool, due to lack of a large sets of specific assays available for porcine transcripts as well as the relatively high cost per gene and biological sample assayed. Recently, however, a large number of validated QPCR assays have become publicly available (http://www.ars.usda.gov/Services/docs.htm?docid=6065) through the efforts of a collaborative group headed by Harry Dawson and Joan Lunney at USDA-ARS-Beltsville. This group has developed and tested 474 QPCR assays based on the Taqman fluorescent dual-labeled probe technology (PIN database v. 3.7 as of October 2006; H. Dawson, personal communication). These assays focus on genes in immune and nutrition/metabolic pathways, but include assays for many signaling molecule of general interest. Dawson, Lunney and co-workers have used such high-throughput QPCR to assay expression of many genes responding to pathogenic parasites 29, PRRSV vaccination 30, and have contributed nearly 150 QPCR assays to recent microarray confirmation studies by our group [31, 32, Wang et al., submitted] described below. Other groups have used a significant number of QPCR assays to validate microarray data, including Passerini et al. 33 (27 genes); Ponsuksili et al. 34 (10 genes); and Okomo-Adhiambo et al. 35 (10 genes); these papers are discussed in the microarray section. The QPCR technique can, of course, be used to measure gene expression independently of microarray confirmation. Examples where 4 or more genes were quantitatively assayed with or without microarray data include measurement of response to LPS 36, to infection 26, 29, 37, 38, or to islet graft rejection 39; measurement of promoter activity in vitro 40, 41; assessing expression of gene family transcript isoforms 42; measuring responses during parthenogenesis 43, 44 or oocyte/conceptus development 23, 45; brain response to early weaning/isolation 46, 47; multi-tissue response to carnitine treatment 48 or liver response to dietary treatments 28, 49; differences across specific muscle types 50; and differences between stented and unstented arteries 51. Microarray Hybridization Studies have Dramatically Expanded our Knowledge of the Porcine Transcriptome The use of miniaturized arrays of individual gene sequences to survey or “profile” the expression levels of hundreds (later; thousands) of transcripts within a cell or tissue was first published by Schena et al. 52, although reports of larger dimension element arrays (dot blots, etc.) had already been published, including early work on porcine muscle EST expression patterns 3. The main technological advances by Schena and colleagues were 1) the use of robotics to place very small amounts of gene sequences very closely together, minimizing reagent use and allowing the use of standard size glass microscope slides as the solid substrate, and 2) the independent labeling, by fluorescent substrates, of the of the RNA samples to be compared, allowing for simultaneous measurement of hybridization of labeled cDNA to each probe on the array. Initially, the criteria for finding DE genes in the comparison at hand was limited to an intuitive but statistically unsupported two-fold difference in fluorescence levels. Later, additional experimental designs and data analyses became statistically rigorous. Such analytical methods have become a significant field unto themselves 53. Much of the data and publications described below are in the very early stages of understanding RNA expression profiles, as the field of transcriptional profiling is relatively new, and especially so for work in the livestock species. Many early papers were limited in scope, with relatively few elements and scarce biological replication, which is critical for adequate statistical rigor. Initial insights as to new aspects of tissue and cellular function have been gleaned from pig microarray data, and connections between datasets are made where possible and appropriate. However, an increased depth of data as well as more comprehensive tools will be required to fully comprehend and integrate the massive amounts of data already collected across many different tissues, systems and perturbations. Different platforms for porcine expression profiling have been used; these are listed in Supplementary Table 2 and briefly described below. Most of this work used two-color analyses, labeling RNA samples using fluorescent Cy3 or Cy5 conjugated nucleotides and performing a dye-swap, in which samples are labeled with each dye and used in separate hybridizations to eliminate dye effects. Cy3 and Cy5 have different excitation/emission wavelengths, allowing for independent collection of hybridization data from each sample on the same array element. Further, most work described below used a normalization method called LOWESS, (Locally Weighted Scatterplot Smoothing), which corrects for intensity-dependent variation in data. These methods will serve as the default in the discussions below and are not described; however, if some other system and analysis is used, it is mentioned below. Statistical methods to identify DE genes across treatments vary, although ANOVA methods and corrections for multiple testing using false discovery rate calculations appear to becoming the standard. As these methods greatly affect the results, they are noted in the descriptions of experiments. Finally, annotations of DNA elements on the array are an important and growing area of research. Such annotations are most often based on sequence similarity to genes whose function is known or has been inferred, in turn, from other gene with similar sequence in another species. Functional annotations can also be obtained from curation of available biological data as complied for genes and pathways at NCBI (http://www.ncbi.nlm.nih.gov/Entrez/query.fcgi?db=gene) or at Mouse Genome Informatics; http://www.informatics.jax.org/). At the Gene Ontology (GO) Consortium site (http://www. geneontology.org/), annotation of a gene's function, where in the cell it performs this function, and its involvement in specific biological processes, can be found. Software to assign GO annotation to gene sequences, such as the DAVID software at NIH (http://david.abcc.ncifcrf.gov/), are very useful and many are freely available. Such functional annotation can be especially helpful in annotating a set of gene sequences that respond similarly to specific treatments; such sequence groups or “clusters” can be identified using a number of clustering algorithms; see the review by Quackenbush 53 for a discussion of current work in this area. In the database section below, we discuss some of the methods being used to evaluate the biological knowledge available on specific genes and groups of genes that are found to be co-expressed. Specific methods used by investigators are also noted in the descriptions below. Every effort was made to include work published that used microarrays to determine porcine expression profiles. Seven different groups also graciously provided summary details of fourteen different unpublished microarray projects (see Supplementary Table 2). These unpublished projects include studies on reproduction (5); muscle traits (3); genetic effects on expression (2), and host response to infection (2), antibody treatment (1), or stress (1). 3. Current Expression Profiling Results in Porcine Tissues and Cells using Microarrays Muscle Expression Profiling Microarray studies have been used in several species to better understand the changes in gene expression during livestock muscle growth and development 54, and the pig is no exception. One of the first large-scale profiles of porcine skeletal muscle was published by Moody and colleagues 55. They used a nylon membrane spotted with human cDNA fragments and hybridized this membrane with human and porcine skeletal muscle cDNA. They found that approximately 48% of the spots provided no signal; however, they also established that the hybridization signals were reproducible within sample for both human and pig RNA, and that concordance of results between species high. As this was a test of cross-hybridization, no expression contrasts were reported, but the authors concluded that cross-hybridization appeared promising 55. Another group has used a cross-species approach to investigate porcine skeletal muscle gene expression. Lin and Hsu 56 used a human cDNA microarray to identify genes differentially expressed (DE) between Duroc and Taoyuan Longissimus dorsi (LD) tissue. They identified 6,400 DE genes with signal 2.5 fold higher that background for both breeds. They used correlation analysis to show a high level of reproducibility as had been found by Moody et al. Among these expressed genes, 117 were found to differ between breeds (p < 0.05), less that one would find by chance (~ 460 genes). Nevertheless, of the eight genes selected for verification by QPCR, all were confirmed. The majority of DE genes had higher expression in Duroc pigs, including genes for ribosomal proteins, heat shock proteins and myofibrillar proteins, and genes involved in transcription/translation and metabolism-related genes. The authors indicated that the higher proportion of proteins involved in myofibrillar structure in DE genes in Duroc over Taoyuan could explain Duroc higher muscle growth. They further showed that SLIM1, a known myogenesis control factor, was over-expressed in Duroc, providing further clues as to the increased muscle growth rate in Duroc 56. A different pioneering approach for skeletal muscle profiling was reported by Bai and others 50, where a porcine microarray, consisting of un-sequenced cDNA clones from fetal and neonatal LD tissue libraries, was used for expression profiling of LD (as a white muscle type) as compared to psoas muscle (as a red muscle type). Genes identified as having muscle-type specific expression were defined as those having a normalized psoas/LD ratio of > 2.0 or < 0.7, and these cDNAs were then sequenced and annotated. Genes with a high psoas/LD ration (70 clones) included 16S and 18S ribosomal RNAs and NADH dehydrogenase subunits 3 and 6, fructose-1,6 bisphosphatase and members of the casein kinase 2 complex or targets of that pathway. Many genes highly expressed in LD (50%) included many of the fast isoforms of muscle fiber proteins and about 25% included genes involved in glycolysis such as GAPDH. In the last 25%, the authors highlighted the tumor suppressor gene, bin1, as it has been linked to myoblast differentiation in culture. Four genes predicted to be DE (GAPDH, bin1, MyHC2b and one novel gene) were tested by QPCR and all were confirmed, although bin1 was higher in LD in only 3 or 4 pigs tested. The authors propose that this method of arraying non-sequenced clones is useful in finding genes DE in species with little genome characterization 50. However, as this lack of information eases, this advantage becomes less important and the inefficiencies on spotting duplicate unknown cDNAs become significant. Members of this group used the same array to investigate the molecular changes in response to dietary restriction (20% less protein, 7% less energy, termed the LPE diet) in porcine LD and psoas muscles 57. In both muscle types, the LPE diet resulted in higher intra-muscular fat and caused twice as many genes to show higher expression as compared to the control diet treatment. The authors sequenced these DE genes and identified genes involved in turnover of protein, fat and carbohydrate, as well as genes involved in growth, mitochondrial function, translation and glycolysis. QPCR analysis verified the microarray data for several genes and indicated several systems were involved: the increase in intramuscular lipid (phytanoyl-CoA- hydroxylase and delta 9 desaturase); energy (creatinine kinase); fiber type specificity (MyHC2a, 2b); and muscle growth (cbl-b, kc2725, P311). The P311 gene, increased by LPE diet in both muscle types, is interesting as the authors further showed that P311 was increased during C2C12 myoblast differentiation 57. The use of sequenced porcine ESTs to create an expression profiling tool was first published by Zhao and others 58. This group initially arrayed known ESTs and tested for detectable hybridization signal using muscle RNA from two fetal ages [day (d) 75 and d105] and two postnatal ages (1 week and 7 week) as targets. The ESTs that provided a clear signal were re-arrayed to generate the test array, which was then used to profile the fetal (d75) and postnatal (1 week) targets. A mixed linear model was employed to identify 28 genes with stage-specific expression (defined as two fold difference in expression between stages with P < 0.01). Of these, 19 genes matched known human genes, and included elongation factor 1 alpha, a number of ribosomal proteins (12), GAPDH, and structural proteins such as vimentin and tubulin. Five of these genes were tested and confirmed by Northern blot analysis of RNA representing seven stages of muscle development, from d45 to 7 week postnatal. The results with intermediate filament (IF) protein family members vimentin and desmin (tested as an additional control) confirmed protein-level data 59 that vimentin expression dramatically decreases during prenatal muscle development while muscle-specific IF protein desmin increases. The authors conclude their membrane and radioactive-labeling approach is a low cost and accurate expression profiling method to find DE genes of biological interest 58. Two papers that appeared recently studied myogenesis by using an array containing cDNAs selected for their roles in myogenesis, energy metabolism, and myofibrillar structure, as well as additional skeletal muscle ESTs. The first paper 60 reported the expression pattern for these genes during seven stages of gestation, d14 to d91, covering the two stages of myogenesis in the pig (d30-60 and d54-90). Taking advantage of known annotations, genes were divided into functional groups for analysis. Genes both stimulating and inhibiting muscle differentiation as well as muscle structural genes had a peak of expression at d35, and declined thereafter. Genes in the glycolysis pathway decreased early then had a peak of expression at d77. The number of genes showing expression from d14 to d91 increased for the myofibrillar group, while the number of expressing myogenesis-affecting genes and differentiation-stimulating genes decreased from d14 to d49, then increased from d49 to d91. The expression pattern of five genes presenting the major annotation groups (myogenesis: EPO-receptor, beta-catenin, and TGF beta2; energy metabolism: GAPDH; and structural: COL3A1) was tested and confirmed by QPCR, although statistical analysis of these data was not described 60. In a second paper, Cagnazzo and collaborators used the same microarray and tissue sampling to compare expression between the Duroc and Pietrain breeds 61. The authors found that energy metabolism genes were consistently higher in Pietrain over Duroc for all ages except d35 in which this ratio was reversed. On the other hand, fatty acid metabolism genes had the opposite profile: higher levels in Duroc from d14 to d49, then higher in Pietrain at later ages. They found myogenesis apparently initiated earlier in Duroc pigs, as the expression of proliferation and differentiation genes was higher at d14 and d35 in this breed; this was reversed at later stages. Muscle structural genes also showed higher expression in early Duroc embryos, and after d49 expression was higher in Pietrain samples. The same five genes as assayed in te Pas et al. 60 were tested by QPCR and all results were consistent with the microarray data, although, as for the te Pas paper, no statistical analysis was discussed. The authors conclude that, as a delay in fiber formation has been associated with higher fiber numbers in other species, their finding that Pietrain myogenesis is delayed relative to Duroc may explain the Pietrain higher adult muscle mass 61. Two reports have been published looking at porcine heart gene expression as a model for human disease. Lahmers and co-workers 62 used a novel array consisting of 50 mer oligonucleotides representing each of the 363 exons of the enormous human titin gene. They investigated rabbit, mouse, rat, and pig myocardial gene and protein expression from adult left ventricular tissue. In preliminary validation using pig soleus adult tissue, they found that 92% of the exons expressed in human soleus were also expressed in the corresponding pig muscle. For pig myocardium, the authors reported ~20 exons were fetal-specific and these were very similar to human fetal-specific exons 62. In the second report, Lai et al. 63 used a human cDNA microarray to determine genes whose expression changes during atrial fibrillation. Nearly 500 genes showed DE, with genes annotated in cell signaling and cell communication over-represented, while the second largest group were genes involved in gene regulation. A single gene, MLC-2V, was tested and confirmed by QPCR. The authors indicate that many of these genes had not previously been associated with atrial fibrillation, and thus further study on these genes is warranted 63. Reproductive Tissue Expression Profiling A number of papers have been published in the past two years on the use of microarray to study reproductive tissues, primarily the ovary (3) uterus (1) and testis (1), but also on the developing embryo/conceptus (2). As described above in the DD section, Gladney and colleagues 20 have studied the expression profile in ovarian follicles collected from animals selected for improved litter size and in a random-bred control population. In addition to DD, microarray analysis of pooled follicles from these two populations was performed by using two different Incyte UniGEM human cDNA microarrays. Based on two-fold differences only, they identified 33 and 21 DE genes using these microarray platforms, and found little overlap in the two lists of genes. Northern blot analyses of three genes (follistatin (FST); early growth response 1, nuclear family receptor 4A1(NR4A1)) confirmed the microarray results for FST and NR4A1 20. This group has built on this early work, using a porcine cDNA microarray to profile the expression pattern of ovary and follicle tissue from the select and control lines of pigs 64. A mixed-model analysis identified 72 ovary and 59 follicle genes DE between these two lines. In contrast to the data obtained with the human microarrays, 32 genes (about half of the total) were found DE in both tissues, providing a measure of confidence in these results. Northern hybridization using probes for calpain light subunit I and cytochrome p450scc (P450scc) genes were used to confirm the microarray data for both genes. Overall, the lists of DE genes contained both genes known to be expressed during folliculogenesis as well as genes not previously associated with this process. The authors highlighted DE genes involved in steroid biosynthesis (Collagen type I receptor, P450scc, STAR, 3betaHSD, CYP17 and CYP19) and tissue remodeling (PAI1) that could provide mechanistic clues for the difference in ovulation rate seen between these lines, and they indicate further ontogenic studies of these candidates would be helpful in such interpretations 64. Gene profiling was used successfully to study the process of luteinization by comparing pre-estrous preovulatory ovarian follicles to luteinized follicles on day 2 of estrous prior to ovulation 65. A custom porcine cDNA microarray was created from clones selected from the ovary EST library project described above 9. Microarray data collected on these two types of follicle was produced using a reference design; the reference was an RNA mixture from all follicular stages, and data normalization and the Welsh t-test were used to identify 150 DE genes (P < 0.01) between follicle stages. The authors reported the major classifications for the 107 genes decreased from pre-estrous to luteinized post-estrous follicles were cytoskeletal structural and regulatory proteins, chromatin component and nucleic acid-binding proteins, metabolic enzymes, oxidative response proteins, cAMP receptor pathways proteins, and cell proliferation/differentiation pathways proteins. For the 43 genes increased in the same comparison, the major annotation groups were cell adhesion; migration, growth inhibition; and angiogenesis. Some functional categories were observed in both lists, including different genes involved in steroidogenesis, proteolysis and metabolism. In validation work, four genes (CYP17A1, 3beta-HSD, LHCGR, and PLANH1) were tested by QPCR and results showed similar patterns to the microarray data, although statistical evidence was not discussed. The authors note that many genes, whose function and expression patterns were known previously to be involved in folliculogenesis, were identified in this analysis and this further validates their results. Many of the identified genes were novel with respect to luteinization, however, and are interesting candidates for future study of this process 65. Two manuscripts have been published on expression profiling in the developing porcine embryo. Lee et al. 66 used a custom cDNA membrane array to initiate microarray studies on the elongating, peri-implantation embryo, a morphological transition critical for successful implantation. Concepti from four stages of this transition (small spherical, large spherical, tubular and filamentous forms) were collected and profiled using the above array. The minimal quantities of embryo tissue available required an amplification step; preliminary hybridizations with un-amplified and amplified material indicated amplification did not introduce bias. Nine genes were declared as DE using three criteria. Two genes met the stringent Bonferonni test for multiple testing, four genes were DE at p <0.001, while the remaining three genes had p < 0.01 and at least a two-fold difference in expression between stages. Four genes covering these three categories (STAR, TGFbeta3, interleukin 1 beta (IL1B), and thymosin beta 10) were selected for QPCR confirmation and all four showed statistically significant differential expression in both QPCR and hybridization data. The IL1B results confirm reported SSH 23 and EST frequency data 7, 8 and the thymosin beta 10 results agreed with data reported by Smith et al. 7. Further, the STAR gene has been shown to increase in expression by SAGE data 14 and by follow-up Northern, QPCR and protein level analysis by the same group 15. These confirmations verify the results of the Lee et al. study 66, which contributed additional genes for understanding of the rapid transition spherical to filamentous conceptus which is nearly unique to the pig. Whitworth and others 67 developed a custom cDNA array containing genes collected from cDNA libraries from ovary, embryo, oocyte, oviduct, uterus, conceptus and fetus, to determine genes DE across several stages and modes of embryogenesis. These included in vivo produced germinal vesicle oocytes (pgvo), four cell embryos (p4civv) and blastocysts (pblivv), as well as in vitro produced four cell embryos (p4civp) and blastocysts (pblivp). An ontogenic study from oocyte to blastocysts (all in vivo samples) showed ~ 2,000 to 4,500 DE genes. A comparison of in vivo to in vitro embryos identified 1,409 (p4civv versus p4civp) and 1,696 (pblivv versus pblivp) DE genes. False discovery calculations for the latter two comparisons eliminated nearly all of the DE genes, and a condition tree clustering did not show evidence that in vitro expression patterns could be easily distinguished from the in vivo patterns. In the ontogeny study results, similar clustering showed each stage correctly clustered together and that, as expected, the two earlier stages clustered more closely together. QPCR validation of seven genes from these comparisons was performed and of the 21 pair-wise comparisons available in the microarray data for these seven genes, 20 were confirmed by QPCR. The authors found that genes induced in the transition from germinal vesicle to four cell embryo were similar to those reported for the same transition in the mouse, and included nuclear structural and functional proteins, while genes decreased during this transition included those involved in cell adhesion receptor activity, mitotic cell cycle, transporter activity and M-phase microtubule activity. Genes increased in blastocyst as compared to 4 cell embryos included ribosome, hydrogen ion transport and cation transport activities. Many examples were discussed showing much of this data is similar to that reported in the mouse, although many differences between pig and mouse profiles were observed. The authors correctly note that this analysis is just the beginning of understanding the transcriptional program during early embryogenesis 67. This group has also used this array to investigate gene expression changes during the estrous cycle and pregnancy in the porcine endometrium 68. They collected RNA at seven times during the estrous cycle (0, 3, 6, 10, 12, 14, and 18 days post-estrous), and used a reference design to identify genes DE across the estrous cycle. The reference RNA consisted of a mixture of RNA from the tissues used to make the array, see above. Genes were identified as DE using ANOVA with Benjamini-Hochberg correction to control for multiple testing, with the added criteria that the gene must be at least twice as abundant as seen in the reference. The number of DE genes varied by day of estrous, with a minimum of 118 genes on day 0 and a maximum of 542 genes on day 12. Clustering of these genes using k-means methods identified six main patterns of expression across the estrous cycle. Several of these patterns (i.e., DE on day 0; DE on days 3 and 6; DE on days 10-14) are coincident with specific known functions of the endometrium: a) sperm maturation; b) blastocyst growth and position; c) conceptus development and attachment, respectively. They also used EASE software to connect GO terms to these clusters, where, for example, in the day 0 cluster immune cell markers and cytokine genes predominated; within the day 10-14 cluster, many DE genes were annotated with tyrosine receptor kinase activity function. Using cluster and EASE results as a guide, seven genes were selected for QPCR based validation; all resulting QPCR data was consistent with the microarray results, although statistical significance of the QPCR results was not discussed. The authors conclude that these data on the global expression patterns with the cycling uterus will guide transgenetic and cell transfer approaches to improve reproduction efficiency 68. Finally, Stewart and colleagues have reported 69 the use of a human cDNA microarray to assess testicular expression patterns in boars differing in steroidogenesis levels. The RNA from animals (n=4) with known high plasma estrone was directly compared to RNA from low plasma estrone boars in paired hybridizations. Genes with statistically significant differences between the two states were identified using Student's t-test and Benjamini-Hochberg correction for multiple testing. Seven genes were found by this method to be more highly expressed in the high estrone boars (p< 0.05). Five of the genes tested were verified as DE using QPCR, and included CyB5, Cyp19A1, SAT, FTL, and DNASE1L1; the authors note that the fold changes were much higher than those observed in the microarray data 69. Immune Response Expression Profiling The immune response is highly complex, with multiple tissues and cell types communicating information that is contextual and changes during early inflammatory stages as well as during inflammatory resolution and the adaptive immune response, and changes can also be long-term through immune memory. Hammamieh et al. 70 have developed useful data on variability among pigs within peripheral blood mononuclear cells (PBMC). Using a human cDNA membrane array, they measured the PBMC expression profiles of ten Yorkshire piglets. They found 19 DE genes, which were annotated as stress response, immune response, and genes involved in transcription. This group then used this same array in a follow-up paper comparing expression profiling of human PMBC exposed in vitro to staphylococcal enterotoxin B (SEB) to profiling data of PBMC from piglets intoxicated with SEB to LD95 in vivo. Using a supervised learning procedure for class prediction to compare these two datasets and by using a separate training dataset of gene expression profiles of human PBMC infected with eight separate pathogens, they were able to identify a set of 16 genes predictive for SEB exposure 71. A separate group has also reported 72 the development of tools for assaying gene expression responses in immune cells. This group selected 20 cytokines, 11 chemokines and 12 receptors relevant to immune response, and developed a cDNA array containing these genes. They then validated the resulting array by interrogating RNAs isolated from control or phorbol ester/ionomycin stimulated PBMC from pigs. Expression patterns obtained through hybridization of radioactively labeled cDNAs to the array followed closely the expected expression patterns published previously for most of these genes, indicating this focused immune array can be used for profiling the porcine immune response 72. Afonso and co-workers 73 have used a cDNA microarray constructed from a swine macrophage library to compare expression profiles from macrophage cultures infected with two forms of African swine fever virus (ASFV). One infection used a parental ASFV (Pr4) while the alternate infection used a mutant virus (Pr4delta35) deficient in two genes that control growth of ASFV in macrophages. RNA was collected at 3 and 6 hours post infection (hpi), and t-test analysis of the 3 hpi hybridization data indicated a total of 38 genes were found to be up-regulated 2 fold or more in the Pr4delta35 infections as compared to the Pr4 infections, while 133 genes were down-regulated. The up-regulated genes, responding only to the gene-deleted virus, included many interferon-regulated genes. Control experiments with mock-infected cultures, as well as Northern blots and or QPCR of 14 test genes, showed this data to be reliable. Interferon regulation of response to Pr4delta35 virus was further confirmed by the expression profile observed for an inhibitor of IFN-induced antiviral protein, which was decreased in Pr4delta35 infected cells 73. Another analysis of blood cell response to infection has been published by Moser et al. 74. This group is interested in finding genes that may control susceptibility to bacterial disease. A microarray containing cDNAs from ten immune cell types and tissues was used to interrogate expression in peripheral blood leukocytes isolated from the two most extreme responders to an Actinobacillis pleuropneumoniae (A.p.) infection. The RNA was collected from all 18 pigs in the infection study just before (T=0) and 24 hpi (T=24). The experimental design was a reference design and used either uninfected leukocyte RNA or a 1:2 mixture of muscle and leukocyte RNA. A sophisticated mixed model analysis was used to identify DE genes from T=0 to T=24 for each pig. The authors reported a list of 128 genes was decreased during infection in the resistant animal and increased in the susceptible animal, and a second cluster of 179 genes with the reverse pattern. The origin of the genes in the first cluster was predominately a cDNA library created by subtraction of liver expressed genes from genes expressed in lymphocytes, indicating an immune system origin. The cDNAs found in the second cluster primarily arose from a library created from A.p. infected leukocyte RNA subtracted by uninfected leukocytes. No specific genes were mentioned; however, the apparent immune response indicated genes relevant to this infection were identified 74. Several papers have been recently published that use microarray analysis to investigate the transcript profile of the intestine and associated lymph tissue. Dvorak and colleagues 75 describe the generation of over 3,000 ESTs from a cDNA library of Peyer's Patch, a mucosal tissue of the intestine, and the development of a custom cDNA microarray from these clones. The microarray was used to assess expression across total library cRNA from un-stimulated Peyer's Patch tissue compared to stimulated Peyer's Patch total cRNA. The latter cRNA was a pool of RNAs from cell cultures stimulated for 3 hours in tissue culture chambers with SC, lipopolysaccaride + cholera toxin, or Phorbol ester + homocyclic AMP + concanavalin A. Additional preliminary hybridizations comparing the normal un-stimulated library, the stimulated library and the pooled and subtracted library were run in duplicate with dye swap. A number of genes were found to be differentially represented among these three libraries, indicating the pooling and subtractions was successful 75. This group then used this microarray to examine expression differences between juvenile Peyer's Patch (JPP) and adult Peyer's Patch tissue 76. The hybridization analyses used a reference design, with the reference being pooled mesenteric lymph nodes (MLN) from young pigs. The Student's t-test with Benjamini-Hochberg correction was used to identify DE genes in JPP compared to reference. Using as minimum fold change value of 1.5 (based on JPP:JPP self hybridization analysis) as an additional criteria, 24 genes were identified as DE, and all but 2 of these were more highly expressed in JPP compared to MLN. Hierarchical clustering of these DR genes across the four individual pig samples indicated significant variation among animals. Seven of these ESTs and thirteen additional non-differentially expressed ESTs were tested by QPCR in JPP, MLN and adult PP. Of 18 ESTs with QPCR data for JPP, there was a positive correlation (correlation coefficient = 0.48) between the JPP/MLN rations for QPCR and microarray. Several genes (AECC, GW112, SPAI, PSP) had much higher rations in the QPCR data (100 to 1,000 fold) than seen in the microarray data. In QPCR comparison between JPP and adult PP, several genes were highlighted. Five genes, MARCKS, CIDE-B, GW112, PSP, and PROLI, were much more highly expressed in JPP as compared to adult PP, while MHCII-DR was more highly expressed in adult PP. The genes CIDE-B, GW112, and PSP are interesting as these are annotated as growth and apoptosis-regulating genes associated with responses to normal and/or abnormal intestinal microflora 76. Additional microarray-based analysis of the intestinal response to bacteria, in this case Escherichia coli and S. enterica serotype Typhimurium (ST), has recently been published by Niewold and collaborators 77, 78. In these reports, this group used a model of in vivo infection, called small intestinal segment perfusion (SISP). They created a cDNA array using jejunal intestine ESTs and have used this array to measure expression in jejunal tissue after infection with an enterotoxigenic E. coli (ETEC) strain, or mock-infected pigs 77. Fifteen genes were declared DE when the fold change in expression between uninfected and infected tissue (within a SISP loop in the same animal) was >4 or <-4 and the false discovery rate, as calculated by significance of microarray, was < 0.02. One of these, PAP, was thirty-fold higher in infected tissue; the other genes were not identified. The PAP expression pattern was confirmed by Northern blot analysis 77. In a second report 78 using the same array and SISP technique, the same group looked at the response to ST. Jejunal RNA collected from several locations in the jejunum after mock-infected or ST infected jejunum at 2, 4 or 8 hpi was used in microarray hybridizations; they pooled tissue samples from 4 animals so unfortunately there was no biological replication. Spotfire software was used to find DE genes with fold differences <-1.58 or >1.58 and p < 0.025 relative to mock-infected jejunal RNA. Seven genes were found DE between either the 4 or 8 hpi samples as compared to mock-infected tissue. Three genes (MMP1, PAP, and STAT3) were DE at both time points, while IL8 and TM4SF20 were DE at only the 4 hpi time, while THO4 and an unknown EST were increased at 8 hpi only. No down-regulated genes were found. One gene (PAP) was tested using QPCR and was confirmed although the data was not shown. The most interesting result was that STAT3, known to function in an immunosuppressive pathway involving SOCS3, was up-regulated and thus could indicate resolution of inflammation was occurring by 4 hours. Alternatively, the authors suggest that immunosuppression by ST may also be occurring, and that such immunosuppression could be a reason for the very low numbers of identified DE genes by this model. They also acknowledge that, as they collected whole jejunal tissue, the DE gene signal may be swamped by the large number of other cell types in the tissue isolated 78. Another microarray was recently used to study the transcriptional response to SC in the gut. Zhao and others 31 used a novel broad-coverage oligonucleotide array (further discussed below in the Tissue Expression Pattern section) to investigate changes in lung expression at 24 and 48 hpi relative to uninfected pigs. A loop design was used in array hybridizations to identify DE genes across these time points. Fifty-seven genes showed DE (P< 0.001, maximal FDR 27%) between uninfected and infected lung. Cluster analysis of these 57 genes showed that, of the 33 genes with annotation, 17 (52%) were related to immune response, apoptosis or tumorigenesis, clearly indicating that the array was useful in identifying relevant genes for this infection. Several genes showed dramatic increases in expression level relative to controls; 25 genes showed >4.5 fold greater expression at 48 hours. These included INDO, IRF1, HSPA6 (known immune response genes), and GBP1, GBP2 and GBP3 (a known interferon-inducible gene family). Interestingly, TGM1 and TGM3, members of a transglutaminase gene family with possible roles in apoptosis and/or antigen processing, increased 31 fold and >1,000 fold, respectively. This gene family has not directly been implicated in the immune response to bacteria in any species, but the known role in apoptosis for TGM genes indicates this pathway is important in the response of the lung to infection. A large number of genes, 61, were chosen for QPCR for two purposes: a) to validate microarray expression patterns (33 genes) and b) to characterize more fully the transcript response to SC (28 genes). The QPCR confirmed 23 of the 33 DE genes tested, and identified six additional DE genes, validating the microarray data from this oligonucleotide array. Overall, the QPCR results showed a strong T helper 1 cell type response in the lung to Salmonella (exemplified by induction of IFNG, IL15, INDO, IRF1, SOCS1, TNF, and WARS). These results also demonstrated a strong apoptotic response (exemplified by TGM3, TNFRSF5, TNFSF6, and CASP1 induction) and an antigen processing response (exemplified by MHC2TA, PSMB8, TAP1, TAP2 induction). The classical complement pathway (C1s, C1r) and type 1 interferon pathway (GBP1, GBP2) were were also confirmed as strongly induced, while decrease of T helper 2 cell type response genes (IL4, TPS1, IL13) by 48 hpi was confirmed 31. These data clearly show the value of the microarray, as well as extensive QPCR analyses, to determine expression profiles during infection that inform us of the conserved and potentially pig-specific biological pathways involved. A first generation Affymetrix porcine GeneChip® with probe sets to assay over 23,000 transcripts, was recently used to study host mesenteric lymph nodes (MLN) transcriptional response to ST [Wang et al., submitted]. Animals were infected with ST and tissues collected after 8 hpi, 24 hpi, 48 hpi, or 21 days post-infection. RNA was used in standard Affymetrix analyses to produce profiling data analyzed by MAS5.0 and a mixed model ANOVA with false discovery rate control to identify genes DE across stages within infection. Results showed that 848 genes changed their MLN expression level across one or more pair-wise time-point comparisons in the ST infection (p<0.01, fold change >2, q <0.24); about 100-150 genes were found DE at each time relative to uninfected pigs. Of interest was the finding that, in contrast to the strong transcriptional response to SC seen in the lung by 48 hpi 31, a limited induction of genes at 24 hpi with ST was observed. In fact, from 24 hours to 48 hours post infection, many genes decreased their expression. To study this further, cluster analysis and analysis of specific pathways were used to reveal common expression patterns for sets of genes, and identify specific features of the host response to ST infection. In Figure 1, the cluster analysis of all genes found DE in the ST infection is shown; one particular cluster (#4) is highlighted. This cluster, the genes within which on average rose in expression by 24 hours and then decreased by 48 hours, contains a large number of cytokine genes and NFkB-dependent genes known to be involved in the inflammatory response. This result indicates that, in parallel to the clinical features of ST infection where inflammation is mild and peaks at 24 hpi, the NFkB pathways appears to be activated between 8 and 24 hpi, and then suppressed thereafter. The expression profile of 22 genes (seven of which are present in cluster #4 in Figure 1) was analyzed by Q-PCR, and 95% showed statistically significant confirmation of the expression pattern observed in the microarray data [Wang et al., submitted]. Toxoplasma gondii is a protozoan parasite that infects a significant portion of the world-wide human population. A secondary host reservoir for this parasite is the pig, and the host interaction between T. gondii and porcine cells was recently investigated using cDNA microarrays 35. The ESTs on the array were selected from immune libraries for likely involvement in immune response, and this selected set was used to assay expression in uninfected kidney epithelial cells (PK13) or after T. gondii infection at eight time points after infection (1 hour to 72 hours). Relative to uninfected cells, a total of 263 genes DE genes identified using Student's t-test were found to be induced, and 48 more were decreased, at one or more times after infections. The majority of the induced genes responded to infection within the first 4 hours, and 12 different functional classes of DE genes responded, including transcription and signaling, metabolism, immune response, cell cycle, and apoptosis. Eleven DE genes were tested and confirmed by QPCR. The authors concluded that this approach identified many classes of genes that will be useful candidates to study cell-mediated responses, especially apoptosis and NFkB-dependent pathways, to this parasitic infection 35. Using Microarrays to Determine Tissue-Selective Gene Expression Patterns and Microarray Applications in Other Research Areas Most of the microarrays discussed above were developed to ask specific questions in biology. When tools for a specific tissue were not available, some researchers tested the utility of human arrays to profile porcine expression (55; see additional papers in Supplementary Table 2). While these cross-species tools can be successful, large-scale porcine-specific tools useful in studying a wide variety of biological questions are available. As discussed above in the infection section, a first-generation porcine oligonucleotide set, representing 13,297 cDNAs and ESTs with broad coverage across tissues, has been designed by Qiagen-Operon in collaboration with researchers in the Swine Sub-committee of the USDA-NRSP8 research project. Zhao and collaborators 32 validated the novel 70-base oligonucleotides on the array by hybridized with targets from porcine adult liver, lung, muscle, or small intestine. A loop design were utilized to collect transcriptome data for each tissue and to identify DE genes across tissues. Using available negative controls (average signal of five Arabidopsis gene oligonucleotides) to calculate a true background level, the large-scale transcriptome for each adult tissue (from 8,358 in muscle to 10,556 in lung) was established as those genes with greater than 3 fold background signal (q < 0.01). Using a criteria of P < 0.001 and q < 0.003, tissue-selective gene lists were produced, from a low of 147 genes in small intestine to a high of 405 genes in liver. Clustering results of the expressed genes identified a number of patterns across these four tissues that are useful in annotating the array ESTs. QPCR analysis of 11 selected genes across the four tissues was used to verify tissue expression and oligonucleotide specificity (multiple members of gene families tested) and showed statistically significant confirmation of all but 2 genes. These results demonstrated that this porcine oligonucleotide array is informative and the oligonucleotide specificity is high, thus the Qiagen-Operon–NRSP8 porcine array can be used for porcine functional genomics analysis 32. A small number of papers have been published recently on explorations of porcine brain (2), liver (2), and adipose (1) tissues. Nobis and coauthors are interested in the expression pattern of genes in the porcine brain and have submitted ESTs from a brain cDNA library constructed from pooled brain regions 79. In addition, they constructed a brain cDNA microarray, and demonstrated with self-hybridizations of normal pooled brain cDNA (and appropriate statistical corrections) that the microarray can provide quality data when a fold cut-off of 2.0 is used to control false positives 79. Members of this group have also used this microarray to investigate frontal cortex expression profiles in early-weaned (EWC) pigs compared to nonweaned (NWC) pigs as well as weaned/nonweaned pigs isolated socially (EWI, NWI) in a 2x2 factorial design 46. Using a replicate loop design to collect data for all comparisons, they found 103 DE genes using a modified t-test that accounts for multiple testing (P < 0.05, fold change > 1.25). Of these genes, they found 24 of 42 annotated genes had relevant brain functions. Six DE genes were selected for QPCR validation based on DE in specific contrasts, including social isolation (NWI vs. NWC; 14-3-3, CPE, and PEA-15), or social isolation in early weaned pigs (EWI vs. EWC; DBI and ARP2/3) and early weaning in the presence of social isolation (EWI vs. NWI; OAZ2). The QPCR results confirmed the first five genes, and indicated that social isolation in either weaned or nonweaned pigs has effects on neuronal gene expression, but that they did not detect gene expression differences due to weaning alone. The authors postulate this latter result is potentially due to the small study size 46. A custom oligonucleotide microarray was used to study porcine adipose tissue and stromal/vascular (SV) gene expression 80. The SV cultures were collected from three stages of growth and development (90-day and 102-day fetal stages, and 5-7 day neonatal stage), and adipose tissue was collected from 105 day fetal and 5-7 day neonatal stages. They identified expression above background for 200 genes in SV cultured cells and 160 genes in the adipose tissue samples. Many of the these genes have been reported to be expressed in adipose cells previously; however the authors highlighted eleven genes not previously so reported, including relaxin, chromogranins A and B, INSL3, FGF12, IGFBP7, GDF9B, BDNF, IL12, and APOR). Three genes were reported as more highly expressed in fetal SV cells as compared to neonatal adipose tissue (IL4, IFNG and IGFBP5) using a t-test of normalized spot intensities 80. Several papers have reported changes in liver gene expression under a number of treatments. To profile the response to fasting and to treatment with a peroxisome proliferator-activated receptor alpha (PPARG) ligand (Clofibric acid, CA), Cheon and collaborators 49 used a pig skeletal muscle EST array to measure the liver transcriptional profile in normal fed pigs compared to fasted or CA-treated pigs. Liver RNA was pooled within treatment thus no biological replication was available. Genes DE between treatments were identified by as criteria: a) P < 0.1, b) minimum fluorescent intensity of 1,000 units, and c) 2 fold or greater difference between treatments. In contrast to results from rodents, there was little evidence from the microarray data that fatty acid oxidation related genes were highly induced by either CA treatment or fasting. QPCR assays showed statistically significant increased expression for ABCD3, CAT, CYP4A, ACOX1, and EHHADH by CA treatment, and all but the last two genes were also up-regulated by fasting. The QPCR data showed glucokinase was decreased only in the fasting treatment, while glycogen synthase was increased by both treatments. In fatty acid metabolic pathways, both QPCR and microarray data showed that steroyl-CoA desaturase and fatty acid desaturase 1 were decreased by fasting, but these two genes, as well as fatty acid desaturase 2, were increased by CA treatment. The authors conclude that there are major species differences in the liver RNA response to peroxisome proliferator ligands like CA, although they provide additional evidence that the response to fasting, especially for mitochondrial fatty acid oxidation and ketogenic pathways, is more similar across species 49. A recently published study 34 combined microarray analysis, SNP detection within expression candidates, and association and physical mapping analyses to find liver genes affecting carcass traits. These authors intended to capture expression differences related to phenotypic traits (and minimize other genetic differences) by using discordant sib pair analyses. They identified sibs within an F2 population from a Duroc x German miniature pig cross that were distinctly different for eye muscle area and backfat thickness. To profile the liver expression in these sib pairs, the microarray contained well-annotated genes, including liver genes from earlier work 18, 19 described above, as well as genes known to be important in hepatic metabolism. The liver mRNA from 4 sib pairs (high performing (HP) and low performing (LP)) was pooled by trait, and hybridized to this array. Overall, nine genes were identified by t-test as DE at least two fold between HP and LP groups; four were higher in HP (TBG, PIGSPI, DBI, and SLC01B3) and five were lower (PLA2G6, CPS1, PTN, NAN1, and BHMT1). Ten genes, including six DE genes, were tested by QPCR on liver RNA from the eight individual pigs from the pooled samples. Although good agreement with the microarray data was seen for most genes, only four (TBG, SLC01B3, PEDF and APOH) were statistically different between HP and LP individuals. Two confirmed genes, TBG (higher in HP) and APO (higher in LP) were screened for SNPs by re-sequencing of cDNAs for all eight HP/LP animals. Of three SNPs found in the TBG cDNA sequence, one (A>C at nt 778) was found associated (P < 0.0001) with variation in three fat traits, with the C allele associated with higher fat. As the SNP at nt 718 changes the amino acid at this position, the authors also checked TBG protein expression and function. Available serum data on TBG and TBG bound metabolite levels showed an association of TBG concentration (P < 0.12) and function (P < 0.06 to 0.11) with TBG genotype. Thus this group was able to use discordant sib pair microarray expression analysis to identify expression candidates, identify SNPs in one candidate that was associated with quantitative phenotypes for fatness and with relevant biochemical differences in serum. The combination of evidence at the genetic, expression and functional levels makes a strong case that TBG variation directly controls part of the variation in fatness in this population 34. 4. But What Does It All Mean? Pig Expression Bioinformatics and Databases At this juncture, available swine transcriptomic data, especially for microarray projects, is somewhat fragmented and sparse. Many different platforms are being used and the data is not always being submitted to a common repository. The recent public disclosure of nearly a million additional ESTs from 97 different non-normalized libraries by the Sino-Danish consortium (Gorodkin et al., submitted) will certainly improve the accuracy of EST frequency data as an estimate of expression level. To become more efficient at drawing biological meaning out of such data, more attention needs to be paid to public sharing of data and integration of that data so that an increase in power is possible. In this section, we discuss the available public resources for pig microarray and other transcriptomic data and discuss some of our efforts to integrate these platforms and data sources. The following information does not include several storage/analysis efforts on genomic data that includes Sus scrofa as one of the species. These include mapping databases such as ArkDB and databases that link traits to phenotypes without expression data, such as OMIA. Links to these and other sites of interest to pig genomics researchers can be found at the U.S. Pig Genome Coordination website (http://www.animalgenome.org/pigs/). Two groups have databases available for both expression and sequence information. The National Center for Biotechnology Information (NCBI; http://www.ncbi.nlm.nih.gov) and the European Bioinformatics Institute (EBI; http://www.ebi.ac.uk) have multiple databases, each designed for a specific purpose. At NCBI, UniGene takes the sequence data in the general EST and nr databases and clusters them into single gene units by species (http://www.ncbi.nlm.nih.gov/UniGene/UGOrg.cgi?TAXID=9823). The expression database at NCBI is Gene Expression Omnibus (GEO; http://www.ncbi.nlm.nih.gov/geo/), which uses a specific format for loading and exporting data. As of October 2006, GEO currently includes 18 porcine experiments, and the platform designs for 19 platforms, including the two large commercial platforms: Affymetrix 24k Porcine GeneChip® and several versions of spotted platforms using the Qiagen-Operon-NRSP8 13k oligonucleotide set. Of the 18 experiments, the majority were performed on microarrays, though three used SAGE. The user interface at NCBI for text or BLAST-based searches is advanced, allowing searches to be refined to search specific sub-databases or with other limits. In contrast, at EBI all the sequences are in a central repository and there is no species-specific clustering database as at NCBI/UniGene. The expression database at EBI is called ArrayExpress (http://www.ebi.ac.uk/arrayexpress/), and uses the Minimal Information About Microarray Experiments (MIAME) standards created by the Microarray Gene Expression Data Society (MGED). ArrayExpress, as of October 23, 2006, has a single porcine expression study, the main bulk of the data being human and mouse. Another comprehensive EST database is the Dana Farber Cancer Institute Computational Biology and Functional Genomics Laboratory's Pig Gene Index (PGI; http://compbio.dfci.harvard.edu/tgi/). The PGI database is the TIGR Gene Indices information that was brought to Dana Farber by John Quackenbush. The PGI database is similar to UniGene in that it clusters ESTs together, but goes a step further and provides a tentative consensus sequence for each cluster. Also, unlike UniGene which tries to group alternative spliced genes together, PGI attempts to separate them into their own, unique clusters. However, it does provide a link that clusters the tentative alternatively spliced sequences together. The DF database also provides Gene Ontology (GO) annotations, metabolic pathway information, and predicted 70-mer oligonucleotide sequences and SNP information for each of their TCs. Both UniGene and DFGIP data collections can be searched using BLAST, and an expression summary of each of the EST's based on the libraries, and EST count within the libraries, is provided. Both data collections can also be downloaded for local investigator use. Recently the Sino-Danish group has begun to release their EST sequence information, including alignments into clusters, which is based on over one million ESTs (http://pigest.kvl.dk/index.html). They have submitted their sequences to the Trace Archive at NCBI, but these data have not, as of October 2006, been integrated into dbEST. They have made their clusters available online, and have provided a bulk download website for both the ESTs and the consensus sequences of their clusters. Expression profiles for all the genes are also provided, based on EST frequency within the 97 non-normalized libraries. They predict SNPs based on sequence discrepancies within alignments, as the libraries were created from tissues from multiple breeds. They provide some online tools, including viewing the aligned and clustered sequences as well as searches against the database using BLAST. The website indicates these data have been submitted for publication (Gorodkin et al., submitted). There are non-comprehensive EST databases, which focus on a select number of tissues. Among these is the Michigan State Center for Animal Functional Genomics (CAFG; http://gowhite.ans.msu.edu/public_php/showPage.php). This database contains tissue specific porcine EST libraries: brain, adipose, and skeletal muscle, the use of which was discussed above in the microarray section 12, 46, 79. It also contains a mixed tissue porcine library. To annotate the ESTs, the database uses homology searches to various NCBI databases, including RefSeq and Gene. To search the libraries, keywords based on this information transferred from GenBank can be searched; there is also an option to use BLAST to search the different libraries. Clones and clusters can be viewed to determine which libraries have sequences donating to the clusters, and clones are available for order. In addition to the EST information, there is a private internal section for microarrays. Another focused database is the Pig Expression Data Explorer (PEDE; http://pede.dna.affrc.go.jp/). PEDE also uses homologues in RefSeq and UniGene to annotate the EST sequences that primarily are derived from tissues of interest to porcine immunology research (81 and references therein). They identified the full-length cDNAs within the ESTs. Their online interface allows a user to search their EST/cDNA clusters using keywords, library, their accession numbers, and the corresponding human chromosome. It allows filtering the results based on evidence of SNP in a specific breed, and provides suggested primers for detection of the SNP. Like the Sino-Danish EST database and the PGI database, it provides a visual alignment of the ESTs to each other and to the cluster consensus sequence. A unique feature of this database is a more in-depth analysis of artiodactyl-specific antigens for furthering the development of xenotransplantation. Several databases are also in production at the moment, and at various stages of development. One of these is under development at the University of Minnesota (http://gnomix.ansci.umn.edu/bioinf.htm). Their database focuses on sequence data currently, though they indicate they plan on integrating expression data in the future, and currently have annotation provided for the Qiagen-Operon-NRSP8 13K array. Most recently a new database annotating the new Swine Protein Annotated oligonucleotide Microarray (SPAM) has become available (http://www.pigoligoarray.org/). The database describes the set of 18,254 oligonucleotide sequences in the SPAM, and provided the best Ensembl protein matches and Gene Ontology (GO) annotations for each sequence. Another database is being developed at the Advanced Food and Materials in Canada (http://www.afmnet.ca/index.php?fa=Research.myProject&project_id=77&page=1). Their aim is to integrate various types of expression data, including microarray, proteomics, and metabolomics, to compare genetically modified foods. Their first goal is to compare the EnviroPig to the Yorkshire. A fourth database currently under construction is at Iowa State University (URL pending). Our focus is storage and analysis of data from the Affymetrix platform, although Qiagen-Operon-NRSP8 platform data is also curated. One specific interest is using expression data to help identify tissue-selective genes and across-species expression comparison of such genes to recognize evolutionarily conserved regulatory modules of interest to pig genome scientists. Here we describe some of our efforts in this area; integration and comparison of data from the two broad-coverage platforms that currently exist for the pig; the Qiagen-Operon-NRSP8 13K oligonucleotide array (hereafter abbreviated the Operon array) and the Affymetrix 23K Porcine GeneChip® (abbreviated the Affymetrix chip), both of which were discussed above. To integrate the expression data for these two platforms, we need first to determine which Operon oligonucleotide probes have a sufficient sequence similarity to the consensus sequences used to create the probesets on the Affymetrix chip so that the two elements will recognize the same transcript(s). The Operon 70-mer probe sequences were used as a BLAST query against the Affymetrix porcine consensus sequences (Figure 2). We set a cutoff criterion of an alignment length ≥ 67 nucleotides with an alignment identity of ≥ 97% for pairings between the 70-mer probes and the consensus sequences. This resulted in a total of 8,317 cross platform mappings (Table 2), which fall on six different match classes: (1) an Operon probe not matching any Affymetrix consensus sequence, (2) an Affymetrix consensus sequences without an Operon probe match, (3) the ideal case when a single Operon probe matches a single Affymetrix consensus sequences, (4) two or more Operon probes map to a single Affymetrix consensus sequence, (5) a single Operon probe maps to multiple Affymetrix consensus sequences, and (6) multiples of each match creating a cluster of probes and probe sets (Figure 2, Table 2). The different classes most likely result from the difference in the known amount of porcine sequence during the development of the probes/probe sets; the Operon probes being designed in 2002, while the Affymetrix probe sets were designed in 2004. This helps explains the different classes: Class 4 from the merging of old clusters, Class 5 from Affymetrix designing separate probe sets for alternative transcripts (which might not have been known in 2002) or that the Operon probe is no longer unique (for example, now targeting multiple genes in the same family), and Class 6 from the joining of different consensus sequences into single genes, but then having alternative splicing. Now that we had the ability to directly compare platform elements, we determined the extent to which the two platforms agree on assaying for the presence or absence of expression for these genes in a common tissue. Two Affymetrix experiments and one Operon experiment were used to calculate the agreement of the presence/absence calls of the platform pairings created by BLAST. All three experiments were performed on normal liver tissue; the first Affymetrix experiment (A1) collected data from the same RNA on six different chips; the second Affymetrix experiment (A2) data was collected using four separate RNA samples on four different arrays. The Operon (O) experimental data on an additional six RNA samples was available from our previous study 32. MAS5 provides the present/absent calls for the Affymetrix GeneChip®, and for the Operon array we used presence/absence as calculated by Zhao et al. 32. To take a step beyond simple agreement of presence/absence, we also calculated the correlation of expression level for each gene as estimated by each platform. We ranked each gene by expression level and used the Spearman's Rank correlation to determine the extent of agreement for relative expression of each gene within the common set of genes. Across the two platforms, we saw a significant amount of agreement in declaring a gene expressed (Table 2). Of the different classes, Class 4 had both the highest agreement (89%) and highest correlation (r2 = 0.77 for Operon to either array and r2 = 0.97 between Affymetrix experiments). The other three classes had similar correlations: r2 = 0.62 for the Operon to A1 experiment, r2 = 0.55 to 0.61 for the Operon to A2 experiment, and r2 = 0.94 to 0.97 for the A1 to A2 comparison. However, they differed in their agreement: 71% for Class 5, 82% for Class 3, and 84% for Class 6. When all pairs are considered, we found an r2 = 0.71 to 0.72 for the cross-platform comparisons and r2 =0.97 for the within platform comparison, with an present/absent agreement of 82%. Since Class 5 has the lowest agreement and correlations (equal to Class 3 for cross platform correlations, but with a lower within platform correlation), this result indicates that the Operon probes could cross-hybridize to either alternative splice variants or gene family members with close sequence homology that the Affymetrix platform was designed to assay separately. Further investigation is needed to see if the Operon probe is present, while the Affymetrix probe sets are absent, or if there are multiple Affymetrix probe sets present while the single Operon is absent. The results for Class 4, which have the highest agreement and the best correlation, lends support to the proposal that multiple Operon probes target the same gene product—likely by the clustering of sequences from the time the Operon chip was developed to the time the Affymetrix chip was developed. While these results already show good agreement between platforms, it will be important to update the sequence comparisons on a regular basis, especially with the Sino-Danish data, as well as the genome sequence, coming online. As part of our database we plan to develop the means to regularly develop a consensus sequence for each gene from all available sequences to map the various probes and probe sets to each other. Beyond EST and microarray databases, as noted above, the USDA Beltsville group has set up a database for quantitative real-time PCR assays for genes related to nutrition and immunity (http://www.ba.ars.usda.gov/nrfl/nutri-immun-db/nrfl_query1.html). All assays are based on the Taqman technology, so sequences for both a primer pair as well as the dual-labeled probe are presented. Of the almost 3,000 genes identified, they have validated assays for 474 available, of which 237 are known to be cross-reactive in humans, and another 771 candidate assays are being validated. Many of these assays have been published by this group and their collaborators 29, 31, 32. In addition to creating databases for porcine transcriptomics, groups are using available stored information for addressing biological information and additional annotation. In a purely bioinformatic analyses of porcine expressed sequences, Jiang and collaborators 82 used 33,308 human gene sequences and mapped, by using BLAST against the est_other database at NCBI, nearly 14,000 of them to a porcine EST in either embryonic or reproductive tissue. Of these, 2,167 were found only in the embryo and 4,552 were only in reproductive tissues, while 7,243 were found in both tissues. Therefore they found a total of 9,410 ESTs present in embryos and 11,795 in reproductive tissues 82. 5. Conclusions Functional genomics data, primarily at the RNA level currently, is accumulating rapidly for the pig species. Excellent, sensitive and broadly useful tools are already available and more will be become available within the next year. Annotation of the draft porcine genome sequence, expected in late 2007 and into 2008, will allow rapid integration of the gene expression data discussed above with gene sequences, potential splice sites, and gene families within the draft sequence. Advances in other areas of investigation in pig genetics and genomics can be anticipated. One such area would be the ability to find common regulatory sites within flanking DNA of co-expressed/co-regulated genes; leading to the identification of critical regulatory proteins in common with, or distinct from, those found in other species. Such information will reinforce the discovery of pathways through gene list annotations, improve pathway understanding through differentiation of direct targets from indirect targets of transcriptional signals, and would identify targets for manipulation of complete pathways and systems. We can also anticipate the comprehensive integration of linkage mapping and expression profiling of the same population, termed eQTL studies. Such integration of functional and structural genomic data will dramatically improve our understanding of the genetic architecture controlling quantitative traits in pigs. eQTL analyses may lead to the first application of “systems biology” to genetic improvement in the pig through the identification of cis-regulatory variation controlling an economically important phenotype. As we move toward such a “systems biology” approach in animal genomics, access to and integration of these data sources will become critical. Thus a continuing need is for bioinformatics to integrate the structural and functional data we are generating to inform our investigations 83. It is unfortunate that many datasets are not being submitted to public repositories; it is our hope and expectation that journals will both facilitate and require such public submissions, as is more the norm in the biomedical fields. Many groups are working on this bioinformatics effort and therefore the near future in pig genomics is especially exciting. The long-term goal of the application of genome data to improve pig genetics will be reached when we can apply a more robust understanding of pig genetic pathways to identify variation at genes controlling important traits of interest in the pig. Supplementary Material Supplementary Table 1 Click here for additional data file. Supplementary Table 2 Click here for additional data file.

J_Neurooncol-4-1-2295256

The influence of low-grade glioma on resting state oscillatory brain activity: a magnetoencephalography study

Purpose In the present MEG-study, power spectral analysis of oscillatory brain activity was used to compare resting state brain activity in both low-grade glioma (LGG) patients and healthy controls. We hypothesized that LGG patients show local as well as diffuse slowing of resting state brain activity compared to healthy controls and that particularly global slowing correlates with neurocognitive dysfunction. Patient and methods Resting state MEG recordings were obtained from 17 LGG patients and 17 age-, sex-, and education-matched healthy controls. Relative spectral power was calculated in the delta, theta, upper and lower alpha, beta, and gamma frequency band. A battery of standardized neurocognitive tests measuring 6 neurocognitive domains was administered. Results LGG patients showed a slowing of the resting state brain activity when compared to healthy controls. Decrease in relative power was mainly found in the gamma frequency band in the bilateral frontocentral MEG regions, whereas an increase in relative power was found in the theta frequency band in the left parietal region. An increase of the relative power in the theta and lower alpha band correlated with impaired executive functioning, information processing, and working memory. Conclusion LGG patients are characterized by global slowing of their resting state brain activity and this slowing phenomenon correlates with the observed neurocognitive deficits. Introduction Low–grade glioma (LGG) patients constitute 25% [1] of the glioma patient population and have a survival of 5–10 years [2]. The optimal treatment for this patient group is still a matter of debate [3–5]. The first clinical feature of this disease is epilepsy in two-third of the patients [1, 4]. The majority of LGG patients also suffer from cognitive deficits, which tend to have a global character and cannot only be explained by the tumor localization alone [6, 7]. Higher cognitive functions depend on the integrated activity of several specialized brain areas. Functional imaging techniques such as electroencephalography (EEG) and magnetoencephalography (MEG) are used to characterize oscillatory activity in patients with neurodegenerative diseases, including Alzheimer’s [AD; 8–10] and Parkinson’s disease [PD; 11–13]. Oscillations are a feature of neuronal brain activity and the synchronization of the oscillatory activity (reflecting the interactions of neuronal activity) is a likely mechanism for neuronal communication. For an overview of the relation between oscillatory brain activity and neurocognitive function [see 14, 15]. Magnetoencephalography (MEG) is a relatively novel way to capture the dynamics of the electromagnetic fields within the brain, and combines excellent temporal and spatial resolution. Comparison of MEG in AD patients and healthy controls showed more delta and theta activity in the patient population especially in the temporo-parietal region [16]. Another study compared spectral power in four different groups (severe AD, moderate AD, Lewy Body dementia, and healthy controls) and showed that the healthy control population had the largest rhythmic activity in the alpha band. The moderate AD population and the patients with Lewy Body dementia had the rhythmic predominance in the pre-alpha frequency band (7–9 Hz), whereas the severe AD showed a shift towards the slow-band (3–7 Hz) [17]. In PD patients, MEG has been used by Kotini [18], who showed slowing of the alpha rhythm in the patient population compared to healthy controls, and Bosboom [19], who demonstrated slowing in the theta, beta, and gamma frequency band in non-demented PD patients with a further slowing (and involvement of delta and alpha band) in the demented PD patient population. It is suggested that the increase in slow wave activity corresponds with neurocognitive deficits in AD as well as in PD [13, 16, 19, 20]. MEG-studies have previously been performed in brain tumor patients. Oshino [21] found an increase in delta and theta band activity especially in the vicinity of the tumor and in the surrounding regions of edema. In the majority of these patients, the increase correlated with the clinical symptoms of these patients, assessed with a routine neurological examination, but neurocognitive function were not formally investigated. Although the focal lesion in brain tumor patients contrasts with the generalized brain degeneration in AD and PD, there is evidence that brain tumors can cause functional disturbances in areas remote from the tumor [22–25]. De Jongh [22] found focal clusters of delta activity near the tumor area but also in the contralateral hemisphere. An evaluation of fast MEG waves in brain tumor patients showed dipoles located in the parietal and occipital area, whereas the tumors were located in the parietal, temporal, and frontal area [23]. In this study, these dipoles were found remote from the tumor area and even in the contralateral hemisphere. Bartolomei [24, 25] found differences in resting state functional connectivity in the same brain tumor population within several frequency bands when compared to healthy controls. Interestingly, in agreement with de Jongh’s results, these differences were not limited to the tumor area and were more obvious in those with a tumor in the left hemisphere. Since higher cognitive functions are presumed to depend on the integrated activity of several specialized brain areas, it is suggested that neurocognitive deficits may have a stronger correlation with diffuse alterations in resting state brain oscillatory activity than focal abnormalities in the brain tumor population. In the present MEG-study, power spectral analysis of oscillatory brain activity was used to compare resting state brain oscillatory activity in both LGG patients and healthy controls. We hypothesize that (1) LGG patients show, in addition to the local abnormalities at the tumor site, global slowing of the resting state brain activity compared to healthy controls which will vary between patients with a tumor in the left or right hemisphere (2) changes in resting state brain oscillatory activity reflect an intermediate level between the impact of tumor and tumor-related treatment on the one hand (‘input’) and the neurocognitive deficits (‘output’) on the other hand and that diffuse slowing is correlated with higher neurocognitive dysfunction. Materials and methods Patients and controls Twenty-three LGG patients were asked to participate in this study. Patients were eligible if: (a) they had a suspected or histologically confirmed LGG; (b) there was no radiological (confirmed by MR or CT scan) and/or clinical tumor progression in the previous 6 months; (c) they did not use medication possibly interfering with neurocognitive function, other than anti-epileptic drugs (AEDs). Patients were recruited from the VU University Medical Center (VUmc) and the Academic Medical Center (AMC), both tertiary referral centres in Amsterdam for brain tumor patients, after the institutional ethical review boards of both participating hospitals approved the study protocol. Relatives of the patients were asked to participate as healthy controls. Healthy controls were eligible if they: (a) did not suffer from any neurological disease; (b) did not use any medication that might influence cognitive function. For patients who could not provide a healthy control participant, VU University Medical Center staff members were included. Magnetoencephalography MEG recordings were obtained using a 151-channel whole-head MEG system (CTF systems; Port Coquitlam, British Columbia, Canada) while participants were seated inside a magnetically shielded room (Vacuumschmelze GmbH, Hanau, Germany). Magnetic fields were recorded during a no-task, eyes-closed resting state. Metal artefacts were avoided as much as possible. A third-order software gradient [26] was used with a recording passband of 0.25–125 Hz and a sample frequency of 312.5 Hz. At the beginning, middle and end of each recording, the head position relative to the coordinate system of the helmet was recorded by leading small alternating currents through three head position coils attached to the left and right preauricular points and the nasion on the subject’s head. Head position changes up to approximately 1.5 cm during a recording condition were accepted. For this study, 149 of the 151 channels could be used. MEG recordings were converted to ASCII files. From these ASCII files four artefact free epochs of 13 s per subject (4096 samples) were carefully selected by visual analysis by one of the authors (IB). Magnetic field frequencies ranging from 0.5 to 80 Hz were recorded. The MEG data were digitally filtered off-line in the following frequency bands: delta (0.5–4 Hz), theta (4–8 Hz), lower alpha (8–10 Hz), upper alpha (10–13 Hz), beta (13–30 Hz), gamma (30–50 Hz). The MEG channels were grouped into 10 distinct regions, respectively left and right frontal region, left and right temporal region, left and right parietal region, left and right occipital region and left and right central region, as shown in Fig. 1. Fast Fourier transformation was separately applied for every patient and control on the four epochs in the above mentioned frequency bands. The results of the four epochs were averaged for each participant and the mean relative power for each MEG region were used for the analysis. Fig. 1Distribution of MEG regions Neurocognitive assessment Patients and controls participated in neurocognitive assessment (see Table 1). The total duration of the assessment varied between 1 and 2 h. Individual patient’s test scores were converted to z-scores, using the means and standard deviations of the age-, sex- and education-matched healthy controls as a reference. Table 1Description of neuropsychological test batteryLetter-digit substitution test [27] This test provides a measure of psychomotor performance that is relatively unaffected by intellectual prowess and is suitable for groups with an age range exceeding 60 years. The number of items written down in 90 s is registered, as in the decrease in performance when graphomotor speed is involvedVisual verbal learning test [27] This version of the Rey Auditory Verbal Learning Test calls for various aspects of verbal learning and recall. Measures used for analysis are memory performance on trial 1 as indicator of immediate recall, total recall after five trials, delayed recall and recognition after 20 min as indicators of memory consolidation into long-term memory, and a delta score as a measure of learning capacityStroop color-word test [27] This test is a selective attention task aiming at measuring interference susceptibility and consists of three subtasks with increasing task complexityCategoric word fluency [28]A simple task requiring the generation of words from semantic categories (animals) within a limited timeConcept shifting test [29]This test, which has two conditions of complexity, predominately measures functions associated with executive function, especially visual scanning and conceptual tracking. The motor component of this task is measured by three dummy conditions in which no cognitive capacity except for graphomotor speed is requiredMemory comparison test (MCT) Selective attention, mental concentration, memory and information processing To reduce data, individual scores on these tests were summarized into six cognitive domains, namely information processing speed, psychomotor function, attention, verbal memory, working memory, and executive functioning. Construction of these domains has previously been reported [30], and was based on a Principal Component Analysis using Varimax rotation with Kaiser normalization performed on the z-scores of a large group of healthy controls [31]. The domains found are commonly used in neurocognitive practice and research. Statistical analysis Differences between both groups in the distribution of age, sex, and education were analyzed by means of chi-square tests. Mann–Whitney nonparametric U-tests were used to investigate whether patients’ standardised z-scores on neurocognitive tests in the overall measure of cognition differed significantly from healthy control z-scores. Because of the non-normal distribution of the relative power, Mann–Whitney nonparametric U-tests were used to determine possible differences between the patient population and healthy controls. To assess the association between relative power and cognition within the patient group, Spearman’s correlation coefficient (ρ) was calculated between the relative power in the different MEG regions and z-scores of the six neurocognitive domains for all separate frequency bands. Results Patient characteristics From the initial patient group, six patients were excluded, four patients due to metal artefacts on the MEG measurements, one due to severe epileptic seizures, and one due to tumor progression at the time of registration. The final analyses were performed on a sample of 17 patients and 17 matched healthy control participants. Due to the matching procedure, there were no significant differences between patients and healthy controls in age (M = 42.7, SD = 11.2 in patients, M = 42.6, SD = 12.7 in healthy controls, p = 0.99), and educational level (M = 5.2, SD = 1.8 in patients, M = 5.5, SD = 1.8 in healthy controls, p = 0.64). The male–female ratio between the two groups did not differ significantly (p = 0.37). Fourteen of the 17 LGG patients had a histologically confirmed LGG, clinically and radiologically stable for more than six months before inclusion, whereas the other 3 patients were suspected of having LGG, and stable for more than six months. The patients in our study were diagnosed several years ago (mean 8 years, range: 1–19 years). One year after MEG registration two of the three patients with suspected LGG were operated on because of increasing epilepsy frequency. Eleven of those 16 patients underwent debulking, whereas 3 patients underwent a stereotactic biopsy and another two patients had an open biopsy. Of the 16 patients with a histologically confirmed LGG, the pathological diagnosis was grade II astrocytoma in ten patients, oligodendroglioma grade II in four patients and oligoastrocytoma grade II in two patients. Seven of the seventeen patients underwent radiotherapy with prior chemotherapy in two patients (1 patient with 5 cycles of PCV and 1 patient with 2 cycles of PCV and 3 cycles of temozolomide). Eleven patients had left hemisphere tumors and 6 patients had right-sided tumors. The specific localisation of the tumor is shown in Table 2. Table 2Tumor lateralization and localizationLeft hemisphereRight hemisphereTumor locationNo. of patientsTumor locationNo. of patientsLeft frontal4Right frontal2Left parietal3Right frontoparietal3Left temporal3Right insular region1Left parieto-occipital 1Total11Total6 In the patient group, all but one patient used AED mono- or poly-therapy. Six of the 16 patients on AED were free of seizures, while the other 10 patients were still having seizures. Differences between patient and healthy control group Neurocognitive functioning As expected, patients performed poorer than healthy controls on the neurocognitive test battery (controls M = 0.00). More specifically, patients showed a significantly lower psychomotor function (M = −0.50, SD = 0.75 versus SD = 0.55, p = 0.044), working memory capacity (M = −1.43, SD = 1.62 versus SD = 0.91, p = 0.002), information processing speed (M = −0.85, SD = 0.88 versus SD = 0.97, p = 0.018) and attention (M = −1.92, SD = 3.87 versus SD = 0.68, p = 0.003). Patients’ performance on the other two cognitive domains did not differ significantly from controls (Fig. 2). Fig. 2Patients’ z-scores on the six neurocognitive domains and on total neurocognitive functioning. Note: * p < 0.05, ** p < 0.01. A = attention, EF = executive functioning, IPS = information processing speed, PF = psychomotor function, VM = verbal memory, WM = working memory. Performance is relative to that of age-, sex-, and education-matched healthy controls (represented by the 0-line). A higher score (i.e. approaching 0) means better performance Spectral analysis Global spectral analysis. The mean relative power of the LGG patient population showed a significant decrease (Mann–Whitney U, p = 0.034) in the gamma band (30–50 Hz) compared with healthy controls. Spectral analysis of frequency bands within each MEG region. An increase (Mann–Whitney U; p = 0.014) in relative power in the LGG patients compared to the healthy controls was found in the left parietal region within the theta frequency band (4–8 Hz). A decrease in relative power was seen in the left and right central region (Mann–Whitney U; both p = 0.024) and left and right frontal region (Mann–Whitney U; both p = 0.005) of the LGG patient population compared to the healthy controls within the gamma band (30–50 Hz), as shown in Table 3 and Fig. 3. Table 3Significant differences in relative power between patients and controls per frequency band and accompanying statisticsPatientsControlspMSDMSDThetaLeft parietal0.1310.0530.0920.0260.014GammaLeft central0.0640.0270.0950.0410.024Right central0.0660.0330.1020.0470.024Left frontal0.0480.0270.0880.0430.005Right frontal0.0500.0300.0830.0370.005Significant higher relative power is depicted in bold and in italicsFig. 3The significant differences in relative power between the patient group and the healthy controls within the different frequency bands. Green area: significant higher relative power in the patient group compared to the healthy controls. Red area: significant lower relative power in the patient group compared to the healthy controls Influence of tumor lateralization on relative power. Eleven patients had left hemisphere tumors and 6 patients had a tumor in the right hemisphere. Patients with a tumor in the left hemisphere showed an increase in theta activity in the left and right central region (Mann–Whitney U; respectively p = 0.001 and p = 0.014), left and right parietal region (respectively p = 0.004 and p = 0.029), and left and right temporal region (respectively p = 0.022 and p = 0.027) when compared to the healthy control population. Patients with left-sided tumors also showed a decrease in relative power in the gamma band in the left and right central region (respectively p = 0.023 and p = 0.048) and left and right frontal region (respectively p = 0.011 and p = 0.019). Patients with a tumor in the right hemisphere showed a significant increase in relative power of delta activity in the right parietal region (Mann–Whitney U; p = 0.050) compared to the healthy control population. These patients also showed a decrease in relative power in the gamma band in the left and right frontal region (respectively p = 0.005 and p = 0.025), right temporal region (p = 0.042), and left and right occipital region (respectively p = 0.025 and p = 0.014), as shown in Table 4 and Fig. 4. Table 4Significant differences in relative power between patients with a tumor in the left or right hemisphere and controls per frequency band and accompanying statisticsPatientsControlspMSDMSDLeft hemisphereThetaLeft central0.1410.0390.0950.0180.001Right central0.1330.0410.0960.0220.014Left parietal0.1490.0570.0920.0260.004Right parietal0.1410.0650.0890.0290.029Left temporal0.1070.0490.0720.0190.022Right temporal0.1000.0430.0710.0230.027GammaLeft central0.0630.0260.0950.0410.023Right central0.0690.0320.1020.0470.048Left frontal0.0480.0270.0880.0430.011Right frontal0.0530.0310.0830.0370.019Right hemisphereDeltaRight parietal0.4240.1440.3030.1370.050GammaLeft frontal0.0500.0280.0880.0430.005Right frontal0.0450.0310.0830.0370.025Right temporal0.0420.0220.0700.0450.042Left occipital0.0400.0290.0770.0320.025Right occipital0.0480.0360.0990.0370.014Significant higher relative power is depicted in bold and in italicsFig. 4The significant differences in relative power between the patients with a tumor in the left or right hemisphere compared to the healthy controls within the different frequency bands. Green area: significant higher relative power in the patient group compared to the healthy controls. Red area: significant lower relative power in the patient group compared to the healthy controls Since these results could be influenced by differences in tumor volume, we evaluated the tumor size of participating patients. We used T1-weigthed MRI and defined tumor size as the product of the two largest perpendicular diameters of tumor hypointensity [32]. We found no significant differences in the tumor size between left- and right-sided tumors. Associations between patient’s neurocognitive functioning and relative power in distinct MEG regions In the patient population, negative associations between neurocognitive functioning and relative power were found in the theta and lower alpha band. An increase in the relative power in the theta band within the left frontal (−0.606; p = 0.010) and the right frontal region (−0.490; p = 0.046) was correlated with a poorer executive functioning. In the same frequency band (4–8 Hz), an increase in theta band power was correlated with poorer information processing within the left central and left frontal region (respectively −0.515; p = 0.034 and −0.592; p = 0.012). Within the lower alpha band in the right temporal region, an increase in relative power was correlated with a disturbed working memory (−0.508; p = 0.037). Discussion The primary goal of the present study was to evaluate whether LGG patients show (in addition to the well known MEG slowing around the tumor) diffuse slowing in resting state brain activity. The secondary goal of the study was to investigate whether this slowing is correlated with neurocognitive dysfunction. By means of MEG-registrations we have demonstrated that LGG patients have slowing of the resting state brain activity when compared to healthy controls. The decrease in relative power was mainly found in the gamma frequency band in the bilateral frontocentral MEG regions. Regarding the low frequency bands, an increase in relative power was found in the theta frequency band in the left parietal region. Correlations of neurocognitive functioning with the relative power in the patient population showed clear associations in the lower alpha and theta band, increased slowing correlating with poorer performance. We chose to use the relative power instead of the absolute power, because relative power is less influenced by the distance between the MEG sensor and the underlying neural populations. The distance is variable due to head position in the helmet and the thickness of the skull. Lower variance of power values in subject groups are expected by using the relative power instead of the absolute power. A second reason for us to use this method is the fact that by using this method we are able to compare our results with those of other patient cohorts at our MEG center [19, 20]. This includes comparisons with other brain tumor data sets but also the comparison of our results with neurodegenerative diseases (including AD and Parkinson’s disease (PD)). This will give us an opportunity to learn more about differences in resting state oscillatory brain activity in the different diseases and at different disease stages. A limitation of this method of analysis is that an increase in one frequency band can cause a decrease in another band, although this might not be the observed pattern in absolute measures. In previous studies, MEG has been used for power analysis in patients with neurodegenerative diseases, such as AD and PD. They showed similar results, namely a rhythmic predominance in the lower frequency bands, as was found in our study with LGG patients [16–20]. There are only few studies performed in brain tumor patients, however. De Jongh [22] demonstrated that both a higher tumor malignancy and larger tumor volume were associated with higher signal powers in the delta band. In our study, we did not find increased oscillatory brain activity in the delta band. It might be that the conflicting results are due to differences in methodology of both studies. In contrast with De Jongh’s study [22], we analysed the relative power, which means that the power is influenced by the power in the other frequency bands and therefore group differences in a specific frequency band could results from changes in that specific band or in the absolute power of the other frequency bands. Furthermore, we evaluated diffuse delta activity instead of clusters of dipoles in the vicinity of the tumor. Apart from these technical differences, De Jongh’s patient population consisted of a more heterogeneous group of brain tumor patients. They found a higher delta power in the high-grade gliomas compared to those patients with a low-grade glioma and a delta signal power decrease after surgery. Our patient population consisted exclusively of patients with a low-grade glioma, and all but three MEG registrations were done after surgery, which might explain the lack of changes in the delta band in the present study. It also suggests that changes in the gamma and theta band may be more sensitive indicators of brain dysfunction in LGG. In another study of De Jongh [23], the authors found focal clusters of fast MEG waves outside the tumor area and even in the contralateral hemisphere. They concluded that the asymmetry possibly reflects features of normal background activity, since asymmetry has already been demonstrated in healthy subjects for the alpha activity which is usually dominant in the right hemisphere. Since De Jongh used a single dipole analysis instead of our analysis of diffuse power changes and, again, analysed a group of patients with diverse brain tumors, it is difficult to compare the results of both studies. Oshino and colleagues [21] used synthetic aperture magnetometry (SAM) to analyze oscillatory activity recorded by the MEG. SAM is an alternative method investigating the distribution of reconstructed sources over different cortical regions. In contrast to other source localization methods, beamforming does not rely on averaging and therefore allows analysis of evoked and induced brain activity. In a group of 15 patients with various types of primary or secondary brain tumors, they found an increase in delta and theta band activity especially in the vicinity of the tumor and in the surrounding regions of edema. In the majority of these patients, the increase in the delta band correlated well with the clinical symptoms, assessed with a routine neurological examination. The results were not correlated with neurocognitive functioning because they stated that there is much individual variation in brain activity related to neurocognitive function and they expected it difficult and sometimes impossible for these patients to undergo the neurocognitive battery. In contrast to their results, we only found an increase in theta activity and decrease in the gamma frequency band. As mentioned before we analyzed the relative power and also our patient population with only LGG patients is not comparable to the patients with various brain tumors in the study of Oshino. Although, different methods to reconstruct sources can show significant discrepancies, we also found spectral changes in the contralateral hemisphere. By evaluating separately those patients with a tumor localisation in the left or the right hemisphere, we found more significant differences in patients with a tumor in the left hemisphere compared to healthy controls. In agreement with our previous study [24], in which we evaluated the functional connectivity (functional connectivity is a statistical correlation between time series of brain activity recorded over distinct regions which are assumed to reflect interactions between the brain regions) in patients with diverse primary brain tumors, we also found that patients with a tumor in the left hemisphere showed more differences in this functional connectivity compared to those with right-sided tumors. Why those patients with a tumor in the left hemisphere do show more differences in oscillatory activity is not known. It is possible that the left hemisphere (usual the dominant hemisphere) acts differently to the tumor compared to those located in the right hemisphere. It is also possible that patients with a tumor in the left hemisphere are in a different stage of their disease compared to those with a tumor in the right hemisphere. Since our sample size is rather small, this could be reflected on the observed results. On the other hand is it possible that our results reflect hemispheric asymmetry which has been found in the healthy population previously. Gootjes et al. [33] found hemispheric differences in functional connectivity in healthy young adults during resting state. The neurocognitive domains of psychomotor function, working memory, information processing speed and attentional tasks were impaired in LGG patients. These results corroborate previous studies indicating a decline in neurocognitive functioning in LGG patients [6, 34–38]. Several research groups have suggested that the increase in slow wave activity corresponds with neurocognitive deficits in AD and PD. Fernandez [16] found right parietal delta activity to be significantly associated with variability in the Cambridge Examination for Mental Disorders in the elderly (CAMCOG) score, whereas the left temporal theta activity predicted the variability of Mini Mental State Examination (MMSE) results. Sinanovic [13] found a positive correlation between diffuse changes in spectral power and MMSE scores in the PD patients with dementia. Bosboom [19] found a negative correlation in the non-demented PD patients between theta power and CAMCOG scores in both occipital and right temporal region but expected these associations, given the large number of relations that were analyzed, likely to be caused by coincidence. In the demented PD group they found no significant correlations. Stoffers [20] showed a negative association between lower alpha power and the performance on perseveration-related tasks in the non-demented PD patients. Evidently, this study has its limitations. First of all, the patient group is rather small as mentioned before. Secondly, although our patient population consisted of only LGG patients, they did not all receive the same treatment. In our study we hypothesized changes in oscillatory brain activity to be the intermediate between the impact of the tumor and its treatment on the one hand and the neurocognitive deficits as the output on the other hand. This study showed that irrespective of the different treatment options, changes in oscillatory brain activity can be found in brain tumor patients which is associated with neurocognitive function. To get informed on the influence of the different tumor treatments on brain activity and its relation with neurocognitive function is very interesting. A longitudinal study of these effects (e.g. surgery, radiotherapy and chemotherapy) on brain activity is currently under way. The goal of that study is to explore correlations between changes in neurocognitive function and changes in functional brain dynamics during the disease course. Our study is the first to correlate the relative power with neurocognitive functioning in brain tumor patients. The observed correlations were quite strong and showed that an increased activity in the theta and lower alpha band is correlated with impaired executive functioning, information processing and working memory. This is a first step in unraveling the underlying mechanisms of neurocognitive dysfunction in brain tumor patients. MEG power analysis gives us an interesting tool to assess functional alterations in the patient’s brain in the course of disease and to evaluate its relationship with neurocognitive functioning.

Matern_Child_Health_J-2-2-1592151

What Obstetrician-Gynecologists Think of Preconception Care

Objectives: To describe obstetrician-gynecolog-ists’ opinions of preconception care (PCC) and ascertain patient uptake for this service. Methods: A questionnaire was mailed to 1105 ACOG members in August 2004. Results: There was a 60% response rate. Most physicians think PCC is important (87%) and almost always recommend it to women planning a pregnancy (94%); 54% do so with women who are sexually active. Around a third (34%) thought their patients usually do not plan their pregnancies and 49% said very few pregnant patients came in for PCC. Of those who obtain PCC, they were believed to do so more likely to assure a healthy pregnancy (83%) than because of an elevated risk for birth defects (20%). Of 11 issues presented, cigarette smoking and folic acid supplementation were rated the most important for PCC counseling; exercise and environmental concerns were the least important. Conclusions: Physicians are willing to provide PCC but few patients are accessing such services. Introduction When poor pregnancy outcomes occur, they frequently have been set in motion long before the first prenatal visit. Fortunately, many of the factors contributing to less than ideal birth outcomes can be managed and brought under control prior to conception. Preconception care (PCC) consists of the identification of those conditions that could affect a future pregnancy or fetus and that may be amenable to intervention [1]. ACOG recommends that all health encounters during a woman's reproductive years, particularly those that are a part of PCC, should include counseling on appropriate medical care and behavior to optimize pregnancy outcomes [1]. Through PCC, physicians can advise patients on numerous behavioral and lifestyle changes, ranging from exercise and weight control to the use of prescribed medications and illicit drugs. They may also address carrier screening for heritable genetic disorders and the options available for helping to avoid an affected birth. For example, it is well known that women with a chronic disease such as diabetes have an increased risk of congenital abnormalities in their offspring, and they are known to have improved birth outcomes when they plan their pregnancies and utilize PCC [2]. Another prime example of the success of PCC efforts is the reduction in the number of pregnancies affected by neural-tube defects following national efforts to increase consumption and supplementation of folic acid by women of childbearing potential [3] through education and food fortification. Despite such improvements, fewer than 50% of women are aware of the benefits of taking folic acid supplements around the time of conception [4], the majority of women with diabetes do not plan their pregnancies [5], many women continue to drink alcohol regularly during pregnancy [6] and almost 50% of pregnancies in the U.S. are unplanned [7]. PCC has had its successes, but clearly more work needs to be done. The importance of PCC in promoting maternal and fetal health has long been recognized, and national recommendations and guidelines for PCC are being developed [8]. One aspect of successful development of PCC guidelines involves knowledge of obstetrician-gynecologists’ current practices and opinions regarding PCC and what they perceive to be barriers to successful implementation of PCC. The purpose of this study was to assess the practices and opinions of obstetrician-gynecologists regarding PCC, and how frequently they perceive their patients to avail themselves of such services. Materials and methods Questionnaires were mailed in August 2004 to 1105 American College of Obstetricians and Gynecologists (ACOG) Fellows and Junior Fellows in Practice. Of these subjects, 605 were members of the Collaborative Ambulatory Research Network (CARN), and 500 were a computer-generated random sample of ACOG members who had not received a survey from ACOG during the previous two years (Non-CARN). Members of CARN are practicing obstetrician-gynecologists who have volunteered to participate in survey studies on a regular basis. CARN was established to facilitate assessment of clinical practice patterns and aid the development of educational materials. Two reminder mailings were sent to non-respondents and questionnaires returned by December 31, 2004 were included in the survey. This protocol has typically resulted in a total sample size of >450, which is sufficient to detect differences between groups of <0.5 standard deviation with power of 80% and significance at the 0.05 level [9]. The survey recorded demographic details of physicians and their patient population, and assessed practices and opinions regarding PCC. A questionnaire concerning screening for aneuploidy was included in the same mailing; both questionnaires shared a demographics section containing 9 questions on one page. The PCC questionnaire contained 11 questions, some multi-part, on two pages. Both questionnaires were brief and resulted in a final document of 3 double-sided pages, a standard length for our full-length questionnaires. Seven of the 11 questions on PCC involved rating the frequency of an activity (e.g., always, usually, occasionally, never) or degree of agreement with a statement, and three questions were of a multiple-choice format. The questionnaires were developed in consultation with medical specialists and were pilot tested on a sample of practicing obstetrician-gynecologists prior to final distribution. The data were analyzed using a personal computer-based software package (SPSS® 12.0, SPSS Inc., Chicago, IL). Descriptive statistics were computed for the measures used in the analyses, which are reported as mean ± SEM. Student's t test was used to compare group means of continuous variables. Factor Analysis was conducted on two sets of rating variables (Agreement with 6 statements regarding PCC, and Importance of counseling on 11 issues during PCC and during routine care), and factors with eigenvalues greater than one were treated with a varimax rotation. Differences in ratings on the Importance scales were examined with multivariate analysis of variance. Differences on categorical measures were assessed using χ2. Group differences on ordinal measures were assessed using the Mann-Whitney U test or Kruskal Wallis χ2. Related-sample differences on ordinal measures used the Wilcoxon signed ranks test. Correlations including an ordinal measure used the Spearman's rho coefficient. All analyses were tested for significance using an alpha of 0.05. Results Demographics A total of 670 questionnaires were returned. Data from 18 respondents were judged invalid (physician retired, returned to sender), resulting in a valid response rate of 60% (652/1087), 432 from CARN members (72.1% response rate) and 220 from Non-CARN (45.1% response rate). Physicians responded from every state of the United States except Maine, as well as the District of Columbia, Puerto Rico, Canada, and overseas military installations. The respondents’ mean age (47.32±0.39) and the proportion of males to females (males = 55%) closely matched those of the larger population to whom the survey was sent (46.99±0.32, males = 55%) and of ACOG Fellows and Junior Fellows in Practice as a whole (47.64, males = 58%). The remaining analyses are limited to the 88.8% (579) of respondents who indicated that their primary medical specialty was gynecology or general obstetrics and gynecology. This selection was made to best assess the practices of non-subspecialist obstetricians and gynecologists, rather than those in maternal fetal medicine, reproductive endocrinology, or other subspecialties, to whom patients may be referred after pregnancy has already been attempted. Of these physicians, 83.9% (486) practice obstetrics (ObGyn), and 16.1% (93) do not (GynOnly); the term ‘physicians’ is used when referring to both ObGyns and GynOnlys. See Table 1 for physician demographics. CARN and Non-CARN differed on only one non-demographic item (CARN were more likely than Non-CARN to discuss carrier screening with all patients who present for PCC. See results.); thus data were collapsed across these two groups. Table 1Physician demographicsTotal (n=579)Gender (%) Males53.9 Females46.1CARN67.5Non-CARN32.5Age in years, mean (SEM)*47.12 (0.42) Males50.88 (0.54) Females42.76 (0.54)Years in practice, mean (SEM)15.22 (0.41)Deliveries in 2003, mean (SEM)129.65 (3.13)Practice location (%) Urban, inner city10.7 Urban, non-inner city29.3 Suburban32.1 Mid-sized town19.4 Rural6.9 Other1.6Practice type (%) Ob/Gyn partnership/gp51.4 Solo practice22.5 Multi-specialty10.4 University full-time faculty and practice8.5 Other7.2Patient ethnicity—mean (SEM) % of patients Non-Hispanic white62.29 (1.10) African-American16.78 (0.79) Hispanic13.35 (0.78) Asian/Pacific Islander3.83 (0.29) Native American1.29 (0.23)*P < 0.001: Males older than females; CARN older than non-CARN. Defining preconception care Physicians were asked whether they would define PCC as specialized or routine care. The great majority (86.7%) defined it “As specialized pre-pregnancy care that focuses on issues not typically addressed during a routine exam which are specific to ensuring an optimal pregnancy outcome.” Far fewer (13.3%) defined it as “The same as routine well-woman care that occurs during the reproductive years, prior to a pregnancy.” Those who defined PCC as routine tended to be in practice longer than those defining it as specialized (17.55±1.16 yrs versus 14.84±0.45 yrs. t(555)=2.22; P < 0.05), independent of gender. Recommending PCC The vast majority of physicians (97.3%) indicated that they provide PCC for their patients. Physicians were asked how frequently they recommend PCC to different groups of women of childbearing age (see Table 2). Three quarters (75%) always recommend PCC to patients planning a pregnancy, and 9 in 10 (89.9%) always recommend it to diabetic patients planning a pregnancy. Physicians who defined PCC as routine were more likely than those who defined it as specialized to recommend PCC to women who are sexually active (‘always’ or ‘usually’: routine = 63.4%, specialized = 52.1%. M-W U=13989.5; P<0.05) and to women who are using birth control (‘always’ or ‘usually’: routine = 48.6%, specialized = 32.8%. M-W U=13002; P < 0.01). Table 2Percent of physicians indicating how frequently they recommend PCC to different groups of womenHow frequently do you recommend preconceptioncare to the following women of childbearing age?AlwaysUsuallyOccasionallyNeverDiabetic women planning a pregnancy89.97.91.70.5Women who are planning a pregnancy75.019.25.00.8Obese women planning a pregnancy61.527.09.52.0Women indicating they want children in the future38.543.516.02.0Women who are sexually active19.134.938.67.4Women who are using birth control11.524.145.918.5 Opinions about PCC Physicians were asked their level of agreement with several statements regarding PCC (see Table 3). The majority agreed (4–5 on 5 point scale) that PCC is an important issue (87.3%) and that it has a positive effect on pregnancy outcomes (83.5%), though only 20.7% agreed that it is a high priority in their workload. Three quarters (76.8%) thought they have appropriate training to provide PCC. Half agreed that there is not enough time to provide preconception care visits to all women of childbearing age (51.4%), and that time devoted to PCC is not reimbursed (49.8%). Based on factor analysis, the first three items were combined to produce a single mean score for the ‘positive’ aspects of PCC (Factor I), and the second set of three items for the ‘negative’ aspects of PCC (Factor II). There were significant positive correlations between agreement on Factor I and frequency of recommending PCC in all groups of women (all P’s < 0.001); there were significant negative correlations between agreement on Factor II and frequency of recommending PCC in all groups of women (all P’s < 0.01). In other words, the more strongly they agreed that PCC was important/positive/high priority, the more frequently they recommended PCC. Looking at the individual components of the two factors, lack of reimbursement for PCC was the only item that did not correlate with frequency of recommending PCC in any group of women. Table 3Percent of physicians indicating how strongly they agree with several statement regarding PCCStronglyStronglyagreedisagreeMean on 5 pt Six statements regarding opinions about PCC54321scale (SEM)Factor I: positive aspects Preconception care is an important issue47.439.910.51.30.94.32 (0.034) Preconception care has a positive effect on pregnancy outcomes44.738.814.31.11.14.25 (0.035) Preconception care is a high priority in my workload6.813.942.327.99.12.82 (0.043)Factor II: negative aspects Time devoted to preconception care is not reimbursed24.725.130.814.05.53.50 (0.051) There is not enough time to provide preconception care visits to all women of childbearing age22.628.819.718.910.03.35 (0.056) I do not have appropriate training to provide preconception care3.05.214.735.741.41.92 (0.044) Counseling in PCC Almost a third (31.9%) of physicians said they discuss screening for aneuploidy with All patients who present for preconception counseling. Of those not selecting All, 80.7% said they did so with patients at risk for aneuploidy. Almost one-in-ten (9.8%) said they did not discuss aneuploidy. The majority (54.4%) of physicians said they discuss carrier screening for heritable genetic disorders (e.g., CF, sickle cell anemia) with All patients who present for preconception counseling, and, of those not selecting All, 84.8% said they did so with patients who have a family history of heritable genetic disorders or other risk factors. Only 4.3% said they did not discuss carrier screening. Physicians were presented with 11 health-related issues (e.g., exercise, nutrition, over the counter drug use) and were asked to rate how important counseling was for each issue during a routine office exam and during preconception care (see Table 4). (Note: Physicians were not asked to rank order the issues in terms of relative importance, nor were they asked to compare the issues in terms of routine or PCC; such calculations were made based on analyzing the mean ratings for each issue.) Physicians rated all issues as more important in PCC than in routine care, although the difference for Exercise was marginal (Wilcoxon signed ranks test. Exercise: P=0.052. All other P’s < 0.001). All issues were rated as very important for PCC, with cigarette smoking and folic acid supplementation rated the most highly important issues in PCC. Cigarette smoking and illegal drug use were rated the most important issues for counseling in routine care. Whereas folic acid supplementation ranked second in importance for PCC, it ranked tenth for routine care. Table 4Percent of physicians indicating how important they think counseling is on several issues for patients who (a) come in for a routine office exam, and (b) those who are planning a pregnancy(Rank of meanImportanceVery importantNot importantMean (SEM) within a or b)Neutral [3–5]on 7 pt scaleFactor I Folic acid supplements:  (a) routine exam(10)36.852.510.74.88 (0.07)  (b) preconception care(2)96.43.4.26.83 (0.03)Factor II Cigarette smoking:  (a) routine exam(1)89.210.60.26.57 (0.03)  (b) preconception care(1)98.41.60.06.86 (0.02) Illegal drug use:  (a) routine exam(2)83.215.21.66.40 (0.05)  (b) preconception care(3)94.55.10.46.74 (0.03) Alcohol consumption:  (a) routine exam(7)49.146.54.45.40 (0.06)  (b) preconception care(4)88.511.50.06.57 (0.03)Factor III Obesity:  (a) routine exam(3)76.123.70.26.11 (0.04)  (b) preconception care(7)82.717.30.06.30 (0.04) Exercise:  (a) routine exam(4)67.031.41.65.87 (0.05)  (b) preconception care(10)68.331.00.75.94 (0.05) General Nutrition:  (a) routine exam(6)51.246.72.15.51 (0.06)  (b) preconception care(8)82.217.50.46.28 (0.04)Factor IV Chronic diseases:  (a) routine exam(5)63.834.91.25.79 (0.05)  (b) preconception care(6)88.310.80.96.45 (0.04) Family health history (inherited disorders):  (a) routine exam(8)51.543.94.65.38 (0.06)  (b) preconception care(5)90.69.20.26.55 (0.04) Over the counter and prescription drug use:  (a) routine exam(9)42.051.46.55.08 (0.06)  (b) preconception care(9)81.617.70.76.27 (0.04) Environmental concerns:  (a) routine exam(11)19.966.313.94.25 (0.06)  (b) preconception care(11)56.139.14.85.47 (0.06) The 11 issues were grouped into 4 factors, for both routine care and PCC (see item labels, Table 4). How physicians defined PCC had a significant effect on importance ratings overall, controlling for years in practice [F(8, 539)=3.437; P < 0.005]. However, the only individual factor on which ratings were distinctly different was folic acid supplementation during routine care: physicians who defined PCC as routine were more likely than those defining it as specialized to rate it as very important (49.3% vs 34.8%) during routine care [t(556)=2.53; P < 0.02]. Gender, controlling for years in practice, also had an overall effect on ratings [F(8, 541)=4.066; P < 0.001]: females tended to rate most items as more highly important than did males. Patient use of PCC Of the physicians who provide PCC, a third (32.3%) said that patients ‘rarely’ present for PCC, almost half (48.5%) said they ‘sometimes’ do, and 17.3% said patients ‘frequently’ present for PCC. Over a third (34.7%) agreed that their patients usually do not plan their pregnancies. Almost half (49.1%) of ObGyns said ‘few’ or ‘none’ of their pregnant patients came in for PCC prior to the pregnancy, and almost two thirds (63.1%) said ‘most’ or ‘all’ of their pregnant patients initially made contact with them once they were already pregnant. See Table 5. The frequency with which patients reportedly present for PCC was positively correlated with physician agreement that their patients plan their pregnancies (P < 0.001), as well as with how frequently physicians recommend PCC to the 6 different patient groups listed in Table 2 (all P’s < 0.001, except ‘Diabetic,’ P=0.01). Table 5Patient use of pre-pregnancy planningPercent of physicians selecting a particular response to questionnaire items:How frequently do you have patients present for PCC?Never or rarelySometimesFrequently32.348.517.3Women that I see usually do not plan their pregnancies Agree (1–2)Neutral (3)Disagree (4–5) (on a 5 point scale)34.733.831.6How many of your pregnant patients came in for NoneFewSomeHalf or More preconception care before they became pregnant? (OB)3.345.838.312.6How many of your pregnant patients initially made AllMostManyHalf or Fewer contact with you once they were already pregnant? (OB)9.453.718.018.9 Physicians were asked to indicate how many of the patients who presented for PCC did so for each of three reasons (7 point scale: None Few Some Half Many Most All). Over four-fifths (82.7%) of physicians indicated that more than half of the patients do so to ensure a healthy pregnancy (mean 5.63±0.062 on 7 pt scale). In contrast, 41.9% said more than half do so because of difficulties conceiving (mean 4.05±.058), and 20.0% said more than half do so because of an elevated risk of a birth defect or developmental disorder (mean 3.13±0.064). Discussion The purpose of this study was to examine obstetrician-gynecologists’ opinions and beliefs about PCC, and how frequently they perceived their patients to be utilizing PCC. Our findings document that most physicians think it is an important issue, that it has a positive effect on pregnancy outcomes, and that they are appropriately trained to provide it (Table 3). However, respondents reported that few of their patients seek PCC. Explanations for this lack of patients seeking PCC likely include the fact that almost 50% of pregnancies in the U.S. are unplanned [7], the lack of third party reimbursement for PCC visits, and patients’ rather poor understanding of the potential benefits of PCC to the lifetime health of future offspring. We did find an association between the frequency with which physicians recommend PCC and the frequency with which patients present for PCC. It is also possible that some women receiving routine health care prior to a first pregnancy see an internist, family physician, or nurse practitioner for such care, and only seek the care of an obstetrician-gynecologist once they become pregnant. Thus, it is possible that our study of obstetrician-gynecologists does not reflect the full extent to which women may be seeking PCC. However, a recent study of primary care providers (not obstetrician-gynecologist) also found that few patients were receiving PCC [10]. Stumbling blocks to increasing utilization of PCC may include physicians’ beliefs that time devoted to PCC is not reimbursed and that there is not enough time to provide PCC to all women of childbearing age. Regarding this latter point, opinions differ within the medical community as to who the target population for PCC should be. Three general target groups include women at high risk for poor birth outcome, women planning a pregnancy, and all women of childbearing age [11]. In this study, the vast majority of physicians said they frequently recommend PCC to diabetic and obese women planning a pregnancy, as well as to women generally who are planning a pregnancy (see Table 2). Far fewer obstetrician-gynecologists said they frequently recommend PCC to women described simply as sexually active (54%) or to those using birth control (36%). This suggests that our survey sample population views women who have consciously decided to plan a pregnancy as the most appropriate target group for PCC. Almost 9 in 10 (87%) physicians defined PCC as specialized pre-pregnancy care focusing on issues not typically addressed during a routine exam, and the remaining 13% defined it as routine well-woman care occurring during the reproductive years, prior to a pregnancy. Those who defined it as routine were more likely to offer such counseling to women who were sexually active or using birth control than were those defining it as specialized care. One aspect of PCC includes counseling women on topics that encourage a healthy pregnancy, although many such topics are clearly relevant to the general health of any woman. Physicians were asked to rate eleven topics in terms of their importance for counseling during PCC and routine care, such as exercise, nutrition, and over the counter drug use (see Table 4). All were considered very important topics for counseling in PCC, with cigarette smoking and folic acid supplementation rated the most highly important for PCC. For counseling in routine care, folic acid supplementation received the second lowest score of all, with under two-fifths (37%) of physicians rating it as important. This points to the need for continued national attention to increase consumption and supplementation of folic acid. Our study has limitations that should be acknowledged. The response rate was 60%, and our findings are based on the responses of 579 non-subspecialist obstetricians and gynecologists. While our study may be subject to non-response bias, we believe that our findings are reliable. The typical response rate in these survey studies is approximately 35–60%, and our response rate was at the high end of expected participation. In addition, the responses were derived from geographically diverse locations and from physicians from different practice types reflecting the influence of physician location and practice type. Whereas physicians who were more interested in the topic of the survey may have been more likely to respond, a subset of our subject pool was comprised of CARN members who respond to several questionnaires a year covering a wide variety of topics; it is unlikely that Preconception Care is a topic of greater interest to this group than to the group of randomly selected ACOG members. CARN members differed significantly from non-CARN subjects on only one non-demographic response, and the mean age and male to female ratio of respondents closely matched those of the larger group to whom the survey was sent and of ACOG Fellows and Junior Fellows as a whole, all of which suggests that response bias was minimized. Recent studies have recommended offering PCC “opportunistically,” incorporating information relevant to preconception care into routine well-woman visits [10, 12]. Most topics relevant to preventing congenital birth defects are also important to the overall health of women, pointing out the compatibility of the two counseling contexts. The one exception to this in our questionnaire was folic acid supplementation, and it would need greater emphasis in routine care to take advantage of opportunistic counseling. We found that those physicians who defined PCC as “the same as routine well-woman care” differed from those defining it as “specialized” in several ways, including increased importance of counseling on folic acid supplementation during routine care. A recent study found that topics such as vitamin supplements and alcohol use were rarely discussed during routine care by a high proportion of obstetrician-gynecologists, and by an even higher proportion of family practitioners [13]. They also found that physician recommendations influenced women's decisions to take folic acid. Obstetrician-gynecologists recognize the importance of preconception care and provide this care for their patients. However, lack of third party reimbursement, lack of time during office visits due to competing demands, and lack of consumer awareness pose barriers to effective implementation of PCC. The fact that almost half of all pregnancies in the U.S. are unplanned poses an even greater challenge. Continued efforts are needed to raise awareness of the importance of PCC by consumers, health care providers, third party carriers, and policy makers.

Planta-4-1-2270913

Identification of an OsPR10a promoter region responsive to salicylic acid

Orysa sativa pathogenesis-related protein 10a (OsPR10a) was induced by pathogens, salicylic acid (SA), jasmonic acid (JA), ethephon, abscisic acid (ABA), and NaCl. We tried to analyze the OsPR10a promoter to investigate the transcriptional regulation of OsPR10a by SA. We demonstrated the inducibility of OsPR10a promoter by SA using transgenic Arabidopsis carrying OsPR10a:GFP as well as by transient expression assays in rice. To further identify the promoter region responsible for its induction by SA, four different deletions of the OsPR10a promoter were made, and their activities were measured by transient assays. The construct containing 687-bp OsPR10a promoter from its start codon exhibited a six-fold increase of induction compared to the control in response to SA. Mutation in the W-box like element 1 (WLE 1) between 687 and 637-bp from TGACA to TGAAA completely abolished induction of the OsPR10a promoter by SA, indicating that the WLE 1 between −687 and −637 of OsPR10a promoter is important in SA-mediated OsPR10a expression. We show for the first time that the W-box like element plays a role in SA mediated PR gene expression. Introduction Plants have developed defense mechanisms to recognize pathogens and subsequently activate defense-related genes, such as pathogenesis-related proteins (PR proteins). The major families of PR proteins have been grouped into at least 14 different classes, primarily on the basis of their amino acid sequences (Van Loon and Van Strien 1999). Although the biological and/or biochemical functions of many PR proteins remain unclear, PR2 (β-1, 3-glucanase activity) and PR3 (chitinase) proteins have been shown to inhibit fungal growth (Woloshuk et al. 1991; Sela-Buurlarge et al. 1993). These responses are not limited to pathogen attack and can be induced by defense signaling molecules such as SA, JA and ET (Dempsey et al. 1999; Pieterse and van Loon 1999). To study the defense signaling in plants, many groups have isolated promoters of PR proteins in several plant species, such as Arabidopsis, tobacco, pepper, and rice (Malnoy et al. 2003; Hong et al. 2005; Li et al. 2005; Liu et al. 2005a; Lee and Hwang 2006). Expression profiles of PR proteins, such as OsPR1, OsPR10 and OsPR1b were reported. Originally, OsPR10a was known to be induced by probenazole and thus, was called a probenazole-inducible gene, PBZ1 (Midoh and Iwata 1996). Later, PBZ1 was renamed as OsPR10a because it shares a similar sequence with (has sequence similarity to) PR-10 proteins. The investigators reported that OsPR10a is only induced by probenazole but not by ethephon, NAA, SA, NaCl or mannitol in rice leaves. In contrast, Rakwal et al. (2001) reported that OsPR10a is induced by JA, SA, and ABA but not by IAA or GA in light. Ryu et al. (2006) found similar results using RT-PCR. Chen et al. (2006) reported that the elicitor derived from Magnaporthe grisea induces OsPR10a. However, there has been only one study on OsPR10a and OsCHNIII promoters, even though many reports are available for the expression profile of PR genes in rice (Rakwal et al. 2001; Hashimoto et al. 2004; Chen et al. 2006; Ryu et al. 2006). The authors reported that OsPR10a and OsCHNIII promoters are induced by an elicitor derived from Magnaporthe grisea by a transient assay in vitro. However, the cis elements were not analyzed. Most promoters induced by pathogens or SA contain the W-box, GCC box, RAV1 AAT, or ASF1 motif, etc. (Li et al. 2005; Lee and Hwang 2006; Sohn et al. 2006). Their cis-elements have been identified by series deletion of the promoter and site directed mutagenesis of its plausible site. Maleck et al. (2000) analyzed the transcriptome of Arabidopsis under defense inducing conditions, and they studied induced promoters such as PR1. The W-boxes ((T)TGACC/T) are enriched in the PR1 regulon promoter. They also described that the W-box like element ((T)TGACA) is also enriched in PR-1 regulon promoters even though there is no evidence that WRKYs bind to this motif (Maleck et al. 2000). Transcription factors that can recognize the cognate cis element were identified by methods such as the gel-mobility shift assay, yeast-one hybrid, transient assay in plant. WRKY, ERF, RAV, bZIP, MYB, etc. have been shown to be involved in the defense signaling (Rushton et al. 1996, 2002; Eulgem et al. 1999; Kirsch et al. 2001; Heise et al. 2002). The interaction of a transcription factor to its cognate cis element is a key step in the process of defense signaling. Among transcription factors, WRKY proteins are the most extensively studied in defense signaling (Eulgem et al. 1999; Robatzek and Somssich 2001; Shimono et al. 2007). Asai et al. (2002) reported that AtWRKY22 and AtWRKY29 regulate FLS2-mediated defense signaling. The complex of TGA factor and NPR1 binds to the LS7 in the PR-1 promoter of Arabidopsis (Johnson et al. 2003). Furthermore, the TGA/NPR1 complex is as well conserved in rice as in Arabidopsis (Fitzgerald et al. 2005). Recently, there are three reports that OsWRKY45, OsWRKY71 and OsWRKY03 regulate the defense signaling in rice, respectively (Liu et al. 2005b, 2006; Shimono et al. 2007), implicating that WRKYs also binds to the W boxes in rice as it does in Arabidopsis. In this study, we analyzed the expression profile of OsPR10a. We isolated its promoter and analyzed its cis-elements. We also identified the WLE1 (TGACA) controlling induction of OsPR10a promoter by SA. This is the first report that the W-box like element actually plays a role in SA-mediated defense signaling. Material and methods Plant materials Rice seedlings (Oryza sativa cv. Hwachung; seeds from Dr. Wan-He Ye, NIAST, Suwon, South Korea) were grown in a greenhouse at 28°C for 3 weeks. Three-week-old rice seedlings were washed, incubated in tap water for 2 days, and then treated with SA, JA, ethephon, ABA, or NaCl at 1 mM, 100 μM, 100 μM, 100 μM, and 200 mM, respectively. Rice leaves were harvested at the times indicated in the figures. For bacterial inoculations, a strain of Xanthomonas oryzae pv oryzae KXO98 (Xoo; obtained from Korean Agricultural Culture Collection, KACC, Suwon, South Korea) incompatible to O. sativa cv. Hwachung was grown in PSA medium (10 g peptone, 10 g sucrose, 1 g sodium–glutamate, and 15 g agar per L) for 2 days and then resuspended in 1 mM MgCl2 to a final OD600 of 0.5. Xoo was sprayed on 3-week-old rice seedlings. After inoculation, plants were incubated in a humidity chamber for 24 h. Samples were taken at the times indicated in the figures and were immediately frozen in liquid nitrogen and stored at −80°C until further analysis. Isolation of OsPR10a promoter Based on an annotation of the rice genome, a −1000 bp fragment of the OsPR10a promoter was obtained by PCR from rice genomic DNA using an OsPR10a gene-specific primer sets. These primers were designed from the Genbank sequence AL845342. PCR was performed for 30 cycles under the following condition: 94°C for 30 s, 53°C for 1 min, and 72°C for 1 min, followed by a final extension at 72°C for 7 min. The primer sets are as follows: 5′-AAAAAGCAGGCTTGTTTTGAATGCTGGAATGATAA-3′, and 5′-AGAAAGCTGGGTCACTGAAGATATAATCTA-3′. The underlined sequences match the attB1 and attB2 sites for the Gateway cloning system (Invitrogen, Carlsbad, CA, USA). A 1000 bp amplified PCR product was cloned into pDONR221 to make an entry clone by BP clonase (Invitrogen); successful insertion was confirmed by sequencing. Promoter-LUC constructs The reporter constructs used in the transient expression assays in this study were prepared according to the following procedure. For a 1.0 kb OsPR10a promoter, 1000-bp upstream from the start codon of OsPR10a was cloned by BP reaction into pDONR221 to make the −1000 bp PR10a promoter entry clone described in the previous section. −1000-OsPR10a:LUC was created by LR reaction with the −1000-PR10a entry clone and promoter destination vector (attB1-ccdB-Cmr-attB2-LUC, unpublished results; Invitrogen). On the basis of W-boxes involved in the activation of defense genes in plants to construct the deleted OsPR10a:LUC construct, we amplified the OsPR10a promoter region using these sense primers: −818: 5′-AAAAAGCAGGCTCGTGACATCAGATTGAGTAT-3′−687: 5′-AAAAAGCAGGCTCGATAAAGGGTATTTGTTTA-3′−637: 5′-AAAAAGCAGGCTACCTATCATCTAAAAGCATT-3′. We also used the antisense primer 5′-AGAAAGCTGGGTCACTGAAGATATAATCTA-3′. PCR was performed for 30 cycles under the following condition: 94°C for 30 s, 53–55°C for 1 min, and 72°C for 1 min, followed by a final extension at 72°C for 7 min. The sequences underlined match attB1 and attB2 sites in the Gateway cloning system. These −818, −687, and −637 bp amplified PCR products were cloned into pDONR221 to make entry clones by BP clonase and confirmed by sequencing. 818-, 687-, and 637-OsPR10a:LUC were created by LR reaction with the 818-, 687-, and 637-OsPR10a entry clones and the promoter destination vector (attB1-ccdB-Cmr-attB2-LUC, unpublished results). Site directed mutagenesis The mutagenized reporter constructs used in the transient expression assays in this study were prepared according to the manufacturer’s instruction (Stratagene, La Jolla, CA, USA). For the mutagenized 687 bp-OsPR10a promoter, −1000-OsPR10a entry clone (20 μg) was added to 1 μL of 10× reaction buffer, 1 μL (1 ρmole) phosphorylated specific mutagenic primer sets, 0.5 μL 10 mM dNTPs, 0.6 μL Quick solution; and the volume was adjusted to 10 μL with sterile deionized water. The solution was added to 0.3 μL of Pfu Turbo DNA polymerase (2.5 units/μL). The reaction matrix mix was used for subsequent PCR. Specific PCRs were performed for 18 cycles under the following conditions: 95°C for 50 s, 60°C for 50 s, and 68°C for 1 min, followed by a final extension at 68°C for 7 min. After generation of the mutgenic double stranded plasmid containing staggered nicks, the product was treated with Dpn I and incubated at 42°C for 60 min. The nicked plasmid incorporating the desired mutations was purified with phenol and chloroform extraction and ligated with T4 DNA Ligase at 16°C for overnight. Five microliters of the mutated plasmid was transformed into E.coli (DH5α) cells. After transformation, the plasmid DNA was isolated from the mutagenic transformant and confirmed by sequencing. For PCR of mutant strand synthesis reaction, the following mutagenic primer pairs were used: 5′-TGAAATGTAGTCGTACCTATCA-3′ and 5′-TGCTCTGAGATGGGTCTAAACA-3′. Particle bombardment and transient expression assays Leaf bombardments were performed in a Biolistic PDS-1000/He particle delivery system using 1100-p.s.i. rupture disks (BioRad, Hercules, CA, USA). Plasmid DNAs for particle bombardment were prepared as described by the manufacturer’s instructions. For reporters, −1000-bp OsPR10a:LUC and 818-, 687-, 637-OsPR10a:LUC and m687-OsPR10a:LUC were used; 35S:RLUC was used as an internal control to normalize LUC activities between samples after bombardments. About 2 cm lengths of one-week old rice seedlings grown in the dark were cut and incubated on a plate in 1/2 MS medium overnight (Murashige and Skoog 1962). Tungsten particles coated with 1000-, 818-, 687-, 637-OsPR10a:LUC, or m687-OsPR10a:LUC, and the internal control were delivered into leaf segments by the particle delivery system (BioRad). Leaf segments were incubated at 28°C for 24 h with buffer (1/2 MS medium) or with 1 mM SA and then harvested. Leaf segments were ground in liquid nitrogen and dissociated in 1× passive buffer. The luciferase activities from protein extracts were measured by a dual luciferase system (Promega, Madison, WI, USA) with a luminometer (Aureon Biosystems, Vienna, Austria). RT-PCR analysis Leaf samples were ground to powder in liquid nitrogen, and total RNA was extracted using the Trizol reagent according to the manufacturer’s instructions (Invitrogen). For reverse transcription, total RNA (1 μg) was added to 1 μL of oligo (dT)16 and 1 μL of gene specific primer sets (0.5 ρmole); and the volume was adjusted to 15 μL with sterile deionized water. The solution was incubated at 70°C for 5 min, then immediately transferred to ice before the addition of 35 μL of reverse transcriptase master mix containing 10 μL 5× buffer, 3 μL 0.1 M DTT, 5 μL 10 mM dNTPs, 1 μL (200 units/μL) M-MLV RTase (Promega) and 0.2 μL (40 units/μL) RNasin (Promega). The reaction was incubated at 42°C for 90 min before heat inactivation at 65°C for 10 min. Two microliters of each reverse transcriptase reaction was used for subsequent PCR. Gene specific PCRs were performed for 35 cycles under the following conditions: 94°C for 30 s, 53°C for 1 min, and 72°C for 1 min, followed by a final extension at 72°C for 7 min. Samples were visualized on 1.2% agarose gels. For RT-PCR analysis of OsPR10a genes in rice, the following primer pairs were used: OsPR10a (D38170)5′-GCTACAGGCATCAGTGGTCA-3′ and 5′-GACTCAAACGCCACGAGAAT-3′,OsActin (XM469569) 5′-TCCATCTTGGCATCTCTCAG-3′and 5′-GTACCCGCATCAGGCATCTG-3′. Generation of transgenic Arabidopsis, induction with SA, and fluorescence microscopy OsPR10a:GFP was constructed by LR reaction with pBGWFS7 (Gateway™; Department of Plant Systems Biology, VIB-Ghent University, Belgium) and the OsPR10a promoter entry clone described in the previous section, and then transformed into Agrobacterium tumefaciens GV3101 for Arabidopsis. Arabidopsis (Columbia ecotype) was transformed with A. tumefaciens GV3101 carrying OsPR10a:GFP::GUS and 35S:GFP(35S:pBGWFS7::GUS) as an internal control. A bacterial suspension of A. tumefaciens GV3101 carrying OsPR10a:GFP::GUS and 35S:GFP::GUS was sprayed on the unopened flowers of Arabidopsis. T1 plants were screened by 0.3% Barstar spray (Misung, Daejeon, South Korea). Samples were taken from independent T1 plants for RT-PCR analysis. T2 plants were also screened by 0.3% Barstar spray. Three individual T2 plants in each line were used for induction with 1 mM SA at 28°C for 72 h. After induction, transgenic plants carrying OsPR10a:GFP and 35S:GFP were examined by fluorescence microscopy using an Olympus SZX-RFL3 (Olympus Optical Co., LTD, Tokyo, Japan). Excitation and emission filters SZX-FGFP and SZX-FGFPA were used for GFP and GFPA (Ex 460–490/Em510- for GFP and Ex460–490/Em510–550 for GFPA). Images were captured with a JP/FV300 camera (Olympus). Results Expression patterns of OsPR10a in response to different stimuli Several research groups have reported some discrepancy for the expression patterns of OsPR10a to some stimuli. Here we looked at the expression of OsPR10a by RT-PCR, which is a more sensitive method than reported previously. First, we tested whether OsPR10a is induced by a pathogen, as reported previously (Midoh and Iwata 1996; Ryu et al. 2006; Fig. 1a). Induction of OsPR10a started at 6 h and reached a maximum at 48 h after Xoo infection, as reported previously (Ryu et al. 2006). We also determined whether OsPR10a was induced by biotic elicitors such as SA, JA, and ethephon. For the fist time, we show that OsPR10a is induced by ethephon (Fig. 1b). Witzh regard to abiotic stress treatments, OsPR10a was induced by NaCl and ABA (Fig. 1c). Taken together, we conclude that OsPR10a is induced by the pathogen Xoo, SA, JA, ethephon, NaCl, and ABA. Fig. 1Expression pattern of OsPR10a in rice leaves treated with Xoo and five compounds. a Three-week-old rice seedlings were infected with Xoo and were harvested at 0, 6, 12, 24, and 48 h. b, c Three-week-old rice seedlings were treated with SA, JA, ethephon, ABA, or NaCl and were harvested at 0, 6, 12, 24, and 48 h. Total RNA was isolated from each sample, and RT-PCR was performed using OsPR10a specific primer pair. Transcript levels of OsActin show that equal amounts of RNA were used in the RT-PCR samples The OsPR10a promoter is induced by SA treatment as shown in transient-assay system We analyzed the expression of the OsPR10a gene to pathogens and various phytohormones. We focused on the SA-mediated response of the OsPR10a gene. To investigate how the OsPR10a gene was transcriptionally regulated by SA, we isolated the OsPR10a promoter in a 1.0 kb genomic DNA fragment upstream from the start codon of the OsPR10a gene by PCR. To analyze whether the 1.0 kb OsPR10a promoter was activated by SA as expected by its expression pattern, we carried out a transient assay using particle bombardment. The 1.0 kb fragment of OsPR10a promoter was used to make a reporter construct (OsPR10a:LUC). Its schematic diagram is shown in Fig. 2a. OsPR10a:LUC was introduced into rice leaves by particle bombardment; leaf segments were then treated with either buffer or SA. Protein extracts were prepared from samples after 24 h post-treatment, and their relative luciferase activities were measured. OsPR10a promoter activities were expressed as relative luciferase activities. Figure 2b shows a representative graph out of more than three independent experiments. The absolute values from each experiment were different, but the relative ratios from each sample were similar. Luciferase activity in the SA-treated sample was about two-fold higher than in non-treated (control) samples (Fig. 2b). This result indicates that the OsPR10a promoter is activated by SA, based on its expression profile. Fig. 2OsPR10a promoter activity in response to SA: a schematic representation of the OsPR10a promoter in the reporter construct. b A transient assay showing the OsPR10a promoter in response to SA. OsPR10a:LUC was bombarded into rice leaves, which were then incubated in MS medium or MS medium containing 1 mM SA at 28°C for 24 h. Protein extracts were made by dissociation in passive lysis buffer as described in “Materials and methods”. Relative luciferase activity is the ratio of the value obtained with the SA-treated OsPR10a:LUC divided by the value obtained with the buffer-treated OsPR10a:LUC. Bars indicate the standard error of three replicates The OsPR10a promoter is induced by SA treatment in stably transformed Arabidopsis We further investigated whether SA, as seen in the transient assay, activates the OsPR10a promoter using a transgenic approach. A 1.0 kb fragment of the OsPR10a promoter was cloned into a promoter-less GFP::GUS expression vector to make a OsPR10a:GFP::GUS construct, and was then introduced into Arabidopsis by Agrobacterium-mediated transformation (Fig. 3a). Transgenic Arabidopsis plants (T1) were screened by spraying with 0.3% Barstar and then with self-crossing. Induction of the OsPR10a promoter by SA was analyzed by GFP fluorescence in T2 transgenic Arabidopsis seedlings treated with either SA or buffer (Fig. 3b). In these GFP filter images (>510 nm), transgenic Arabidopsis carrying OsPR10a:GFP::GUS exhibited an orange fluorescence in the SA-treated sample because the green fluorescence from GFP was mixed with the red fluorescence from plants themselves (Fig. 3b, middle panel). The green fluorescence from GFP is shown more clearly using a GFPA filter (510–550 nm; bottom panel of Fig. 3b). As shown in Fig. 3b, the OsPR10a promoter was clearly activated by SA in transgenic Arabidopsis. Fig. 3Fluorescence images of Arabidopsis transgenic plants carrying OsPR10a:GFP: a schematic diagram of OsPR10a:GFP::GUS fusion construct. b Induction of OsPR10a promoter by SA. OsPR10a:GFP::GUS was introduced into Arabidopsis by Agrobacterium-mediated transformation. Transgenic Arabidopsis seedlings carrying the OsPR10a:GFP::GUS was examined using fluorescence microscopy after SA treatment at 72 h. Non-transgenic Arabidopsis seedling was used as a control (left panel at mock and SA treatments). Shown are the bright-field images (upper panel Bright), the green fluorescent images using GFP filter (middle panel GFP) and the GFPA filter (bottom panel GFPA). Images are representatives from two independent experiments. The experiments were repeated at least twice Analysis of cis-elements of OsPR10a promoter In order to find a cis-acting element of the promoter in response to SA, an analysis was done using the PLACE program (a database for PLAnt Cis-acting Elements located at http://www.dna.affrc.go.jp/cDNA/place) (Fig. 4). Among many putative cis elements, we only indicated cis-elements in boxes known to be related to defense inducers, such as the pathogens, SA, JA, and ethephon, of the OsPR10a gene shown in Fig. 1 (Shinshi et al. 1995; Eulgem et al. 1999; Kagaya et al. 1999). The OsPR10a promoter analyzed by the PLACE program contains four W-boxes, whose detail sequences are different; there are one canonical W-box ((T)TGACC/T) and three W-box like elements (WLE 1) containing TGAC core (TGACA). There would be more W-box like elements in defense gene regulon promoters. Therefore, we decided to name TGACA as the W-box like element 1 (WLE1). In addition, there are three RAV1AAT elements, and one ASF1 motif element (Fig. 4). The W-box, RAV1AAT, and ASF1 motif are known to be cis-elements of the WRKY, RAV1, and bZIP proteins, respectively (Abe et al. 1997; Chen and Chen 2002; Yamamoto et al. 2004). The WRKY, RAV1, or bZIP proteins might be involved in the response of the OsPR10a promoter to SA. In addition to them, there are many cis-elements involved in ABA responsiveness, even though they are not indicated in Fig. 4. These elements might be involved in the induction of OsPR10a by ABA as shown in Fig. 1c. Interestingly, there is no cis-element, such as the JA responsive element (JERE) (AGACCGCC) or the ethylene response element (ERE) (AGCCGCC), which is the binding site for ethylene response element binding proteins (EREBP), despite the fact that OsPR10a was induced by JA and ethephon. Fig. 4Putative cis-acting elements in 1.0 kb OsPR10a promoter. The putative cis-elements are indicated in boxes and its name is given above each element. Arrows indicate the direction of the cis-element. W-box WRKY transcription factor binding site; RAV1AAT RAV transcription factor binding site; ASF1 motif bZIP factor binding site; WLE1 putative WRKY transcription factor binding site Deletion analysis of the OsPR10a promoter to identify the regions responsible for the induction by SA To identify the region of the OsPR10a promoter involved in the response to SA, we made serial deletions of the OsPR10a promoter by PCR (Fig. 5a). Deletions, beginning with the locations −818, −687, and −637, were fused to the LUC coding sequences and 3′ nopaline synthase gene terminator (Fig. 5a). These four constructs were tested for SA inducibility of the OsPR10a promoter by introducing them into rice leaves using particle bombardment and then treating them with either buffer or SA for 24 h. Protein extracts were made from the bombarded leaves and their luciferase activities were measured (Fig. 5b). In the case of the 1.0 kb OsPR10a:LUC construct, luciferase activity was increased up to two fold over the control with SA treatment but not in the 818:LUC construct, indicating that there is a weak positive cis-element in region I between −1000 and −818 bp of OsPR10a promoter (Fig. 5a). One ASF1 motif was found in region I. The exact positive element in this region has not yet been identified. Luciferase activity in the 687:LUC construct was increased up to sixfold with SA treatment, indicating that there is a negative element in region II between −818 and −687 bp of the OsPR10a promoter. There is only one WLE1 with the TGAC core (TGACA) and one RAV1AAT element in region II that is known to be bound by transcription factors associated with the defense signaling (Fig. 4). Besides this, there are many putative cis-elements in region II (data not shown). Therefore, the exact negative element has not yet been determined. In the 637:LUC construct, there was only about a two-fold increase in luciferase activity with SA treatment, indicating that there is at least one positive element between −687 and −637 bp (region III) and another one between −637 and 1 bp (region IV) of the OsPR10a promoter. There is only one WLE1 containing the TGAC core (TGACA) in region III, suggesting that this element may play an important role in the strong inducibility of the 687:LUC construct by SA (Table 1). The W-box, RAV1AAT, and WLE1 are found in region IV of the OsPR10a promoter, and at least one of them can act as a weak positive element. Fig. 5Deletion analysis of OsPR10a promoter: a schematic diagrams of serial deletion constructs of OsPR10a promoter. The numbers to the left of each construct indicate the distance from the start codon ATG. The predicted cis-elements ( W-box, RAV1AAT, and ASF1motif) are indicated by their respective abbreviations. The start codon, ATG, is written in bold. b Luciferase activity in deletion constructs of the OsPR10a promoter. Each deletion construct OsPR10a:LUC was bombarded into rice leaves, which were incubated in MS liquid medium or MS medium containing 1 mM SA at 28°C for 24 h. Protein extracts were made by dissociation in passive lysis buffer as described in “Materials and methods”. Bars indicate the standard error of three replicates. The values are the ratio of the value obtained from each deletion constructs of OsPR10a promoter treated with SA or buffer divided by the value obtained from 1.0 kb OsPR10a promoter construct treated with bufferTable 1The list of putative cis-acting elements of the OsPR10a promoter in region IIIRegionaPositionbcis-Elements (#)/putative factor cIII−687 to −637GT1CONSENSUS(1)/GT-1, WLE1 (1)/WRKYaOsPR10a promoter was divided into four regions depending on the presence of W-box. The region III between −687 and −637 bp of OsPR10a promoterbindicates the distance of upstream from the start codon of OsPR10acPutative cis-acting elements in region III of OsPR10a promoter were analyzed using PLACE (a database for PLAnt Cis-acting Elements located at the web site (http://www.dna.affrc.go.jp/cDNA/place) Mutation of a W-box like element in OsPR10a promoter abolished its SA inducibility SA inducibility of the 687:LUC construct is the highest among deletion constructs, and only one WLE1 with the TGAC core is present in region III. To further verify this, the WLE1 in region III was mutagenized from TGAC to TGAA (Fig. 6a). Eulgem et al. (1999) reported that the WRKY protein couldn’t bind to a TGAA sequence; therefore, this mutation prevents the association of WRKY to the WLE1 of the OsPR10a promoter. Interestingly, SA inducibility of 687 bp-OsPR10a promoter was completely abolished in the mutagenized 687 bp-OsPR10a promoter, indicating that this WLE1 is involved in the SA inducibility of OsPR10a. Fig. 6The effect of the mutation in the WLE1 of OsPR10a promoter region III. a Sequences of the WLE1 (the −659 to −644 bp) in the OsPR10a promoter and the mWLE1 with the TGAAA instead of TGACA. The WLE1 sequence is underlined and bolded. The asterisk represents the mutated base in the WLE1. b Luciferase activity in 687:LUC and m687:LUC in rice leaves. Bar indicates the standard error of the three replicates Discussion Plant defense mechanisms to pathogen attack have been extensively studied in Arabidopsis; however, it is not well studied in rice. To study the defense mechanisms in rice, we tried to understand the transcriptional regulation of OsPR10a because OsPR10a has been used as a marker of induction for the defense response in rice (Ryu et al. 2006; Chen et al. 2006). OsPR10a was originally cloned by Midoh and Iwata (1996). They reported that OsPR10a was induced by Magnaporthe grisea and probenazole but not by ethephon, NAA, SA, NaCl, mannitol, and wound. More recently, Rakwal et al. (2001) reported that OsPR10a was induced by various phytohormones, such as SA, JA, and ABA, but not by IAA and GA. There were discrepancies between these two reports. In our study, we have shown that OsPR10a was induced by Xanthomonas oryzae pv. oryzae, phytohormones, such as JA, SA, ethephon, ABA and NaCl, but not by IAA and GA. In the case of JA, our result is consistent with the previous report (Rakwal et al. 2001). However, in the case of SA, our result is consistent with the findings by Rakwal et al. (2001), but not with that one by Midoh and Iwata (1996). There might be some differences in the method of SA treatment. We treated rice seedlings with SA by the soil drenching method because our previous result, based on the expression of OsPR1 gene, indicated that spraying rice leaves with SA does not reliably induce the defense response. In the case of ethephon, our data are also not consistent with the results from Midoh and Iwata (1996). Their data on the expression of OsPR10a were generated by Northern blots, whereas our results were generated by a more sensitive method, RT-PCR. Our result is the first report on the response of OsPR10a to ethephon. Our data suggest that OsPR10a is induced by three different defense signaling transducers (SA, JA, and ethephon). For abiotic stress treatments, OsPR10a was induced by NaCl and ABA. In the case of NaCl, our result also differs from the data shown by Midoh and Iwata (1996). We think that there is a sensitivity difference due to the detection methods of OsPR10a mRNA between RT-PCR and Northern hybridization as in the case of ethephon. In the case of ABA, our result is consistent with a report by Rakwal et al. (2001). Taken together, we conclude that OsPR10a is induced by the pathogens, SA, JA, ethephon, NaCl, and ABA. In this study, we focused on SA mediated induction of OsPR10a because SA mediated defense signaling is the most well studied in Arabidopsis. The OsPR10a promoter was isolated to study the transcriptional regulation of OsPR10a gene. Gene activity was induced by SA in a transient assay system as expected by its expression profile. Chen et al. (2006) reported that it was induced by an elicitor derived from Magnaporthe grisea as shown in a transient assay system. SA might be involved in elicitor-mediated defense signaling, yet there was no report on the activity of OsPR10a promoter in plants. Our data now have shown that the OsPR10a promoter was also activated by SA in stably transformed plants, as we have seen in a transient-assay system. The cis-acting elements of the OsPR10a promoter were analyzed to find the elements responsible for its induction by SA. It resulted in many putative cis-acting elements. The transcription factors which play an important role in defense signaling are WRKY, ERF, bZIP, MYB, RAV1, etc. (Ruston and Somssich 1998; Singh et al. 2002; Sohn et al. 2006). Therefore, we searched binding sites in the OsPR10a promoter for WRKY, ERF, bZIP, MYB, and RAV1. We found W-box, RAV1AAT element, and ASF1 motif element. The W-box, RAV1AAT, and ASF1 motif are known to be the binding sites of the WRKY, RAV1, and bZIP proteins, respectively (Abe et al. 1997; Chen and Chen 2002; Yamamoto et al. 2004). This suggests that the WRKY, RAV1, or bZIP proteins might be responsible for the induction of OsPR10a promoter by SA. We also found several ABRE sequences that are known to be responsible for ABA responsiveness of the gene (Shinozaki and Yamaguchi-Shinozaki 1996). This element might be involved in the induction of OsPR10a by ABA. There is no cis-element, such as JERE or ERE even-though OsPR10a was induced by JA and ethephon. Induction of OsPR10a by JA and ethephon appears to occur indirectly through some other transcription factors bound to the OsPR10a promoter. Based on cis-elements found in OsPR10a promoter, three different deletion constructs (818:LUC, 687:LUC, and 637:LUC) were made. Induction of the OsPR10a promoter by SA was completely abolished using the 818:LUC construct, indicating that at least one weak positive element exists in region I. In the 687:LUC construct, there was approximately a sixfold increase compared to the 1.0 kb OsPR10a promoter construct. This suggests that at least one negative element exists in region II. In the 637:LUC construct, its activity was dramatically reduced compared to the 687:LUC construct, suggesting that there is a positive element in region III. Induction of the promoter by SA was also maintained in the 637:LUC construct, suggesting a positive element is present in region IV. In region III, there were a number of available cis elements in the OsPR10a promoter (Table 1). However, only one WLE1 with the TGAC core was present in region III. Its nucleotide sequence is different from the canonical W-boxes ((T)TGACC/T) (Maleck et al. 2000). However, they also described that the WLE1 (TGACA) is enriched in PR-1 regulon promoters. To verify involvement of the WLE1 in response to SA, its sequences were mutagenized (Eulgem et al. 1999). The mutation of the WLE1 from TGAC to TGAA in region III completely abolished the induction of the 687:LUC construct by SA. This suggests that the WLE1 is important in the expression of the OsPR10a gene in response to SA. This is the first finding that the WLE1 (TGACA) is important in SA mediated PR gene expression. Interaction of transcription factors and cis-acting elements constitute a key step in the defense signaling. The OsTGA factor interacts with OsNPR1 as reported in Arabidopsis (Chern et al. 2001; Yu et al. 2001). These authors suggest that NPR1-mediated defense signaling in Arabidopsis is conserved in rice. However, they did not report the identity of the target gene of this complex. Liu et al. (2005b) reported that OsWRKY12 induces the expression of OsNPR1 and OsPR1b; however, they did not show evidence that OsWRKY12 directly regulates the expression of OsNPR1 and OsPR1b since their experiments utilized transgenic plants over-expressing OsWRKY12. Here, we suggest that WRKY may play a major role in SA-mediated OsPR10a expression in rice. However, we cannot exclude involvement of other transcription factors in SA-mediated expression of OsPR10a. In the near-future, we will carry out electrophoretic mobility assays of the WRKY proteins to the WLE1 described in this study. We will further address what kinds of WRKY proteins regulate the OsPR10a promoter and identify the different partners required for SA-mediated OsPR10a expression.

Plant_Mol_Biol-3-1-2140092

PsRBR1 encodes a pea retinoblastoma-related protein that is phosphorylated in axillary buds during dormancy-to-growth transition

In intact plants, cells in axillary buds are arrested at the G1 phase of the cell cycle during dormancy. In mammalian cells, the cell cycle is suppressed at the G1 phase by the activities of retinoblastoma tumor suppressor gene (RB) family proteins, depending on their phosphorylation state. Here, we report the isolation of a pea cDNA clone encoding an RB-related protein (PsRBR1, Accession No. AB012024) with a high degree of amino acid conservation in comparison with RB family proteins. PsRBR1 protein was detected as two polypeptides using an anti-PsRBR1 antibody in dormant axillary buds, whereas it was detected as three polypeptides, which were the same two polypeptides and another larger polypeptide 2 h after terminal decapitation. Both in vitro-synthesized PsPRB1 protein and lambda protein phosphatase-treated PsRBR1 protein corresponded to the smallest polypeptide detected by anti-PsRBR1 antibody, suggesting that the three polypeptides correspond to non-phosphorylated form of PsRBR1 protein, and lower- and higher-molecular mass forms of phosphorylated PsRBR1 protein. Furthermore, in vivo labeling with [32P]-inorganic phosphate indicated that PsRBR1 protein was more phosphorylated before mRNA accumulation of cell cycle regulatory genes such as PCNA. Together these findings suggest that dormancy-to-growth transition in pea axillary buds is regulated by molecular mechanisms of cell cycle control similar to those in mammals, and that the PsRBR1 protein has an important role in suppressing the cell cycle during dormancy in axillary buds. Introduction In intact plants, the terminal bud predominantly grows, while axillary bud growth is suppressed. This phenomenon is called apical dominance. After terminal bud decapitation, however, the axillary buds grow rapidly and develop in the same manner as the terminal bud. Apical dominance was one of the first developmental phenomena shown to be regulated by plant hormones (Thimann and Skoog 1934). The outgrowth of axillary buds is inhibited by auxin flow from the terminal bud (Booker et al. 2003), and is promoted by cytokinin biosynthesized in the stem after decapitation (Tanaka et al. 2006). Recently, a variety of experimental approaches were used to examine the molecular mechanisms that control shoot branching (Bennett and Leyser 2006; Dun et al. 2006; Shimizu-Sato and Mori 2001). Some mutants were isolated and characterized in Arabidopsis (max1, 2, 3, and 4), pea (rms1, 2, 3, 4, and 5), and petunia (dad1 and 2). Findings from these molecular genetic approaches combined with conventional physiologic studies, such as grafting experiments, suggested that novel signals regulate axillary bud dormancy. The molecular mechanisms of cell cycle control in axillary buds, however, are still not well understood. Most cells in pea axillary buds are arrested at the G1 phase in the cell cycle, and axillary buds have the capacity to undergo multiple cycles of dormancy and growth (Devitt and Stafstrom 1995; Shimizu and Mori 1998). Arrest and progression of the cell cycle in axillary buds are controlled in response to environmental signals and by developmental programs. Thus, we hypothesized that cells in dormant axillary buds are regulated through suppression mechanisms of cell cycle control suppression mechanisms similar to those in mammals. In mammals, arrest and progression of the cell cycle are also controlled during the G1 phase in response to environmental signals and by developmental programs (Sherr 2004). This process is governed by both positive- and negative-regulatory factors. One of the most important key regulators is the product of the retinoblastoma tumor suppressor gene (RB). The human RB gene encodes a 110-kDa nuclear phosphoprotein (pRB). pRB activities are regulated by phosphorylation in a cell cycle-dependent manner. Phosphorylation states of pRB regulate the interactions of multiple cellular proteins. In the G1 phase, hypo-phosphorylated pRB (hypo-ppRB) binds to a transcriptional factor, E2F family protein, and suppresses its transcriptional activity. E2F family proteins regulate the transcription of several genes (e.g., dihydrofolate reductase, thymidine kinase, DNA polymerase α, and proliferating cell nuclear antigen [PCNA]) whose products are required for either G1/S transition or DNA replication. Thus, by negatively regulating E2F family proteins, pRB negatively controls cell cycle progression. pRB is highly phosphorylated by several types of cyclin and cyclin-dependent kinase (CDK) complexes in response to environmental signals or developmental programs to progress the cell cycle. E2F family proteins are released from hyper-ppRB and promote the transcription of genes for G1/S transition or DNA replication. Thus, pRB regulates the arrest and progression of the cell cycle based on its phosphorylation state. In mammals, there are pRB-related proteins, p130 and p107 (Graña et al. 1998). RB family proteins regulate the progression of the cell cycle as well as the entry/exit transition of the cell cycle. RB family proteins have the same pocket protein feature; that is, they interact with a multitude of cellular proteins in a phosphorylation-dependent manner. RB family proteins are, however, modulated by differential regulation of phosphorylation states and protein levels in quiescent cells re-entering the cell cycle. In quiescent cells and in cells at the G0/G1 transition, pRB is in a hypo-phosphorylated form, and p130 is detected in non- and low-phosphorylated forms, whereas p107 is not detectable (Lin and DeCaprio 2003). In mid-G1/S transition, RB family proteins are phosphorylated by cyclin/CDK complexes. Hypo-ppRB and non- and low-phosphorylated p130 change to hyper-ppRB and high-phosphorylated p130, respectively. p107 is synthesized de novo, followed by phosphorylation to a high-phosphorylated form. Protein levels of pRB remain relatively constant throughout the cell cycle, compared to either p130 or p107. Although there is abundant p130 in the G0/G1 phase, it is dramatically down-regulated in the mid-G1/S phase. p130 down-regulation appears to be mediated by a cell cycle stage-dependent proteasome degradation pathway (Tedesco et al. 2002). These observations suggest that RB family proteins have overlapping functions, as well as different functions in the cell cycle regulation of quiescent cells. In plants, knowledge of cell cycle control has increased in recent years (Gegas and Doonan 2006). Some basic mechanisms that regulate the cell cycle appear to have been conserved throughout eukaryotic evolution (Dewitte and Murray 2003). To date, plant cDNAs encoding RB-related proteins have been reported in Zea mays (Ach et al. 1997; Grafi et al. 1996; Xie et al. 1996), Nicotiana tabacum (Nakagami et al. 1999), Chenopodium rubrum (Fountain et al. 1999), Arabidopsis thaliana (Kong et al. 2000), Populus tremula × Populus tremuloides (AF133675), Cocos nucifera (AY117036), and Oryza sativa (AP004592). The amino acid conservation between animal and plant RB-related proteins suggests that the proteins have similar biochemical properties. Like animal RB family proteins, plant RB-related proteins also interact with various cellular proteins, such as E2F family proteins, D-type cyclin, mammalian viral oncoproteins (SV40 large-T antigen, adenovirus E1a, and HPV E7), and the plant virus proteins (wheat dwarf virus RepA, and the tomato golden mosaic virus replication factor AL1) (Ach et al. 1997; Grafi et al. 1996; Huntley et al. 1998; Nakagami et al. 1999; Ramirez-Parra et al. 1999; Sekine et al. 1999; Xie et al. 1996). Because the functions of the mammalian RB family proteins depend on their phosphorylation state, the phosphorylation states of plant RB-related proteins were analyzed. ZmRBR1 protein undergoes changes in phosphorylation states concomitant with endoreduplication in maize (Grafi et al. 1996). The human G1/S protein kinases cyclinD/CDK4, cyclinE/CDK2, and cyclinA/CDK2 can phosphorylate ZmRBR1 protein in vitro (Huntley et al. 1998), and NtRBR1 protein is phosphorylated by tobacco cyclinD/CDC2 in vitro (Nakagami et al. 1999). The ZmRBR1 kinase activity correlates with the proliferation state in maize leaf (Boniotti and Gutierrez 2001), and NtRBR1 kinase activity is detected only during the mid-G1/S phase in tobacco BY-2 cells (Nakagami et al. 2002). These observations also suggest that plant RB-related proteins have biochemical properties similar to those of mammalian RB family proteins. Plant RB-related proteins seem to control not only cell cycle arrest/progression, but also development and cellular differentiation in endosperm, leaf, and root (Ebel et al. 2004; Desvoyes et al. 2006; Wildwater et al. 2005). Only limited information is available, however, on the development-dependent phosphorylation states of plant RB-related protein. In this paper, we describe the isolation and functional characterization of a pea (Pisum sativum L. cv. Alaska) cDNA encoding an RB-related protein (PsRBR1), which has biochemical properties similar to those of mammalian RB family proteins. PsRBR1 protein undergoes changes in its phosphorylation state concomitant with dormancy-to-growth transition in pea axillary buds. Materials and methods Plant growth and tissue collection Seeds of Pisum sativum L. cv. Alaska were soaked in running tap water for 24 h and sown in trays of rockwool. Plants were grown at 25°C in the dark for 4 days, and then in a 16 h light/8 h dark photoperiod for 3 days. Tissues studied were axillary buds at the second node in 7-day-old seedlings. Plants were decapitated 1 cm above the second node to stimulate outgrowth of axillary buds. PCR and cloning of PsRBR1 cDNA Degenerate oligonucleotide primers were designed for conserved regions of published amino acid sequences of the RB family proteins (sense; 5′-TT(T/C)TT(T/C)AA(T/C)C GNCA(T/C)AT(T/C/A)GA(T/C)CA-3′ and antisense; 5′-AC(T/C)TC(G/A)TT(G/A)TA(G/A)AANGT(T/G/A)AT(T/G/A)AT-3′), where the N in the parentheses indicates all four deoxyribonucleotides. Polymerase chain reaction (PCR) amplification was performed with cDNA from total RNAs of shoot apices. The amplified fragments (237 bp) were cloned into a BSII TSK-plasmid vector (Ichihara and Kurosawa 1993) and sequenced. A pea cDNA library was constructed using poly (A)+ RNA prepared from dormant axillary buds with a HybriZAP-2.1 A Two-Hybrid cDNA Gigapack Cloning Kit (Stratagene, La Jolla, CA). Approximately 1 × 106 phage recombinants derived from the pea cDNA library were screened with the selected PCR products labeled with [32P]-dCTP. Hybridization and subsequent washing were performed as described by Shimizu and Mori (1998). DNA sequencing The cDNAs were sequenced by the dideoxy chain termination method using an automatic DNA sequencer (LIC-4000, LI-COR Inc., Lincoln, NE) according to the manufacturer’s protocol. Both strands were entirely sequenced. Yeast two-hybrid system The vectors and strains were provided in the MATCHMAKER Two-Hybrid System 2 (Clontech, Palo Alto, CA). DNA manipulations and yeast methods were performed according to the manufacturer’s instructions. PsRBR1 cDNA was introduced into the pAS2-1 vector, and another cDNA was introduced into the pACT2 vector. Saccharomyces cerevisiae Y187 was transformed with combinations of these plasmids. For quantitative liquid assay, β-galactosidase activity was assayed with O-nitrophenylgalactoside as a substrate. Antibody preparation and immunoblot analysis For an anti-PsRBR1 antibody, full-length PsRBR1 protein was expressed in Escherichia coli BL21 pLysS as a 6xHis-tagged protein in pET32a (Novagen, Madison, WI). The antigen was purified with TALON Metal Affinity Resin (Clontech) according to the manufacturer’s protocol. The antigen was injected into a mouse using conventional procedures. Proteins from axillary buds were extracted by grinding with aluminum oxide in extraction buffer (20 mM Tris–HCl [pH 8.0], 100 mM NaCl, 1 mM EDTA, 0.5% NP-40, 50 μM NaF, and 1 μM Na3VO4, and the proteinase inhibitor cocktail Complete [Roche, Penzberg, Germany]). Insoluble materials were removed by microcentrifugation. Protein concentrations were determined with the DC Protein Assay (Bio-Rad, Hercules, CA). Proteins were separated using sodium dodecyl sulfate—polyacrylamide gel electrophoresis (SDS-PAGE, 10% acrylamide gels [acrylamide:bis-acrylamide = 37.5:1]), or 10% low bis-acrylamide gels [acrylamide:bis-acrylamide = 142:1]), and blotted onto a nitrocellulose membrane (BA-S 85, Schleicher & Schuell, Dassel, Germany). The membrane was blocked with 5% skim milk and 0.05% Tween 20 in Tris-buffered saline (50 mM Tris–HCl [pH 8.0], and 150 mM NaCl), and incubated with the anti-PsRBR1 antibody at a dilution of 1:1000. Horseradish peroxidase-conjugated goat anti-mouse IgG (Pierce, Rockford, IL) was used as the secondary antibody at a dilution of 1:2000. The signals were detected with the SuperSignal ULTRA (Pierce). In vitro-coupled transcription and translation A cDNA fragment corresponding to PsRBR1 cDNA, encoding the full-length protein, was cloned into the pBluescript SK+ vector (Stratagene). The template was digested with a restriction enzyme for linearization. In vitro-coupled transcription-translation was performed with the digested template (1 μg) in a TnT® T3 Coupled Wheat Germ Extract System (Promega, Madison, WI) in the presence of [35S]-methionine according to the manufacturer’s protocol. As a control, the in vitro-coupled transcription-translation reaction was performed without the DNA template. Samples were separated by SDS-PAGE. The gel was dried and exposed to a Fuji Imaging Plate (Fuji Photo Film, Tokyo, Japan) overnight. The radiographic image was visualized with a BAS2000 Image Analyzer (Fuji Photo Film). Dephosphorylation Protein extracts were incubated in reaction mixture (50 mM Hepes [pH 7.5], 2 mM MnCl2, 0.1 mM EDTA, 5 mM dithiothreitol, 0.01% Brij35, and the proteinase inhibitor cocktail Complete) with 1 unit of lambda protein phosphatase (λ-PPase, New England Biolabs, Beverly, MA) for 30 min at 30°C. Samples were immunoblotted by the anti-PsRBR1 antibody as described above. RNA extraction and blot analysis Total RNA was isolated from the tissues by SDS-phenol extraction, followed by LiCl precipitation. Formaldehyde agarose gel electrophoresis of total RNA was performed using standard procedures. The RNAs were blotted onto a Hybond N+ membrane (Amersham Biosciences), and hybridized with PsRBR1, PCNA, and Histone H4 cDNA labeled with [32P]-dCTP, respectively. Hybridization was performed in 0.25 M NaH2PO4 buffer (pH 7.2), 0.25 M NaCl, 7% SDS, 10% dextran sulfate (Amersham Biosciences), and 1% polyvinylpyrrolidone K30 at 65°C for at least 16 h. The membrane was washed in 2xSSPE (1xSSPE is 0.15 M NaCl, 10 mM NaH2PO4, and 1 mM EDTA [pH 7.4]) and 0.1% SDS at 50°C twice for 15 min each, and visualized by autoradiography at −80°C using Kodak XAR-5 film and an intensifying screen (Kodak, Rochester, NY). Phospholabeling and immunoprecipitation [32P]-inorganic phosphate (Amersham Biosciences, Piscataway, NJ) was applied to axillary buds (10 μCi per bud) 1 h before collection of the buds. Protein extraction was performed as described above. Protein extracts were incubated with both preimmune serum and Protein A Sepharose 4 Fast Flow (Amersham Biosciences) for 1.5 h at 4°C, and then non-specific associated materials were removed by microcentrifugation. The anti-PsRBR1 antibody was added to the clarified extracts and incubated for 1 h on ice. Protein A Sepharose 4 Fast Flow was added, and incubated for 30 min at 4°C. The complex of immunoprecipitates was washed three times with extraction buffer with 1 mg/ml bovine serum albumin, resuspended in SDS sample buffer (20 mM Tris-HCl [pH 6.8], 40% glycerol, 2% SDS, 2% β-mercaptoethanol, and bromophenol blue), and separated by SDS-PAGE. Dried gels were placed on a Fuji Imaging Plate, and the radiographic images were analyzed using the BAS2000 Image Analyzer. Results Isolation of a pea cDNA encoding RB-related protein To analyze the molecular mechanisms of cell cycle control in axillary buds, we focused on the function of plant RB-related proteins during dormancy-to-growth transition. Based on the amino acid sequences of reported RB-related proteins, we designed degenerate primers and performed reverse transcription PCR with total RNA prepared from pea shoot apices. Using the amplified DNA fragments as a probe, we screened the cDNA library prepared from pea dormant axillary buds. We obtained 14 positive clones by screening, and partially sequenced their 5′ ends. The clone that had the longest 5′ sequence (3540 bp) was used for further analysis because the DNA sequences of the other clones corresponded to those of the longest clone. This cDNA contained a single open reading frame capable of encoding a 1026-amino acid residue protein, PsRBR1. The molecular mass of PsRBR1 protein was estimated to be 114-kDa. The PsRBR1 cDNA contained a stop codon (TAG) four amino acids upstream of the putative initiation methionine codon within the reading frame, indicating that the PsRBR1 cDNA contained the entire coding region. The alignment of the A and B conserved domains and the N-terminal leucine-rich domain of PsRBR1 protein with maize ZmRBR1 protein and human RB family proteins is shown in Fig. 1(A). The A and B conserved domains of the PsRBR1 protein share 43% identity and 71% similarity, respectively with the A and B conserved domains of the ZmRBR1 protein. The conserved cysteine residue corresponding to Cys 706 of human pRB was present as Cys 798 in PsRBR1 protein (shown by an asterisk in the B conserved domain). This conserved cysteine residue is essential for the function of mammalian and plant RB proteins (Ach et al. 1997; Grafi et al. 1996; Kaye et al. 1990). Fig. 1Comparison of deduced amino acid sequences of pea PsRBR1 with selected homologous sequences. (A) Sequence alignments of the A and B conserved domains, and the N-terminal leucine-rich domain of PsRBR1 protein and ZmRBR1 protein with human RB family proteins. Black boxes indicate conserved amino acid residues, and dashes indicate gaps in the sequences. The asterisk in the B conserved domain denotes a conserved cysteine residue (798 of PsRBR1 protein), essential for RB function in mammalian cells. (B) Phylogenic analysis of the RB-related proteins under neighbor-joining analysis, and schematic of the salient features of RB-related proteins. The phylogeny is based on the amino acid sequences from the A conserved domain of RB-related proteins. Branch lengths are indicated above each branch. Bootstrap values from the results of 1000 replicates are shown below a node if the node is present in more than 50% of the bootstrap replicate analyses. The three conserved regions (N, N-terminal leucine-rich domain; A, A conserved domain; B, B conserved domain) between plants and mammals are indicated in boxes. Bars represent potential CDK phosphorylation sites. Numbers refer to protein length in amino acids, and numbers in parentheses show potential CDK phosphorylation sites. The accession numbers are AtRBR1 (AC069472), CnRBR1 (AY117036), CrRBR1 (AJ011681), NtRBR1 (AB015221), OsRBR1 (AP004592), PsRBR1 (AB012024), PtRBR1 (AF133675), ZmRBR1 (AF250050), p107 (L14812), p130 (X76061), and pRB (AF551763) The phylogenic analysis of plant RB-related proteins reported so far and human RB family proteins is shown in Fig. 1(B). The scheme on the right in Fig. 1(B) shows a diagram of the characteristic features of RB-related proteins. PsRBR1 protein was most similar to Populus tremula × Populus tremuloides PtRBR1 protein. PsRBR1 protein has 16 potential phosphorylation sites (Ser-Pro or Thr-Pro) for CDKs, which are highly clustered in regions flanking the A and B conserved domains. These conserved primary structures strongly suggest that RB-related proteins, including PsRBR1 protein, have similar biochemical properties in plant and animals. Genomic Southern blot analyses were performed to estimate the copy number and closely related genes of PsRBR1 in pea. Full-length PsRBR1 cDNA was used as a probe. Hybridization at both high- and low-stringencies yielded only signals derived from PsRBR1 (data not shown). These results suggested that PsRBR1 gene exists as a single-copy gene in pea. Interactions of PsRBR1 protein with both plant D-type cyclins and a mammalian oncoprotein Mammalian and plant RB proteins bind to a large number of cellular and viral proteins, such as D-type cyclin and SV40 large-T antigen. To confirm that PsRBR1 protein acts in the same manner as the mammalian and various plant RB proteins, we tested the ability of PsRBR1 protein to interact with these proteins using a yeast two-hybrid system. The genes for the tested proteins were fused to the GAL4 DNA-binding domain or the GAL4 activation domain. The data shown in Table 1 are the averages of three independent transformants. We verified that PsRBR1 protein alone did not activate the reporter gene. PsRBR1 protein interacted with both pea and Arabidopsis D-type cyclin (Pissa;CycD3;1 and Arath;CycD3;1). Both D-type cyclins contain the canonical RB-binding motif, LXCXE, in their N-terminus (Shimizu and Mori 1998; Soni et al. 1995). PsRBR1 protein interacted with SV40 large-T antigen, which is a mammalian DNA tumor virus oncoprotein. The similarities in primary structures and the coincidence of interacting partners of RB-related proteins suggest that animal and plant RB-related proteins, including PsRBR1 protein, have similar functions in cell cycle regulation. Table 1PsRBR1 protein binding in the two-hybrid assayPlasmidaβ-galactosidase activitybPsRBR10.11 ± 0.01PsRBR1 + Pissa;CycD3;113.9 ± 0.2PsRBR1 + Arath;CycD3;110.6 ± 0.1PsRBR1 + SV 40 large-T antigen31.5 ± 0.1aSaccharomyces cerevisiae Y187 was transformed with the indicated plasmidsb1 unit of β-galactosidase is defined as the amount that hydrolyzes 1 μmol of O-nitrophenylgalactoside to O-nitrophenyl and d-galactose per min. LacZ assays were done in triplicate and the data represent mean ± standard deviation Immunochemical detection of PsRBR1 protein in pea axillary buds To investigate protein levels and phosphorylation states of PsRBR1 protein in axillary buds during dormancy-to-growth transition, we prepared an anti-PsRBR1 antibody. The specificity of the anti-PsRBR1 antibody against PsRBR1 protein was examined (Fig. 2A). Protein extracted from growing axillary buds was detected by immunoblotting with the anti-PsRBR1 antibody. The anti-PsRBR1 antibody cross-reacted with three polypeptides (Fig. 2A, lane 1, filled circle, empty circle, and arrowhead), with molecular masses of approximately 150, 140, and 114-kDa. The pre-immune serum did not cross-react with any polypeptides in the protein extracts (Fig. 2A, lane 2). The anti-PsRBR1 antibody was incubated with an excess of purified recombinant PsRBR1 protein as a competitor, and then used to detect PsRBR1 protein in pea extract (Fig. 2A, lane 3). In this case, no polypeptides were detected with the anti-PsRBR1 antibody. These results indicate that the anti-PsRBR1 antibody specifically cross-reacts with PsRBR1 protein in axillary bud extracts. Fig. 2Immunochemical detection of PsRBR1 protein in pea axillary buds. (A) The terminal bud was cut 1 cm above the second node. After 24 h, the growing axillary buds were collected. Proteins in the protein extracts (30 μg) prepared from axillary buds were detected by immunoblotting with the anti-PsRBR1 antibody (lane 1), pre-immune serum (lane 2), and anti-PsRBR1 antibody preincubated with excess recombinant PsRBR1 protein (lane 3). Molecular mass markers are indicated at the left of the lane. The anti-PsRBR1 antibody cross-reacted with three polypeptides, indicated by the filled circle, empty circle, and arrowhead. (B) Comparison of molecular mass of authentic PsRBR1 protein with that of PsRBR1 synthesized by in vitro-coupled transcription-translation reaction. The plasmid DNA containing the T3 promoter and the PsRBR1 coding sequence was transcribed, and the resultant RNA was translated. The translation products labeled with [35S]-methionine (lane 2) were immunoprecipitated by the anti-PsRBR1 antibody (lane 3). Proteins in the protein extracts prepared from growing axillary buds were detected by immunoblotting with the anti-PsRBR1 antibody (lane 4). The in vitro-coupled transcription-translation reaction was performed without the plasmid DNA (lane 1). Molecular mass markers are indicated at the left of the lane. The product synthesized in vitro was a 114-kDa polypeptide, indicated by the arrowhead (lane 2), which is consistent with the calculated values based on the deduced amino acids sequences of PsRBR1 cDNA. (C) Dephosphorylation of authentic PsRBR1 protein. Protein extracts prepared from growing axillary buds were dephosphorylated with λ-PPase treatment, separated by SDS-PAGE in either a standard acrylamide gel (left) or low bis-acrylamide gel (right), and proteins were detected by immunoblotting with the anti-PsRBR1 antibody. Numbers in parentheses indicate the ratio of acrylamide to bis-acrylamide. Lane 1, untreated samples; lane 2, samples incubated without λ-PPase; lane 3, samples incubated with λ-PPase Comparison of molecular mass of authentic PsRBR1 protein in pea with that of in vitro-synthesized PsRBR1 protein Three polypeptides were detected in protein extracts from pea axillary buds by immunoblotting with the anti-PsRBR1 antibody. The molecular mass of PsRBR1 protein was estimated to be 114-kDa based on the deduced amino acid sequences of PsRBR1 cDNA. To clarify these differences in the molecular mass of PsRBR1 protein, we compared the molecular mass of authentic PsRBR1 protein in pea with that of in vitro-synthesized PsRBR1 protein (Fig. 2B). PsRBR1 protein was synthesized using an in vitro-coupled transcription-translation system with [35S]-methionine. In vitro-synthesized PsRBR1 protein produced a major 114-kDa polypeptide (Fig. 2B, lane 2, arrowhead). The molecular mass of the PsRBR1 protein was consistent with that predicted from the deduced amino acid sequences of PsRBR1 cDNA. The 114-kDa polypeptide was not detected in protein products prepared from the in vitro-coupled transcription-translation system without the PsRBR1 DNA template, which was used as a negative control (Fig. 2B, lane 1). To confirm whether the 114-kDa polypeptide was PsRBR1 protein, in vitro-coupled transcription-translation products were immunoprecipitated with the anti-PsRBR1 antibody (Fig. 2B, lane 3). The 114-kDa polypeptide was immunoprecipitated with the anti-PsRBR1 antibody, indicating that in vitro-synthesized PsRBR1 protein is 114-kDa and corresponds to the smallest polypeptide detected by anti-PsRBR1 antibody in growing axillary buds (Fig. 2B, lane 4). These results suggested that the smallest polypeptide (114-kDa) detected by the anti-PsRBR1 antibody was non-modified PsRBR1 protein in axillary buds, and the two larger polypeptides of PsRBR1 proteins were modified by post-translational regulation in axillary buds. Dephosphorylation of authentic PsRBR1 protein The activity of mammalian RB family proteins is regulated by their phosphorylation states. Plant RB-related proteins are also phosphorylated by cyclin/CDK complexes (Huntley et al. 1998; Nakagami et al. 1999). PsRBR1 protein has some putative CDK phosphorylation sites (Fig. 1B). Thus, we expected that the modification of authentic PsRBR1 protein in the axillary buds caused its phosphorylation. To examine this, we performed in vitro dephosphorylation reactions against PsRBR1 protein using (λ-PPase. Proteins were separated by two types of SDS-PAGE gels, because some phosphorylated proteins were more clearly separated by low bis-acrylamide gel (Fig. 2C, right) than by standard acrylamide gel (Fig. 2C, left) (Iwasaki et al. 2002). Immunoblotting with anti-PsRBR1 antibody indicated that there were three polypeptides corresponding to PsRBR1 protein in growing axillary buds (Fig. 2C, lane 1), but after λ-PPase treatment of the protein extract, PsRBR1 protein was detected as the smallest of the polypeptides (Fig. 2C, lane 3). On the other hand, the molecular mass of PsRBR1 protein did not change after treatment without λ-PPase added to the protein extract as a negative control (Fig. 2C, lane 2). In addition, the low bis-acrylamide gel separated the protein more clearly than did the standard acrylamide gel. These results suggested that PsRBR1 protein was controlled by post-translational modification, and exists in non-phosphorylated form of PsRBR1 protein, lower-molecular mass form of phosphorylated PsRBR1 protein (Lp-form), and higher-molecular mass form of phosphorylated PsRBR1 protein (Hp-form) in growing axillary buds. Protein levels and phosphorylation states of PsRBR1 protein in axillary buds during dormancy-to-growth transition Because suppression of the cell cycle is regulated by the phosphorylation states of mammalian RB family proteins, we evaluated the phosphorylation states of PsRBR1 during dormancy-to-growth transition in pea axillary buds by immunoblotting using the anti-PsRBR1 antibody (Fig. 3A). During dormancy, PsRBR1 protein was detected as two polypeptides, which are non-phosphorylated form of PsRBR1 protein and the Lp-form (Fig. 3A, 0 h after decapitation). In addition to these non-phosphorylated form of PsRBR1 protein and the Lp-form, the Hp-form was detected 2 h after decapitation (Fig. 3A, indicated by the arrow). These results suggested that PsRBR1 protein was shifted from non-phosphorylated form of PsRBR1 protein and the Lp-form to the Hp-form; that is, the PsRBR1 protein was more phosphorylated after decapitation. The protein levels of these three polypeptides were increased 12 h after decapitation. Non-phosphorylated PsRBR1 protein and the Lp-form existed during late stage (12 to 24 h) after decapitation. PsRBR1 mRNA level was increased 12 h after decapitation (see Fig. 3B). It seems that non-phosphorylated PsRBR1 protein and the Lp-form were synthesized and phosphorylated de novo after decapitation. Moreover, because the axillary bud cell cycle is not completely synchronized, the non-phosphorylated PsRBR1 protein and the Lp-form were included in the protein extracts prepared from growing axillary buds. Fig. 3Protein levels and phosphorylation states of PsRBR1 protein in axillary buds during the dormancy-to-growth transition. (A) Proteins were extracted from axillary buds after terminal bud decapitation. Numbers below each lane indicate the hours after decapitation. Proteins in the protein extracts (30 μg) were detected by immunoblotting with the anti-PsRBR1 antibody. Molecular mass markers are indicated at the left of the lane. The bottom panel shows the Coomassie Brilliant Blue (CBB) staining around 100 kDa as a loading control. The filled circle, empty circle, and arrowhead indicate higher-molecular mass form of phosphorylated PsRBR1 protein (Hp-form), lower-molecular mass form of phosphorylated PsRBR1 protein (Lp-form) and non-phosphorylated PsRBR1 protein, respectively. The arrow shows more phosphorylated PsRBR1 protein (Hp-form) appearing 2 h after decapitation. (B) Northern blot analyses of total RNA isolated from pea axillary buds before and after decapitation. Plants were decapitated 1 cm above the second node. After decapitation, axillary buds were collected at several time points, as indicated below each lane. Total RNA (10 μg) was separated by denaturing formaldehyde gel electrophoresis, and transferred to a Hybond N+ membrane. The membranes were hybridized with the cDNA of PsRBR1, PCNA, and histone H4. The bottom panel shows the ethidium bromide-stained RNA gel as a loading control Accumulation patterns of PsRBR1 transcripts in axillary buds during dormancy-to-growth transition To investigate the relationship between different phosphorylation states of PsRBR1 protein with the transcriptional control of cell cycle regulators, we examined the accumulation patterns of PsRBR1, PCNA, and Histone H4 transcripts during dormancy-to-growth transition in axillary buds. The results of Northern blot analyses using PsRBR1, PCNA, and histone H4 cDNAs (Shimizu and Mori 1998) are shown in Fig. 3B. PsRBR1 transcripts were detected at 3.6 kb. The mRNA levels of the three genes were very low during dormancy in axillary buds. When the axillary buds were stimulated to grow by terminal bud decapitation, mRNA levels of these genes increased markedly. PsRBR1 and histone H4 mRNA increased 12 h after decapitation. Histone H4 was specifically expressed during the S phase. The accumulation patterns of PsRBR1 mRNA are similar to that of histone H4 mRNA, suggesting that PsRBR1 is predominantly expressed at the S phase in pea axillary buds at least during the first 24 h after terminal bud decapitation. Because PsRBR1 protein was increased 12 h after decapitation (see Fig. 3A), the expression patterns of PsRBR1 are coincident with those of PsRBR1 protein. The mRNA expression of PCNA seems to be regulated by RB-related proteins. PCNA mRNA increased significantly 4 h after decapitation, and PCNA mRNA accumulated after the appearance of the Hp-form. Detection of phosphorylated PsRBR1 protein in axillary buds after decapitation We confirmed that PsRBR1 protein was phosphorylated in the axillary buds just after decapitation with in vivo labeling followed by immunoprecipitation using the anti-PsRBR1 antibody (Fig. 4). [32P]-inorganic phosphate was incorporated into growing axillary buds. Many proteins were phosphorylated in growing axillary buds after decapitation (Fig. 4A, lane 1). Immunoprecipitation with the anti-PsRBR1 antibody revealed a [32P]-labeled 150-kDa phosphoprotein (Fig. 4A, lane 3, arrow). The size of the [32P]-labeled PsRBR1 protein was consistent with the Hp-form detected by immunoblotting analysis. The signal was not detected by immunoprecipitation with preimmune serum used as a negative control (Fig. 4A, lane 2). Fig. 4Detection of phosphorylation of PsRBR1 protein in axillary buds during dormancy-to-growth transition. (A) Eight hours after decapitation of the terminal bud, [32P]-inorganic phosphate was applied to axillary buds to be incorporated in vivo. One hour after the application of [32P]-inorganic phosphate, proteins were extracted from the axillary buds. The protein extracts were immunoprecipitated with the anti-PsRBR1 antibody (lane 3) or preimmune serum (lane 2), and separated by SDS-PAGE. Total proteins without immunoprecipitation were loaded in lane 1. The dried gel was exposed to a Fuji Imaging Plate. Molecular mass markers are indicated on the left of the lane. (B) [32P]-inorganic phosphate was applied to axillary buds 1 h before collection of the buds. The axillary buds were collected at 0, 2, and 4 h after terminal bud decapitation (lanes 1 to 3, respectively), and proteins were extracted from the axillary buds. Other details are as described in (A) To confirm whether PsRBR1 protein is more phosphorylated in axillary buds just after decapitation, we examined dormant and outgrowing axillary buds after decapitation. No signal was detected before decapitation (Fig. 4B, lane 1), suggesting that PsRBR1 protein was not highly phosphorylated de novo before decapitation. Two hours after decapitation, the 150-kDa polypeptide was detected (Fig. 4B, lane 2) and the signal was increased 4 h after decapitation (Fig. 4B, lane 3, arrow). These results suggested that the level of phosphorylation of PsRBR1 protein increased immediately after terminal bud decapitation, producing the Hp-form. Discussion In intact plants, the pea shoot apex inhibits growth; i.e., cell elongation and cell division of the dormant axillary buds. The cell cycle of dormant axillary buds is arrested at the G1 phase. Therefore, from the point of view of cell cycle control, the transition from dormancy to outgrowth in axillary buds seems to be from the G1 phase to the S phase. In animal cells, RB family proteins function as negative controllers in G1 arrest, and the activities of RB family proteins are controlled by their phosphorylation states. Here, we investigated phosphorylation states of the pea RB-related protein, PsRBR1, during the dormancy-to-growth transition in axillary buds. The results of the present study provide a basis for further studies of the molecular mechanisms underlying apical dominance and cell cycle control. Mammalian and plant RB proteins have several highly conserved regions, which are the N-terminal leucine rich, A and B conserved domains, multiple potential CDK phosphorylation sites, and a conserved cysteine residue. PsRBR1 protein, like other plant RB-related proteins, has the ability to interact with both plant D-type cyclins and a mammalian oncoprotein. Plant RB-related proteins have the same properties and activities as animal RB family proteins. For example, ZmRBR1 protein binds to human and Drosophila E2F, and inhibits transcriptional activation of human E2F (Huntley et al. 1998; Ramirez-Parra et al. 1999; Sekine et al. 1999). Geminivirus DNA replication is reduced in plant cells transfected with plasmids encoding human p130 (Xie et al. 1996). Human cyclinD/CDKA phosphorylates NtRBR1 protein in vitro (Nakagami et al. 1999). Thus, plant RB-related proteins can function not only as plant factors, but also as animal factors. These results strongly suggest that the functions of plant RB-related proteins, including PsRBR1 protein, are similar to those of the mammalian RB family proteins in G1/S transition, and that plant and animal cells might use similar regulatory proteins and pathways for cell cycle control during the G1 phase. Therefore, the maintenance of this high degree of similarity throughout evolution (in plants, insects, and vertebrates), further emphasizes the critical roles of RB-related proteins. Three human RB family proteins (pRB, p130, and p107) have individual functions in cell cycle regulation and development. Two RB-related genes were isolated from Drosophila melanogaster. To date, plant RB-related proteins have been reported in Zea mays, Nicotiana tabacum, Chenopodium rubrum, Arabidopsis thaliana, Populus tremula x Populus tremuloides, Cocos nunucifera, and Oryza sativa (Ach et al. 1997; Fountain et al. 1999; Grafi et al. 1996; Kong et al. 2000; Nakagami et al. 1999; Xie et al. 1996). Of these, Arabidopsis thaliana and Oryza sativa might contain only one copy of the RB-related gene in the whole genome based on the similarity of amino acid sequences of RB-related proteins. Here, our genomic Southern blot analysis suggested the presence of only one copy of the PsRBR1 gene in pea. The dicot RB-related gene seems to contain only one copy in the genome. By contrast, maize contains at least three RB-related genes. ZmRBR3 protein also functions to promote endosperm development (Sabelli et al. 2005), suggesting that maize RB family proteins have special roles in endosperm development. When the protein extracts in growing axillary buds were treated with λ-PPase, the Lp- and Hp-forms should have been dephopshorylated and stoichiometrically shifted to non-phosphorylated forms. As shown in Fig. 3(C), however, it appears that there was no increase in the non-phosphorylated form (lane 3) compared to the other lanes. The reason for this might be that a 40-kDa polypeptide was also detected by immunoblotting with the anti-PsRBR1 antibody (data not shown). The 40-kDa polypeptide was probably derived from in vitro-dephosphorylated PsRBR1 protein by degradation during λ-PPase treatment. In vitro-dephosphorylated PsRBR1 protein might be more unstable than in vivo-dephosphorylated PsRBR1 protein, despite the presence of proteinase inhibitors, though the reason for this is not clear. The results of the immunoblotting experiments indicated that the migration patterns of mammalian RB family proteins differ depending on their phosphorylation states. According to the reports (Farkas et al. 2002), the faster migrating proteins were the non- and hypo-phosphorylated forms in the G0/G1 phases. The slower migrating protein was the hyper-phosphorylated form during the G1/S transition. In plant, immunoblotting with the anti-ZmRBR1 antibody revealed the ZmRBR1 protein to be a diffuse protein band of approximately 110-kDa during endoreduplication in maize endosperm, and there are some polypeptides of approximately 110-kDa in maize leaf (Grafi et al. 1996; Huntley et al. 1998). The diffused protein bands of ZmRBR1 were confirmed by the use of λ-phosphatase to result from phosphorylation. (Grafi et al. 1996). In the present paper, using λ-PPase treatment and in vivo labeling followed by immunoprecipitation with the anti-PsRBR1 antibody, we showed that PsRBR1 protein exists in three phosphorylated forms in axillary buds. Although we could not determine the number of phosphorylation sites in the Lp- and Hp-forms, the Lp- and Hp-forms might correspond to the hypo- and hyper-phosphorylated forms of PsRBR1 protein, respectively. Plant E2F isolated from tobacco and wheat has transcriptional activity and interacts with RB-related protein (Ramirez-Parra et al. 1999; Sekine et al. 1999). There are E2F-binding sites in the promoter region of PCNA, whose transcripts are induced in the late-G1/S phase (Kodama et al. 1991). These observations suggest that plant E2F, like that in animal, induces gene expression of PCNA. Further phosphorylation of PsRBR1 protein was induced 2 h after terminal bud decapitation, and PCNA transcripts were induced 4 h after terminal bud decapitation in pea axillary buds. Although it is unclear whether the transcriptional activation of plant PCNA is regulated by E2F, E2F released from RB-related protein might promote PCNA mRNA expression, as in mammals. This possibility is supported by the observation that ZmRBR1-associated kinase activity peaks at the G1/S transition in synchronized tobacco BY-2 cells (Boniotti and Gutierrez 2001). It is likely that increased phosphorylation states of PsRBR1 protein have an important role in the regulation of E2F-mediated gene expression involved in the G1/S transition and DNA replication during dormancy-to-growth transition in pea axillary buds. In some plants, the gene expression of cyclins and CDK activities are regulated by plant hormones, e.g., cytokinin, auxin, and gibberellin (Miao et al. 1993; Sauter 1997; Soni et al. 1995). In Arabidopsis suspension-cultured cells, D-type cyclin mRNA is rapidly induced by the addition of cytokinin or glucose, and repressed by the addition of auxin (Soni et al. 1995). Cytokinin is biosynthesized at the stem after decapitation and moves into the axillary buds for outgrowth (Tanaka et al. 2006). It is possible that the increasing cytokinin levels caused by decapitation induces the expression of D-type cyclin. D-type cyclin forms a complex with CDKs and immediately phosphorylates RB-related protein. The Hp-form induces gene expression for the G1/S transition in axillary bud cells. It is likely that RB-related protein controls the cell cycle in axillary buds during the dormancy-to-growth transition. Further analyses will provide more insight into the regulation of dormancy-to-growth transition in axillary buds.

J_Med_Internet_Res-4-2-1761932

Using the Internet for Surveys and Health Research

This paper concerns the use of the Internet in the research process, from identifying research issues through qualitative research, through using the Web for surveys and clinical trials, to pre-publishing and publishing research results. Material published on the Internet may be a valuable resource for researchers desiring to understand people and the social and cultural contexts within which they live outside of experimental settings, with due emphasis on the interpretations, experiences, and views of `real world' people. Reviews of information posted by consumers on the Internet may help to identify health beliefs, common topics, motives, information, and emotional needs of patients, and point to areas where research is needed. The Internet can further be used for survey research. Internet-based surveys may be conducted by means of interactive interviews or by questionnaires designed for self-completion. Electronic one-to-one interviews can be conducted via e-mail or using chat rooms. Questionnaires can be administered by e-mail (e.g. using mailing lists), by posting to newsgroups, and on the Web using fill-in forms. In "open" web-based surveys, selection bias occurs due to the non-representative nature of the Internet population, and (more importantly) through self-selection of participants, i.e. the non-representative nature of respondents, also called the `volunteer effect'. A synopsis of important techniques and tips for implementing Web-based surveys is given. Ethical issues involved in any type of online research are discussed. Internet addresses for finding methods and protocols are provided. The Web is also being used to assist in the identification and conduction of clinical trials. For example, the web can be used by researchers doing a systematic review who are looking for unpublished trials. Finally, the web is used for two distinct types of electronic publication. Type 1 publication is unrefereed publication of protocols or work in progress (a `post-publication' peer review process may take place), whereas Type 2 publication is peer-reviewed and will ordinarily take place in online journals. Identifying issues for qualitative research As the most comprehensive archive of written material representing our world and people's opinions, concerns, and desires (in industrialized countries), the Internet can be used to identify `issues' for qualitative (descriptive) research and to generate hypotheses. Material published on the Internet may be a valuable resource for researchers desiring to understand people and the social and cultural contexts within which they live--outside of experimental settings--with due emphasis on the interpretations, experiences, and views of `real world' people. Reviews of information posted by consumers on the Internet may help to identify health beliefs, common topics, motives, information, and emotional needs of patients, and point to areas where research is needed. Comparing recommendations found on the Web against evidence-based guidelines is one way to identify areas where there is a gap between opinion and evidence, or where there is a need for clinical innovation. The accessibility of information for analysis and the anonymity of the Internet allow researchers to analyse text and narratives on Web sites, to use newsgroups as global focus groups, and to conduct interviews and surveys via e-mail, chat rooms, Web sites, or newsgroups.Topics suited to qualitative research include: Analysis of interactive communications (e.g. e-mail). Study of online communities (virtual self-help groups, newsgroups, mailing lists). Investigation of communication processes between patients and professionals. Study of consumer preferences, patient concerns, and information needs. Exploration of the `epidemiology of health information' on the Web [1-2]. The Internet population is unrepresentative of the general population, restricting the use of the Internet for quantitative studies (i.e. studies focusing on measurement). Qualitative studies, however, do not require representative samples:`In qualitative research we are not interested in an average view of a patient population, but want to gain an in-depth understanding of the experience of particular individuals or groups; we should therefore deliberately seek out individuals or groups who fit the bill' [3]. Three different research methodologies for qualitative research on the Internet may be distinguished: Passive analysis: For example, studying information on Web sites or interactions in newsgroups, mailing lists, and chat rooms--without researchers actively involving themselves. Active analysis: Also called participant observation; the researcher participates in the communication process, often without disclosing their identity as researcher. For example, they may ask questions in a patient discussion group implying that she or he is a fellow patient. Such studies often involve elements of deception, unless the researcher is a sufferer him- or herself. Interviews and surveys: See below. Examples of these three types of qualitative research on the Internet are available elsewhere [1]. Using the Internet for surveys Using the Internet for surveys requires an awareness of methodologies, selection bias, and technical issues. Methodological issues Internet-based surveys may be conducted by means of interactive interviews or by questionnaires designed for self-completion. Electronic one-to-one interviews can be conducted via e-mail or using chat rooms. Questionnaires can be administered by e-mail (e.g. using mailing lists), by posting to newsgroups, and on the Web using fill-in forms. When e-mail is used to administer questionnaires, messages are usually sent to a selected group with a known number of participants, thus allowing calculation of the response rate. Surveys posted to newsgroups may request that the completed questionnaire is posted back to the researcher, but it is impossible to know who and how many people read the questionnaire. If Web-based forms are used, questionnaires can be placed in a password-protected area of a Web site (i.e. participation by invitation or registration only), or alternatively they may be open to the public (i.e. any site visitor can complete the survey). The latter option makes calculation of a response rate more difficult but not impossible: the number of people who access (without necessarily completing) the questionnaire is counted and used as the denominator. Web-based surveys have the advantage that the respondent can remain anonymous (as opposed to e-mail surveys, where the e-mail address of the responder is revealed). Furthermore, they are very convenient for the researcher, as responses can be directly stored in a database where they are immediately accessible for analysis. Electronic interviews and surveys (`e-surveys') are emerging scientific research methodologies, pioneered by communication scientists, sociologists, and psychologists, although their use for health-related research is still in its infancy [4-10]. Examples of health-related research include: A Web-based survey on the effects of ulcerative colitis on quality of life [11]. Collection of clinical data from atopy patients [12]. A Web-based survey looking at complementary and alternative medicine use by patients with inflammatory bowel disease and Internet access [13]. A survey of dentists regarding the usefulness of the Internet in supporting patient care [14,15]. Guidelines for Web-based surveys Scenarios that may be suitable for a Web-based survey Respondent features: Respondents are already avid Internet users; e-mail addresses known for reminder messages. Respondents are enthusiastic form fillers; will not require monetary incentives. Need for respondents covering a wide geographical area (e.g. rare clinical special ties, diseases). Respondents are known to match non-respondents and even non-Internet users on key variables. Survey features: Need for complex branching, interactive questionnaire or multimedia as part of the survey instrument. Survey content will evolve fast (e.g. Delphi method surveys use repeating rounds of revised questionnaires delivered over a short period, incorporating aggregate results from previous rounds until convergence is achieved). Intent is to document bizarre, rare phenomena whose simple occurrence is of interest. No need for representative results: collecting ideas vs. hypothesis testing. Investigator features: Limited budget for mailing and data processing, but good in-house Web skills. Precautions can be taken against multiple responses by same individual, password sharing. Web survey forms have been piloted with representative participants and demonstrate acceptable validity and reliability with most platform, browser, and Internet access provider combinations. Data is required fast in a readily analysed form. Scenarios that are unsuitable for a Web-based survey Respondent features: Target group is under-represented on Internet; e.g. the underprivileged, elderly people. Target group is concerned, however unreasonably, about privacy aspects. Target group requires substantial incentives to complete the survey. Need for a representative sample. Survey features: Need for very accurate timing data on participants (inaccuracies in the range of seconds are added due to network transmission times, unless JavaScript or Java applets are used; see Glossary) or observational data on participants. An existing paper instrument has been carefully validated on target group. Need to capture qualitative data or observations about participants. Wish to reach the same group of participants in the same way months or years later. Investigator features: Limited in-house Web or Java expertise but existing desktop publishing and mailing facility. E-surveys may be part of a qualitative research process, but results can be analysed quantitatively as long as researchers are aware of potential bias (see below). In addition to gathering data, the Internet may also be used in the course of developing questionnaires, as it allows rapid prototyping and pilot testing of instruments, e.g. to evaluate the effect of framing the questions differently [16]. Several studies have checked the validity of Web-based surveys by comparing the results of studies conducted on the Web with identical studies in the real world. These seem to suggest that the validity and reliability of data obtained online are comparable to those obtained by classical methods [4,5,17-19]. However, issues of generalizability (mainly due to selection bias, discussed in detail below) remain important considerations, and the researcher should select his or her research question and interpret the results with care.The benefits and problems of Web-based surveys have been summarized by Wyatt, who suggests guidelines for when they may be appropriate (see Box 1) [20]. Selection bias In `open' surveys conducted via the Internet where Web users, newsgroup readers, or mailing list subscribers are invited to participate by completing a questionnaire, selection bias is a major factor limiting the generalizability (external validity) of results. Selection bias occurs due to: The non-representative nature of the Internet population. The self-selection of participants, i.e. the non-representative nature of respondents, also called the `volunteer effect' [21]. The non-representative nature of Internet demographics was briefly considered earlier. Considering whether the topic chosen for study is suitable for the Internet population is the first and probably the most important step in minimizing bias, thus maximizing response rates and increasing the external validity of the results [20]. For example, targeting elderly homeless alcoholics is unsuitable for an Internet survey and the results are likely to be heavily skewed by hoax responses. Self-selection bias originates from the fact that people are more likely to respond to questionnaires if they see items which interest them, e.g. because they are affected by the items asked about, or because they are attracted by the incentives offered for participating. As people who respond almost certainly have different characteristics than those who do not, the results are likely to be biased.This kind of selection bias is more serious than the bias arising from the non-representative nature of the population, because the researcher deals with a myriad of unknown factors and has little opportunity to interpret his or her results accordingly. Such bias may be exacerbated via loaded incentives (e.g. typical `male' incentives such as computer equipment). Evidence suggests women are generally more interested in health topics and exhibit more active information-seeking behaviour [22], so are more likely to volunteer participation in health questionnaires. For Web surveys, the potential for self-selection bias can be estimated by measuring the response rate, expressed as the number of people completing the questionnaire divided by those who viewed it (cf. the participation rate, expressed as the number of site visitors viewing the questionnaire divided by the total number of site visitors). Technical issues Although a detailed analysis is beyond the scope of this chapter, a synopsis of important techniques and tips for implementing Web-based surveys provides some insight into the difficulties faced by survey designers (see Box 2). Technical issues in implementing Web-based surveys Use of `cookies' Cookies can assign a unique identifier to every questionnaire viewer, useful for determining response and participation rates, and for filtering out multiple responses by the same person. As cookies may be regarded with suspicion, we recommend that researchers openly state that cookies will be sent (and the reasons for this); set the cookie to expire on the day that data collection ceases; and publish a privacy policy. Measuring response time The time needed to complete a questionnaire can be readily calculated by subtracting the time a form was called up by the browser from the time it was submitted using an automatic time-stamp. The response time may be used to exclude respondents who fill in the questionnaire too quickly: this may identify hoax responses, where respondents don't read the questions. Avoiding missing data Forms can be configured to automatically reject incomplete questionnaires and point out missing or contradictory items. Checks can be made on the client (p. 9) prior to submission, or following submission to the server (where incomplete responses can also be analysed, e.g. during a questionnaire pilot). Maximizing response rate The number of contacts, personalized contacts, and contact with participants before the actual survey are the factors most associated with higher response rates in Web surveys [23]. Incentives increase the risk of selection bias (see text), but less so if cash is offered. Perhaps the best incentive (and the easiest to deliver via the Internet) is the promise of survey results or personalized answers (e.g. a score). The option to complete questionnaires anonymously avoids wariness associated with requests for personal information (e.g. an e-mail address), but increases the risk of hoax responses. Researchers should be open about who is behind the study, what the aim is, and provide opportunities for feedback. Although postal surveys are superior to e-mail surveys with regard to response rate, online surveys are much cheaper [24,25]. Schleyer [15] estimated that the cost of their Web-based survey was 38 percent less than that of an equivalent mail survey and presented a general formula for calculating break-even points between electronic and hard-copy surveys. Jones gave figures of 92 p per reply for postal surveys, 35 p for e-mail, and 41 p for the Web [24]. Randomizing items Scripting languages may be used to build dynamic questionnaires (as opposed to static forms) that look different for certain user groups or which randomize certain aspects of the questionnaire (e.g. the order of the items). This can be useful to exclude possible systematic influences of the order of the items upon responses. Ethical issues The ethical issues involved in any type of online research should not be forgotten [1,26-31]. These include informed consent as a basic ethical tenet of scientific research on human populations [32], protection of privacy, and avoiding psychological harm. In qualitative research on the Web, informed consent is required when: Data are collected from research participants through any form of communication, interaction, or intervention. Behaviour of research participants occurs in a private context where an individual can reasonably expect that no observation or reporting is taking place, except when researchers do research `in public places or use publicly available information about individuals (e.g. naturalistic observations in public places, analysis of public records, or archival research)' [33]. The question therefore arises of whether researchers analysing newsgroup postings enter a `public place', or whether the space they invade is perceived as private. In the context of research, the expectation of the individual (whether he/she can reasonably expect that no observation is taking place) is crucial. Different Internet services have different levels of perceived privacy (in decreasing order of privacy: private e-mail; chat rooms; mailing lists; newsgroups;Web sites).The perceived level of privacy is a function of the number of participants, but also depends on other factors such as group norms established by the community to be studied. For example, in a controversial paper, Finn studied a virtual self-support group where the moderator was actively discouraging interested professionals who were not sexual abuse survivors from joining the group [34]. In those cases, obtaining informed consent (or seeking an ethical waiver, if the research could not practicably be carried out were informed consent to be required) is mandatory. In practice, obtaining informed consent, especially for passive research methods, is difficult, as researchers usually cannot post an announcement to a mailing list or newsgroup saying that it will be monitored and analysed for the next few months, as this may greatly influence or even spoil the results, and because the mere posting of such a request may disrupt the community, and therefore be considered unethical. Researchers should therefore first obtain consent from a group moderator in order to explore whether even a request for permission is felt to be disruptive to the group process. If the moderator or person responsible for the list has no objections, one may then post a message to a newsgroup or mailing list explaining the purpose of the research, explaining that one will observe the community, assuring all participants of anonymity, and giving them the opportunity to withdraw from the newsgroup or mailing list or to exclude themselves from the study by writing to the researcher. The fundamental problem is that this may influence the communication process and may even destroy the community. Besides, participants who later join the group need to get the same information. An alternative would be to analyse the communication retrospectively and to write individual e-mails to all participants whose comments were to be analysed or quoted, asking for permission to use them; this technique has been used by Sharf [35]. In any case, researchers should make themselves familiar with the virtual community they are approaching; i.e. read the messages in a newsgroup for some time (`lurking'). Under no circumstances should researchers blindly spam (p. 31) or cross-post requests for research participation to various newsgroups. Informed consent may also play a role when researchers report aggregate (collated and hence anonymous) data on usage patterns, such as a log-file analysis (reporting data on what Web sites have been accessed by a population). Crucial here is an appropriate privacy statement stating that these data may be analysed and reported in aggregate [28]. Note that aggregate data are exempt from the registration requirements of the UK's Data Protection Act of 1998. In conducting surveys researchers may obtain informed consent by declaring the purpose of the study; disclosing which institutions are behind the study; explaining how privacy will be assured; and detailing with whom data will be shared and how it will be reported, before participants complete the questionnaire. When reporting results, it is obvious that the total anonymity of research participants needs to be maintained. Researchers have to keep in mind that, by the very process of quoting the exact words of a newsgroup or mailing list participant, the confidentiality of the participant may already be broken as Internet search engines may be able to retrieve the original message, including the e-mail address of the sender. It is essential, therefore, to ask participants whether they agree to be quoted whenever there may be a retrievable archive, pointing out the risk that they may be identifiable. Problems can also potentially arise from just citing the name of the community (e.g. of a newsgroup), which may damage the community being studied. Finding methods, protocols, and instruments For laboratory `bench work', researchers often need a protocol for a specific assay method. In addition to the possibility of searching literature databases, there are also specialized services on the Web that can assist in this research, such as MethodsFinder and the `Technical tips online' database at BioMedNet: MethodsFinder (BIOSIS): http://www.methodsfinder.org/ BioMedNet: http://www.bmn.com/ Sometimes asking a specific question on the right newsgroup or mailing list is also very effective. Clinical researchers may be more interested in instruments to measure patient outcomes.An excellent guide to selecting quality-of-life instruments is the Quality of Life Instruments Database at the Mapi Research Institute: http://www.qolid.org/ Online statistical analysis tools are available at the Simple Interactive Statistical Analysis (SISA) Web site, while background information is available within the online book Statistics at square one: SISA (Daan Uitenbroek): http://home.clara.net/sisa/ Statistics at square one (British Medical Journal Publishing Group): http://www.bmj.com/collections/statsbk/ Protocols of clinical trials, which may be useful for researchers developing their own protocols, can be found in some of the clinical trial databases available on the Web, as described below. Clinical trials and the Web The Web is being used to assist in the identification and conduction of clinical trials. Identifying trials To prevent unintended duplication of clinical research, detect underreporting of research, and ease the work of systemic reviewing, it has been suggested that we should prospectively register clinical trials [36-39]. It is, however, unlikely that there will ever be one complete centralized multinational database. Instead, multiple resources set up by numerous different organizations will exist [40]. Internet technology will play a central role in linking these databases and making this information available to researchers and patients. Some scenarios in which a search of trial databases may be useful: A researcher wants to conduct a randomized controlled trial and wants to know whether anyone else is already running one on the same topic. A physician has a patient who is asking about available trials. A patient is looking for ongoing trials. A researcher is looking for possible participants for his trial. A researcher doing a systematic review is looking for unpublished trials. Information about ongoing and completed clinical trials is increasingly being published on the Internet, and searches on the Web may be a useful means of complementing traditional bibliographic searches if authors of systematic reviews wish to find ongoing or unpublished trials [41]. Researchers use their personal or department home pages to announce their interest in a certain research area or to recruit patients [42]. Journals like The Lancet have begun to publish research protocols on their Web site [43], and more and more researchers will also publish `pre-prints' (p. 239) of their findings on the Web [44]. Consumers and patient organizations also have an interest in disseminating information about ongoing trials; e.g. the National Alliance of Breast Cancer Organizations: http://www.nabco.org/ Government and funding agencies react to this need by establishing trial databases for consumers; e.g. the US National Institutes of Health searchable database [45]: http://ClinicalTrials.gov Commercial enterprises also help researchers to recruit patients, or help patients to find clinical trials. For example: CenterWatch Clinical Trials Listing Service (CenterWatch, Inc.): http://www.centerwatch.com/ ClinicalTrialFinder.com (Clinical Data Technologies Ltd): http://www.clinicaltrialfinder.com/ Current Controlled Trials (BioMed Central): http://www.controlled-trials.com Pharmaceutical companies and industry associations have likewise begun to recognize that openness and access to information on clinical trials and new drug developments can improve patient care and are part of social responsibility [46]. For example: Clinical Trials Register (GlaxoSmithKline): http://ctr.glaxowellcome.co.uk/ Search for Cures (Pharmaceutical Research and Manufacturers of America): http://www.phrma.org/searchcures/ Finally, information or databases on ongoing clinical trials can often also be found on disease-specific sites. For example: Canadian HIV Trials Network: http://www.hivnet.ubc.ca/ctn.html CancerNet (National Cancer Institute): http://cancernet.nci.nih.gov/ Conducting trials on the Web The Web is increasingly being used in the course of conducting large-scale multi-centre clinical trials (e.g. for remote randomization and data entry), and in the distribution of information on trial progress or protocols [47,48]. Trial centres may enter patient data using Java applets (see Glossary) that encrypt data and send it to the data centre via the Internet [49-52], where the data are stored and randomized, returning for example a study number and randomization information. Pre-publishing and publishing research Traditional publication is a well-defined event, whereas `publication' in the electronic age is much more of a continuum [53], reflecting and occurring during the entire research process from hypothesis formulation to data gathering, interpretation, and the presentation and discussion of the final results. In order to distinguish online collaborative `work in progress' from `final' peer-reviewed publication we may term the former `Type 1' and the latter `Type 2' electronic publication [54]. Here, peer review is not the distinguishing characteristic: in Type 1 publication a `post-publication' peer review process takes place. Type 2 publication will ordinarily take place in online journals. The following scenarios illustrate how researchers might use Type 1 electronic publication on the Internet: Sending and discussing preliminary results on mailing lists. Publishing drafts of scientific papers on pre-print/e-print sites (p. 239) in order to solicit comments and to improve the manuscript. Publishing data and information in databases; e.g. nucleotide sequences in the EMBL/Genbank databases. Publishing clinical trial protocols and raw data in a `trial bank' [55]. Current awareness services Electronic editions of paper journals and `stand alone' e-journals typically offer subscriptions to `TOC alerts', where users receive a table of contents by e-mail as soon as a new issue appears.The more sophisticated systems allow users to specify their interests using a controlled vocabulary, enabling the system to screen each newly published article for certain keywords or citations. Examples of current awareness services include: Customised @lerts (British Medical Journal): http://bmj.com/cgi/customalert/ JournAlert (Doctors.net.uk): http://www.doctors.net.uk/ Journal Watch (Massachusetts Medical Society): http://www.jwatch.org/

Purinergic_Signal-3-4-2072918

Monitoring the expression of purinoceptors and nucleotide-metabolizing ecto-enzymes with antibodies directed against proteins in native conformation

Following their release from cells, ATP and NAD, the universal currencies of energy metabolism, function as extracellular signalling molecules. Mammalian cells express numerous purinoceptors, i.e., the nucleotide-gated P2X ion channels and the G-protein-coupled P2Y receptors. Signalling through purinoceptors is controlled by nucleotide-metabolizing ecto-enzymes, which regulate the availability of extracellular nucleotides. These enzymes include ecto-nucleoside triphosphate diphosphohydrolases (ENTPD, CD39 family) and ecto-nucleotide pyrophosphatase/phosphodiesterases (ENPP, CD203 family). Investigation of these receptors and enzymes has been hampered by the lack of available antibodies, especially ones that recognize these proteins in their native conformation. This study reports the use of genetic immunization to generate such antibodies against P2X1, P2X4, P2X7, ENTPD1, ENPTD2, ENPTD5, ENPTD6, ENPP2, ENPP3, ENPP4, ENPP5, and ENPP6. Genetic immunization ensures expression of the native protein by the cells of the immunized animal and yields antibodies directed against proteins in native conformation (ADAPINCs). Such antibodies are especially useful for immunofluorescence and immunoprecipitation analyses, whereas antibodies against synthetic peptides usually function well only in Western-blot analyses. Here we illustrate the utility of the new antibodies to monitor the cell surface expression of and to purify some key players of purinergic signalling. Introduction Following their release from cells, the universal currencies of energy metabolism, ATP and NAD, function as extracellular signalling molecules [1–5]. Molecular cloning has identified 7 nucleotide-gated ion channels (P2X purinoceptors) and >8 nucleotide-activated G-protein-coupled receptors (P2Y purinoceptors) [6–10]. Signaling through purinoceptors is effectively controlled by nucleotide-metabolizing ecto-enzymes, which regulate the availability of extracellular nucleotides [11]. Prominent roles are played by members of two families of ecto-enzymes: the ecto-nucleoside triphosphate diphosphohydrolases (ENTPD, CD39 family) [12] and the ecto-nucleotide pyrophosphatase/phosphodiesterases (ENPP, CD203 family) [13]. The mammalian P2X family encompasses seven members. All P2X purinoceptors contain cytosolic N- and C-terminal tails, two membrane spanning domains, and an extracellular domain of approximately 280 amino acids containing the ligand-binding site and 10 conserved cysteine residues that likely form five intrachain disulfide bonds [14, 15]. The extracellular domains of the respective mouse and human P2X orthologues show high (ca. 80–90%) sequence identity; the paralogues within a given species show lower (40–50%) identity. The mammalian ENTPD family encompasses eight members [16]. ENTPD1–4, 7 and 8 have a structure similar to that of P2X purinoceptors, i.e., cytosolic N- and C-terminal tails, two transmembrane spanning domains, and an extracellular domain of approximately 440 amino acids, encompassing the ligand-binding site and four conserved cysteines likely engaged in intrachain disulfide bridges [12]. ENTPD1–3 and 8 are expressed as cell surface enzymes, ENTPD4 and ENTPD7 as luminal enzymes of intracellular organelles. ENTPDases share structural similarities with the actin/HSP70/sugar kinase superfamily [12]. ENTPD5 and ENTPD6 lack the second transmembrane region. Moreover, in ENTPD5, the region corresponding to the first transmembrane domain of the other family members is N-terminal and functions as a signal peptide, resulting in the secretion of ENTPD5 into the lumen of the endoplasmic reticulum. ENTPD6 is a type II transmembrane protein. ENTPD orthologues show high (ca. 80–90%) sequence identity; the paralogues within a given species show lower (30–50%) identity. The ENPP family encompasses seven members [13]. ENPP1–3 are type II transmembrane proteins with a structure analogous to that of ENTPD6, i.e., an N-terminal cytosolic tail, a single transmembrane domain and an extracellular catalytic domain of approximately 400 amino acids. ENPP4–7, in contrast, are type I membrane proteins. A hydrophobic N-terminus acts as a signal peptide, the extracellular catalytic domain is followed by a single transmembrane domain and a short C-terminal cytosolic tail. ENPPs 4–7 consist of catalytic domains only, the catalytic domains of ENPPs 1–3 are flanked by genetically fused upstream somatomedin B and downstream nuclease-like domains [13]. ENPPs share structural similarities with the alkaline-phosphatase superfamily [17, 18]. The catalytic domain of ENPP orthologues shows high (ca. 80–90%) sequence identity; the paralogues within a given species show lower (40–50%) identity. Despite their relevance for purinergic signalling, investigation of these receptors and enzymes has been hampered by the lack of available antibodies, especially ones that recognize the enzymes in their native conformation. Antibodies raised by immunization with synthetic peptides derived from the known amino acid sequence of a protein of interest generally work well in Western-blot analyses but often fail to recognize the native protein on the cell surface (Fig. 1). Such antibodies can be used successfully for monitoring the overall expression level of the protein of interest in a population of cells but cannot be used to assess cell surface expression on individual cells. Genetic immunization, in contrast, ensures that the protein of interest is expressed in native conformation by the cells of the immunized animal and yields antibodies directed against proteins in native conformation (ADAPINCs) [19, 20]. Such antibodies function well in immunofluorescence, FACS, ELISA, and immunoprecipitation analyses, i.e., assays in which antipeptide antibodies often fail. We have previously employed this technique to generate highly specific polyclonal and monoclonal antibodies against ecto-ADP-ribsoyltransferases and against murine P2X7 [20–23]. Here we report the generation and characterization of similar antibodies against other key players of purinergic signalling and illustrate the utility of these antibodies for monitoring the cell surface expression of their cognate antigens. Fig. 1DNA immunization vs. peptide immunization Materials and methods Cloning of expression vectors and cell transfections Clones encoding full-length open-reading frames representing members of the P2X, ENTPD, and ENPP families were purchased from the mammalian genome collection (www.rzpd.de). Clones that were not already in an expression vector (pCMV-SPORT6) were subcloned into the pCDNA6 expression vector. Further expression constructs were kindly provided by James Wiley, Sydney (human P2X7), Richard J. Evans, Leicester (human P2X1), François Rassendren, Montpellier (mouse P2X4), Ruth Murrell-Lagnado, Cambridge (rat P2X4), and Astrid Kehlen, Halle (hENPP2). An expression construct for nuclear green fluorescent protein was generated by cloning the DNA-binding domain of the LKLF transcription factor [24] as a C-terminal fusion protein into the pEGFP-N1 vector (Clontech). Expression constructs were transfected into CHO and HEK cells (5 μg for purinoceptors and ecto-enzymes and/or 0.5 μg eGFP-LKLF per T25 flask containing 2 × 106 cells) with the JetPei transfection reagent (Q-Biogen). For stable transfections, plasmids were linearized with Ssp1 or Sca1 before transfection, and stable transfectants were selected by cultivating cells in the presence of 12 μg/ml blasticidin (Sigma). Genetic immunization and purification of antibodies Genetic immunizations were performed as described [20, 22]. In brief, rabbits or rats were immunized four times by ballistic DNA immunization with plasmid-conjugated gold particles. Serum samples were obtained 10 days post-immunization. For monoclonal antibody production, animals received a final boost with transiently transfected HEK cells (2 × 106 cells in 200 μl PBS i.v.) 3 days prior to sacrifice. Splenocytes were fused to Sp2/0 myeloma cells and screened for production of specific antibodies by immunofluorescence analysis of CHO cells that had been co-transfected 20–42 h earlier with constructs encoding eGFP-LKLF and the protein of interest. Antibodies were purified by affinity chromatography on Protein G Sepharose (Pharmacia), and conjugated to Alexa488 according to the manufacturer’s (Molecular Probes) instructions. Antibodies directed against the C-terminal peptides of P2X purinoceptors were purchased from Caltag (rat P2X1, Cal), ABCAM (rat P2X4, Abc), and Alomone (mouse P2X7, Al). The L4 anti-human P2X7 mAb was kindly provided by Dr. Ian Chessel at Glaxo Wellcome (Middlesex, UK). Immunofluorescence analyses CHO cells were co-transfected with expression constructs for purinoceptors or ecto-enzymes and nuclear GFP (5 and 0.5 μg/106 cells, respectively). Cells were transferred onto 96-well plates (4 × 105 cells/well) 6 h post-transfection. Then, 20 h after transfection, cells were fixed for 10 min in 2% paraformaldehyde (PFA) and then stained with serial dilutions of immune serum for 30 min at room temperature. To detect ENTPD5, which is not bound to the cell surface, cells were fixed in 4% PFA and subsequently permeabilized with 0.5% IGEPAL (Sigma) before staining. Bound antibody was detected with PE-conjugated antirat or antirabbit IgG (1:100) (Dianova). Cell nuclei were counterstained with Hoechst 33342. Cells were analyzed with a Zeiss Axiovert 25 microscope equipped with the Canon PowerShot G2 digital camera and appropriate filters for visualizing Hoechst stain (Zeiss filter set 01), green fluorescent protein (Zeiss filter set 10) and PE fluorochrome (Zeiss filter set 15). Photographs were taken at 40× magnification with fixed camera settings (digital zoom: 4.3×, aperture: 5.0, exposure time: 2 s for GFP and PE, 1/8 s for Hoechst 33342). GFP and PE images were merged using Adobe Photoshop software. FACS analyses For FACS analyses, cells were harvested by brief trypsinization (5 min at room temperature) and washed in serum-containing medium. Cells (5 × 105/100 μl) were stained either directly with Alexa-488-conjugated antibodies (1 μg) or with serially diluted immune serum followed by PE-conjugated antirat or antirabbit IgG (1:100, Dianova). Gating was performed on living cells on the basis of propidium iodide exclusion. Immunoprecipitation analyses Antibodies were conjugated to Protein G Sepharose beads (Pharmacia) (1 μg antibody/20 μL beads in 200 μL PBS containing 1% Triton X-100) by incubation on a roller for 30 min at room temperature. Beads were washed twice in PBS/1% Triton X-100 before use for immunoprecipitation. Cells (107/ml) were lysed in PBS, 1% Triton-X100, 1 mM AEBSF (Sigma) for 20 min at 4°C. Insoluble material was pelleted by centrifugation, and the cleared lysates (2 × 106 cells in 200 μL) were subjected to immunoprecipitation using 20 μL antibody-conjugated beads by incubation on a roller for 30 min at room temperature. Beads were washed four times in PBS/1% Triton X-100. Proteins were eluted from the beads by incubation in SDS-PAGE sample buffer for 15 min at 70°C. Eluted proteins were size-fractionated on precast SDS-PAGE gels (Invitrogen) and blotted onto PVDF membranes. Purinoceptors and ecto-enzymes were detected with serially diluted serum followed by peroxidase (PO)-conjugated antirabbit IgG (1:5,000) using the ECL system (Amersham). Immunoblot analyses Cells (107/ml) were pelleted by centrifugation and lysed by resuspension in SDS-PAGE sample buffer (Invitrogen). DNA was sheared by brief ultrasonication on ice (two 15-s pulses with an MSE ultrasonicator at maximum power setting). Insoluble material was pelleted by centrifugation, and soluble proteins (5 × 105 cell equivalents/lane) were size-fractionated on precast SDS-PAGE gels (Invitrogen) under reducing conditions and blotted onto PVDF membranes. Purinoceptors and ecto-enzymes were detected with serially diluted serum followed by PO-conjugated antirabbit IgG (1:5,000) using the ECL system (Amersham). Results and discussion Use of transiently transfected CHO cells to assess the specificity and titres of antisera against purinoceptors and ecto-enzymes Antibodies raised by genetic immunization of rabbits and rats with cDNA expression constructs for purinoceptors, ENTPDs, and ENPPs were assayed for reactivity and specificity by indirect immunofluorescence staining of CHO cells, 24 h after cotransfection with a construct encoding nuclear GFP and the cDNA expression construct used for immunization (Fig. 2 and Table 1). By this time, transfected cells were clearly distinguishable from untransfected cells by virtue of their green fluorescent nuclei (Fig. 2). Specific antibodies were detected on the basis of bright cell surface staining of transfected cells and lack of reactivity with untransfected cells. Antipeptide antisera against the same antigens, in contrast, did not show any detectable reactivity in this assay (Table 1). Fig. 2Immunofluorescence analyses of CHO cells following transient cotransfection with expression constructs for nuclear green fluorescent protein (GFP) and purinoceptors (a), CD39-like ENTPDs (b), and CD203-like ENPPs (c). CHO cells were cotransfected with nuclear GFP and expression constructs for the indicated purinoceptors or ecto-enzymes. Twenty hours after transfection, cells were stained with serial dilutions of the indicated antisera followed by PE-conjugated antirabbit IgG. To detect non-membrane-bound ENTPD5, intracellular staining was performed by fixing and permeabilizing cells. Cell nuclei were counterstained with Hoechst 33342. Cells were analyzed with a Zeiss Axiovert 25 microscope equipped with filters for visualizing Hoechst dye, GFP, and PE fluorochrome. Panels show merged images for Hoechst dye and GFP (b, c, left panels) or for GFP and PE (a–c, right panels)Table 1Titres, specificities, and applications of antibodies against members of the families of P2X purinoceptors, CD39-like ENTPDs, and CD203-like ENPPsAntibodyHostTitreSpecificityFACS/IFTWestern blotP2X family9316rb1:800hP2X1+++(+)RG2rt1:800hP2X1+++++CR29rb1:1600hP2X4+++++CR30rb1:1600mP2X1+++++116rb1:800mP2X4RG22rt1:1600mP2X4++++++K1Grb1:1600mP2X7+++−RH23rt1:1600mP2X7++−Hano43rtmAbmP2X7++−3428rb1:1600rP2X7++−Caltrb1:2000rP2X1−+++Abcrb1:2000rP2X4−+++Alrb1:5000mP2X7−+++ENTPD family5441rb1:1600hENTPD2+++−RG10rt1:1000h/mENTPD2+++ndRG10-B28rtmAbhENTPD2+++nd5446rb1: 400hENTPD5+++−CR44rb1:6,400mENTPD1++++++RG9rt1:1600mENTPD1+++ndRG9-A59rtmAbmENTPD1+++−CR45rb1:1600mENTPD2++++CR63rb1:1600mETNTPD2+++ndCR46rb1:400mENTPD5+++−CR47rb1:200mENTPD6+++−ENPP familyK2Grb1:200hENPP2+++nd7619rb1:800hENPP2+++ndCR65rb1:800mENPP4+++ndCR66rb1:400mENPP5+++ndCR67rb1:800mENPP6+++nd7620rb1:3200rENPP3+++nd Monitoring cell surface expression of purinoceptors and ecto-enzymes by flow cytometry The utility of the ADAPINC antisera for monitoring cell surface expression was assessed further by flow cytometry (FACS analyses) of transiently transfected HEK cells (Fig. 3). The results confirm that the antisera react specifically with appropriately transfected cells and that they do not react with untransfected HEK cells. Importantly, the antisera recognize the respective purinoceptor/ecto-enzyme on the cell surface of living cells. In contrast, antipeptide antisera against the same antigens did not show any detectable reactivity with the transfected cells in these assays. Further, the use of the ADAPINC antisera for FACS analyses permits an assessment of the relative expression level of the cognate protein on the cell surface of individual cells as well as an assessment of the number of transfected vs. untransfected cells in the cell population. These parameters cannot be quantified appropriately by Western-blot analyses, since for these analyses, lysates are prepared from cell populations. We further used the FACS-based assay system to test whether the antisera crossreact with other members (paralogues) of the receptor/enzyme family and/or the same receptor/enzyme (orthologue) in other species (human, rat, or mouse). The results reveal that most antisera are specific for the receptor/enzyme used for immunization and do not cross-react with other members of the same protein family. Some antisera, however, did react with the orthologous protein, in particular at low dilutions, e.g., antiserum CR30 raised against mouse P2X1 shows some reactivity with the human P2X1 orthologue (Fig. 3a). Fig. 3FACS analyses of purinoceptor (a) and ecto-enzyme (b) expression by transfected HEK cells. HEK cells were transfected with expression constructs for the indicated purinoceptors and ecto-enzymes. Then, 24 h post-transfection, cells were harvested by brief trypsinization and single cell suspensions were stained with serial dilutions of the indicated antisera followed by PE-conjugated antirabbit IgG before FACS analyses. Control stainings were performed with untransfected cells (shaded histograms in a, dotted lines in b) and with cells transfected with the species orthologue (red lines in a) Some of the antisera recognize the denatured cognate antigen in Western-blot analyses We next tested the utility of the ADAPINC antisera for monitoring overall protein expression levels in immunoblot analyses (Fig. 4). To this end, we size-fractionated the proteins from lysates of untransfected and of transiently transfected HEK cells by SDS-PAGE followed by immunoblot analyses. The results show that only some of the ADAPINC antisera detected bands of the expected size. For comparison, we again performed parallel analyses with antisera raised against the same proteins by peptide immunization. The results show that the antipeptide sera generally detected bands of the expected size. The C-terminal peptide is commonly used for peptide immunization. In the case of P2X7, this sequence is highly conserved between rat, mouse and human orthologues, and the corresponding antipeptide serum recognized the respective orthologues (not shown). In contrast, in the case of P2X4, the mouse and human sequences differ in two and five positions, respectively, from the rat peptide used for immunization, and this antipeptide serum shows only weak cross-reactivity with these orthologues (Fig. 4a, Abc α-mP2X4). Fig. 4Immunoblot analyses of purinoceptor (a) and ecto-enzyme (b) expression by transfected HEK cells. Untransfected (u) or HEK cells transfected with mouse (m) or human (h) purinoceptors or ecto-enzymes were solubilized 24 h post-transfection with SDS-PAGE sample buffer. Proteins in cell lysates were size-fractionated by SDS-PAGE and subjected to immunoblot analyses using the indicated antisera. Bound antibodies were detected with peroxidase-conjugated secondary antibody and the ECL system Affinity purification of purinoceptors and ecto-enzymes Finally, we tested the utility of the ADAPINC antisera for immunoprecipitation of the cognate proteins from cell lysates (Fig. 5). To this end, we prepared lysates of untransfected and of transiently transfected HEK using the nondenaturing nonionic detergent Triton-X-100. Lysates were incubated with ADAPINC antibodies immobilized on protein-G sepharaose beads, and proteins bound to washed beads were subsequently analyzed by SDS-PAGE followed by immunoblot analyses with antipeptide antibodies. The results show that all of the ADAPINC antisera efficiently precipitated proteins of the expected size from cell lysates. This underscores the utility of these antibodies for purifying native receptors and ecto-enzymes from lysed cells. Fig. 5Immunoprecipitation of purinoceptors by antibodies raised via genetic immunization. Untransfected (u) and transfected (t) HEK cells were solubilized 24 h post-transfection with PBS/1% Triton X-100. Purinoceptors were immunoprecipitated from cell lysates by incubation with the indicated antisera bound to Protein G Sepharose. Precipitates were subjected to immunoblot analysis as in Fig. 4, using the indicated antipeptide antisera for detection Conclusion Using genetic immunization, we have raised highly specific polyclonal and monoclonal antibodies against key players of purinergic signalling, i.e., P2X1, P2X4, and P2X7 purinoceptors and ENTPD1, ENPTD2, ENPTD5, ENPTD6, ENPP2, ENPP3, ENPP4, ENPP5, and ENPP6 enzymes. Our findings underscore the utility of genetic immunization for generating highly specific polyclonal and monoclonal antibodies directed against proteins in native conformation (ADAPINCs) [19, 20]. These antibodies are valuable tools for assessing the expression levels of the native protein by immunofluorescence analyses and flow cytometry, i.e., assays in which antipeptide antibodies often fail. The antibodies described here provide useful tools for further characterization of the structure and function of these purinoceptors and ecto-enzymes.

Int_J_Colorectal_Dis-3-1-2077921

The artificial bowel sphincter for faecal incontinence: a single centre study

Background and aims Faecal incontinence (FI) is a socially devastating problem. The treatment algorithm depends on the aetiology of the problem. Large anal sphincter defects can be treated by sphincter replacement procedures: the dynamic graciloplasty and the artificial bowel sphincter (ABS). Introduction Faecal incontinence (FI) is a complex problem. The resulting social isolation is a major concern, which results in a reduced quality of life [1]. The real prevalence is unknown, but studies show a higher prevalence than expected [2–5]. Most patients are females with one or more vaginal deliveries in the past. Direct trauma to the anal sphincter complex can give immediate problems or problems later in life [6, 7]. The initial therapy should be conservative, e.g. diet modifications, medication, biofeedback physiotherapy or retrograde irrigation. Surgical intervention is indicated when conservative treatment fails. An anal repair is usually the first choice of treatment for a minor sphincter defect. Satisfactory results are achieved in a tension-free repair in 47–100% of the cases [8]. Long-term results are less satisfying [9]. Sacral nerve modulation (SNM) has proven to be effective for treating faecal incontinence in patients with an intact sphincter complex [10]. Sphincter replacing therapy is indicated in patients with large sphincter defects or completely disrupted sphincters and in case of SNM failure. The sphincter replacement procedures are grossly divided in the dynamic graciloplasty (DGP) [11, 12] or the artificial bowel sphincter (ABS). The first artificial bowel sphincter for faecal incontinence was a urinary prosthesis (AMS 800, AMS) placed by Christiansen in 1987 [13]. Modifications had to be made to suit the anal sphincter for use in patients with faecal incontinence. Until 1997, patients with faecal incontinence due to large anal sphincter defects were treated with DGP in our institution [16]. Since then, the ABS was introduced in our institution for the same indication. Because the operating technique is similar, there was no learning curve to be dealt with. Is this study, the results of the ABS implantations for the treatment of feacal incontinence in a large volume centre are presented. Materials and methods This study is a non-randomised, non-controlled, prospective single-centre study. Thirty-four patients with persisting or recurrent end-stage FI were included between 1997 and 2006.The majority of patients had large (>33% of circumference) anal sphincter defects. A sufficient length of the perineum was a prerequisite for ABS implantation. Previous sphincter replacement surgery was no exclusion criterion for implantation of an ABS. All patients underwent a full preoperative examination consisting of a defaecography, endo-anal ultrasound (SDD 2000, Multiview, Aloka, Japan, 7,5 Mhz endo-anal transducer), pudendal nerve terminal motor latency measurement (St Mark’s pudendal electrode) and anal manometry using a Konigsberg catheter (Konigsberg Instrument, Pasadena CA, USA) connected to a polygraph (Synectics Medical, Stockholm, Sweden). An Acticon artificial bowel sphincter (ABS, American Medical Systems, Minneapolis, MN, USA) was used in all patients. The Williams incontinence score was used to classify the symptoms. Anal manometry was routinely performed during the follow-up and used to objectivity ABS function. The follow-up appointments were scheduled at 1, 3, 6, 12 months and annually. Infection necessitating explantation was a primary endpoint. A re-intervention was a secondary endpoint. The system implantation has been described extensively elsewhere [14, 15], but will be summarised here. The ABS implant consists of three parts: an inflatable balloon, a cuff and a pump. Under strict systemic and local antibiotic prophylaxes, the cuff is placed around the anus using two lateral incisions. The pump is placed in the labia majora or scrotum, and the pressure-regulating balloon is placed in cavum Retzii. Care is taken not to perforate the rectum. If a perforation occurs, the procedure is stopped. After proper wound healing, the patient is eligible for another implantation procedure. Data are expressed as the mean with the range between parenthesis. Data were analysed using the commercially available GraphPad Prism 4.00 software (GraphPad Software, San Diego, USA). The Wilcoxon signed rank test was used for non-parametric paired values. Statistical significance was set at p < 0.05. Results The patient population existed of 25 women and nine men. The aetiology of the faecal incontinence is shown in Table 1. Three patients were previously treated with a DGP. The average age was 55.3 (23.8–75.6) years. The mean period of faecal incontinence before the placement of the ABS was 11.0 (1.0–48.0) years. One patient had a rectum perforation during the initial surgery, and placement of the ABS was abandoned. She awaits a second implant attempt. Thirty-three patients were implanted. The mean follow-up was 17.4 (0.8–106.3) months. The mean procedure time was 68.1 min (38.0–105.0). In 24 patients, the length of the cuff was 11 cm, in three patients 10 cm, in two patients 13 cm, in two patients 12 cm, in one patient 14 and in one patient 9. The width of the cuff was in all, but one patient, 2.9 cm. There was one patient with a cuff off 2.0 cm. All patients received a pressure-regulating balloon of 91–100 cm H2O. The mean postoperative hospital stay was 3.5 (2.0–12.0) days. Table 1Aetiology and previous surgical treatmentNumberSexEtiologyPrevious treatment1FHysterectomy, cervix carcinoma, radiotherapy2MAnal atresiaDGP3FTwo breech deliveries: ruptureAnal repair, SNM4FEpisiotomy, hysterectomyPNE5MPelvic trauma: urethra/rectum ruptureRepair and colostomy6MTrauma, partial spinal cord lesionPNE7FDelivery trauma: total rupture, hysterectomyAnal repair, SNM8FDelivery trauma: rupture9MAnal atresia10MClassical hemorroidectomy11FEpisiotomy, hysterectomyPNE12FDelivery trauma: ruptureTwo anal repairs, PNE13FDelivery trauma: total ruptureTwo anal repairs, 14FDelivery trauma: ruptureAnal repair15FDelivery trauma: rupture, cauda syndromeAnal repair16FDelivery trauma: rupture, hysterectomyAnal repair17FDelivery trauma: rupture, hysterectomyAnal repair18MAnal atresiaDGP19FDelivery trauma: ruptureAnal repair20FDelivery trauma: rupturePre-/post-anal repair, PNE21FDelivery trauma: rupture, hysterectomyAnal repair, PNE22FDelivery trauma: rupturePost-anal repair, SNM22MLow anterior resection T2NOM023FDelivery trauma: rupture24MPelvic crush trauma: urethra/rectum ruptureRepair and colostomy25FDelivery trauma: rupture26FDelivery trauma: total rupture12 anal repairs27FDelivery trauma: ruptureAnal repair, colostomy28FDelivery trauma: ruptureAnal repair29MPelvic trauma30FDelivery trauma, uterus extirpatie DGP31FDelivery traumaPNE32FClassical hemorroidectomySECCA33FDelivery trauma, total ruptureAnal repair, Thiersch wire34FDelivery trauma, hysterectomyPNEF Female, M male, DGP dynamic graciloplasty, SNM sacral neuromodulation, PNE percutaneous nerve evaluation The mean preoperative Williams score of 4.8 (4–5) decreased significantly after ABS placement to 2.1 (1–5; Fig. 1). The mean preoperative anal resting pressure was 58.1 (17.0–128.0) mmHg. This was not significantly altered after implantation (60.3 (21.0–93.0 mmHg; p = 0.89). The mean preoperative squeeze pressure was 80.1 (25.0–149.0) mmHg, which increased to 120.5 (65.0–154.0) mmHg after implantation (p = 0.003; Fig. 2). Fig. 1Mean pre- and postoperative Williams score (1 = continent, 2 = incontinent to flatus, 3 = incontinent to liquid stool, 4 = occasional incontinence to normal stool <1, 5 = fully incontinent)Fig. 2Baseline resting pressure versus deflated ABS pressure and baseline squeeze pressure vs inflated ABS pressure pre- and postoperatively (at last follow-up) Thirteen patients (39%) complained about a rectal evacuation problem. In 12 patients, this could be managed conservatively. One patient had a revision of the system with placement of a wider anal cuff. Seven patients (21.2%) had an infection of the system, which led to seven explantations. One of these patients has been implanted successfully with a new ABS (Fig. 3). Fig. 3Flow chart of implanted patients In one patient, the ABS was successfully converted to a dynamic graciloplasty. In two patients, a colostomy was performed. The other three patients had no other interventions. One patient was explanted due to persisting peri-anal pain without an infection. She received a colostomy. Twenty-six reinterventions (including explantations) had to be performed. This means 0.79 re-intervention per implanted patient. Discussion There is a large experience in our institution with the DGP [16]. However, since 1997, the ABS is also performed in our institution for the same area of indications as the DGP. When a patient qualifies for a sphincter replacement procedure, he or she can decide whether an ABS or DGP will be performed. Nonetheless, sufficient perineal length is a prerequisite for ABS implantation in a female patient. We believe that the risk for late erosion of the ABS is higher in the case of severe, cloaca-like malformations of the perineum. In these cases, a DGP is the preferred procedure. All patients in this study had an adequate perineal length. In the beginning, the initial infection rate of the DGP was a problem, but improved as a result of technical modifications and the introduction of systemic and local antibiotic prophylaxis. The same prophylaxis protocol was used for the implantation of ABS. However, despite meticulous application of the antimicrobial protocol, the infection rate of the ABS implantations in our patient population remains high and is comparable with other series [14, 15]. We believe that this infection rate is likely to remain a serious problem in every attempt to place a corpus alienum around the anus through peri-anal incisions. To overcome this problem of infection, Finlay et al. [17] have developed a new prosthetic sphincter, which is placed above the pelvic floor musculature by means of a laparotomy. It was hypothesised that this sphincter will function as a new puborectal sling in this position. Till now, 12 patients are implanted. Infectious complications, however, occurred in three patients (25%), with subsequent removal of the system. Technical problems occurred in five of the nine remaining patients during follow-up. Technical failure is also one of the main problems of the ABS. Twelve of our patients had some sort of technical failure. This is also known from other studies concerning the ABS [18]. Only limited data on long-term follow-up of a sufficient number of ABS sphincters are available. There is one multicentre study with disappointing long-term data where the initial data were promising [19]. The anal manometry data of this patient population suggest poor action of the ABS. The authors conclude that the ABS acts as a passive barrier causing a rectal outlet obstruction. Our manometry data contradict with this conclusion. We strongly believe that the ABS acts as an active sphincter. In our experience, the patients need to deflate the anal cuff to defecate. Nevertheless, constipation can be a problem. Thirteen of our patients complained about constipation. This could be solved in the majority of patients by conservative means. One patient needed a wider anal cuff to treat an outlet obstruction. The indications for sphincter replacement surgery are decreasing in our institution since the introduction of SNM. The relative numbers of DGPs and ABSs decreased, while the number of SNM has increased. This implicates that ABS and DGP are reserved for the more severe complicated cases of faecal incontinence. A higher complication rate is therefore expected. However, the placement of an ABS remains an alternative to a colostomy in the well-informed and motivated patient even if a DGP has failed. Conclusion The artificial bowel sphincter is an effective treatment option for severe faecal incontinence. Even in experienced hands, the risk of infection, explantation and system malfunctioning remain high. In well-informed and motivated patients, it is worthwhile to proceed to implantation, as the alternative is a colostomy. Our data suggest that the ABS acts as an active sphincter and not as a passive barrier.

Eur_J_Health_Econ-_-_-1388081

The “Health Benefit Basket” in France

The French “Health Benefit Basket” is defined principally by positive lists of reimbursed goods and services; however, global budget-financed hospital-delivered services are more implicitly defined. The range of reimbursable curative care services is defined by two coexisting positive lists/fee schedules: the Classification Commune des Actes Médicaux (CCAM) and the Nomenclature Générale des Actes Professionnels (NGAP). The National Union of Health Insurance Funds has been updating these positive lists since August 2004, with the main criterion for inclusion being the proposed procedure’s effectiveness. This is assessed by the newly created High Health authority (replacing the former ANAES). In addition, complementary health insurers are consulted in the inclusion process due to their important role in French healthcare financing. The French health insurance system, a mix of explicit and implicit regulations, offers wide-ranging reimbursement in the fields of preventive, curative, rehabilitative, and palliative care. The present analysis of the health benefit basket in France is being carried out during a period of reforms. This contribution describes the structure of the statutory health insurance system, followed by a description of the entitlements and benefits and of the actors involved in the decision-making process with its criteria for services of curative care, HC1. Finally, we analyze the main changes which may affect the health basket in France. The structure of the French statutory health insurance system The financial management of health care in France is undertaken mainly by the statutory health insurance system as a branch of the wider social security system. The health insurance system’s current structure is based on its founding text, the Ordinance of 4 October 1945 and subsequent legislative measures. This system has covered the entire population of France since 1 January 2000 when the Couverture Maladie Universelle (CMU, Universal Health Coverage Act) extended basic health insurance cover to all legal residents of France. The French statutory health insurance system is a compulsory scheme and covers all households regardless of health status, income, number of persons, etc. It provides a somewhat uniform field of reimbursement, with the “basket of goods and services” covered by the insurance funds being identical for all the statutory schemes with the reimbursement rate being the same for the three main insurance schemes since 2000. The exceptions are the Alsace-Moselle region’s local scheme and certain public sector schemes [1]. Membership in one of the health insurance funds of the statutory health insurance depends on the profession of each person. In the context of the CMU, however, participation depends on legal residence in France and on the level of income. Any dependants of the insured person are also covered by his/her health insurance. The three main health insurance schemes are as follows: (a) The General Scheme (Régime général) covers employees in commerce and industry and their families (about 84% of the population) as well as persons receiving CMU, estimated in late 2003 at about 1,500,000 person (2.4% of the population) [2]. (b) The Agricultural Scheme (Mutualité sociale agricole) covers farmers and agricultural employees and their families (about 7.2% of the population). (c) The Scheme for the Non-agricultural Self-Employed (CANAM) covers craftsmen and self-employed persons, including self-employed professionals such as lawyers (about 5% of the population). Another ten work-related schemes cover specific sections of the population. Statutory health insurance funds three-quarters of the health expenditure and therefore leaves considerable scope for complementary sources of funding. An estimated 85% of the population have complementary health insurance. Taking into account recipients of Couverture Maladie Universelle Complémentaire (Complementary Universal Health Coverage 6% of the population), about 91% of the French population are covered by the complementary health insurance scheme, which covers the same health basket as the statutory schemes and not other goods and services [3]. The most important benefit catalogues for France and their underlying criteria are displayed in Tables 1 and 2 while the following sections are confined to curative health care services.Table 1 Catalogues and implicit regulationGBRCCAMNGAPLSAC ATUMed. dev.NABM Sp. reg.HC.1 Curative care services  1. Inpatient curative care++–––––  2. Day-patient curative care++–––––  3. Outpatient curative care+–––––    3.1. Basic medical and diagnostic services++–––––    3.2. Outpatient dental care++–––––    3.3. All other specialized health care++–––––    3.4. All other outpatient curative care (paramedical)––+––––    3.5. Services of curative home care+–+––––HC.2 Services of rehabilitative care  1. Inpatient rehabilitative care+–––––+  2. Day cases of rehabilitative care+–––––+  3. Outpatient rehabilitative care+–+––––  4. Services of rehabilitative home care+–+–––+HC.3 Services of long-term nursing care  1. Inpatient long-term nursing care+–––––+  2. Day cases of long-term nursing care+–––––+  3. Long-term nursing care: home care+–+––––HC.4 Ancillary health care services  1. Medical analysis laboratory+––––+–  2. Diagnostic imaging++–––––  3. Patient transport and emergency rescue +–––––+HC.5 Medical goods dispensed to outpatients  1. Prescribed medicines+––+–––  2. Other medical nondurables+–––+––  3. Therapeutic devices and other medical durables+–––+––HC.6 Prevention and public health services––––––+GBR General Benefit Regulation, CCAM list of physicians’ and dentists’ technical procedures, NGAP list of physicians’ consultations and other health professionals’ activity, LSAC ATU positive lists of drugs, Med. dev. positive lists of medical devices, NABM positive lists NABM biology procedure, Sp. reg. specific regulationTable 2 Benefit-defining laws/decrees, catalogues and implicit regulationGBRCCAM, NGAPLSAC ATUMed. dev.NABM Sp. reg.Legal status: law, decreeLawUNCAM decisionMinisterial orderMinisterial orderUNCAM decisionAdministrative documentDecision makerParliament UNCAM, on the advice of HAS and UNOCMinisters of health and social security, on the advice of the Transparency CommissionMinisters of health and social security on the advice of the ad hoc CommissionUNCAM, on the advice of HAS and UNOCOriginal purpose: entitlements, reimbursement, target settingReimbursementPositive listPositive listPositive listPositive listPositive listTariffsFee schedulePrices or reference pricesReference pricesFee schedulePositive-negative definition of benefitsPositivePositivePositivePositivePositivePositiveDegree of explicitnessa3 (except inpatient care: 1)33332 or 3If itemized: goods/procedures only; linked to indicationsMainly goods and procedures; linked to indicationsProcedures, sometimes linked to indicationsPharmaceutical products, linked to indicationsGoods linked to indicationsProcedures linked to indicationsMainly indicationsUpdating–RegularlyRegularlyRegularlyRegularly–GBR General Benefit Regulation, CCAM list of physicians’ and dentists’ technical procedures, NGAP list of physicians’ consultations and other health professionals’ activity, LSAC ATU positive lists of drugs, Med. dev. positive lists of medical devices, NABM positive lists NABM biology procedure, Sp. reg. specific regulationa 1, “all necessary”; 2, “areas of care”; 3 “items” Organizational structure and actors involved in the definition of the benefit basket for curative services In France the general conditions of the reimbursement system are established by law. Health benefit catalogues are drawn up at national level with the whole range of goods and services reimbursed by the statutory scheme being specified in Article L.321-1 of the Social Security Code (SSC) [4]. The reimbursement of goods and services depends on their inclusion in positive lists, according to Articles L.162-1-7, L.162-17, and L. 165-1 of the SSC. Until 2004 positive lists were officially enforced by ministerial orders detailing the inclusion of new goods and services. Ministers made their decisions upon the advice of ad hoc scientific commissions and agencies, for example, the National Agency for Accreditation and Evaluation in Health Care (ANAES). The inclusion of all the listed procedures depends on ANAES advice which considers the effectiveness and/or safety of these procedures and the conditions under which they need to be performed. Reimbursement is legal only when goods or services are provided in an appropriate medical context. Reimbursement for all goods and all paramedical procedures depends on the provision of a prescription, which serves as confirmation of the necessity of such goods and services. For some types of treatment, such as physiotherapy and spa treatment, the prescription from a physician does not provide the needed status for reimbursement. Coverage by statutory health insurance is subject to the prior authorization (entente préalable) of the physicians advising the health insurance funds, after examination of the patient’s case history and a possible interviewing of the patient. According to Article L.322-2 of the SSC, the insured person’s copayment is fixed by a decision of Union Nationale des Caisses d’Assurance Maladie (UNCAM, National Union of Health Insurance Funds) within rate limits defined by a Council of State decree. The copayment can be either a percentage of the charges or a lump sum. The patient’s contribution to the total cost of treatment varies according to the type of treatment, being higher for outpatient care and drugs compared to hospital treatment. The copayment must be paid by the insured person or, where applicable, by their complementary health insurance fund. Council of State decrees also define copayment exemption conditions (Article L.322-3 of the SSC). The most important exemptions are linked to the health status, especially in the case of one person suffering from one of 30 specified long-term illnesses, for example, diabetes, AIDS, cancer, and psychiatric illness, and if the patient is suffering from one or several incapacitating diseases. Other health status-based exemptions concern, for example, disabled persons under the age of 20 years and pregnant women during the final 4 months of pregnancy. The health benefit basket is explicitly defined by positive lists of goods for both the public and the private sectors. Positive lists only apply to services delivered by private sector professionals in their own practices or in private for-profit hospitals. Conversely, services dispensed in public hospitals or private not-for-profit hospitals were mainly the subject of implicit definition since they were paid for by a global budget, which means in practice that basically every dispensed service was reimbursed. The situation is currently undergoing an important reform. HC1 services qualifying for reimbursement by the health insurance system include: (a) inpatient care and treatment in public or private healthcare institutions; (b) outpatient care provided by general practitioners, specialists, dentists, and midwives; and (c) diagnostic services and care prescribed by physicians and performed by laboratories and paramedical professionals, such as nurses, physiotherapists and speech therapists. In July 2005, two health benefit catalogues for curative services exist: (a) The Classification Commune des Actes Médicaux (CCAM), which revises the previous medical services catalogue, has been adopted in 2005 with its application still only partial [5]. (b) The general fee schedule (Nomenclature Générale des Actes Professionnels, NGAP), which is the medical procedures positive list for health professionals in private practice, in their own surgeries and consulting rooms and in private-for-profit hospitals, remains in force until the CCAM is fully implemented. Inpatient curative service The Social Security Act of 18 December 2003 (Loi de financement de la sécurité sociale, LFSS) changed the inpatient acute care funding rules, but implementation is still in progress. The situation will continue to change during the next few years for at least two reasons. Firstly, the positive list for physicians’ procedures, the CCAM, applies to both private and public hospitals (Article 162-1-7 of the SSC), which had not been the case before. Secondly, the implementation of the per-case payment reform that will lead to the result of both sectors being brought into line. These situations before and after the reforms of 2004-2005 are shown in Table 3.Table 3 Definition of benefit catalogues for inpatient care Before 2004–2005After 2004–2005Funding rules depend upon hospital status: PH and PNFPH, vs. PFPHNew funding rules making the basket more explicitRange of reimbursed medical procedures defined at least for PFPH but not for other services, e.g., nursing careSame list of medical procedures for both private and public hospitals (Article 162-1-7 of the SSC),No positive list for PH, an implicit coverage for all services in PH and PNFPH mainly financed by the global budgetSame prospective per-case payment system for PH and PFPH for all medical, surgery and obstetrics services, based on DRG-type classification of GHS, but tariffs still differ from PH to PFPH.Basket of reimbursable drugs defined for PH and PFPHBasket of reimbursable drugs defined for PH and PFPHPH public hospital, PNFPH private non-for-profit hospital, PFPH private-for-profit hospital, DRG diagnosis-related group, GHS homogeneous stay group As seen in Table 3, there has been no positive list to define reimbursable services in public and in private not-for-profit hospitals, in which inpatient services were regulated by implicit coverage. However, the implementation of the per-case payment system will simplify the situation as all medical, surgery and obstetrics services in all hospitals will be included in it. The reform will also change the remuneration schemes of inpatient and outpatient care. Services provided in inpatient or outpatient acute care will be financed through a payment-per-case system. This is based on a diagnosis-related group type of classification of 700 Groupes Homogènes de Malades (GHM), considering comorbidities. A nationally fixed tariff (Groupe Homogène de Séjours, GHS, Homogeneous Group of Stays) is applied to each GHM [6]. Outpatient procedures will be paid on a fee-for-service basis, and organ retrieval and emergency services by annual lump sum payments. This payment system also includes earmarked funding (Missions d’intérêt général et Aide à la contractualisation, MIGAC) to finance other activities, such as research and education, but also certain healthcare activities promoted by contracts between hospitals and regional health agencies. The new GHS payment system has been in operation since 1 March 2005 for private-for-profit hospitals. Procedures carried out by physicians are always paid separately and directly to the physicians concerned on a fee-for-service basis, while in public hospitals, tariffs include specialists’ salaries. Since July 2005, the CCAM is used as the general fee schedule and can be considered as the positive list for private for hospitals. Health insurance funds will not finance any stay if one of these procedures does not apply. CCAM is currently available only for technical procedures such as surgery, medical imaging, and radiotherapy performed by physicians. Paramedical services, including, physiotherapy, which are linked to inpatient medical procedures, are reimbursed only if they are an essential part of the medical treatment. Regarding the definition of the health basket, the rules for day-case patient care are the same as those applicable to inpatient care. GHS tariffs refer to day hospital treatment. Outpatient curative care Patients who need health care have until now been free to choose which physician to consult and have also been allowed to refer themselves to specialists. Only few general practitioners have begun playing the role of gatekeepers, although a few of them have been making some attempts in this regard since 1996. Recent health insurance reforms are about to change this situation. Since 1 July 2005 all those benefiting from health insurance coverage in France are obliged to choose their main physician (médecin traitant). As a result there will be higher user charges if a person chooses to consult a specialist directly without being referred by her/his “gate-keeping” médecin traitant general practitioner [7]. The basket of reimbursable outpatient curative services is defined by a positive list which is also used as a fee-schedule for health professionals in private practice. Until March 2005 the NGAP was the only list and applied to all services. This will be replaced by the new CCAM classification. In a rather lengthy process CCAM is currently being developed. At this stage CCAM specifies only technical procedures such as diagnosis, surgery, and radiology performed by specialists and dentists. CCAM is also being used in public hospitals to charge for outpatient care and to specify the reimbursement rate. This classification is not yet used for specialist consultations or procedures carried out by other healthcare professionals who are still in use of the NGAP. The CCAM is fully comprehensive in content as it contains details of all medical procedures, even those that are not reimbursable. Each procedure corresponds to only one label and one code, eliminating any ambiguity and making it easy to use. The classification is structured according to the anatomical classification and specialities. There are 17 chapters which are based on the organ system to which the procedures refer. The CCAM is based on the rule of procedural totality, meaning that each label implicitly contains all the operations necessary for the performance of the medical procedure. It was drawn up as a resource-based relative value scale (RBRVS) by the main health insurance fund (CNAMTS) with the collaboration of health professionals, and the ANAES being involved in the selection of effective procedures. Since the reform of 13 August 2004 the UNCAM, which includes representatives of the three main sickness funds, is in charge of preparing and updating the positive lists. UNCAM will be assisted in its decision making by the advice of the two newly created bodies, the High Health Authority (Haute Autorité de Santé), which replaced the ANAES and the National Union of Complementary Health Insurance Organizations (Union Nationale des Organismes d’Assurance Maladie Complémentaire, UNOC). The same criteria used for the NGAP will be used for the taxonomy of medical procedures leaving room for improvements in medical services and achieving cost-containment at the same time [8]. Nevertheless, the ministries of health and social security retain the right to reject UNCAM’s decisions and to include or exclude goods and services in or from the list, especially if public health issues are concerned. A commission comprising healthcare professionals’ unions and representatives of UNCAM has been created to determine the general rules for drawing up the RBRVS and then for validating the scale proposed by the health insurance fund services. In addition, UNCAM is responsible for negotiating the tariffs of medical procedures with healthcare professionals’ unions and for determining the levels of copayment and coinsurance. Other outpatient services such as auxiliary services provided by nurses, physiotherapists, orthoptists, speech therapists and chiropodists must be provided on medical prescription in order to be reimbursed. In the area of complementary and alternative medicine (CAM) special recognition is given to acupuncture and homeopathy. These two therapies are thus recognized and may be legally practiced, but only by physicians or licensed physiotherapists on medical prescriptions for acupuncture. A few osteopathy and chiropractic procedures have recently been included in the CCAM in addition to certain vertebral manipulations which were already listed in the NGAP. The range of services covered by statutory health insurance does not include cosmetic surgery and some other treatments and services of uncertain effectiveness such as spa treatments. The issue of the allocation of scarce resources means that choices must be made which may result in the nonreimbursement of certain procedures, for example, bone density measurement when performed in the private sector as a preventive measure, and the imposition of limits on the frequency for which they can be reimbursed, such as mammography for cancer screening purposes. Home care services The entitlements laid down in the SSC include services provided by independent physicians (home visits) and by other health professionals such as nurses and hospital-at-home services. Regarding palliative care at home, according to Article R 162-1-10 of the SSC, a multidisciplinary team of health professionals can be organized at the special request of a person whose state of health demands it, to provide palliative home care. The remuneration can be paid on a fee-for-service basis or as a lump sum. Discussion This study faces a number of limitations which may affect the analysis of the French health benefit basket. Firstly, the study was carried out in a period of changes related to the redefinition of the medical procedures for both inpatient and outpatient care. This will make some statements in the previous catalogue become obsolete. Furthermore, the introduction of the new CCAM catalogue has been postponed several times due to conflicts of interest between health professionals, especially those between the physicians’ unions and the health insurance funds. The catalogue deals with the two issues of the remuneration of healthcare professionals and the state regulation. Secondly, the recent health insurance reform (Health Insurance Act, 13 August 2004) affects all actors involved in the definition of the benefit catalogue and thus the decision criteria. The High Health Authority, for instance, is expected to have considerable powers. At the same time, UNCAM is also playing an important role in the drawing up of the positive lists of procedures. However, the ministers of health and social security still retain their right to reject the UNCAM’s decisions. It could be asked whether the change regarding the delegation of the task of drawing up benefit catalogues to a so-called “self-governing body” is fully effective, as this is the case in Germany. It is too early to assess the consequences that these new regulations will bring about, but it can be anticipated that it might well be a case of plus ça change, plus c’est la même chose, as both the French government and the physicians have always played a major role in the healthcare decision-making process. Thirdly, the French system offers a mix of explicit regulation for ambulatory care, pharmaceuticals and medical devices, dental care, etc., with benefit catalogues and positive lists and of implicit regulation for a large part of inpatient care [9]. Altogether the package of health care services covered is comprehensive and wide-ranging although that the additional billing is rather high for certain types of goods and services. As a consequence of the implicitness of the coverage with particular respect to inpatient care, we have noted some differences between principles and practice, indicating that not all services are actually covered to the same extent. The interviews carried out at the Ministry of Health confirmed this finding. Finally, the drawing-up of the positive lists still remains hotly contested among those at the health and social security ministries who were interviewed for this study. Current French health policy combines the harmonization of regulation with the reduction in health inequalities on the basis of improved knowledge of public health needs. However, this policy is being implemented in a context of increasing health expenditure and especially of increasing user charges. It is a contradictory situation in which things that have been received by the one hand have again been taken away by the other one.

Exp_Brain_Res-4-1-2315690

Eye–hand coordination during manual object transport with the affected and less affected hand in adolescents with hemiparetic cerebral palsy

In the present study we investigated eye–hand coordination in adolescents with hemiparetic cerebral palsy (CP) and neurologically healthy controls. Using an object prehension and transport task, we addressed two hypotheses, motivated by the question whether early brain damage and the ensuing limitations of motor activity lead to general and/or effector-specific effects in visuomotor control of manual actions. We hypothesized that individuals with hemiparetic CP would more closely visually monitor actions with their affected hand, compared to both their less affected hand and to control participants without a sensorimotor impairment. A second, more speculative hypothesis was that, in relation to previously established deficits in prospective action control in individuals with hemiparetic CP, gaze patterns might be less anticipatory in general, also during actions performed with the less affected hand. Analysis of the gaze and hand movement data revealed the increased visual monitoring of participants with CP when using their affected hand at the beginning as well as during object transport. In contrast, no general deficit in anticipatory gaze control in the participants with hemiparetic CP could be observed. Collectively, these findings are the first to directly show that individuals with hemiparetic CP adapt eye–hand coordination to the specific constraints of the moving limb, presumably to compensate for sensorimotor deficits. Introduction Cerebral palsy (CP) is an umbrella-term describing a group of disorders of movement and posture leading to activity limitations, and that are caused by damage to the fetal or infant brain (Bax et al. 2005). In this study, we focused on the most common subtype of cerebral palsy, spastic hemiparesis, which typically occurs after unilateral lesions to the cerebral cortex or corticospinal pathways (Kwong et al. 2004). This condition is characterized by impaired control of muscle tone and spasticity in the upper and lower limbs of the contralesional, “affected”, side of the body (Albright 1996), generally progressing from proximal to distal (Freund 1987; Steenbergen et al. 2000b), and often accompanied by sensory deficits of proprioception and tactile perception (Cooper et al. 1995). These symptoms induce limitations on manual actions performed with the affected hand (AH), in particular related to fine motor skills such as fingertip force control (Eliasson et al. 1991). These limitations are further amplified by more central deficits in the integration of different modalities, such as matching proprioceptive and visual spatial information (Wann 1991), and integrating sensory information with motor output (Gordon et al. 2006). While the less affected hand (LAH) is usually spared from severe sensorimotor impairments (but see, Steenbergen and Meulenbroek 2006 for subtle deficits at the less affected body side), there is converging evidence for more high-level deficits in prospective control, or motor planning, in particular, in individuals with right hemiparesis (Mutsaarts et al. 2005, 2006; Steenbergen et al. 2000a, 2004; te Velde et al. 2005). These studies on manual control have advanced our knowledge about the different levels at which the motor system affects CP as well as the behavioural reorganisation during the performance of actions (e.g., van Roon et al. 2005b). Yet, up to date, eye–hand coordination has not been investigated, despite preliminary evidence suggesting that it may contribute to deficits in action performance in individuals with CP (van der Meer et al. 1995; van der Weel et al. 1996). Even indirect evidence regarding the role of visual feedback for manual action performance in CP is scarce. Wann (1991) investigated visuo-proprioceptive integration in adolescents with CP. Comparing arm postures in a bimanual position-matching task under different conditions of sensory feedback, he found that, in contrast to adult controls, participants with CP employed very similar end postures (joint angle configurations) with the left and right arm in conditions in which vision of one or both arms was prevented. This was interpreted as evidence that they relied more strongly on purely proprioceptive information rather than on a shared visuospatial-proprioceptive representation and that the encoding of visual and proprioceptive information into a common egocentric frame may be problematic in these individuals. This would suggest that individuals with CP more strongly depend on continuous visual feedback for guiding manual actions to visual targets, compared to neurologically healthy adults. Moreover, as the severity of the motor impairment, and concomitant proprioceptive impairment, increases from proximal to distal (e.g., Steenbergen et al. 2000b), it may be expected that actions with the affected-hand involving more fine motor control, such as grasping, need closer visual monitoring to provide additional feedback. In line with this, Steenbergen and colleagues (1996), studying bimanual actions in participants with hemiparetic CP, made the anecdotal observation that (overt) visual attention seemed to be drawn to the affected side of the body. However, in contrast to the reasoning above, a recent study that scrutinized the effect of removing visual information of the moving limb on end-point accuracy in a straight-line drawing task (van Roon et al. 2005a) found no specific adverse effect of this manipulation in participants with tetraparetic cerebral palsy when using their relatively less-impaired hand compared to control participants. Thus, up to date, there is no consensus on the role of visual guidance for the control of manual actions in CP. Besides the sensorimotor impairments of the AH, more high-level aspects of action control, such as motor planning and prospective control, have been shown to be compromised as well in hemiparetic CP. In research on motor planning in individuals with CP, the LAH is used, as this hand has no (or small) movement restrictions, as is often the case with the AH. Hence, the higher-order process of motor planning can be studied without the confounding effects of possible movement restrictions. Atypical performance with the LAH was found for instance in tasks involving planning of the hand’s posture at the end of the action (Steenbergen and Gordon 2006; Steenbergen et al. 2000a), in particular, in individuals with right hemiparesis (Steenbergen et al. 2004). In these tasks, individuals have to pick up objects and place them in another orientation on a designated target. Ample evidence in controls suggest that individuals pick-up objects with a grip that allows them to end the task with a comfortable posture, even when this means that they have to sacrifice comfort of the start grip (see Steenbergen and Gordon 2006). This “end state comfort” effect was not present in individuals with CP, indicating a higher-level impairment of movement planning. Prospective control in individuals with hemiparetic CP was recently investigated in a collision avoidance task (te Velde et al. 2005). This study found later, hand-movement initiation in individuals with right as compared to left hemiparesis, again indicating reduced anticipation of action requirements. The extent to which deficits in motor planning or prospective control may be related to deviations in visual monitoring, with its typically anticipatory role for action control (Johansson et al. 2001) has not been investigated yet in individuals with CP. In the present study, we examined eye–hand coordination in an unconstrained object prehension task in individuals with hemiparetic CP and neurologically healthy controls. To manipulate task difficulty, an obstacle was present in half of the trials, inducing a more complex transport trajectory. Participants performed the task with each hand separately. Based on the registration of eye and hand movements, we calculated several measures of temporal and spatial aspects of eye–hand coordination. Besides giving a first descriptive account of gaze patterns during object manipulation in a population with congenital hemiparesis, we addressed two hypotheses. Our first and main research question concerned the way in which individuals with hemiparetic CP adapt eye–hand coordination to the sensorimotor impairments of their affected hand. We hypothesized that individuals with CP would use increased visual monitoring when they perform the task with their AH as compared to performance with the LAH and to performance of control participants. Second, we examined eye–hand coordination when the task was performed with the LAH in participants with CP. Based on the previously established deficits in action planning, we hypothesized that gaze patterns are less anticipatory in general, in individuals with CP. Given the limited knowledge on visumotor control in CP and on the role of eye-movements for action planning, this second research question was much more exploratory than the first. Methods Participants In total, 16 individuals took part in the study. Six participants with hemiparetic CP constituted the experimental group and ten participants with no known history of neurological disorders served as controls. Control participants were all right handed (Dutch version of the Edinburgh handedness inventory, Oldfield 1971) University students (9 females, 20–25 years). The six participants of the experimental group (5 females, 14–19 years) were students at the Werkenrode Institute (Groesbeek, The Netherlands), where they followed an adapted educational programme. They had been asked to take part in the study based on their condition (viz. congenital spastic hemiparesis). Additional requirements were no known oculomotor disturbance (such as nystagmus), and having the motor, cognitive and attentional capacities to perform the experimental task as assessed by initial screening. All were able to walk and sit independently, and able to read and write with their LAH. Consent was obtained prior to the experiment, both in verbal form from the participants during the initial screening session, as well as in written form from their parents or caregivers. In a separate session taking place after the experiment, the participants with CP performed two standardised dexterity tests, the Box and Block test for gross dexterity (Mathiowetz et al. 1985) and the Purdue pegboard test for fine dexterity (Tiffin 1968). One of the participants was not available for the manual dexterity testing afterwards. Table 1 shows the results of these tests as well as other participant information. For the Purdue–Pegboard test, the norm score for individuals at the age of 16–17 years is 49.5 for the preferred hand. Standardised norm scores for the Box and Block test are not available for individuals younger than 20 years. For the youngest age group available, individuals between 20 and 24, the norm score is 88.2 for the preferred hand. The performance of the participants with CP was consistently below these norms, even when using the LAH. Table 1Participant information for the participants with cerebral palsyParticipantAgeParetic sideB and Ba, cPPb, cCP115Right15/380/33CP217Left24/520/42CP317Left16/326/27CP415Right15/430/33CP514Left18/423/33CP619RightNAdNAdaBox and Blocks testbPurdue–Pegboard testcAffected hand/less affected handdNot available for testing All participants received the same reimbursement (€6 per hour) for taking part. The study was approved by the local ethics committee and performed in accordance with the standards laid down in the 1964 Declaration of Helsinki. Setup and procedure The experimental setup is depicted in Fig. 1. Participants were comfortably seated on a chair approximately centered at the midline of the table, which constituted the working region for the experiment. Two objects (blue and green), two target regions, and an optional obstacle were placed on the table, as well as a fixation target (see Fig. 1 and caption for details). Two loudspeakers, placed below the table, were used to verbally instruct the participant through a pre-recorded voice. The experimenter sat facing the participant at a separate table. From this position he was able to supervise the experiment by instructing the participant, starting and terminating the trials, and monitoring participants’ performance (directly and via a computer monitor that showed the eye movement recording system’s output in real time). Fig. 1Experimental setup (top view). The participant is seated at the table, with the right hand resting at the starting position. Positions of objects (blue and green cylinder, 6 cm in height, 3 cm in diameter, with an enlarged stabilizing basis of 6 cm in diameter), target regions (blue and green disc, 10 cm in diameter, contralateral to the corresponding object), optional obstacle (cylinder, 15 cm in height, 6 cm in diameter), and the pre-trial fixation target (red ball 2 cm in diameter mounted on a frame at about eye height) are shown. The experimenter was seated at a separate desk facing the participant. The dashed line represents the object transport movement from object to target region in a “blue” trial. The x- and z-axis of the Optotrak system were approximately aligned to the axes indicated here, the y-axis was orthogonal to both (vertical axis) For each participant, the experiment consisted of eight conditions in a 2 × 2 × 2 factorial design, with the factors hand (AH/LAH for the participants with CP, NPH/PH for controls), obstacle (present, absent) and side (blue, green). In each of these conditions, five trials were performed, yielding a total of 40 trials per subject. The factors hand and obstacle were blocked while side was randomized within blocks. The order of the blocks was partially counterbalanced between participants. Prior to each trial, the participant was instructed to rest the “inactive” hand in his/her lap, such that it did not interfere with the task. Each trial started with the task-hand resting at the starting position, and with gaze directed to the fixation target. After a variable delay (500–1,500 ms), the pre-recorded instruction “green” or “blue” (in Dutch) was played, upon which, the participant was to grasp the corresponding object and transport it to and place it on the contralateral target region of the same color. When an obstacle was present, the object had to pass in front of it, that is, between obstacle and participant (see Fig. 1, dashed arrow-line). After placing the object on the target region, the participant moved his/her hand back to the starting position and the experimenter put the cylinder back to its original location. When a trial failed, for instance because the cylinder slipped out of the participant’s hand, it was immediately repeated. This never occurred more than twice in a single participant. Data acquisition During the experiment, hand and head movements were recorded using an Optotrak 3020 system (sampling rate set at 125 Hz). Four Optotrak markers were located on the Eyelink helmet (see below) to monitor position and orientation of the head. Hand movements were recorded by means of a marker placed on the back, each of the participant’s hands near the head of the second metacarpal. For two participants with CP, this position did not guarantee sufficient visibility of the marker on the AH due to individual grasping patterns (excessive arm pronation leading to a thumb-down grasping posture). In these instances, the marker was placed more laterally on the hand. As we were mainly interested in the temporal aspects of the movements kinematics, this adjustment did not affect the measurement in a relevant way. The coordinate frame of the Optotrak system was oriented such that x, y and z axes roughly corresponded to the horizontal, vertical and posterior–anterior dimension with respect to the participant’s position (see Fig. 1). Gaze direction was assessed using a head mounted Eyelink II system (sampling frequency 500 Hz), which was calibrated twice during the experiment, before each of the main blocks that corresponded to the obstacle present/absent conditions. To calibrate the system, the subject was asked to look at 10–15 points covering the frontal working plane. These points were manually indicated by the experimenter by using a stick on the end of which both a fixation target (small red ball with a diameter of 13 mm) and an Optotrak marker were attached. This “interactive” calibration method was chosen because pilot recordings revealed that two participants with CP failed to follow the instructions of the standard calibration procedure (employing a calibration board with LEDs). In addition, the “interactive” method allowed for constant monitoring of the Eyelink system’s output by the experimenter (see “Setup and procedure”). The position of the fixation target and the position of the eye relative to the Eyelink helmet were used to compute gaze direction. Gaze direction was subsequently transformed to a 2D signal by projecting it to the xy-plane. Linear regression between this direction and the raw Eyelink data was used to calibrate the system. Although calibration error (deviation between actual and reconstructed gaze direction for the calibration targets) was relatively large—on average, 5° of visual angle, increasing towards the periphery—it was well within the precision requirements needed to answer our research questions. Particular care was taken in the analysis of eye movements (see “Eye movements”) to minimize the effect of this inaccuracy. Preprocessing The raw Optotrak data were partially interpolated with cubic spline interpolation (up to ten successive samples, corresponding to 80 ms) and low-pass filtered (third order Butterworth filter with cut-off frequency of 10 Hz). Hand velocity was computed by 3-point numerical differentiation. The raw Eyelink signal was transformed to relative gaze direction and integrated with the helmet position and orientation to yield absolute gaze direction. Approximate fixations were computed as the intersection of gaze direction with the plane parallel to the xy-plane (see Fig. 1) and containing the centers of objects and target regions. Only data from one eye (selected by visual inspection for each individual participant) were used for further analysis. Gaze data were partially interpolated (up to four successive samples, corresponding to 8 ms, using cubic spline interpolation) and low-pass filtered (third order Butterworth with cut-off frequency of 20 Hz). Subsequently, gaze velocity, acceleration, and jerk (second time differential of velocity) were computed by repeated 3-point numeric differentiation. Data reduction Hand movements Hand movements were identified using an absolute velocity threshold (0.20 m/s) and a direction criterion (start of a new movement indicated by a reversal of horizontal direction). These criteria provided the algorithms for the semi-automatic custom-written selection routines for the segmentation of the trials. Based on this segmentation, the grasp time (interval between the hand reaching and leaving the object region) and hand movement duration (interval between the hand leaving the object and reaching the target region) were determined. Analysis of eye–hand coordination was confined to the object transport phase that is from onset of the hand movement away from the object region till the end, reaching the target region. The duration of this movement was determined. Eye movements Saccades were detected using a jerk (second time differential of velocity) criterion, which has been shown to be more reliable than methods based purely on gaze velocity or acceleration, especially for head-free eye movement recordings. A saccade was scored whenever jerk exceeded 200.000°/s3, with an additional velocity requirement (50°/s) (see Wyatt 1998, for similar analaysis). Saccade termination was determined as the first moment after saccade onset at which velocity dropped below 50°/s. The total number of saccades registered in the experiment was 4,613. Saccade initiation/termination time was determined when none of the first or last eight samples (16 ms) were missing. Saccade peak velocity was determined, when in addition no more than four samples were missing in a row throughout the entire saccade. According to these criteria, saccade initiation and termination time could not be determined in 5.5% of all saccades, and peak velocity could not be determined in 10.5% of all saccades. To test for potential group differences in basic eye movement kinematics, saccade main sequences were plotted, based on the saccades for which the peak velocity could be determined. Main sequences showed, for both groups, the typical linear relationship between amplitude and duration (e.g., Carpenter 1988), and the log-linear relationship between amplitude and peak velocity for all participants of both groups. No significant group differences were found in gradient or intercept of linear regressions of these relationships (all p > 0.05, Welch’s unpaired t-test). Subsequently, for each trial the “object-leaving” and the “target-reaching” saccade were determined (in many trials these saccades coincided since there was a single gaze shift from object to target). Automatic detection of these saccades was complicated by the fact that precision of gaze data was not sufficient for a procedure based on landmark regions around object and target. Therefore, the following semi-automatic two-step procedure was adopted. First, fixations to object and target were determined automatically, assuming that these occurred before (up to −600 ms) the beginning and at the end of object transport. Fixation periods were defined as intervals of at least 200 ms, during which, the standard deviation of absolute gaze fixation was smaller than 5 mm. Note that this automatic procedure did not use absolute landmark regions but relative position information—this was possible because the object and target were placed at the lateral extremes of the working region of the experiment. For the obstacle, an analogous procedure was not possible. Choices of this first step were inspected trial by trial and manually corrected if necessary. Obstacle fixations were not taken into account since it was not always possible to reliably distinguish them from other, frequently occurring intermediate fixations between object and target. Second, an automatic routine was used to detect the object-leaving saccade, defined as the first saccade, at the end of which, gaze had moved more than 10° horizontally relative to the object fixation. Based on an analogous 10° criterion, the target-reaching saccade was determined at the end of object transport. Thus, small (<10°) corrective saccades at the end of object transport were ignored in the definition of the target-reaching saccade. Eye–hand coordination Visual monitoring during object transport was quantified by the number of intermediate fixations, that is, the number of gaze shifts minus one occurring between the object-leaving and the target-reaching saccade. Note that this number does not include potential small corrective saccades at the end of object transport, since the target-reaching saccade was defined as the last saccade of horizontal amplitude >10° reaching the target region. To examine temporal coordination between eye and hand at the beginning and end of object transport, two variables were computed: the movement onset asynchrony (MOA) and the movement termination asynchrony (MTA). The MOA was defined as the interval between the start of the object transport movement and the onset of the object-leaving saccade, with larger (positive) values corresponding to later gaze departure. Similarly, the MTA was determined by the delay between the termination of the object transport movement and the end of the target-reaching saccade, with smaller (negative) values corresponding to earlier target fixation. To eliminate a potential confounds with hand kinematics (in particular in the comparison of AH and LAH in participants with CP), both MOA and MTA were normalized with respect to hand movement duration (expressed as percentage of the latter). With respect to our experimental hypotheses, we predicted that closer visual monitoring would be reflected in greater gaze-hand proximity (smaller average angular distance), larger number of intermediate gaze fixations, and delayed gaze departure from the object region (that is, a longer MOA). Regarding prospective control, we assumed that less anticipatory gaze patterns would be reflected in delayed departure from the object region (longer MOA) and delayed arrival on target (shorter/less negative MTA). Statistical analysis Data from one control participant were excluded due to insufficient quality of the Eyelink data, leaving data from six experimental and nine control subjects for statistical analysis. For each of the five variables of hand kinematics and eye–hand coordination introduced in “Data reduction”, the means averaged across all replications in each condition were submitted to a three-way repeated measures ANOVA with one between-subject factor (group: CP/control) and two within-subject factors (hand: AH/LAH, NPH/PH; obstacle: present/absent). Subsequent analyses of simple effects were performed using paired t-tests for within-subject factors, and Welch’s unpaired t-test (no equal variance assumed) for the between-subject factor. To correct for multiple (five) ANOVAs, the critical significance level was set to 0.01. Results Of the total 15 × 40 = 600 trials, seven trials were entirely rejected due to insufficient data quality (no more than two trials in any individual participant). Sample trials Figures 2 and 3 show representative trials from participants with CP and control participants. Horizontal components of head and hand position and gaze fixation (see “Data reduction”) are plotted on the vertical axis, as a function of time. The intervals and durations of the (non-normalized) MOA and MTA are indicated by vertical arrows and numerically in the plots. To allow comparisons between the plots, all data shown are from obstacle present trials. Visual inspection indicates qualitatively similar patterns (saccadic gaze shifts to future hand targets) in all conditions, and for both participant groups. Complete visual tracking of the trajectory of the hand was found only in a few trials (detected by visual inspection) in two participants with CP when using their AH. For these trials, MOA and/or MTA could not be determined. Fig. 2Sample trials of two control participants, using their PH (a, c) and NPH (b, c) in trials with the obstacle present. Interpolated data points are represented by dashed lines. Vertical arrows at the top and bottom of the plots specify the intervals defining the MOA and MTA (beginning/end of hand and eye movement), with numerical values of these measures addedFig. 3Sample trials of two participants with CP, using their LAH (a, c) and AH (b, d) in trials with the obstacle present. Note the delayed gaze departure (longer MOA) and increased number of intermediate fixations for the AH compared to the LAH and to both hands of control participants (Fig. 2). Interpolated data points are represented by dashed lines. Vertical arrows at the top and bottom of the plots specify the intervals defining the MOA and MTA (beginning/end of hand and eye movement), with numerical values of these measures added The representative trials shown in Fig. 2 (control participants using their PH and NPH) and Fig. 3 (participants with CP using their LAH and AH) illustrate some general trends when participants with CP use their AH. These are a prolonged MOA, a larger number of intermediate fixations and a greater gaze-hand proximity compared to the other combinations of group and task hand. These observations are in agreement with the hypothesis of increased visual monitoring (see Eye–hand coordination”), and this is corroborated by the statistical analysis presented below. In contrast, no clear pattern with respect to the MTA is obvious from these data plots, nor from the statistical analysis. Dependent variables Values for the different dependent variables are summarized in Table 2. Due to missing samples, some of the measures reported below could not be determined in a number of trials. This number did not exceed 40 out of 600 trials (never more than five out of ten in a single condition and participant) for any of the variables. Table 2Mean and standard deviation (between participants) of dependent variablesDependent variableParticipants with CPControl participantsLAHaAHbPHcNPHdNo obstacleObstacleNo obstacleObstacleNo obstacleObstacleNo obstacleObstacleHand movement duration (s)  Mean0.811.051.051.600.610.83 0.650.87 SEM0.030.050.030.150.010.020.020.03Object grasp time Mean0.220.251.071.160.0630.0710.0860.094 SEM0.030.040.140.160.0090.010.0170.015Number of intermediate fixations Mean0.240.820.76 1.300.06 0.73 0.07 0.61 SEM0.090.340.241.210.040.220.060.26Movement onset asynchrony (% hand movement duration) Mean−6.881.5313.4 14.62−4.45 0.02−3.143.84  SEM5.814.964.372.124.844.214.844.37Movement termination asynchrony (% hand movement duration) Mean−76.8 −61.2 −49.5 −51.4 −75.7 −65.6 −71.5 −62.2  SEM5.396.872.843.644.393.784.643.56aLess affected handbAffected handcPreferred handdNon-preferred hand Grasp time showed main effects of group, hand as well as a group–hand interaction (ANOVA, all p-values < 0.0001). Both, with the AH and the LAH participants with CP showed a longer grasp time than controls using either hand. The simple effect of hand was only present in participants with CP (longer grasp time with the AH compared to the LAH). Hand movement durations for object transport showed main effects of group, hand, and obstacle, as well as a group–hand interaction (ANOVA, all p-values < 0.0001). Simple effect analysis revealed that the effect of hand was present in both groups, but more pronounced in participants with CP. Moreover, the movement duration was longer in participants with CP and in the presence of an obstacle (Fig. 4). Fig. 4Mean ± SEM of the movement onset asynchrony (MOA) as a function of participant group, task hand and obstacle presence. MOA was normalized with respect to hand movement duration Regarding the number of intermediate fixations during object transport, a main effect of obstacle [F(1,13) = 15.06, p < 0.002] was found, as well as a group–hand interaction [F(1,13) = 12.36, p < 0.005]. Analysis of simple effects showed that the number of intermediate fixations was larger in the obstacle present condition than in the obstacle absent condition. With respect to the group–hand interaction, significant differences were found between participants with CP using their AH and control participants using either hand (higher number of intermediate fixations for CP using AH), but not between the AH versus LAH of participants with CP. Temporal eye–hand coordination at the beginning of object transport, as quantified by MOA (see Fig. 4), showed main effects of hand [F(1,13) = 17.25, p < 0.002] and a group–hand interaction [F(1,13) = 12.55, p < 0.005]. The simple effect of hand was present in participants with CP, with longer MOA in the AH condition, but not in control participants. Moreover, there was a significant difference (prolonged MOA) between participants with CP using their AH and control participants using either hand. With respect to MTA, a main effect of hand [F(1,13) = 25.68, p < 0.0005], a group–hand interaction [F(1,13) = 14.189, p < 0.005] were found, as well as a marginally significant group–hand–obstacle interaction [F(1,13) = 8.19, p = 0.013]. The simple effect of hand was present only in participants with CP, with a shorter (less negative) MTA when using the AH. Also the difference between participants with CP using the AH and control participants using either hand was significant (shorter MTA in participants with CP using the AH). Discussion The main question we pursued in the present study was whether and in which way individuals with hemiparetic CP adapt eye–hand coordination to their sensorimotor impairments, in particular, when actions are performed with the affected hand (AH). We hypothesized that participants with CP would more closely monitor actions performed with their AH, compared to actions performed with their less affected hand (LAH) and compared to neurologically healthy control participants using either hand. Qualitatively, eye–hand coordination patterns were similar among control participants and participants with CP, regardless of the hand used to perform the task. That is, anticipatory saccadic gaze shifts were used to fixate future “action sites”—such as object, target, or intermediate fixations—in advance. Notwithstanding these qualitative resemblances, a more fine-grained analysis of temporal and spatial aspects of eye–hand coordination did confirm our hypothesis of increased visual monitoring when moving with the AH. Depending on the measure employed, this effect was present when comparing performance with the AH to control participants using either hand (increased number of intermediate fixations), or in addition for comparing performance with the AH to performance with the LAH in participants with CP (delayed gaze departure from object and hand, i.e., a longer MOA). It is important to note that our main result regarding visual monitoring is an interaction effect of group and task hand (effect of task hand was present only in the experimental group). Therefore, it cannot be explained by potential group differences in oculomotor control or gaze data quality. Moreover, despite previous findings that did not find a strong relationship between gaze-exit time relative to hand action and task speed (Flanagan and Johansson 2003), we decided to normalize our measures of temporal eye–hand coordination in order to exclude any potential confounding effects of hand kinematics. This normalization may be considered rather conservative. Without normalizing, the effects of task and participant group on MOA may have even been stronger. A second, and more speculative hypothesis of the present study was that in general, gaze patterns might be less anticipatory in participants with CP, contributing to deficits in prospective control and action planning that have been reported in this population (Steenbergen et al. 2000a, 2004; te Velde et al. 2005), in particular, individual with right hemiparesis. The present study found no evidence in support of this hypothesis. Although participants with CP did show a prolonged MOA, that is, a delayed gaze departure that may indicate less anticipation, this effect was restricted to the AH. No significant differences in the MOA were found between actions of participants with CP using their LAH and control participants using their PH. Similarly, for our measure of temporal eye–hand coordination at the end of object transport (the movement termination asynchrony, MTA), a delayed gaze arrival on target was only present in participants with CP using their AH, not as a general group effect. In addition, preliminary analyses had shown no difference between individuals with left and right hemiparesis with respect to any of our dependent variables. Thus our results suggest that deficits in prospective action control found in CP are not directly related to atypical eye–hand coordination. However, this statement is made with caution as we only tested a small number of participants (n = 6 in total, n = 3 for each left and right hemiparesis). An unexpected finding of the present study was that gaze frequently departed from the object region only after the hand movement had started, in particular, in the presence of the obstacle. This was also the case in control participants and is in contrast to the results of Johansson et al. (2001), who reported that in the majority of the trials gaze departure from the grasp site occurred even before the fingertips had reached the object. These contrasting findings may in part be explained by differences in the experimental set-up. In the present study the distance between initial object position and the target was much larger compared to the one used by Johansson et al. (2001), leading to longer hand movement durations. This potential confounding effect of hand movement duration needs to be addressed in future research, for instance, by systematically manipulating hand movement distance in order to disentangle components of online and prospective control in the MOA. Also, task complexity might play a role. In a block-stacking task, Flanagan and Johansson (2003) found gaze to depart on average just before hand movement onset, which is more consistent with our present findings. In sum, we found evidence in support the hypothesis of increased visual monitoring of manual actions of individuals with hemiparetic CP performing actions with their AH, both at the beginning and during object transport. This may reflect a strategy of visual compensation for sensorimotor deficits (see also, Mulder et al. 2001). However, since the functional significance of the observed gaze patterns was not addressed in the present study, the question whether and by which neurophysiological mechanisms vision enhances manual performance requires further research. Generalizability of our results is certainly limited by the fact that our experimental group was relatively small (n = 6) and highly selective, considering the high variability in the CP population regarding intellectual and motor abilities. In particular, the null-findings on anticipatory control and the comparison of individuals with left and right hemiparesis should be considered preliminary and require further investigation. On the other hand, the fact that we did find significant effects with respect to online visual control (number of intermediate fixations and MOA) despite the small number of participants suggests that the observed effects are robust.

Purinergic_Signal-2-2-2096640

E-NTPDases in human airways: Regulation and relevance for chronic lung diseases

Chronic obstructive lung diseases are characterized by the inability to prevent bacterial infection and a gradual loss of lung function caused by recurrent inflammatory responses. In the past decade, numerous studies have demonstrated the importance of nucleotide-mediated bacterial clearance. Their interaction with P2 receptors on airway epithelia provides a rapid ‘on-and-off’ signal stimulating mucus secretion, cilia beating activity and surface hydration. On the other hand, abnormally high ATP levels resulting from damaged epithelia and bacterial lysis may cause lung edema and exacerbate inflammatory responses. Airway ATP concentrations are regulated by ecto nucleoside triphosphate diphosphohydrolases (E-NTPDases) which are expressed on the mucosal surface and catalyze the sequential dephosphorylation of nucleoside triphosphates to nucleoside monophosphates (ATP → ADP → AMP). The common bacterial product, Pseudomonas aeruginosa lipopolysaccharide (LPS), induces an acute reduction in azide-sensitive E-NTPDase activities, followed by a sustained increase in activity as well as NTPDase 1 and NTPDase 3 expression. Accordingly, chronic lung diseases, including cystic fibrosis (CF) and primary ciliary dyskinesia, are characterized by higher rates of nucleotide elimination, azide-sensitive E-NTPDase activities and expression. This review integrates the biphasic regulation of airway E-NTPDases with the function of purine signaling in lung diseases. During acute insults, a transient reduction in E-NTPDase activities may be beneficial to stimulate ATP-mediated bacterial clearance. In chronic lung diseases, elevating E-NTPDase activities may represent an attempt to prevent P2 receptor desensitization and nucleotide-mediated lung damage. Introduction Airway epithelia constitute an essential protective barrier against lung infection, coordinating luminal and interstitial responses to inhaled pathogens through signals (growth factors, cytokines and nucleotides) provided by epithelial, inflammatory and immune cells [1]. It is now widely accepted that extracellular nucleotides provide an elaborated cell communication system in mammalian tissues [2, 3] including the airways [4, 5]. Each signaling event constitutes a brief ‘on-and-off’ switch mechanism allowing the target cells to perceive the subsequent signal. The major source of extracellular nucleotides in normal airways is the epithelium, releasing ATP under resting conditions and in response to mechanical or osmotic stress [6]. While basal ATP concentrations maintained under resting conditions are insufficient to activate surface receptors, the levels reached near the site of stimulated release initiate a variety of P2 receptor-mediated responses [7]. Two P2 receptor families have been identified: Fast-acting (ionotropic) P2X ligand-gated cation channels and slow-acting (metabotropic) G protein-coupled P2Y receptors [8]. Although ATP activates members of both families, P2X receptors generally respond to higher concentrations (EC50 = 1–10 µM) than P2Y receptors (EC50 = 0.1–1.0 µM). Each signal is promptly terminated by surface conversion of ATP to adenosine [9, 10] and dispersal into the interstitial fluid. Adenosine also initiates cellular responses through P1 (A1, A2A, A2B or A3) surface receptors [11]. In the airways, nucleotide- and adenosine-mediated communications are involved in wound healing [4], bacterial clearance [12] and inflammatory responses [1, 3, 13, 14]. Cell surface nucleotide concentrations are regulated by three families of ectonucleotidases in mammalian tissues: Ecto-nucleotide pyrophosphatase/phosphodiesterases (E-NPPs: ATP → AMP), alkaline phosphatases (APs: ATP → AMP → AMP → adenosine) and ecto-nucleoside triphosphate diphosphohydrolases (E-NTPDases: ATP → AMP → AMP) [15]. This review describes the importance of regulating extracellular nucleotide concentrations for airway homeostasis and the impact of chronic lung diseases on E-NTPDases. Few studies have addressed the activity or expression of airway E-NTPDases. In this review, the available information is complemented by original data. Nucleotide-mediated bacterial clearance The importance of purine signaling in the airways is well illustrated in nucleotide-mediated mucociliary clearance (MCC), which constitutes the first line of defense against bacterial infection [16]. Mucosal epithelial surfaces are protected by an airway surface liquid (ASL) layer containing a thin mucin sheath kept above the cilia by a periciliary liquid (PCL) layer (Fig. 1). Inhaled bacteria adhere to the mucin and are continuously displaced upward by coordinated cilia beating activity [17, 18]. Comparative analysis of the bioelectric properties of airway epithelia from normal donors and CF patients revealed that PCL height is regulated by a delicate balance between the activities of the epithelial sodium channel (ENAC) and the cystic fibrosis transmembrane regulator (CFTR) [19]. Cystic fibrosis is characterized by mutations in the gene encoding CFTR [16]. The lack of ion secretion through CFTR is associated with Na+ hyperabsorption by ENAC and water influx, resulting in the depletion of the PCL layer essential for cilia beating activity. The cilia become collapsed under a thick layer of mucus filled with bacteria and leukocytes, leading to severe inflammatory responses and progressive loss of lung function. Fig. 1Impact of chronic obstructive lung diseases on mucociliary clearance. The polarized epithelium is composed of columnar (ciliated or mucin-secreting) cells facing the lumen and basal cells facing the serosal compartment. Bacterial clearance in healthy lungs is maintained by coordinated cilia beating activity within the PCL layer, moving upward the overlying mucus and pathogens. In chronic obstructive lung diseases, cilia are collapsed under a thick layer of mucus containing bacteria and leukocytes. Extracellular nucleotides and nucleosides regulate CFTR and ENAC [20]. The basal activity of CFTR is maintained by adenosine (A2B) receptors, which induce G protein-coupled adenylate cyclase, followed by cAMP-dependent activation of type II protein kinase A [5]. In contrast, ATP stimulates Cl− secretion through Ca2+-activated Cl− channels. This signaling pathway involves P2Y2 receptors, leading to G protein-coupled phospholipase C activation and cytosolic Ca2+ mobilization. Water efflux is also stimulated by P2Y2 receptor-mediated inhibition of Na+ absorption by ENAC using Ca2+-dependent signaling pathways [21]. Members of the P2X receptor family also support luminal Cl−/water efflux in human airway epithelia, as demonstrated in primary cultures and cell lines (Beas2B, 16HBE14o−, Calu-3, (ΣCFTE-29o−) by Ussing chamber and patch-clamp experiments [22]. The receptors identified in airway epithelia by RT-PCR were P2X2, P2X4 and P2X5. These non-selective cation channels allow Ca2+ influx, which then stimulates Ca2+-activated Cl− channels. Interestingly, P2X7 receptors are only detected in CF airway epithelia, primarily in nasal polyps [22]. Because they are known to induce apoptosis in response to high ATP concentrations [23], P2X7 receptors may contribute to disease-associated epithelial damage. The central role of ASL nucleotides for MCC was further substantiated by the observation that P2Y2 receptor activation constitutes the most potent stimulus for cilia beating activity [24–26] and mucin secretion [27, 28]. In cultured CF airway epithelia, P2Y2 receptor activation restores normal PCL height and rotational mucus transport [29]. Based on these findings, aerosolized nucleotides were proposed for the treatment of chronic obstructive lung diseases [16]. However, clinical studies showed that patients with asthma or chronic obstructive pulmonary disease exposed to aerosolized metabolites of ATP (AMP and adenosine) experience coughing and bronchoconstriction [30, 31]. Furthermore, aerosolized ATP (not adenosine) induced similar reactions from healthy subjects [32, 33]. In developing nucleotide-mediated MCC therapies, the equally potent P2Y2 receptor agonist, UTP, was evaluated [6]. Airway clearance in sheep, measured with nebulized technetium-labeled albumin, was transiently increased by UTP [34]. Aerosolized UTP also improved MCC in patients with mild chronic bronchitis [35], CF [36, 37] and primary ciliary dyskinesia [38]. A major obstacle to the treatment of chronic lung diseases with aerosolized UTP is the rapid disappearance of therapeutic concentrations (0.1 mM) from the ASL layer (<5 min) [10, 29]. Incidentally, the metabolically stable UTP analogs INS365 (diquafosol) and INS37217 (denufosol) developed by Inspire Pharmaceuticals Inc. [39] were reported to improve cough induced-sputum clearance in CF patients [40, 41]. However, long-term treatments with these drugs may require dose optimization to minimize P2Y2 receptor desensitization [42]. P2 receptors and airway inflammation In vivo studies conducted on the injurious effects of mechanical ventilation demonstrated that endogenous ATP may reach ASL concentrations sufficiently high to trigger inflammatory responses and lung damage in the absence of infection [43]. The bronchoalveolar lavage fluid of rats subjected to positive-pressure mechanical ventilation contains significantly higher protein, ATP and cytokine (IL-6, TNFα) levels than control animals. Instillation of an equivalent ATP concentration increased lung fluid volume, supporting the existence of P2 and/or P1 receptor-mediated lung edema [43]. Analysis of bronchoalveolar lavage fluid nucleotide content by etheno-derivatization revealed an increase in AMP + adenosine/ADP + ATP ratio [44], suggesting that mechanical ventilation up-regulates airway ectonucleotidases. Real-time PCR on total lung tissue indicated that mechanical ventilation increases A2B, but decreases P2X7, receptor expression. Rats subjected to both mechanical ventilation and positive end-expiratory pressure exhibited normal bronchoalveolar lavage composition and receptor expression. These findings suggest that patients subjected to positive end-expiratory pressure during large-volume ventilation may avoid ATP-mediated lung injuries. On the other hand, the fact that adenosine levels and A2B receptor expression were raised, whereas ATP levels and P2X7 receptor expression were decreased, by mechanical ventilation supports a dynamic role for ectonucleotidases in the regulation of purine signaling in the airways. The role of P2 receptors in innate defense against bacterial infection was investigated using mice deficient in P2Y1, P2Y2 or both receptors [45]. All these mice exhibited lower survival and lower cytokine levels in lung homogenates than wild-type animals measured 24 h after intranasal instillation of P. aeruginosa. These results suggest a protective role for ATP-induced inflammation [45]. In human airways, P2Y2 receptor activation amplified inflammatory responses to bacterial infection. Primary cultures of bronchial epithelial cells exposed to UTP and sputum from CF patients exhibited an eight-fold increase in IL-8 secretion, compared to two-fold with UTP alone [46]. Furthermore, P2X7 receptors are known to support several processes relevant to inflammation, including LPS neutralization [47], pathogen killing, nitric oxide production [1] and cytokine release from mast cells, leukocytes, dendritic cells and neurons [3, 14]. In human blood, LPS-induced IL-1β release is increased 5-fold by P2X7 receptor activation [48]. Altogether, these studies entail that P2 receptor activation may not only induce innate immunity but also amplify pathogen-induced inflammatory responses in chronic infectious lung diseases. Too much of a good thing.. Excessive inflammatory response to recurrent bacterial infection is a major component in the pathogenesis of chronic lung diseases, irreversibly damaging the airways and leading to bronchiectasis and respiratory failure [49–51]. With respect to purine signaling, nucleotide-mediated cell communications may be overwhelmed by massive ATP release from cell lysis of bacteria and damaged epithelia, as well as stimulated leukocytes and epithelia [6]. The sensitivity of certain P2 receptors (P2Y2 [42], P2X1 and P2X3 [52]) to agonist-induced desensitization may reduce the efficiency of nucleotide-mediated MCC [53]. On the other hand, chronically elevated ATP may recruit signaling pathways normally dormant in healthy lungs. For instance, P2X7 receptor activation only induces cytokine release and cell death in response to high ATP concentrations encountered in damaged tissues [23]. On astrocytes, LPS-induced TNFα secretion is increased by P2Y receptor activation at low micromolar ATP, but reduced by P2X7 receptor activation at high ATP concentrations [54]. The authors propose a beneficial effect for the potentiation of TNFα secretion by P2Y receptors during mild or acute inflammation and a protective role for P2X7 receptors in damaged tissues or chronic infectious diseases. These studies suggest that P2X7 receptor-mediated responses may be tissue-specific and/or influenced by bacterial infections. In the respiratory system, P2Y2 receptors are expressed throughout airway epithelia [39] while P2X7 receptors have only been reported in nasal polyps of CF patients [22]. The impact of P2X and P2Y receptor activation on inflammatory responses mediated by airway epithelia under normal and pathological conditions remains to be investigated. E-NTPDases regulate airway ATP Several studies demonstrate the presence of ectonucleotidase activities on human airway epithelia [9, 10, 55–58]. Time-course experiments conducted on polarized primary cultures of human bronchial epithelial cells showed that mucosal ATP is dephosphorylated into ADP, AMP and adenosine (Fig. 2a). Experiments repeated with nasal, bronchial and bronchiolar epithelial cultures indicated that ATP metabolism accelerates toward the alveolar region (Fig. 2b). Such gradient distribution was also reported for non-specific alkaline phosphatase (NS AP) in human airways [9], an ectonucleotidase capable of sequentially dephosphorylating ATP to adenosine [59]. However, complete inhibition of NS AP with 10 mM levamisole [9] only reduced by 25% total ectoATPase activities [10]. Other ATP-regulating ectonucleotidases were identified in mammalian tissues as members of the ENTPDase family [15, 60]. They catalyze the hydrolysis of γ- and/or β-phosphate residues, resulting in the formation of nucleoside monophosphates. Their activities are Ca2+ and Mg2+-dependent and insensitive to inhibitors of P-, F- and V-type ATPases or APs [61]. Four E-NTPDases are localized to cell surfaces: NTPDase 1, 2, 3 and 8. Functional studies showed that NTPDase 1 [62] and NTPDase 3 [63] are inhibited by azide, whereas NTPDase 2 [64] and NTPDase 8 [65] are insensitive. The mRNA expression of NTPDase 1, 2 and 3 in human airways was reported in total lung RNA [66] and cultured bronchial epithelial cells [67]. We now provide evidence that NTPDase 1 and NTPDase 3 are both expressed throughout human airways by real-time PCR on freshly excised epithelial cells (Fig. 3). Their expression distribution followed surface ectoATPase activities (Fig. 2b), exhibiting higher mRNA levels in smaller airways (Fig. 3). The fact that 20 mM azide reduced total ectoATPase activities by 45% on human bronchial epithelial cells [67] suggests that the high affinity (Km < 15 µM) [68] NTPDases 1 and/or 3 play major roles in the regulation of physiological ASL nucleotide concentrations. Fig. 2Ectonucleotidases regulate nucleotide concentrations on airway epithelial surfaces. a Polarized primary cultures of human bronchial epithelial cells were assayed with 100 µM ATP added to the mucosal surface, as we previously described [10]. Buffer sample analysis by high-pressure liquid chromatography (10) shows that exogenous ATP (•) is dephosphorylated into ADP (▪), AMP (≆) and adenosine (○). b Similar experiments repeated on primary cultures of human nasal, bronchial and bronchiolar epithelial cells show that the elimination rate of ATP increases toward alveoli (N = 3, *, P < 0.05; Mann-Whitney test).Fig. 3Expression level of NTPDase 1 and NTPDase 3 along human airways. Total RNA from excised epithelia was analyzed by real-time PCR using SYBR green assays and normalized to the expression level of the house-keeping gene, 18S, as we previously described [89]. The mRNA level of E-NTPDase 1 (□) and NTPDase 3 (▪) increases toward alveoli (N = 4, *, P < 0.05; Mann-Whitney test). The physiological importance of azide-sensitive E-NTPDases for nucleotide-mediated MCC was recently established using an in vitro model of rhythmic breathing. Airway epithelia are continuously subjected to mechanical stress generated by breathing, coughing or chest movement. Since mechanical stress induces epithelial ATP release [6], static culture conditions may underestimate the lungs capacity to regulate MCC in vivo. Rhythmic pressure changes mimicking normal tidal breathing were reproduced in vitro by a system applying cyclic compressive stress (CCS) to the mucosal surface of primary bronchial epithelial cultures [69]. Whereas CF cultures under static conditions exhibit a depleted PCL layer and mucostasis, CCS mimicking normal tidal breathing (20 cmH2O; 15 cycles/min) restored normal PCL height and mucus transport through ATP release and P2Y2 receptor activation. These results also suggest that purine signaling may provide an explanation for the beneficial effects of oscillatory therapeutic devices clinically used to stimulate sputum clearance [70]. Cyclic compressive stress also enhances MCC through a reduction in ASL nucleotide metabolism. We recently demonstrated that CCS decreases the rate of ATP hydrolysis on the mucosal surface of normal and CF bronchial epithelial cultures [67]. More importantly, CCS restored normal ectoATPase activities on CF epithelial surfaces. The inhibitory effect of CCS on ATP metabolism was abrogated by 20 mM azide, indicating that E-NTPDase 1 and/or E-NTPDase 3 are key components in the regulation P2 receptormediated MCC. Perhaps the ability of CF airway epithelia to provide adequate nucleotide-mediated MCC under conditions mimicking normal breathing may contribute to the relatively healthy state of young patients before the establishment of chronic infection [71]. Chronic lung diseases shorten the signals Recent studies demonstrate that chronic lung diseases accelerate all metabolic steps supporting the mucosal conversion of ATP to adenosine [10, 72]. We identified the ectonucleotidases involved using primary bronchial epithelial cultures from healthy donors and patients with primary ciliary dyskinesia, CF or α1-antitrypsin deficiency. All airway diseases were characterized by a two- to four-fold increase in the elimination rate of ATP on the mucosal surface, without significant effect on the serosal surface (Fig. 4). Selective inhibitors of NS AP (10 mM levamisole), NTPDase 1 and NTPDase 3 (20 mM azide) partially restored normal activity levels. However, the lower substrate affinities of airway NS AP (36 µM and 717 µM) [9] suggest that NTPDase 1 and/or NTPDase 3 are responsible for the accelerated clearance of P2 receptor agonists in diseased airways. Fig. 4Chronic lung diseases enhance azide-sensitive E-NTPDases. Primary bronchial epithelial cultures from healthy donors (N) and patients diagnosed with primary ciliary dyskinesia (PCD), CF or α1-antitrypsin deficiency (α AT) were assayed with bilateral 1 mM ATP in the absence (▪) or presence (□) of 20 mM azide, as we previously described [10]. Analysis of buffer samples by high-pressure liquid chromatography showed that azide-sensitive E-NTPDases are concentrated on the mucosal surface under normal and pathological conditions (N = 4, *, P < 0.05; Mann-Whitney test). Pseudomonas aeruginosa is among the most frequently isolated pathogens from the airways of patients with gram-negative infections [51]. In patients with damaged airways from mechanical ventilation, trauma or viral infection, airway colonization by P. aeruginosa is often followed by pneumonia, sepsis and death. Much of the airway inflammation induced by these organisms is caused by released bacterial products adhering to the epithelial surface, such as LPS. The contribution of bacterial infection to the deregulation of E-NTPDase activities in lung diseases was examined using P. aeruginosa LPS. Primary cultures of human bronchial epithelial cells exposed 24 h on the mucosal surface to an optimum LPS concentration (100 ng/ml) [73] exhibited a biphasic response over time (Fig. 5A). Azide-sensitive E-NTPDases were transiently reduced over 8 h, without affecting mRNA expression. In contrast, prolonged exposures (24 h) raised the activities two-fold above control levels and increased NTPDase 1 and NTPDase 3 expression by 4 and 10-fold (Fig. 5B). These opposite effects suggest that two regulatory mechanisms may be involved in the acute and chronic effects of LPS on airway E-NTPDases. Fig. 5Impact of P. aeruginisa LPS on NTPDase 1 and NTPDase 3. A The mucosal surface of primary bronchial epithelial cultures were exposed 0, 1, 8 or 24 h to 100 ng/ml LPS, then assayed for surface activity by high-pressure liquid chromatography, as we previously described [10]. The activities of azidesensitive mucosal E-NTPDases measured with 0.03 mM ATP were transiently reduced by LPS. B Quantification of their mRNA expression by real-time PCR using the house-keeping gene 18S, as we previously described (89). LPS induced a delayed increase in the expression level of both E-NTPDases (N = 5, *, P < 0.05; Mann-Whitney test). Oxidative stress is one of the major causes of epithelial injury in chronic lung diseases [74]. Reactive oxygen and nitrogen species released from airway epithelia or activated leukocytes damage cell membranes by peroxidation of lipids, amino acids and carbohydrates. P aeruginosa LPS was reported to induce their release from airway epithelia [75]. Incidentally, intrahepatic cholestasis induced by intraperitoneal injection of LPS reduced total liver NTPDase 1 activity within 2 h [76]. In a rat model of glomerulopathy, short-term exposure (1 h) to LPS inactivated endothelial NTPDase 1, which was prevented by pre-treatment with the anti-oxidant superoxide dismutase [77]. Oxidative stress induced by membrane depolarization [78] or ischemia/ reperfusion [79] also inhibited endothelial NTPDase 1. Furthermore, short exposures (4 h) to the pro-inflammatory cytokine, TNFα, inhibited endothelial NTPDase 1, which was mimicked by hydrogen peroxide but prevented by superoxide dismutase [80]. Altogether, these studies suggest that oxidative stress may be responsible for the early and transient reduction in azide-sensitive E-NTPDase activities detected in LPS-treated airway epithelia. Since the mRNA expression of E-NTPDases is unaffected by acute oxidative stress [80] or LPS (Fig. 5B), and the inactivation of NTPDase 1 by ischemia/reperfusion was prevented by inhibitors of lipid peroxidation [79], the early loss of surface activity may involve local peroxidation of membrane lipids. The long-term effects of LPS on the azide-sensitive E-NTPDases have not been documented in mammalian tissues. On the other hand, a 24 h exposure to endotoxin stimulated NTPDase 2 on astrocytes [81], as well as NS AP on endothelial, mesengial [82, 83] and bronchial epithelial [73] cells. Additionally, the mRNA expression of NTPDase 1 in rat forebrain was up-regulated seven days after transient ischemia [84]. P. aeruginosa LPS [85, 86] is well-known to influence gene expression through the activation of the transcription factors: Nuclear factor-kappa B and activator protein-1 [87]. Recent studies have shown that CFTR acts as a pattern recognition molecule for P. aeruginosa LPS [88]. Endocytosis of the complex triggers translocation of these transcription factors to the nucleus, which initiates innate immunity through the expression of numerous mediators recruiting and activating inflammatory cells. These signaling pathways could be involved in the up-regulation of NTPDase 1 and NTPDase 3 expression resulting from long-term exposures to P. aeruginosa LPS in human airways. Adaptation of purine signaling to airway diseases In past decade, the therapeutic potential of aerosolized nucleotides for the treatment of chronic obstructive lung diseases has motivated numerous studies on the importance of purine signaling in the airways. This elaborated cell communication system is involved in major defense mechanisms including bacterial clearance and inflammatory responses to bacterial infection. Each signaling event constitutes a brief ‘on-and-off’ switch mechanism allowing the target cells to perceive the subsequent signal. Environmental changes are communicated to target cells in the form of released ATP, which binds surface receptors and triggers an appropriate response. The efficiency of this communication system depends on the cells capacity to rapidly eliminate these local bursts in extracellular ATP concentrations to prevent receptor desensitization. This review establishes the pivotal role of E-NTPDases in the regulation of airway defenses mediated by purine signaling, namely bacterial clearance (Fig. 6). Cyclic compressive stress, mimicking normal breathing or clinical oscillatory devices, stimulates P2 receptormediated MCC through ATP release and reduced azide-sensitive E-NTPDase activities. We also presented original data indicating that NTPDase 1 and 3 activities and expressions are up-regulated in chronic lung diseases by inflammatory mediators, including endotoxin. Interestingly, P. aeruginosa LPS generated a biphasic modulation of airway E-NTPDases which may reconcile previous conflicting reports based on exposure duration. Acute LPS exposures reduced ENTPDase activities, as observed during CCS, most likely to clear a pathogen through P2 receptor-mediated MCC. The signaling pathways activated by P. aeruginosa LPS and the lack of concomitant reduction in ENTPDase mRNA levels support a local inhibitory mechanism that may involve oxidative stress-mediated membrane damage. On the other hand, prolonged endotoxin exposures mimicked the impact of chronic lung diseases on NTPDases 1 and 3, characterized by an up-regulation of both activities and mRNA expression. In chronically infected lungs, an increase in E-NTPDase activities may represent an attempt to prevent lung damage, excessive inflammation and P2 receptor desensitization by chronically elevated ATP concentrations. Fig. 6Purine signaling and nucleotide salvage pathways on human airway epithelia. In normal lungs, basal adenosine levels maintain adequate PCL height for mucus transport through A2B receptor-mediated CFTR activity. Mechanical stimulation of the epithelial surface by an irritant, a pathogen, mechanical ventilation or cyclic compressive stress induces ATP release and elevates ASL concentrations above activation threshold level for P2Y2 receptor activation. This autocrine signal transiently enhances basal MCC through cilia beating, mucin secretion and ion/water efflux into the lumen. The signal is rapidly terminated by NTPDase 1 and/or NTPDase 3 dephosphorylating excess ATP to AMP. Ecto 5′-nucleotidase (ecto 5′-NT; CD73) produces ASL adenosine from the resulting AMP, and excess adenosine is transported back to the cytosol. Chronic lung diseases burdened by considerable tissue damage recruits additional receptors (P2X4, 5, 7) responding to higher ATP cocnentrations. The information provided in this review also exposes the complexity of the signaling interactions taking place between different classes of mediators, including bacterial products, cytokines and extracellular nucleotides, as combinations of mediators may induce different responses than when tested individually. Furthermore, communication networks in the airways may evolve with time during the establishment of a chronic infection, as new target cells (bacteria, leukocytes, lymphocytes) agglomerate on either side of the epithelial barrier. It is proposed that nucleotide-mediated airway functions defined in normal tissues or aseptic culture conditions may require reassessment under conditions mimicking chronic lung diseases before their therapeutic potential may be clearly established.

Hum_Reprod-1-1-2387222

Myeloid ecotropic viral integration site 1 (MEIS) 1 involvement in embryonic implantation

BACKGROUND The HOXA10 homeobox gene controls embryonic uterine development and adult endometrial receptivity. The three-amino-acid loop extension (TALE) family homeobox genes like myeloid ecotropic viral integration site 1 (MEIS) provide enhanced target gene activation and specificity in HOX-regulated cellular processes by acting as HOX cofactors. Introduction The molecular mechanisms responsible for endometrial development and receptivity to embryonic implantation are poorly understood (Cross et al., 1994; Wang and Dey, 2006). In the last few years, therefore, a large number of high-throughput expression profiling studies have been performed on both human and animal model tissues to identify the molecular signaling involved in the endometrial development and embryo-endometrium interactions leading to successful embryonic implantation. These include DNA-microarray studies (reviewed in Horcajadas et al., 2007; see also Chen et al., 2006; Pan et al., 2006), as well as SAGE (serial analysis of gene expression) studies (Blomberg et al., 2005; Ma et al., 2006). In addition, two databases have been set up that contain expression data on genes thought to be involved in endometrium development and function: www.endometrialdatabase.com and endometrium.bcm.tmc.edu. HOX-class homeobox genes are leading candidates for the regulation of endometrium differentiation in preparation for embryonic implantation (Taylor, 2000a; Eun Kwon and Taylor, 2004). HOX genes assign the correct identity to undifferentiated body segments along several developmental axes (Krumlauf, 1994; Carroll, 1995; Moens and Selleri, 2006; Wellik, 2007; Zakany and Duboule, 2007). They are, however, not only typically expressed during embryogenesis but also persistently active in both the mouse and human female reproductive tract (Taylor et al., 1997; Taylor, 2000b). The continued expression of these HOX genes allows the reproductive tract to maintain developmental plasticity and to differentiate appropriately during each menstrual cycle and throughout pregnancy. The development of the para-mesonephric duct in the embryo is dictated in part by the abdominal-B group of HOX genes (Taylor et al., 1997, 1998; Taylor 2000b; Eun Kwon and Taylor, 2004). Specifically, HOXA10 is responsible for proper uterine development in both mouse and man (Taylor et al., 1997; Block et al., 2000). Female Hoxa10 null-mutant mice demonstrate uterine factor infertility (Satokata et al., 1995). These mice ovulate normally and produce normal embryos, but their uteri will not support the implantation of their own or transplanted wild-type embryos. In humans, HOXA10 expression is up-regulated in the mid-luteal phase at the time of implantation, and its expression is driven by estrogen and progesterone (Taylor et al., 1998). HOXA10 is a well-characterized marker of endometrial receptivity, which exerts pleiotropic effects on multiple aspects of adult endometrial development such as stromal decidualization, leukocyte infiltration and pinopode development (Bagot et al., 2001; Daftary and Taylor, 2004; Qian et al., 2005). Pinopodes (also called uterodomes) are the best-characterized structural markers of endometrial receptivity in rodents and humans (Nikas, 2000; Nardo et al., 2002; Nikas and Makrigiannakis, 2003). Defective endometrial HOXA10 expression has been described in several conditions that result in diminished human embryo implantation such as endometriosis, polycystic ovarian syndrome, and hydrosalpinx fluid (Taylor et al., 1999; Daftary and Taylor 2002; Cermik et al., 2003). One of the known downstream target genes of Hoxa10 in endometrial cells is Itgb3 (Daftary et al., 2002). HOX genes perform their function by acting as transcription factors. They bind to the regulatory regions of downstream target genes through their homeobox domain, and thereby activate or repress transcription. HOX proteins bind only weakly to DNA by themselves and/or exhibit a high degree of redundancy in binding site specificity (Pellerin et al., 1994). It has become increasingly evident that HOX proteins usually act as a part of hetero-dimeric (van Dijk et al., 1995; Kurant et al., 1998) or hetero-trimeric complexes (Jacobs et al., 1999; Shen et al., 1999; Schnabel et al., 2000). The TALE (three-amino-acid loop extension) family of homeodomain proteins represents the most important group of HOX cofactors. The unique TALE motif within their homeodomain allows TALE proteins to interact with other homeobox-containing proteins (Bürglin, 1997; Piper et al., 1999). Unlike most other homeobox gene families, TALE genes share sequence homology outside of the homeobox domain. This extensive homology appears linked to their function. Single TALE proteins have only low DNA-binding activity and specificity, but once complexed with a TALE protein from another subfamily and a HOX-class homeobox protein via these homologous domains, they show powerful and specific downstream target promoter regulation (reviewed in Geerts et al., 2003, 2005). Abd-B-like HOX transcription factors, like HOXA10, can form dimers or trimers with TALE proteins that thereby regulate the multiple HOX downstream target genes necessary for the implantation process. The myeloid ecotropic viral integration site 1 (MEIS) genes belong to the TALE homeobox family. MEIS proteins are not just involved in TALE complex DNA binding and complex stabilization on the target DNA, but can also recruit other TALE partner proteins into the nucleus (Rieckhof et al., 1997). Meis1 was first identified as a major integration site for leukemogenic virus in a murine leukemia model (Moskow et al., 1995). Also in human leukemia, high expression of MEIS1 was found in bone marrow cells of acute myelogenous leukemia (AML) patients, where it was always co-expressed with HOXA9 (Smith et al., 1997; Lawrence et al., 1999). HOXA9 expression was found to cause transformation of normal hematopoietic cells to AML cells only when expressed together with MEIS1 (Kroon et al., 1998). Specifically, MEIS1 expression is vital for keeping HOXA9-immortalized myeloid cells in an undifferentiated, proliferating state, refractory to granulocyte-colony stimulating factor-induced terminal differentiation (Calvo et al., 2001). Though co-expression and interaction of MEIS1 with Abd-B-like HOX transcription factors was convincingly demonstrated in leukemia, the data on MEIS1 gene expression and function in the endometrium were still scarce. In the current study, we performed data-mining on gene expression sets in the public domain to investigate the expression of the MEIS genes (MEIS1–3) in the human female reproductive tract. We next characterized the menstrual cycle-specific pattern of MEIS1 in human endometrium. To block or over-express maternal Meis1 expression with Meis1 small interfering RNA (siRNA) or over-expression constructs in a mouse model, we used liposome-mediated gene transfection in the murine uterus, to investigate whether maternal Meis1 expression in the endometrium contributes to implantation. We conclude from these experiments that manipulation of adult Meis1 expression significantly affected fertility in the mouse. Materials and Methods Meis gene expression in normal tissue Affymetrix data for a set of 87 different normal human tissue types (the ‘human body index’ set) representing a total of 504 tissue samples were retrieved from public gene expression omnibus (GEO) data sets on the National Center for Biotechnology Information (NCBI) website (Barrett et al., 2005, 2007). CEL data from the Affymetrix GeneChip Human Genome U133 Plus 2.0 array data sets were downloaded, and intensity values and the accompanying P-values assigned with GeneChip® Operating Software (GCOS) using the MASS5.0 algorithm (Affymetrix, Santa Barbara, CA, USA). Annotations and clinical data for the tissue samples analyzed were available from http://www.ncbi.nlm.nih.gov/geo/query/ through its GEO ID: GSE7303. Human tissue collection Endometrial tissue was collected during routine endometrial biopsy from women with a normal menstrual cycle who were eligible for IVF and embryo transfer because of (partial) tubal blockage. Informed consent was obtained from all patients. The tissue collection met with the conditions described in the ‘Declaration of Helsinki for Medical Research involving Human Subjects’ as described in (http://www.wma.net/e/plicy/pdf/17c/pdf). Also, approval was obtained from the Tongji Hospital research and ethics committee. One half of the tissue sample was immediately frozen in liquid nitrogen and stored at −72°C. The other half was fixed in formalin, embedded in paraffin, sectioned and stained with hematoxylin and eosin. Menstrual cycle dating was determined by menstrual history and confirmed by histological examination using the Noyes criteria (Noyes et al., 1955). The early and late proliferative phase, and the early, mid and late secretory phase were defined as Days 5–9, 10–14, 15–19, 20–24 and 25–28 of the menstrual cycle, respectively. RNA isolation and RT–PCR Total RNA was extracted using the total RNA isolation reagent (TRI reagent, Molecular Research Center, Cincinnati, OH, USA) according to the manufacturer’s instructions. Four microgram total RNA from each sample was denatured at 70°C for 5 min and chilled rapidly on ice. The RNA was then reverse transcribed in a 20 µl reaction mixture containing 4 µl 5 × RT buffer, 2 µl 10 mM dNTP, 1 µl 0.5 µg/µl Oligo(dT15), 0.5 µl 50 U/µl ribonuclease inhibitor, 1 µl 200 U/µl Moloney murine leukemia virus transcriptase (Promega, Madison, WI, USA) with the following reaction condition: 42°C for 60 min; 95°C for 5 min in a Biometra T Gradient Thermocyler (Biometra, Göttingen, Germany). After first strand complementary DNA (cDNA) synthesis, 2 µl of cDNA was amplified in 50 µl of PCR mixture containing 5 µl 10 × PCR buffer, 4 µl 25 mM MgCl2, 1 µl 10 mM dNTP, 1.5 µl (1 U/µl) Taq polymerase (Fermentas, Foster City, CA, USA) and 1 µl (10 pmol/µl) of each primer pair. The primers used were (forward, reverse): for human MEIS1 (5′-TCCATAGCTCTTCACTTC-3′, 5′-GCTTGATGTGACAATTAG-3′), for mouse Meis1 (5′-ACGGCATCCACTCGTTCA-3′, 5′-TGGCTGTCCATCAGGGTT-3′), for mouse integrin β3 (Itgb3) (5′-GCCTTCGTGGACAAGCCTGTA-3′, 5′-GGACAATGCCTGCCAGTCTTC-3′), for human Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (5′-GGTCGGAGTCAACGGATTTGGTCG-3′, 5′-CTTCCGACGCCTGCTTCACCAC-3′) and for mouse β-actin (5′- TTCCAGCCTTCCTTCTTGGG-3′, 5′-TTGCGCTCAGGAGGAGCAAT-3′). The PCR reactions were performed as follows: 94°C for 5 min, followed by a number of cycles of 94°C for 40 s, annealing for 50 s at different annealing temperatures, and 72°C for 1 min, followed by incubation at 72°C for 10 min. Annealing temperatures for human MEIS1, mouse Meis1, mouse Itgb3, mouse β-actin and human GAPDH were 55°C, 50°C, 55°C, 50°C, and 55°C, respectively, and the numbers of cycles were 25, 30, 28, 30 and 25, respectively. The PCR products were separated on 2% agarose gels and visualized by ethidium bromide (EtBr) staining under UV light. The resulting PCR products were 411 bp for human MEIS1 [representing nucleotides (nt) 2261–2671 of RefSeq NM_002398), 405 bp for mouse Meis1 (nt 1106–1510 of NM_010789), 337 bp for mouse Itgb3 (nt 571–907 of NM_016780), 786 bp for human GAPDH (nt 117–902 of NM_002046) and 224 bp for mouse β-actin (nt 850–1073 of NM_007393). Mouse Meis1 primers used in the experiments described yielded a 405 bp PCR product containing endogenous mouse Meis1 as well as human MEIS1 (nt 1115–1519 of NM_002398), resulting from the pcDNA4-encoded transgene. Human MEIS1 or mouse Meis1 and Itgb3 mRNA levels were calculated as a ratio of the densitometric values for MEIS1, Meis1, or Itgb3 to their corresponding GAPDH or β-actin values. Care was taken to ensure a logarithmic signal for the PCR products measured. RT–PCR reactions performed for human MEIS1/GAPDH showed differential MEIS1 and equal GAPDH expression for all samples when RT–PCR was performed for 28, 25 or 22 cycles. For all three cycle numbers used, MEIS1 and GAPDH signal intensities were different between cycle numbers. Similar results were obtained for mouse Meis1/β-actin (35, 30 or 25 cycles) and mouse itgb3/β-actin (30, 28 or 25 cycles) (results not shown). In situ hybridization Diethyl pyrocarbonate-treated Milli-Q water was used in each step. Sections (4 µm thick) were immersed in 30% H2O2 solution for 30 min to eliminate endogenous peroxidase activity. After rinsing in 100 mM phosphate-buffered saline (PBS), sections were treated with pepsin [0.02% (w/v) in 3% citric acid] at 37°C for 10 min. Thereafter, sections were fixed in freshly prepared 4% (w/v) paraformaldehyde in PBS for 20 min at 4°C, rinsed in PBS and immersed in 0.25% acetic anhydride, 100 mM triethanolamine and 0.9% (w/v) NaCl consecutively, each for 10 min at room temperature. The probe for human MEIS1 was a 1:1 molar mixture of single stranded DNA oligonucleotides complementary to the 5′-CACTCGCATCAGTACCCGCACACAGCTCATACCAA-3′ (nt 593–627 of NM_002398) and 5′-ACACCTTATAATCCTGATGGACAGCCCATGGGAGG-3′ (nt 1493–1527 of NM_002398) MEIS1 sequences, labeled with digoxigenin. Sections were hybridized overnight at 40°C in hybridization buffer containing digoxigenin-labeled oligonucleotide probe at a final concentration of 400 mg/l. After hybridization, sections were washed in fresh 2 × SSC (saline sodium citrate) at 37°C for 10 min, 0.5 × SSC at 37°C for 15 min, and 0.2 × SSC at 37°C for 15 min, then incubated in 5% (w/v) bovine serum albumin for 30 min at room temperature, and in biotin-anti-digoxigenin at 37°C for 1 h, and finally washed and incubated in streptavidin–biotin complex (SABC) for 20 min at 37°C. Probe labeling, and digoxigenin, biotin-anti-digoxigenin and SABC incubations were performed using the MK2502 kit (Boster Bio-technology, Wuhan, Hubei, China). The brown precipitate generated at the site of digoxigenin-probe hybridization by reaction of the anti-digoxigenin antibody bound peroxidase with diaminobenzidine tetrahydrochloride (DAB) (AR1022, Boster Bio-technology) was scored as presence of MEIS1 mRNA. Hematoxylin was used for counterstaining. A total of 33 samples were examined. The staining of 100 consecutive cells in five non-adjacent microscopic fields was evaluated to quantify the expression of MEIS1. Western blot analysis Tissues from each mouse uterus were washed twice with PBS and lysed on ice in lysis buffer [50 mM Tris–HCl (pH 8.0), 150 mM NaCl, 0.1% (w/v) sodium dodecyl sulfate (SDS), 0.5% (w/v) sodium deoxycholate, 1% (v/v) Nonidet-P40, 0.02% (w/v) sodium azide and freshly added protease inhibitors (phenylmethylsulphonyl fluoride to 10 µg/μl and aprotinin to 1 µg/ml)]. Solid cellular debris was removed by centrifugation at 1200g for 5 min. Protein concentration was measured by the Coomassie Brilliant Blue G-250 assay. Protein samples (45 µg each) were subjected to 10% SDS–polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride membrane using a Bio-Rad electroblot apparatus (Bio-Rad, Beijing, China). Non-specific binding sites were blocked in 5% (w/v) non-fat dry milk in PBS containing 0.05% (v/v) Tween-20. The primary goat anti-MEIS1 polyclonal antibody (sc-10599, Santa Cruz, CA, USA) was used at 1:100 and incubated at 4°C overnight. This antibody was raised against a C-terminal epitope of human MEIS1, and recognizes both human MEIS1 and murine Meis1, since their protein sequences are identical in that region (also see below, section MEIS1 constructs). As secondary antibody rabbit peroxidase-conjugated anti-goat immunoglobulin (Ig)G (ZDR-5308, Zhongshan Biotechnology, Beijing, China) was used at 1:200, and incubated at 37°C for 1.5 h. Rabbit anti-actin antibody (sc-101616-R, Zhongshan Biotechnology, as a loading control) was used at 1:500 and incubated at 4°C overnight, and the secondary antibody goat anti-rabbit IgG (ZDR-5306, Zhongshan Biotechnology) was used at 1:500, and incubated at 37°C for 1.5 h. Protein bands were visualized by enhanced chemiluminescence (32109, Pierce Biotechnology, Rockford, IL, USA). MEIS1 constructs The complete human MEIS1 coding sequence was isolated by PCR using neuroblastoma tumor cDNA as a template with forward primer 5′-ccggaattccggATGGCGCAAAGGTACGAC-3′ and reverse primer 5′-ccgcgtcgacggCTACTGAGCATGAATGTCCAT-3′. The forward primer consisted of a non-coding flag containing an EcoRI recognition site (underlined) and the ATG initiation codon (in bold capitals) followed by codons 2–6 (in capitals), representing nt 458–475 of the human MEIS1 NCBI RefSeq NM_002398 coding sequence. The reverse primer consisted of a non-coding flag containing a SalI recognition site (underlined) and the TAG stopcodon (in bold capitals) preceded by the final codons 385–390 (in capitals), representing nt 1630–1610 of the NM_002398 coding sequence. This sequence encoded MEIS1B, the most widely expressed MEIS1 splice variant (see also Geerts et al., 2005 and the MEIS1 locus in TranscriptView at http://bioinfo.amc.uva.nl/human-genetics/transcriptviewb/). The derived human MEIS1B protein (NCBI sequence NP_002389, 390 residues) was identical to mouse Meis1B protein (NCBI sequence Q60 954), except for substitution of the Ile residue at position 201 in the human sequence by an analogous Val residue at position 201 in the mouse sequence. The 1196 bp MEIS1 PCR product was restricted with EcoRI and SalI, purified by agarose gel electrophoresis using Qiaex II gel extraction (20021, Qiagen, Venlo, The Netherlands) and ligated into EcoRI–SalI restricted pCI-neo vector (E1841, Promega Benelux, Leiden, The Netherlands) backbone. The MEIS1 sequence was excised from this construct using EcoRI and the 3′ polylinker NotI site and subcloned into the pcDNA4/TO/myc-HisA mammalian expression vector (V1030-20, Invitrogen, Breda, The Netherlands) restricted with EcoRI and NotI, resulting in pcDNA4/MEIS1. The construct was verified by sequencing, and expression of MEIS1 protein by the construct was demonstrated by western blot analysis of transiently transfected HEK293T cells and stably transfected human neuroblastoma cells (Revet et al., manuscript in preparation, see also Fig. 2 in Geerts et al., 2005). The Meis1 siRNA expression constructs were prepared as follows: oligos T14F 5′-gatcccTCCAGAACTGGATAACTTGttcaagagaCAAGTTATCCAGTTCTGGAtttttggaaa-3′ and T14R 5′-agcttttccaaaaaTCCAGAACTGGATAACTTGtctcttgaaCAAGTTATCCAGTTCTGGAgg-3′, targeting the TCAAGAACTGGATAACTTG sequence in MEIS1 mRNA (nt 871–889 in the human MEIS1 RefSeq NM_002398, and 862–880 in the murine Meis1 RefSeq NM_010789) or 16F 5′-gatcccCTTGATGATTCAAGCCATAttcaagagaTATGGCTTGAATCATCAAGtttttggaaa-3′ and 16R 5′-agcttttccaaaaaCTTGATGATTCAAGCCATAtctcttgaaTATGGCTTGAATCATCAAGgg-3′, targeting the CTTGATGATTCAAGCCATA sequence in MEIS1 mRNA (nt 886–904 in NM_002398, and 877–895 in NM_010789) were annealed and ligated into pTER restricted with BglII and HindIII as described (van de Wetering et al., 2003), to create pTER/Meis1-T14 and pTER/Meis1-T16, respectively. A non-targeting pTER construct contained a similar sequence that had no homology to the human or murine genome, and was used as a negative control. The constructs were verified by sequencing and siRNA-mediated MEIS1 protein and mRNA knockdown by both constructs was demonstrated (Revet et al., manuscript in preparation, see also Fig. 2 in Geerts et al., 2005). The T14 and T16 constructs have complete homology to human MEIS1 as well as to murine Meis1, and target all annotated MEIS1 splice variants (i.e. isoforms A-D, see also Geerts et al., 2005). DNA/liposome preparation Plasmid DNA (1–2 µl) was mixed with 50 µl Opti-MEM reduced serum medium (31985, Invitrogen, Carlsbad, CA, USA), added to 4 µl liposome [a 3:1 (w/w) formulation of 2,3-dioleyloxy-N-[2(sperminecarboxamido)ethyl]-N-N-dimethyl-1- propanaminium trifluoroacetate (DOPSA) and dioleoylphosphatidyl ethanolamine (DOPE) (11668-027, Invitrogen)] in 50 µl OptiMEM and incubated for 20 min at room temperature. Final concentrations were 16 µg/ml DNA and 40 µg/ml liposome. In vivo gene transfection Nulliparous female and male Kunming mice of reproductive age were supplied by Center of Experimental Animals, Tongji Medical College (Wuhan, China). Female mice were mated and examined every 8 h until detection of a vaginal plug, which was designated Day 1 of the pregnancy. The mice were anesthetized 30–60 h following plug detection with 1% butaylone given by i.p. injection. A laparotomy was performed to expose the uterus and 25 µl of the DNA/liposome complex was injected into the base of each uterine horn using a 27-gage needle, as described (Bagot et al., 2000; Hsieh et al., 2002). The incision was closed in two layers (peritoneal and cutaneous) with 4–0 vicryl suture. The mice were euthanized by cervical vertebrae dislocation under anesthesia before excision of the uteri. For the experiments studying the effect of MEIS1 expression manipulation on MEIS1 expression, litter size and pinopode formation, the group sizes were five mice each transfected with pcDNA4/MEIS1, empty pcDNA4, pTER/Meis1 siRNA construct or non-targeting pTER, respectively (20 mice in total). For the experiments studying the effect of MEIS1 expression manipulation on Itgb3 expression, 25 mice each were transfected with pcDNA4/MEIS1, empty pcDNA4, pTER/Meis1 siRNA construct or non-targeting pTER, respectively (100 mice in total). The numbers of mice available for Meis1 expression analysis in Fig. 5 were 19, 17, 19 and 19, respectively (the remainder were not pregnant in spite of the detection of the vaginal plug or died of post-surgical complications). The presented work conformed to the ‘Guiding principles in the Care and Use of Animals’ as described in DHEW publication No. (NIH) 85-23, Revised 1985, Office of Science and Health Reports, DRR/NIH, Bethesda, MD 20982, USA, was approved by the ethical committee of Tongji Medical College and was also covered by Chinese animal husbandry legislation. Scanning electron microscropy Uteri were dissected from mice 90 h after vaginal plug detection. The specimens were dehydrated in an acetone series, dried in a critical point drier using carbon dioxide, mounted on the specimen holder, coated with gold and examined under a Stereoscan 520 Scanning Electron Microscope (Hitachi, Tokyo, Japan). From each biopsy, six to eight tissues fragments, measuring ∼2 mm in thickness and 10 mm in length, were evaluated. A minimum of an aggregated 4–5 mm2 of well-preserved epithelial surface was analyzed, to increase the likelihood that the observations were representative by negating the influence of local differences in endometrial morphology. The evaluations were performed with coded tissue and read blindly by an experienced investigator according to Nikas and Makrigiannakis (2003). Immunohistochemistry The expression of mouse Itgb3 protein in the uterus was evaluated by immunohistochemistry using a goat polyclonal antibody to Itgb3 (sc-6626, Santa Cruz). The full-thickness biopsies were obtained at the time of histological evaluation from each uterine horn. The specimens were embedded in paraffin and serial 5 µm sections obtained. The sections were deparaffinized in xylene and ethanol. Endogenous peroxidase was blocked with 3% H2O2 at room temperature for 10 min. After blocking for 45 min with 1.5% normal horse serum in PBS, the sections were incubated overnight at 4°C with primary Itgb3 antibody (1:200) in PBS. The sections were then incubated with biotinylated rabbit anti-goat IgG (ZDR-5308, Zhongshan Biotechnology) as a secondary antibody at 1:50 and incubated at 37°C for 30 min, subsequently with avidin and biotinylated peroxidase at room temperature for 45 min, and finally with DAB (400 mg/ml) at room temperature for 5 min. Hematoxylin was used for counterstaining. The staining of 100 consecutive cells in five non-adjacent microscopic fields was evaluated. Immunoreactivity of Itgb3 was scored as follows: 1+, <25% of cells stained; 2+, >25% but <50% of cells stained; 3+, >50% but <75% of cells stained; and 4+, >75% of cells stained. Overall intensity was scored as weak (1+), moderate (2+) or strong (3+). The sum of cell stained scores and intensity scores was calculated to quantify the expression of Itgb3. Statistical analysis All numerical results are expressed as the mean value ± SD, except for the Affymetrix expression values of MEIS1–3 mRNAs in normal human tissue, which are represented as mean value ± SEM. The data of MEIS1 mRNA in endometrium during the menstrual cycle were analyzed by analysis of variance. Comparisons between the expression of Meis1 in each group of murine tissue, Itgb3 in each group of murine tissue and litter size of each group of mice were performed using two-tailed unpaired t-tests. The differences of pinopode development between each group of murine tissue were analyzed using a chi-square test. P < 0.05 was considered significant in all tests. The Statistical Package for the Social Sciences software package for Windows (Version 13.0) was used for all statistical analyses. Results Expression of MEIS1, MEIS2 and MEIS3 genes in the normal human uterus Affymetrix data for a set of 87 different normal human tissue types (the ‘human body index’ set) representing a total of 504 tissue samples were retrieved from the NCBI website. As shown in Fig. 1, MEIS1 was highly expressed in both uterine tissues present in the data set: myometrium and endometrium. In fact, the MEIS1 expression in these two tissues was the highest observed in all 87 tissue types investigated. In contrast, MEIS2 expression was high in myometrium, although several other tissues show even higher expression, but only moderate in endometrium. Finally, the MEIS3 gene showed overall much lower expression (5–7-fold) than the MEIS1 and MEIS2 genes. The high expression of MEIS1 compared with MEIS2 and MEIS3 in endometrium prompted us to investigate the expression and function of the MEIS1 gene in the reproductive cycle. Figure 1: Expression of MEIS1-3 mRNAs in normal human tissue. Visual representation of MEIS1, MEIS2 and MEIS3 expression in the ‘human body index’ Affymetrix data set in the public domain: 504 samples representing 87 different normal human tissues. For reasons of presentation, only tissues consisting of five or more different samples were shown: 40 tissues representing 329 samples. MEIS1 expression in endometrium and myometrium was higher than in any of the non-shown sets. Average MEIS expression in all 87 tissues was: MEIS1 338.3, MEIS2 516.6 and MEIS3 77.1. For comparison: average expression of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and β-actin in this data set was 9240 and 11 890, respectively. Average MEIS gene expression in endometrium (18 samples) was: MEIS1 1241 ± 83, MEIS2 358 ± 99 and MEIS3 149 ± 16. Average MEIS expression in myometrium (27 samples) was: MEIS1 1330 ± 90, MEIS2 1273 ± 84 and MEIS3 254 ± 26. The average expression over the different samples per tissue is shown. The error bars represent the SEM Expression of MEIS1 mRNA during the human reproductive cycle MEIS1 expression has been previously demonstrated in epithelial cells of the developing female reproductive tract (Dintilhac et al., 2005). To investigate the expression of MEIS1 in cyclic development of the endometrium, the menstrual cycle stage-specific expression was determined by RT–PCR. Representative EtBr stained agarose gel pictures of RT–PCR are shown in Fig. 2. MEIS1 mRNA expression was evident throughout the menstrual cycle. Figure 2: MEIS1 mRNA expression in human endometrium during the menstrual cycle. RT–PCR analysis of MEIS1 expression. A representative agarose gel stained with ethidium bromide (EtBr) picture is shown. MEIS1 was expressed in all stages of the human menstrual cycle investigated; the first half of the proliferative phase (P1), the second half of the proliferative phase (P2) and the first third of the secretory phase (S1), the mid-secretory phase (S2) and the late secretory phase (S3). The sample numbers for the P1, P2, S1, S2 and S3 groups were 5, 7, 7, 8 and 6, respectively. M, size markers. No PCR product was observed in control samples that did not contain RT enzyme during RT synthesis or template DNA during PCR (not shown). RT–PCR analysis was repeated at least three times per sample. MEIS1 expression was analyzed by densitometry and normalized to GAPDH in 33 endometrial samples. The densitometric analysis was repeated three times, with similar results. The data were analyzed using analysis of variance. The average MEIS1/GAPDH ratio in each group was: P1 0.20 ± 0.02, P2 0.26 ± 0.03, S1 0.41 ± 0.07, S2 0.70 ± 0.10 and S3 0.44 ± 0.09. There was no statistical difference within the proliferative phase (P = 0.47), but the difference between the proliferative phase and the secretory phase was significant (P = 0.00001) To assess potential regulation of MEIS1 expression during the reproductive cycle, densitometric analysis was performed on all samples, and the average abundance of MEIS1 during each stage of the menstrual cycle normalized to GAPDH, calculated. No difference was apparent between the levels of expression in the segments of the proliferative phase. In contrast, expression at the mid secretory phase was elevated relative to other stages (see legend to Fig. 2). These results were a first indication that MEIS1 expression varied during the menstrual cycle, markedly increasing at the mid secretory phase (S2), which corresponded to the time of implantation. MEIS1 mRNA is expressed in the human endometrial stroma and glands To investigate the site of MEIS1 expression in the human endometrium, and as a support of the MEIS1 expression detected with RT–PCR described above, in situ hybridization was performed. Fig. 3a and b shows high-power views of MEIS1 expression in proliferative and secretory endometrium. MEIS1 was expressed in both glandular and stromal cells. This supports the results of RT–PCR in Fig. 2, and further suggests that MEIS1 may play a role in embryonic implantation. Figure 3: MEIS1 mRNA expression in human endometrial tissue. In situ hybridization analysis of MEIS1 expression in human endometrial samples. Views of two representative samples demonstrate MEIS1 expression in both the glandular and stromal cells of (a) proliferative phase and (b) secretory phase endometrium. MEIS1 mRNA is stained brown. No staining was observed when the MEIS1 probe was omitted (not shown). Scale bar = 100 µm. Each sample was analyzed at least in triplicate Down-regulation of Meis1 expression affects murine embryo implantation rate To assess whether an up-regulation of Meis1 expression during embryo implantation was essential to this process, the expression of maternal Meis1 was down-regulated in a mouse model by transfection in vivo, as described by (Bagot et al., 2000; Hsieh et al., 2002). A first indication of decreased Meis1 expression in the uterus was obtained by RT–PCR, and this was confirmed by western blot on Day 3 after transfection. Representative results for mRNA are shown in Fig. 4a and b. Densitometric analysis was performed on all samples, and the average abundance of Meis1 mRNA in each group normalized to β-actin mRNA was calculated. Levels of Meis1 mRNA and protein (Fig. 4b and c) were significantly decreased in endometrium transfected with pTER/Meis1 siRNA compared with the control. Figure 4: Meis1 expression manipulation in murine uterine tissue. (a) RT–PCR analysis of Meis1 expression in 20 murine endometria transfected with pcDNA4/MEIS1, Meis1 small interfering RNA (siRNA) constructs or respective controls (five animals in each group). Shown is a representative EtBr picture. (b) Meis1 mRNA expression was analyzed by densitometry and normalized to β-actin as described in Fig. 2, except that data were analyzed using a two-tailed unpaired t-test. No PCR product was observed in control samples that did not contain RT enzyme during RT synthesis or template DNA during PCR (not shown). RT–PCR and quantification was repeated at least three times, with similar results. Average Meis1/β-actin ratios in endometrium transfected with pcDNA4/MEIS1 (U) and control pcDNA4 DNA (Uc) were 0.90 ± 0.09 and 0.48 ± 0.07, respectively. There was a significant increase in Meis1 mRNA in the up-regulation group (U) compared with the control (Uc) (P = 0.0001). Average Meis1/β-actin level in endometrium transfected with Meis1 siRNA (D) and control pTER constructs (Dc) were 0.27 ± 0.05 and 0.40 ± 0.03, respectively. Levels of Meis1 mRNA were significantly decreased in Meis1 siRNA-transfected samples (D) compared with the control samples (Dc) (P = 0.002). (c) Western analysis performed on 20 murine endometria transfected with pcDNA4/MEIS1, Meis1 siRNA or respective controls (five animals in each group). A representative western blot image is shown. (d) Meis1 protein expression was analyzed by densitometry and normalized to β-actin values as described in Fig. 4b. No Meis1 band was detected on western blot in the absence of primary Meis1 antiserum (not shown). Western blot analysis and quantification was repeated at least three times, with similar results. The average Meis1/β-actin ratio in each group was: U 0.94 ± 0.05, Uc 0.57 ± 0.06, D 0.37 ± 0.05 and Dc 0.57 ± 0.08. There was a significant increase in Meis1 protein in endometria transfected with pcDNA4/MEIS1 (U) compared with pcDNA4-transfected controls (Uc) (P = 0.00001). Endometria transfected with pTER/Meis1 siRNA constructs (D) demonstrated a significant decrease in Meis1 protein compared with pTER controls (Dc) (P = 0.01) At Day 9 of pregnancy, mice were killed and the uteri removed and examined for normal histology and for the number of implantation sites per uterus as a measure of litter size. The endometrium appeared histologically normal, and demonstrated adequate decidualization. Also the total number of pregnant mice did not vary significantly between Meis1 siRNA-treated mice and the control. Pups of mice transfected with Meis1 siRNA were born following a normal length gestation of 18–22 days, were of normal size and demonstrated no morphological abnormalities. However, the average number of implantation sites per uterus was 12.2 ± 1.5 in the mice transfected with the control pTER plasmid and 7.6 ± 2.7 in the mice transfected with a pTER/Meis1 siRNA construct (Fig. 5a). Decreased expression of Meis1 therefore resulted in a significant decrease in litter size (see legend to Fig. 5a). We concluded that the naturally occurring up-regulation of Meis1 in the reproductive cycle contributes to normal embryo implantation. Figure 5: Effect of Meis1 expression on mouse litter size. Litter size, as determined by counting the number of implantation sites per uterus examined, at Day 9 of pregnancy. Mice were transfected 1 day after detection of a vaginal plug with pTER/Meis1, control pTER, pcDNA4/MEIS1 or empty pcDNA4. (a) Litter size was reduced by 40% in the group of mice transfected with pTER/Meis1 siRNA compared with pTER controls. This difference was statistically significant according to a two-tailed unpaired t-test (P = 0.00001). (b) Litter size at Day 9 of gestation was not affected by transfection with pcDNA4/MEIS1 compared with pcDNA4-transfected controls (P = 0.08). The experiment was repeated three times, with similar results To evaluate the effect of up-regulation of Meis1 expression in the adult murine uterus, mice were transfected with the pcDNA4/MEIS1 expression construct, or with empty pcDNA4 plasmid (as a negative control). A first indication of increased Meis1 expression in the uterus was obtained by RT–PCR and this was confirmed by western blot. Levels of Meis1 mRNA and protein were significantly increased in endometrium transfected with the pcDNA4/MEIS1 expression construct compared with the control (Fig. 4c and d). As in the experiment described above, uterine histology was not affected by the transfection, and the number of mice pregnant and the weight and morphology of the pups did not differ between the groups. The average number of implantation sites was 13.4 ± 1.1 in the pcDNA4/MEIS1 treated mice and 12.6 ± 1.1 in the control mice (Fig. 5b). In contrast to Meis1 down-regulation, up-regulation of Meis1 expression by pcDNA4/MEIS1 treatment did not significantly alter litter size. Effect of Meis1 on murine pinopode formation For further clues as to a possible function for Meis1 in murine implantation, uterine epithelial morphology was investigated. As described above, no uterine histological changes could be detected at the light microscopy level in mice transfected with Meis1 DNA constructs. To evaluate the effect of Meis1 on epithelial morphology, the uteri of these mice were examined with scanning electron microscopy for evidence of altered formation of pinopodes. Mice which were transfected in vivo as described above were killed 4.5 days after vaginal plug detection, the uteri removed and prepared for electron microscopy analysis. ***Fig. 6a and b shows a typical section of endometrium from a pcDNA4/MEIS1-treated mouse, and an empty pcDNA4-treated control mouse; there was a significant increase in the development of pinopodes in the mouse transfected with MEIS1 over-expression plasmid. Fig. 6c and d shows a typical section of endometrium from pTER/Meis1 and control pTER-treated mouse uteri. None of the sections from Meis1 siRNA-treated mice contained normal pinopodes. The secretory cells were covered with dense microvilli, instead of with pinopodes, and were slightly bulging. In contrast, all sections from control animals contained a normal number of pinopodes. Apparently, decreasing Meis1 expression had a negative effect on efficient pinopode maturation. Absence of a normal amount of functioning pinopods in mice transfected with Meis1 siRNA could be a mechanism contributing to a less efficient embryonic implantation rate and diminished litter size in these animals. Figure 6: Effect of Meis1 expression on murine pinopode formation. Pinopode development in transfected mouse uteri, examined with scanning electron microscopy. (a) Transfection with pcDNA4/MEIS1 demonstrated fully developed pinopodes on most (≥60%) secretory cells. (b) Transfection with empty pcDNA4 plasmid as a negative control showed a wide distribution of developing pinopodes, covering ∼50% of the endometrial luminal surface. However, there was a much lower occurrence (20%) of fully developed pinopodes in this group. (c) Transfection with Meis1 siRNA plasmid resulted in the majority of the endometrial surface expressing few, to no, pinopodes. Instead, the epithelial cells were covered with short and thick microvilli. (d) Transfection with non-targeting pTER plasmid as a control resulted in developing pinopodes on about half of the secretory cells. Scale bar = 10 µm. The experiment was repeated three times. The data were analyzed using a chi-square test Effect of Meis1 on Itgb3 expression in the murine uterus Maternal Hoxa10 has earlier been shown to be involved in pinopode formation in the mouse uterus (Bagot et al., 2001), in a manner resembling the effect of Meis1 on pinopode formation observed in this study. To investigate a possible connection between Meis1 and Hoxa10 function in murine uterine development and embryonic implantation, we evaluated the effect of Meis1 expression manipulation on Itgb3 expression. To obtain a first indication of differential Itgb3 mRNA expression, peri-implantation uteri from mice treated with pcDNA4/MEIS1 or empty pcDNA4 (as a negative control), or with pTER/Meis1 or non-targeting pTER (as a negative control) were analyzed by RT–PCR. Fig. 7a shows a representative EtBr gel picture of the RT–PCR. Densitometric analysis was performed on all samples, and the average abundance of Itgb3 in each group normalized to β-actin was calculated. Levels of Itgb3 mRNA appeared significantly decreased in endometrium transfected with pTER/Meis1 siRNA compared with the control. In contrast, levels of Itgb3 mRNA increased in endometrium transfected with the pcDNA4/MEIS1 expression construct compared with the control (see legend to Fig. 7). Figure 7: Effect of Meis1 expression manipulation on murine uterine integrin β3 (Itgb3) expression. (a) RT–PCR analysis of Itgb3 expression in murine endometria transfected with pcDNA4/MEIS1 (U), empty pcDNA4 (Uc), pTER/Meis1 siRNA (D) or control pTER constructs (Dc). A representative agarose EtBr picture is shown. No PCR product was observed in control samples that did not contain RT enzyme during RT synthesis or template DNA during PCR (not shown). RT–PCR was repeated three times, with similar results. Itgb3 expression was analyzed by densitometry and normalized to β-actin as described in Fig. 4. Average Itgb3/β-actin ratio in each group was: U 0.83 ± 0.17, Uc 0.50 ± 0.07, D 0.28 ± 0.06 and Dc 0.44 ± 0.04. There was a significant increase in Itgb3 mRNA in endometrium transfected with pcDNA4/MEIS1 compared with endometrium transfected with empty pcDNA4 DNA (P = 0.02). Levels of Itgb3 mRNA in endometrium transfected with pTER/Meis1 siRNA were significantly decreased when compared with endometrium transfected with control pTER (P = 0.006). (b) Immunohistochemistry of Itgb3 protein expression in the endometrium of mice transfected with either pcDNA4/MEIS1 or pTER/Meis1 siRNA plasmid. Shown are representative images of each group: uterine sections from mice transfected with pcDNA4/MEIS1 (U), with empty pcDNA4 vector (Uc), with pTER/Meis1 siRNA (D) or with non-targeting pTER vector (Dc). No staining was observed in a negative control without Itgb3 antibody as the primary antiserum (not shown). Scale bar = 100 µm. Statistical analysis of Itgb3 immunoreactivity was performed as described in the ‘Materials and Methods’ section. The data were analyzed using a two-tailed unpaired t-test. The immunoreactivity analysis was repeated three times, with similar results. The average Itgb3 immunoreactivity scores in each group were: U 6.4 ± 0.5, Uc 4.0 ± 0.8, D 2.4 ± 0.9 and Dc 4.8 ± 0.7. Uterine sections from pcDNA4/MEIS1-transfected mice showed significantly increased glandular and stromal Itgb3 expression compared with control mice treated with empty pcDNA4 vector (P = 0.0015). Uterine sections from pTER/Meis1 siRNA-transfected mice showed significantly reduced glandular and stromal Itgb3 expression compared with control mice treated with non-targeting pTER vector (P = 0.03) To support the results of the RT–PCR described above, the effect of Meis1 expression on Itgb3 protein levels was determined by immunohistochemistry on transfected uterine tissue of each group. Representative pictures are shown in Fig. 7b. Itgb3 expression levels appeared higher in mice transfected with pcDNA4/MEIS1 expression construct compared with the pcDNA4 control. In contrast, Itgb3 expression was lower in mice transfected with Meis1 siRNA expression constructs than in those transfected with the non-targeting siRNA construct. Statistic analysis of the Itgb3 immunoreactivity was performed on the samples, and the average abundance of Itgb3 in each group was calculated. Levels of Itgb3 expression were significantly decreased in endometrium transfected with pTER/Meis1 siRNA compared with the control. In contrast, Itgb3 protein levels were increased in endometrium transfected with the pcDNA4/MEIS1 expression construct compared with the control (see legend to Fig. 7). This was consistent with the results of the RT–PCR. Together, these results showed that, in a murine uterine receptivity model, the expression of the known Hoxa10 target Itgb3 could be regulated by the manipulation of Meis1 expression, suggesting Meis1 action through co-operation, or possibly interaction, with Hoxa10. Discussion HOXA10 is essential for normal embryonic uterine development. In the developing mouse embryo, altered Hoxa10 expression as a result of targeted mutation (Satokata et al., 1995) or exposure to diethylstilbestrol leads to abnormal uterine development (Block et al., 2000). Hoxa10 expression persists in the adult uterus and is important for implantation of the embryo. Hoxa10 null-mutant mice or mice in which Hoxa10 has been blocked using antisense oligonucleotides demonstrate decreased litter size due to failure of implantation (Benson et al., 1996; Bagot et al., 2000). HOX genes function as transcription factors and either activate or repress target genes. The pleiotropic effects of HOXA10 on endometrial differentiation and function are likely modulated through the regulation of multiple downstream targets that are necessary for the implantation process (Daftary et al., 2002; Troy et al., 2003). HOXA10 has been demonstrated to be involved in the regulation of pinopode development and expression of other genes involved in endometrial development such as ITGB3 and EMX2 (Bagot et al., 2001; Daftary et al., 2002; Troy et al., 2003). HOX proteins require co-operation with other proteins to bind to their target DNA. MEIS1 in particular aids 5′ HOX proteins, like HOXA10, to gain this specificity. MEIS1 physically interacts with 5′ HOX proteins HOXA9–11 by forming heterodimeric binding complexes on a DNA target containing a MEIS1 site (TGACAG) and an Abd-B-like Hox site (TTTTACGAC) (Shen et al., 1997). Studies demonstrated that MEIS1 gene and HOXA9-13 genes are co-expressed throughout Müllerian duct differentiation (Dintilhac et al., 2005), which suggests that MEIS1 might play a role in embryonic female genital tract development. We have performed data-mining on Affymetrix gene expression database in the public domain showing that MEIS1 was highly expressed in human endometrium as well as in myometrium (Fig. 1), implying it may play a role in reproductive function in the adult human uterus. We therefore investigated the expression pattern of the MEIS1 gene at different stages of the human menstrual cycle, demonstrating that MEIS1 appeared to be expressed in the endometrium at different levels, depending on the menstrual cycle stage. MEIS1 mRNA levels appeared significantly increased in the mid secretory phase, closely resembling the expression pattern of HOXA10. This observation pointed to the possibility that MEIS1 is an active cofactor of HOXA10 in the human endometrium. On the basis of the results described above, we postulated that Meis1 and Hoxa10 might have closely related roles in vertebrate reproduction and in the regulation of endometrial cell proliferation or differentiation. Meis1 null-mutant mice die between embryonic days 11.5 and 14.5, because of internal hemorrhage, liver hypoplasia and anemia (Azcoitia et al., 2005). The embryo lethal phenotype has precluded investigation of the function of Meis1 in the adult reproductive tract of Meis1 mutant mice. Therefore, to investigate the role of Meis1 in adult reproductive function in a mouse model, we altered the expression of the Meis1 gene in the adult murine endometrium. When mouse endometrium was transfected with Meis1 siRNA at the time of implantation, the number of embryos that could implant in the uterus was reduced by 40%. This significant reduction in implantation sites demonstrated that, in normal embryonic development, optimal implantation benefits from maternal Meis1 expression. We also investigate the effect of Meis1 on pinopodes. Pinopode formation is directly regulated by HOXA10 (Bagot et al., 2001). In this study, we examined whether the expression of pinopodes could be also affected by altering Meis1 expression levels. We demonstrated that blocking Meis1 expression reduced pinopode numbers and reduced their ability to differentiate into fully mature forms. In contrast, supernumerary pinopodes could be induced by over-expression of Meis1. To elucidate potential mechanisms by which Meis1 might control endometrial function and receptivity to the implanting blastocyst, we tested the effect of altering Meis1 expression on Itgb3 expression. The ITGB3 gene has been identified as the first known target of HOXA10 gene regulation in the endometrium (Daftary et al., 2002). As a bridging molecule between the endometrium and trophoblast (Kiefer et al., 1989; Van Dijk et al., 1993; Sueoka et al., 1997), ITGB3 plays an important role in embryonic implantation. We found that a blockade of Meis1 expression with Meis1 siRNA was associated with a decrease in Itgb3 mRNA. Itgb3 mRNA levels significantly increased after up-regulation of Meis1 expression with MEIS1 cDNA. It appeared likely that Itgb3 was not regulated by Hoxa10 monomer, but by a heterodimeric or heterotrimeric transcription complex containing Meis1. The interaction of Abd-B-like HOX transcription factors and MEIS1 at the HOX site in the target gene promoter has been demonstrated in myeloid differentiation. The co-activation of HOXA9 and MEIS1 is a common event in AML (Lawrence et al., 1999; Camós et al., 2006), but co-regulation of MEIS1 and HOXA9, and also of HOXA10, might play a role in pathogenesis of acute lymphocytic and acute myeloid leukemias associated with ALL-1 fusions (Rozovskaia et al., 2001; Ferrando et al., 2003; Dik et al., 2005). In addition, Pineault et al. (2004) demonstrated that MEIS1 enhances the effect of a HOXA10 rearrangement releasing HOX expression necessary for leukemia induction. Finally, binding of MEIS1 to HOXA10 could be demonstrated in vitro (Williams et al., 2005). CDKN1A (also known as P21 or CIP1) is a transcriptional target of HOXA10 in differentiating myelomonocytic cells. CDKN1A reporter expression is enhanced after transfection with PBX1 and HOXA10, especially when MEIS1 was co-transfected (Bromleigh and Freedman, 2000). It is likely that a similar regulation occurs with HOXA10 and MEIS1 cofactor in the endometrium. Recently, Taylor et al. have confirmed that this potential protein–protein interaction does occur in endometrial cells. Ablation of the PBX cofactor half-site results in a loss of HOXA10–PBX2 binding to the EMX2 probe, suggesting the importance of the PBX2/HOXA10 protein–protein interaction in endometrium for high affinity HOXA10 binding to its target genes (Sarno et al., 2005). In conclusion, this study shows for the first time that MEIS1 expression contributes to endometrial receptivity in mouse. MEIS1 can function in cancer as an inhibitor of differentiation, and have a positive influence on cellular proliferation (see ‘Introduction’). In addition, the MEIS genes are potent enhancers of early cell proliferation during development (reviewed in Geerts et al., 2003, 2005). MEIS1 could have a similar proliferative function during endometrial development. The expression pattern of MEIS1, its contribution to pinopode formation in mouse and its effect on Itgb3 expression suggest that the effects of MEIS1 on endometrial function occur through affecting the affinity and specificity of HOXA10–DNA interactions, leading to the regulation of downstream targets of HOXA10. The identification of MEIS1 as a potential HOXA10 co-operating gene could lead to elucidation of a partial HOXA10 network in endometrial development, but could also establish MEIS1 in a signaling network in its own right. Author’s Role B.X.: conception and design of experiments, acquisition of data, analysis and interpretation of data; writing and revising of manuscript; final approval of manuscript to be published. D.G.: design of experiments, acquisition, analysis, and interpretation of data; writing and revising of the manuscript; final approval of manuscript to be published. K.Q.: conception and design of experiments; revision of manuscript; final approval of manuscript to be published. H.Z.: conception and design of experiments; final approval of manuscript to be published. G.Z.: interpretation of data; revision of manuscript; final approval of manuscript to be published. Funding Dutch Cancer Society (‘KWF Kankerbestrijding’) (UVA2003–2849, UVA2005–3665 to D.G.); The ‘973’ Program of China (No. 2007CB948104).

Glycobiology-1-1-2430009

Glycation does not modify bovine serum albumin (BSA)-induced reduction of rat aortic relaxation: The response to glycated and nonglycated BSA is lost in metabolic syndrome

The effects of nonglycated bovine serum albumin (BSA) and advanced glycosylation end products of BSA (AGE-BSA) on vascular responses of control and metabolic syndrome (MS) rats characterized by hypertriglyceridemia, hypertension, hyperinsulinemia, and insulin resistance were studied. Albumin and in vitro prepared AGE-BSA have vascular effects; however, recent studies indicate that some effects of in vitro prepared AGEs are due to the conditions in which they were generated. We produced AGEs by incubating glucose with BSA for 60 days under sterile conditions in darkness and at 37°C. To develop MS rats, male Wistar animals were given 30% sucrose in drinking water since weanling. Six month old animals were used. Blood pressure, insulin, triglycerides, and serum albumin were increased in MS rats. Contraction of aortic rings elicited with norepinephrine was stronger. There were no effects of nonglycated BSA or AGE-BSA on contractions in control or MS rats; however, both groups responded to L-NAME, an inhibitor of nitric oxide synthesis. Arterial relaxation induced using acetylcholine was smaller in MS rats. Nonglycated BSA and AGE-BSA significantly diminished relaxation in a 35% in the control group but the decrease was similar when using nonglycated BSA and AGE-BSA. This decrease was not present in the MS rats and was not due to increased RAGEs or altered biochemical characteristics of BSA. In conclusion, both BSA and AGE-BSA inhibit vascular relaxation in control artic rings. In MS rats the effect is lost possibly due to alterations in endothelial cells that are a consequence of the illness. Introduction Reducing sugars including glucose, fructose, trioses, and glyceraldehyde alter proteins by nonenzymatic glycosylation known as glycation or Maillard reaction. This reaction occurs in several steps, the initial ones being relatively fast and reversible, while the latter ones are slower and irreversible originating products known as advanced glycosylation end products (AGEs). AGEs were originally characterized by their yellow-brown fluorescent color and by their ability to cross-link with and between amino groups (Reynolds 1963). However, the term is now used for a broad range of advanced products of the Maillard reaction, including N-carboxymethyllysine (CML) and pyrroline (Reddy et al. 1995; Baynes and Thorpe 1999). AGEs exert their actions by two different mechanisms: (1) modifying structural intra- and extracellular proteins and (2) binding to their receptors (RAGEs) that belong to the immunoglobulin family and are located in the plasma membranes of monocytes, macrophages, endothelial cells, and vascular smooth muscle cells (Brownlee 1995; Basta et al. 2002). When AGEs bind to their receptors they initiate second messenger cascades. They also generate reactive oxygen species which modulate cellular function and can induce inflammatory processes (Krieglstein and Granger 2001; Basta et al. 2002; Ramasamy et al. 2005). Albumin is one of the main proteins undergoing glycation reactions due to its abundance in serum and to the fact that it can be glycated at multiple sites (Wautier and Guillausseau 1998). Both bovine serum albumin (BSA) and AGE-BSA have been reported to have vascular effects. BSA reacts with nitric oxide (NO) therefore modulating its biological actions and increasing endothelial permeability (Predescu et al. 2002). Some forms of AGEs also diminish vascular relaxation by decreasing the NO production of endothelial cells (Hogan et al. 1992; Xu et al. 2003). Nevertheless, there have been recent reports that other forms of in vitro prepared AGEs have different effects from the naturally produced AGEs (Valencia et al. 2004). AGEs have been implicated in aging processes and in the development of chronic complications of diabetes such as renal and vascular damage and could also participate in complications of metabolic syndrome (Brownlee 1995; Wautier and Guillausseau 1998; Valencia et al. 2004). There is a significant association between a high incidence of cardiovascular events and endothelial damage in animal models and diabetic patients and one of the mechanisms underlying this association is the presence of AGEs in plasma (Tan et al. 2002). A variant of fructose-induced hypertension by giving sucrose to rats has been developed in our institution. The fructose-fed rat becomes hypertensive, hypertriglyceridemic, hyperinsulinemic and has insulin resistance; it exhibits what is known as metabolic syndrome (MS) or “Syndrome X” (Hwang et al. 1987). Some of the characteristics of the model developed in our institution have already been described (Baños et al. 1997) and they coincide with those of fructose fed animals, i.e., our rats also develop hypertension, hypertriglyceridemia, hyperinsulinemia, and insulin resistance. They also show increased oxidative stress (Baños et al. 2005). Some of the characteristics of these animals are accentuated by age and have been associated with renal failure and cardiovascular diseases which are related to alterations in the structure and function of endothelial cells (Rubio et al. 2006). AGEs could play a role in the vascular damage found in MS rats. Therefore, the aim of the present paper was to determine the effect of nonglycated BSA and AGE-BSA on aortic rings from control and MS rats. Results Changes in body weight, arterial pressure, triglycerides, glucose, insulin, and serum albumin Table I shows the values of body weight, arterial pressure, glucose, insulin, homeostasis model assessment (HOMA), triglycerides, and serum albumin for control and MS rats. Experimental animals developed metabolic syndrome characterized by arterial hypertension, hypertriglyceridemia, hyperinsulinemia, and insulin resistance. There was a statistically significant increase in the serum albumin values in MS rats. Although there was no significant change in body weight, the animals with MS clearly developed central adiposity. Table I Clinical characteristics and biochemical parameters from control and MS rats Control SM Body weight (g) 558.3 ± 14.9 547.5 ± 38.1 Arterial pressure (mm of Hg) 101.9 ± 1.4 138.9 ± 0.8* Triglycerides (mg/dL) 55.6 ± 4.7 109.1 ± 12.8* Glucose (mmol/L) 5.9 ± 0.3 4.8 ± 0.7 Insulin (μU/mL) 6.5 ± 0.9 24.2 ± 5.7* HOMA 1.02 ± 0.5 5.02 ± 2.1* Albumin (g/dL) 3.3 ± 0.1 3.6 ± 0.1* Values are mean ± SEM, n = 8; *P < 0.01. AGE-BSA preparation It was possible to produce AGE-BSA under the previously described condition. After incubation with glucose, the albumin solution took a yellow-brown color and showed the characteristic fluorescence spectrum of AGEs. Albumin incubated without glucose did not significantly change its appearance and did not have the characteristic fluorescence spectrum of AGEs (Figure 1). Fig. 1 Fluorescence spectrum of AGE-BSA and nonglycated BSA. Both solutions were tested at 2 mg/mL of protein. The peak of fluorescence is not present in the nonglycated BSA. The interrupted line represents the saturation level of the fluorescence detection by the equipment employed. The purification of AGE-BSA on Affi-Gel-Blue is shown in Figure 2; three peaks were observed which corresponded to the highly glycated BSA (first one), the less glycated BSA moderately, and stronger bound to the chromatographic matrix (second and third peaks, respectively). The first peak was employed for subsequent analysis and tested for vasoreactivity. Fig. 2 Purification of AGE-BSA on Affi-Gel-Blue. Three peaks were observed: the first one corresponds to highly glycated BSA which did not bind to the matrix, the second one corresponds to the less glycated BSA which bound moderately, and the third peak corresponds to the stronger bound BSA. The first peak was the only one to be further analyzed. The arrows indicate the buffer used at each stage. The AGE-BSA produced had a protein concentration of 53.8 mg/mL and a specific activity, before its purification in Affi-Gel Blue, of 7.9 AU/mg of protein and of 10.5 AU/mg of protein, after purification. Fluorescence was not significantly elevated when protein concentration of nonglycated BSA was increased (Figure 3). The value of specific activity was similar to that obtained by other authors using similar procedures (Verbeke et al. 1997). Fig. 3 Evaluation of the specific activity (SA) of AGE-BSA. The increase of specific activity obtained by the purification process on Affi-Gel-Blue is observed. Bold circles represent the nonglycated BSA. Bold squares represent the unpurified AGE-BSA and bold triangles represent the purified AGE-BSA. Specific activity increased in approximately 33% by the purification process. Changes in mass and isoelectric point of BSA and glycated BSA Table II shows the results of the analysis of the main glycation sites of the BSA structure. The analysis was done using the sequence of bovine serum albumin reported in the GenBank with number CAA76847 (gi:3336842) and the PyMOL v0.99 program (DeLano Scientific LLC, CA). The disposition of the lysine and arginine residues in the three-dimensional structure was evaluated. The analysis shows that a maximum of 52 residues from the 79 present in the molecule can be glycated. The rest of the residues are in the inside of the structure of BSA. Table II Distribution of the amino acids lysine and arginine in the three-dimensional structure of BSA Amino acid position in the three-dimensional structure Amino acid Totally exposed Partially exposed Nonexposed Lysine 4,12,20,41,51,64,76, 116,127,132,136,180,187,211,224, 261,273,275,285,312,316,322,350,362,375,377,388,396, 465,474,499,504,520,523,524,535,537,556, 563,573. 159,413,431,533. 93,106,114,131, 204,221,232,239, 242,279,294,439, 544. Arginine 194,196, 198. 143,185,409,412, 444. 10,81,98,144,208, 217,256,335,336, 347,427,458,483, 484. Total 43 9 27 The amino acids localized in depression areas are considered as partially exposed. Figure 4 shows a molecular mass of 66,655.58 Da for nonglycated BSA (Figure 4A). After glycation, the molecular mass increased to 74,461.15 Da (Figure 4B). The net increase of 7,805.57 Da observed corresponds to the addition of 48 glucose molecules, considering that the molecular mass of glucose is 180.2 Da and that during each reaction of Amadori product, a water molecule is liberated (18 Da). The amount of glucose bound to BSA in our preparation was 48, very close to the 52 theoretically possible; therefore, we consider that our AGE-BSA is a highly glycated product. Fig. 4 Analysis of BSA and AGE-BSA by mass spectrometry. The mass increase of protein after glycation is observed. The mass of nonglycated BSA was 66,655.58 Da (A) and that of AGE-BSA was 74,461.15 Da (B). The net increase of mass was 7,805.57 Da, corresponding to the addition of 48 molecules of glucose to each BSA molecule. Inset: change of isoelectric point of BSA after glycation. Nonglycated BSA had a pI of 4.2 while AGE-BSA had a pI of 6.3. Line 2 is the broad range pI standard kit (pH 3–10). Nonglycated BSA had an isoelectric point of 4.2. The glycation process changed the isoelectric point to 6.3 (Figure 4, inset). Vascular responses Norepinephrine-induced vasoconstriction was higher in MS rats than in control rats (Table III); nevertheless, vasoconstriction was not significantly modified in aortic rings from control or MS rats in the presence of nonglycated BSA or AGE-BSA. L-NAME significantly increased vascular contraction in both groups (154.41 ± 10.35% versus 186.18 ± 7.58%, respectively) since inhibition of NOS induces an imbalance in vasoconstriction and vasodilation which is greater in MS than in control rats (Figure 5). Fig. 5 Effect of AGE-BSA on vascular contraction in aortic rings from control (solid bars) and MS rats (open bars). The contractions were induced by NE 1 μM and basal tension was normalized to 100% in control and MS rats. Tension values in grams are shown in Table III. Results are the means ± SEM of six independent experiments. *P < 0.05 between MS and control; **P < 0.001 between the Tyrode solution and L-NAME. Table III Values of contraction (g) and relaxation (%) of aortic rings in control and experimental rats Control MS Vascular contraction 1.5 ± 0.2 2.2 ± 0.2* Vascular relaxation 84.4 ± 2.6 64.9 ± 3.3* The results are expressed as relaxation percentage of the initial precontraction level with NE (1 μM). Final relaxation with Ach (1 × 10−6 M) is reported. Values are mean ± SEM, n = 8; *P < 0.001. Basal relaxation was diminished in MS rats when compared to control rats (Table III). Figure 6 shows the effect of nonglycated BSA and AGE-BSA upon vascular relaxation in aortic rings from control and MS rats. In control rings nonglycated BSA and AGE-BSA at 40 AU/mL (3.8 mg/mL) significantly inhibited the endothelium-dependent vasorelaxation in a 35%. This effect was not present when a lower dose was used (10 AU/mL; 0.95 mg/mL). Fresh albumin, fatty acid free, diminished vasorelaxation in aortic rings from control rats in the same proportion as pre-incubated control BSA and AGE-BSA at a concentration of 3.8 mg/mL (data not shown). Fig. 6 Effect of nonglycated BSA and AGE-BSA at 40 AU/mL (3.8 mg/mL) on endothelium-dependent vasorelaxation in aortic rings from control (A) and MS rats (B). Results are expressed as relaxation percentage from the initial precontraction level with NE 1 μM. *P < 0.05 between the Tyrode solution and nonglycated BSA and AGE-BSA; #P < 0.05 between the Tyrode solution in MS and control rats. In MS aortic rings vasorelaxation was not significantly affected by BSA or AGE-BSA at any of the doses tested. Immunofluorescence of RAGEs Results of expression of RAGE are shown in Figure 7. There is an increased RAGE immunoreactivity in aortas from MS rats when compared to control rats. In aortas from C rats fluorescence is observed mainly in the endothelium while in MS aortas both the endothelium and muscular fibers fluoresce. Damage due to the development of the metabolic syndrome can also be appreciated by the presence of buckles and a loss of continuity in elastic fibers. Fig. 7 Immunohistochemistry for RAGE in aortas of control (Panel A) and MS (Panel B) rats. Negative control (Panel C). Magnification 40×. Asterisk indicates internal elastic lamina. There is an increased RAGE immunoreactivity in MS rats. Discussion The aim of the present paper was to determine the effect of nonglycated BSA and AGE-BSA on vascular aortas from control and MS rats in which the condition is induced by a high ingestion of sucrose and where the structure and function of endothelial cells are altered. Both albumin and AGEs have been reported to have vascular effects (Vlassara et al. 1992). However, the vascular effects of AGEs might not be produced by all types of AGEs, since AGEs may have very different structures depending on the reactions generated during the in vitro glycation process of the protein by reducing sugars (Katchalsky and Sharon 1953). Some of the AGEs generated are capable of cross-linking to proteins and others have the capacity to interact with specific receptors (RAGEs) present in some cellular types (Valencia et al. 2004). There have also been recent reports that some of the effects of in vitro prepared AGEs are due to metal ions which are contaminants commonly found in AGE preparations. Various metal ions have been shown to drive inflammatory responses via generation of oxidative stress (Valencia et al. 2004). The use of a chelator such as EDTA to facilitate the reaction could induce some of the effects reported to be produced by AGEs (Al-Abed et al. 1999). Our results show that body weight was not significantly modified in MS rats even if the diet of the sucrose-fed rats was hypercaloric (Baños et al. 1997) (Table I). However, sucrose-fed animals showed increased central adiposity. Insulin was increased in MS rats and this increase accompanies an increase in blood pressure. Triglycerides were also elevated. Therefore the MS rats show many of the characteristics of metabolic syndrome. Serum albumin concentration was significantly higher in MS rats when compared to control rats. However, Rumble et al. (1997) reported that diabetic and hypertensive rats had lower serum albumin levels than the normotensive Sprague Dawley rats. Vascular contraction to norepinephrine increased MS rats in this study. This has been previously described for this model (Baños et al. 1997) and additionally we have previously found that contraction to KCl in MS rats is also increased and that this is due to an elevated response to endothelin (Nava et al. 1999; Rubio et al. 2006). The increase in the circulating levels of albumin in the MS group may contribute to the increased vascular tone observed in this group. Indeed, an increase in circulating albumin levels has been observed in some forms of hypertensive pathology, due to the hemoconcentration induced by edema, which positively correlates with the hypertension levels (Allen and Patterson 1995). AGEs could also participate in the increase in vascular contraction in MS rats since they increase the production of endothelin-1 by endothelial cells (Tan et al. 2002). Our results show that vasoconstriction was not modified by nonglycated BSA or by AGE-BSA (Figure 5). This is in accordance with the results of Xu et al. (2003). Vasorelaxation in MS rats was significantly diminished in comparison to control rats and could be attributed to an imbalance of constricting and relaxing factors as has been previously reported (Baños et al. 1997). Furthermore, as our model of MS rats is caused by the chronic ingestion of sucrose, vascular damage could be a probable consequence of AGEs increased production. AGEs deposit inside smooth muscle and endothelial cells as well as in extracellular matrix proteins increasing rigidity. Hogan et al. (1992) have reported that AGEs react with NO produced by endothelial cells. They also reported that AGEs reduce the antiproliferative effect of NO on aortic smooth muscle vascular and renal mesangial cells. Insulin, which is also increased in our model, stimulates collagen synthesis in vascular smooth muscle cells (Ruiz Torres et al. 1998) and could contribute to diminished vasorelaxation in MS rats. Arterial endothelium-dependent relaxation is also diminished hypertensive rats (Ibarra et al. 1995; Küng and Lüscher 1995; Challah et al. 1997; Freitas et al. 2003; Shipley and Muller-Delp 2005). These authors have proposed that the reduced response to Ach could be a consequence of an impairment of either the generation (synthesis or release) of relaxant factors or the cellular response to them. In our study, both nonglycated BSA and AGE-BSA decreased vasorelaxation in the same proportion in control rats. In arterial rings incubated in the presence of glycated and nonglycated BSA the response to acetylcholine was 35% reduced in comparison to control rings incubated in the absence of BSA (Figure 6A). The reduction in vascular relaxation in the presence of nonglycated BSA is in accordance with the report by Predescu et al. (2002). These authors suggested that BSA reacts with NO therefore modulating its biological actions and increasing endothelial permeability. Although in our experiments we also found a decrease by AGE-BSA, it seems to be due to BSA since there was not a significant difference in the reduction produced by AGE-BSA and by nonglycated BSA. Previously, Xu et al. (2003) studied the vascular effects of AGEs produced in vitro testing vasoreactivity in rings of thoracic aorta from New Zealand white rabbits and in vitro production of NO in endothelial cells from human umbilical veins. These authors reported that there was not a significant effect of AGEs upon vasoconstriction induced by NE and our results are in accordance with this result. However, they found that AGEs diminished the NO production of isolated endothelial cells and that vasodilation was reduced in thoracic aortic rings while we found that AGEs did not modify vasodilation. Our data suggest that it is BSA that exerts the diminished vasorelaxing effect and not AGEs. In their study the authors produced AGEs incubating BSA with glucose 6-phosphate instead of glucose which is a faster reaction. Glucose 6-phosphate is a reducing sugar of great intracellular importance and can glycate easier than glucose since it is present in a higher proportion in the open-chain form (Hogan et al. 1992). Their reaction was facilitated by the use of EDTA which could induce the effect reported by them (Al-Abed et al. 1999). AGEs derived from a long period of incubation of BSA with glucose, as was done in this paper, conduces mainly to fluorescent cross-linking AGE production. Using specific antibodies, these authors found high levels of carboximetillysine (CML) in their preparation and identified CML as the AGE responsible of the NO synthesis inhibition. Therefore, the kind of AGEs used by them could explain the contradiction between their results and what we observed in the present paper. Furthermore, the difference of the species used for testing vasodilation in both papers (New Zealand rabbits and Wistar rats) might account for the contradicting results found. The same authors (Xu et al. 2003) did not find the expected previously reported decrease in vascular relaxation when they added nonglycated BSA. The absence of a vascular effect of nonglycated BSA in the above-mentioned study is probably due to the very low doses assayed, in the order of μg/mL. It is known that albumin has a physiologic effect at higher concentration. Indeed, our own results show that the inhibitory effect of BSA or AGE-BSA was only observed when the protein concentration assayed was about of 3.81 mg/mL, but not when it was only of 0.95 mg/mL. The response to nonglycated BSA and AGEs disappeared in the MS rats (Figure 6B). Alterations in the structure and function of the vascular structures caused by the metabolic syndrome could be the underlying cause of the lack of response in these animals. This physiological adaptation could be considered as a protector effect to counteract the increased basal vasoconstriction observed in metabolic syndrome rats, and must be better studied. We evaluated the presence of AGEs’ receptors (RAGEs) from aortas in MS rats to better explore the mechanism by which AGEs participate in this pathology. Many of the deleterious effects of AGEs are mediated by binding to their receptors which are present in various cellular types. In endothelial cells, the interaction AGE–RAGE modulates the expression of citocynes and adhesion molecules and increases the production of oxygen reactive species (ROS) generating oxidative stress (Ramasamy et al. 2005). Koyama et al. (2005) demonstrated that an alternative form of RAGE, the endogenous secretory RAGE (esRAGE), is associated with the metabolic syndrome or atherosclerosis in the absence of diabetes. Inflammation might be the factor linking the RAGE system with MS. Although a higher number of RAGEs were observed in the aortas from MS rats, AGE-BSA did not have any effect on arterial relaxation in this group. Moreover, the absence of differences in arterial relaxation induced by AGE-BSA or BSA in the control group suggests that the possible effect of AGE-BSA on vascular function in not mediated by RAGEs. The exposition of the vascular tissue to AGE-BSA during the development of MS could also cause the loss of the relaxation capacity of aortas. The changes in the charge and composition of the BSA molecule produced by glycation increase its aggregation capacity. When the isoelectric point is closer to physiological pH, hydrophobic interactions are favored. These changes could increase the cross-linking activity of AGEs to proteins and the capacity to deposit in the extracellular matrix of the tissues. Correspondingly, we observed an increased aggregation capacity of AGE-BSA during alcohol precipitation or desalting. However, the loss of the relaxing capacity due to biochemical changes of AGE-BSA is not favored by our results since there were no differences in the relaxation in the presence of BSA or AGE-BSA. We speculate that the cause could be the altered albumin internalization to endothelial cells and therefore a modified neutralization of NO. This should be further evaluated. The molecular structure and charge of albumin facilitate the cotransport of a number of hydrophobic molecules, enzymes, and hormones across the endothelium and the maintenance of vascular integrity and transvascular oncotic pressure gradient (Metha and Malik 2006). Endothelial permeability to plasma proteins and liquid is increased in inflammation, a condition manifested by protein-rich edema (Metha and Malik 2006). Inflammation is present in metabolic syndrome and therefore alterations in albumin transport across endothelial cells are likely to be present. Bevers et al. (2006) reported that in bEnd.3 endothelial cells, which only produce eNOS and have higher levels of NO than other endothelial cells, incubation with low albumin levels increased eNOs activity and the production of NO. In conclusion both BSA and AGE-BSA inhibit vascular relaxation mediated by NO in control artic rings in the same proportion. This inhibitory effect is likely to be due to the structure of BSA and not to the fluorescent AGEs synthesized by the employed procedure. However, in MS rats this inhibitory vascular relaxation effect is lost, possibly due to alterations in endothelial cells that are a consequence of the illness. Materials and methods Animals and arterial pressure determination Experiments in animals were approved by the Laboratory Animal Care Committee of our institution and were conducted in compliance with our institution's ethical guidelines for animal research. Weanling male Wistar rats aged 25 days and weighing 50 ± 4 g were separated into two groups—group 1: control rats (C) were given tap water for drinking and group 2: MS rats were given 30% sucrose in drinking water during a 24-week period. All animals were fed Purina 5001 rat chow (Purina Mills Inc., Richmond, IN) ad libitum, which provides 14.63 kJ/g, with 23% protein, 12% fat, and 65% carbohydrate. Animals were kept at controlled temperature and a 12:12-h light-dark cycle. Systolic arterial blood pressure was measured in conscious animals using the tail cuff method; the cuff was connected to a pneumatic pulse transducer (Narco Bio-Systems Inc., Healthdyne Co., Austin, TX) and a programmed electrosphyngomanometer. The mean of five independent determinations was calculated. AGE-BSA preparation AGE-BSA were prepared, purified, and characterized in the reproductive biology laboratory of Instituto Nacional de Ciencias Médicas y Nutrición “Salvador Zubirán” following the technique previously described (Makita et al. 1991, 1992). AGE-BSA was prepared by incubating 5% of bovine serum albumin (BSA fraction V; Sigma Chemical, St. Louis, MO) in 200 mM of sodium phosphate buffer (pH 7.4) with 500 mM glucose for 60 days under sterile conditions, in darkness and at 37°C. After incubation, the product was reduced with 100 mM sodium borohydride (NaBH4, Sigma Chemical) for 30 min at room temperature to stabilize the Schiff bases formed (Eble et al. 1983; Fluckiger and Gallop 1984; Verbeke et al. 1997). Then, the unbound glucose and unreacted NaBH4 were removed by extensive dialysis against a 500 mM sodium phosphate buffer (pH 7.4). Less glycosylated BSA was removed from AGE-BSA by affinity chromatography on Affi-Gel Blue (Bio-Rad Laboratories, Richmond, CA) (Travis et al. 1976; Verbeke et al. 1997). The Affi-Gel matrix has a blue dye named Cibacron Blue F3GA, with natural affinity for native albumin. A column with 5 mL of matrix with a binding capacity of 170 mg of BSA was employed. 2 mL of AGE-BSA solution with a concentration of 53.8 mg/mL was applied in each purification process, using a flow rate of 1 mL/min. The chromatogram was performed evaluating a dilution 1/10 of each tube. Protein concentration of AGE-BSA was evaluated by the Lowry method using BSA as standard (Lowry et al. 1951). Finally, AGE-specific fluorescence determination was performed by measuring emission at 440 nm upon excitation at 370 nm using an Aminco fluorescence spectrometer (Aminco-Bowman Series 2 SLM Instruments. Inc., Rochester, NY) (Monnier and Cerami 1981; Makita et al. 1992). Fluorescence value of AGE-BSA was measured at a protein concentration of 1 mg/mL and expressed in arbitrary units (AU). The specific activity (SA) was expressed as arbitrary units of fluorescence per mg of protein (AU/mg). The control BSA sample (nonglycated BSA) was incubated under identical conditions but without glucose. AGE-BSA and its control (nonglycated BSA) were maintained at −70°C until used. Mass increase analysis of glycated BSA The mass change of BSA after glycation was analyzed by matrix-assisted laser desorption ionization mass spectrometry (MALDI-TOF) (Voyager DE-PRO from Applied Biosystems, CA), equipped with a nitrogen laser (337 nm), operating in a positive high-energy linear mode. This analysis was made in the Analysis and Molecular Diagnose Unit of the Insituto Nacional de Salud Pública de México. Aliquots of 500 μL containing 10 mg/mL of BSA or AGE-BSA were diluted to 1.5 mL with deionized water and centrifuged in Amicon Ultracel-10 K (Millipore Corporation, MA) at 2,000 × g for 30 min to a final volume of 500 μL. Washing was repeated three times. A 10 μL aliquot was taken for MALDI-TOF analysis. The matrix solution employed for BSA analysis contained 10 mg/mL 3,5-dimethoxy-4-hydroxycinnamic acid (sinapinic acid) in 30% vol/vol acetonitrile in 0.1% vol/vol aqueous trifluoroacetic acid. For AGE-BSA the matrix solution contained 2,5-dihydroxybenzoic acid (gentisic acid, which is employed for analysis of glycoproteins) to 10 mg/mL in 50% vol/vol acetonitrile. A sample protein solution (1 μL) of a 1:1 vol/vol mixture of 1 mg/mL of total protein and the matrix solution were applied on a sample spot in a steel plate slide and dried in a warm air stream. Bovine's insulin (5,734.59 Da), E. coli's Thioredoxin (11,674.48 Da), and horse's Apomyoglobyn (16,952.56 Da) were used as standard for calibration. Isoelectric point of glycated BSA The change in the isoelectric point of BSA after glycation was analyzed by isoelectric focusing in polyacrylamide gels using a FastSystem (Pharmacia, Uppsala, Sweden). The bands were dyed with silver nitrate, and the isoelectric point was evaluated using the analytical unit of FastSystem employing a broad range standard kit (pH 3–10) (Amersham Biosciences, Buckinghamshire, United Kingdom) as a protein calibrator. Blood sample collection and determination of glucose, insulin, triglycerides, and serum albumin After overnight fasting (12 h), the animals were killed by decapitation and blood was collected. It was spun and the serum was separated by centrifugation at 600 × g during 15 min at room temperature and stored at −70°C until needed. Serum insulin was determined using a commercial radioimmunoassay (RIA) specific for rat (Linco Research, Inc., Missouri); its sensitivity was 0.1 ng/mL and intra- and interassay coefficients of variation were 5 and 10%, respectively. Glucose concentration was assayed using an enzymatic SERA-PAKR Plus from Bayer Corporation (Sées, France). Homeostasis model assessment (HOMA) was used as an index to measure the degree of insulin resistance and was calculated by the formula [insulin (μU/mL) × glucose (mmol/L)/22.5] (Matthews et al. 1985; Pickavance et al. 1999; Nandhini et al. 2005). Triglycerides (TGs) were determined by commercially available procedures (Randox, Laboratories LTD, Antrim, United Kingdom). Serum albumin was evaluated in both groups of animals using a kit commercialized by Hycel of Mexico which uses bromocresol green (IL-test Instrumentation Laboratory Company, Lexington, MA). Samples were read using an Ilab-600 automatic system. Sample preparation and tension recording The animals were killed by decapitation, and aortas were immediately dissected and placed in an oxygenated normal Tyrode solution (mM: 140 NaCl, 5 KCl, 1 CaCl2, 1 MgCl2, 5 Hepes, and 5.5 glucose) pH 7.4. Arteries were carefully cleaned from connective and adipose tissue, taking care not to damage the endothelium. Tension measurements were made as previously described (Nava et al. 1999). Briefly, segments of about 2 to 3 mm long were cut and two 250-μm-diameter S-shaped silver wires (Medwire Corp, Mount Vernon, NY) were inserted into the lumen to measure tension developed transversely by rings of the vessel. One of the silver wires was fixed to the bottom of an in vitro chamber, and the other was attached to a tension transducer that was connected to a Grass polygraph model 79D. The chamber was filled with a Tyrode solution, and termorregulated and bubbled with carbogen (95% oxygen, 5% carbon dioxide). A basal passive tension of 2 g was applied after determination in preliminary tests that this was the optimal resting tension under our experimental conditions. Arteries were allowed to rest for 1 h and the solution was changed every 20 min. Contraction was induced twice by the addition of norepinephrine (NE) (1 μM). Arteries were washed by adding a fresh Tyrode solution to the chamber allowing the rings to return to their basal tension (2 g). The mean contraction value was considered as 100% of response. The vasodilator activity was studied by cumulative concentration–response curves to acetylcholine (Ach) (10−9−10−4 M) on precontracted aortic rings in the following experimental groups: (A) incubated only with the Tyrode solution, (B) incubated during 40 min with AGE-BSA at 10 AU/mL, (C) incubated with AGE-BSA at 40 AU/mL, (D) incubated with nonglycated BSA at a protein concentration equivalent to 10 UA/mL of AGE-BSA (0.95 mg/mL), (E) incubated with nonglycated BSA at a protein concentration equivalent to 40 AU/mL of AGE-BSA (3.81 mg/mL), and (F) incubated with L-NAME at 300 μM, in both control and MS rats. The percentage of the response in each experimental group was calculated in relation to the tension developed by the same ring during its basal contraction. Immunohistochemistry of RAGEs Aortic rings of both MS and C rats were quickly frozen in Tissue-Tek (Sakura Finetek USA, Inc., Torrance, CA). Sections were fixed with acetone and were blocked with PBS/Azide 0.02%/BSA 1% for 30 min. Subsequently, sections were incubated during 2 h at room temperature with a rabbit polyclonal antibody against RAGE (1:50; Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Primary antibodies were detected by using goat anti-rabbit FITC (Jackson ImmunoResearch Laboratories Inc. West Grove, PA), at room temperature for 60 min. Negative controls were prepared by substituting the primary antibody with an irrelevant antibody. Staining was analyzed using fluorescence microscopy. Statistical analysis Results are expressed as mean ± standard errors of the mean (SEM) from 6 to 10 different artery preparations. The percentage of contraction in each experiment was calculated, and the mean was then determined. When applicable (comparisons between two values; MS and controls), statistical analysis was done by Student's t-test. Comparisons between groups were done by analysis of variance (ANOVA) or Anova on ranks followed by Student–Newman–Keuls or Dunn's tests, depending on whether the data were normally distributed or not, using the SigmaStat 2.0 program (Jandel Scientific, San Rafael, CA). Differences were considered statistically significant when P < 0.05.

Biochim_Biophys_Acta-2-1-2428106

Regulation of intestinal hPepT1 (SLC15A1) activity by phosphodiesterase inhibitors is via inhibition of NHE3 (SLC9A3)

The H+-coupled transporter hPepT1 (SLC15A1) mediates the transport of di/tripeptides and many orally-active drugs across the brush-border membrane of the small intestinal epithelium. Incubation of Caco-2 cell monolayers (15 min) with the dietary phosphodiesterase inhibitors caffeine and theophylline inhibited Gly–Sar uptake across the apical membrane. Pentoxifylline, a phosphodiesterase inhibitor given orally to treat intermittent claudication, also decreased Gly–Sar uptake through a reduction in capacity (Vmax) without any effect on affinity (Km). The reduction in dipeptide transport was dependent upon both extracellular Na+ and apical pH but was not observed in the presence of the selective Na+/H+ exchanger NHE3 (SLC9A3) inhibitor S1611. Measurement of intracellular pH confirmed that caffeine was not directly inhibiting hPepT1 but rather having an indirect effect through inhibition of NHE3 activity. NHE3 maintains the H+-electrochemical gradient which, in turn, acts as the driving force for H+-coupled solute transport. Uptake of β-alanine, a substrate for the H+-coupled amino acid transporter hPAT1 (SLC36A1), was also inhibited by caffeine. The regulation of NHE3 by non-nutrient components of diet or orally-delivered drugs may alter the function of any solute carrier dependent upon the H+-electrochemical gradient and may, therefore, be a site for both nutrient–drug and drug–drug interactions in the small intestine. 1 Introduction The human intestinal H+-coupled di/tripeptide transporter (hPepT1 or SLC15A1) is responsible for the transport of a significant proportion of dietary protein across the brush-border membrane of the small intestinal epithelium. hPepT1 is also of pharmacological interest as it can transport orally-active drugs such as β-lactam antibiotics and angiotensin converting enzyme inhibitors [1–4]. More recently a number of amino acid modified pro-drugs such as val-acyclovir, l-DOPA-l-Phe and the anti-hypotensive midodrine have been designed specifically to target hPepT1 and thus to enhance oral bioavailability [3–5]. When expressed heterologously, hPepT1 functions as a Na+-independent, pH-dependent, H+-coupled transporter [6,7]. However, until the mid-1980s, it was considered widely that di/tripeptide transport was via a Na+-coupled transport mechanism. Indeed, dipeptides can stimulate Na+ absorption in the human small intestine and Na+ uptake in human intestinal Caco-2 cell monolayers [8–10]. Ganapathy and Leibach proposed a model to account for the interaction of H+-coupled di/tripeptide transport and Na+ absorption [11] whereby the intracellular acidification caused by di/tripeptide associated H+ influx activated an apical Na+/H+ exchanger and thus Na+ absorption. This model also explains the partial Na+-dependence of di/tripeptide uptake observed in mammalian small intestine and Caco-2 cell monolayers [11,12]. Inhibition of Na+/H+ exchange by removal of extracellular Na+, prevents the enterocyte from maintaining intracellular pH (pHi) during H+/solute influx and, therefore, abolishes the driving force (the transmembrane H+ electrochemical gradient) for further transport. The apical Na+/H+ exchanger specifically activated by di/tripeptide uptake is NHE3 (SLC9A3) which plays a significant role in Na+ absorption in the small intestine [7,10,13]. As hPepT1 is involved in both protein assimilation and drug delivery it is important to elucidate how altering hPepT1 function may affect nutritional status and oral bioavailability. Progress has been made in understanding how intestinal di/tripeptide transport is regulated by hormonal and neural signals as well as by disease states and surgical intervention (for review see [14]). However, little is known about how co-administered drugs or dietary factors affect hPepT1 function or the function of other brush-border membrane transport proteins. Alteration of cAMP levels within enterocytes by hormonal, neural and paracrine signals is a key step in the regulation of many brush-border membrane nutrient and electrolyte transporters. NHE3 is inhibited acutely by increasing [cAMP]i [15]. Therefore, the aim of this study was to identify how dietary and orally-delivered compounds, which act as phosphodiesterase inhibitors (and thus increase [cAMP]i by preventing breakdown), affect hPepT1 function. The dietary phosphodiesterase inhibitors caffeine and theophylline and the orally-delivered drug pentoxifylline all inhibit dipeptide uptake across the brush-border membrane of human intestinal epithelial Caco-2 cell monolayers. This inhibition is not a direct effect on hPepT1 but rather is indirect through inhibition of NHE3. 2 Materials and methods 2.1 Materials Glycyl[1-14C]sarcosine (specific activity 57 mCi mmol− 1)] was obtained from Cambridge Research Biochemicals (Billingham, UK). [3-3H]β-Alanine (specific activity 50 Ci mmol− 1)] was obtained from American Radiolabeled Chemicals (St. Louis, MO, USA). PACAP (pituitary adenylate cyclase-activating polypeptide) was from Bachem (St. Helens, UK). Biotrak cAMP enzyme immunoassay was from GE Healthcare (Chalfont St. Giles, UK). 2′,7′-Bis(2-carboxyethyl-5(6)-carboxyfluorescein) acetoxymethyl ester (BCECF-AM) was from Invitrogen (Paisley, UK). Transwell polycarbonate filters were from Corning (Schiphol-Rijk, The Netherlands). S1611 was obtained from H.J. Lang (Aventis Pharma Deutschland GmbH, Frankfurt/Main, Germany). Forskolin, phosphodiesterase inhibitors, cell culture media and supplements were from Sigma-Aldrich (Poole, UK). All other chemicals were from Sigma-Aldrich or VWR (Lutterworth, UK) and were of the highest quality available. 2.2 Cell culture Caco-2 cells (passages 100–119) were cultured as confluent monolayers on permeable polycarbonate filters, as described previously [16]. Monolayers were used at 14–22 days post-seeding and fed approximately 24 h prior to use. 2.3 Measurement of dipeptide uptake [14C]Gly–Sar (0.5 μCi ml− 1, 0.1 – 10 mM) or [3H]β-alanine (0.5 μCi ml− 1, 0.1 mM) uptake across the apical membrane of Caco-2 cell monolayers was measured, essentially as described previously [16]. Briefly, Caco-2 cells were washed (in 4 × 500 ml) and bathed in modified Krebs' solution (composition (mM): NaCl, 137; KCl, 5.4; MgSO4, 0.99; KH2PO4, 0.34; NaH2PO4, 0.3; CaCl2, 2.8; Glucose, 10) or Na+-free solution (identical except choline chloride replaced NaCl and NaH2PO4 was omitted). Uptake was measured at apical pH 5.5 or pH 6.5 (10 mM MES, adjusted to the correct pH by Tris) with the basolateral solution being pH 7.4 (10 mM HEPES, adjusted to the correct pH by Tris) in all experiments. Uptake was measured over 15 min (37 °C) at the end of which the monolayers were rinsed thoroughly (in 3 × 500 ml, ice-cold, pH 7.4 Krebs' solution). Various compounds were added to the buffers for the duration of uptake (see figure legends for details). Cell monolayer-associated radioactivity was determined by scintillation counting. 2.4 Measurement of pHi Intracellular pH (pHi) was measured in Caco-2 cell monolayers by microspectrofluorimetry using the pH-sensitive dye BCECF, essentially as described previously [10,17]. Cell monolayers were acidified by superfusion with apical Gly–Sar (20 mM, 5 min) in modified Krebs' solution (Na+-free, apical pH 5.5). The basolateral solution was Na+-free, pH 7.4 throughout all experiments. pHi recovery was then measured in Na+-containing pH 7.4 apical solution until pHi returned to baseline. The monolayers were incubated with caffeine (5 mM, apical and basolateral) for 10 min. The cells were then exposed to Gly–Sar once again but in the continued presence of caffeine (thus total exposure time to the compound before pHi recovery was 15 min). pHi recovery was then measured in the continued presence of caffeine. The initial rate of pHi recovery is expressed as H+ efflux calculated from the gradient of the first 30 s of recovery [18]. 2.5 Measurement of intracellular cAMP levels Caco-2 cell monolayers were washed in 4 × 500 ml Krebs' solution (pH 7.4). Monolayers were incubated for 15 min in the presence or absence of caffeine, pentoxifylline, theophylline (all 5 mM), or forskolin (10 μM) at apical pH 6.5 and basolateral pH 7.4. Monolayers were then washed in 3 × 500 ml, ice-cold Krebs' (pH 7.4) and lysed in 500 μl lysis buffer (from the Biotrak enzyme immunoassay kit) for 10 min. [cAMP]i was measured following the kit protocol. 2.6 Statistical analysis Data are expressed as mean ± SEM unless stated otherwise. Statistical comparisons of mean values were made using paired two-tailed Student's t-test or one-way analysis of variance (ANOVA) (using the Tukey–Kramer or Bonferroni's multiple comparisons post-test) as appropriate. Significance was assumed if p < 0.05. Curves were fitted using FigP software. 3 Results 3.1 Inhibition of dipeptide uptake by caffeine is via inhibition of NHE3 Uptake of the dipeptide Gly–Sar (glycylsarcosine) across the apical membrane of Caco-2 cell monolayers was measured over 15 min at an apical pH (pH 6.5) representative of that found at the surface of the proximal small intestine. Incubation of Caco-2 cell monolayers with the dietary phosphodiesterase inhibitor caffeine (5 mM) at both the apical and basolateral surfaces (for the duration of the 15 min uptake period) reduced Gly–Sar uptake by around 50% (from 683.6 ± 18.9 to 351.1 ± 10.7 pmol cm− 2 [15 min]− 1, n = 17–18, p < 0.001, Fig. 1A). Caffeine was equally effective at inhibiting Gly–Sar uptake when added solely to either the apical or basolateral chamber (data not shown). Removal of extracellular Na+ reduces Gly–Sar uptake (as described previously [12]) (Fig. 1A). However, in the absence of extracellular Na+, caffeine had no effect on Gly–Sar uptake (uptake being 244.0 ± 9.1 and 242.0 ± 9.5 pmol cm− 2 [15 min]− 1 in the absence and presence of caffeine respectively, n = 17–18, p > 0.05). Previously we have shown that the Na+-dependence of Gly–Sar uptake can be attributed to a lack of NHE3 activity in the absence of extracellular Na+ [10,12,18]. To test if the lack of effect of caffeine in the absence of extracellular Na+ was due to a requirement for NHE3 activity, dipeptide uptake was measured under other conditions where NHE3 would be inactive (Fig. 1B). At apical pH 6.5, the NHE3-selective inhibitor S1611 (3 μM) significantly reduced Gly–Sar uptake in a manner similar to that caused by the removal of extracellular Na+ (p < 0.001). This concentration of S1611 is sufficient to inhibit NHE3 completely but would not inhibit NHE2 or NHE1 [19]. Previous work has shown that in the absence of Na+, S1611 has no effect on Gly–Sar uptake into Caco-2 cells [12]. In addition, when hPepT1 is expressed in isolation in Xenopus laevis oocytes, dipeptide uptake is not inhibited by S1611 [7]. This observation confirms that the inhibition of dipeptide uptake by S1611 into Caco-2 cells is an indirect effect via inhibition of NHE3. In the presence of S1611, Gly–Sar uptake was no longer inhibited by caffeine (5 mM). The level of inhibition caused by caffeine alone, by S1611 alone and by both compounds together were not significantly different (p > 0.05) suggesting that caffeine and S1611 inhibit hPepT1 activity through a common mechanism (i.e. NHE3 inhibition). Gly–Sar uptake increases with decreasing apical pH, consistent with hPepT1 being a H+/dipeptide symporter. However, NHE3 activity decreases with acidification of the apical solution and is inactive by pH 5.5 due to the unfavourable transmembrane pH gradient [12]. Previous studies have shown that Gly–Sar uptake at apical pH 5.5 is not inhibited by S1611 or by the removal of extracellular Na+ [7,12]. Similarly, Fig. 1B shows that caffeine (5 mM) has no effect on Gly–Sar uptake at apical pH 5.5 (p > 0.05) suggesting again that caffeine is inhibiting Gly–Sar uptake through inhibition of NHE3. 3.2 The inhibition of hPepT1 by caffeine and related compounds is consistent with inhibition of phosphodiesterase activity Phosphodiesterase inhibitors structurally-related to caffeine and either found in diet (theophylline), or used as orally-delivered therapeutics (pentoxifylline) or used as laboratory tools (IBMX, 3-isobutyl-1-methylxanthine) were tested for inhibition of hPepT1 (Fig. 2). Theophylline (5 mM, Fig. 2A), pentoxifylline (5 mM, Fig. 2B) and IBMX (1 mM, Fig. 2C) reduced dipeptide uptake in the presence of extracellular Na+ (p < 0.001) but not in Na+-free conditions (p > 0.05). Previously, we have shown that the enteric neuropeptide PACAP (pituitary adenylate cyclase-activating polypeptide) and the related neuropeptide VIP (vasoactive intestinal peptide) reduce dipeptide uptake across the apical membrane of Caco-2 cell monolayers by inhibition of NHE3 [18]. PACAP activates the VPAC1 receptor which is coupled to adenylate cyclase and, therefore, increases [cAMP]i. Fig. 2C shows that when Caco-2 cell monolayers were incubated with PACAP (5 nM) in the basolateral solution for the duration of uptake, Gly–Sar uptake was reduced. Incubation of the cells with both PACAP and IBMX resulted in a reduction in uptake which was not significantly different from the uptake in the presence of either PACAP or IBMX alone suggesting IBMX, like PACAP, is inhibiting the NHE3-dependent component of Gly–Sar uptake. Phosphodiesterase inhibitors increase the levels of cAMP within cells by preventing cAMP breakdown. However, caffeine can also elicit cell signalling effects such as antagonism of adenosine receptors and activation of calcium release from intracellular stores [20–22]. To confirm that the compounds used in this study inhibit hPepT1 function through increasing cAMP levels, [cAMP]i in Caco-2 cell monolayers was measured. Monolayers were incubated with caffeine (5 mM), theophylline (5 mM), pentoxifylline (5 mM) or the adenylate cyclase activator forskolin (10 μM) which can also inhibit dipeptide uptake [12]. Cells were lysed and [cAMP]i measured by enzyme immunoassay. Caffeine, theophylline and pentoxifylline caused [cAMP]i to more than double from control levels of 950 ± 230 fmol cm− 2 (mean ± SD, n = 3) to 2200 ± 830, 2780 ±300 and 2330 ± 150 fmol cm− 2 (mean ± SD, n = 3), respectively. However, the change in cAMP was modest compared to [cAMP]i detected in the presence of forskolin (92500 ±12990 fmol cm− 2 [mean ± SD, n = 3]). Inhibition of the Na+-dependent (NHE3-dependent) component of Gly–Sar uptake by phosphodiesterase inhibitors was concentration-dependent (Fig. 3). The IC50 values were 1.0 ±0.2 mM for caffeine, 0.6 ± 0.1 mM for theophylline and 0.5 ±0.1 mM for pentoxifylline. The sensitivity of hPepT1 to these compounds is consistent with the effect being through an inhibition of phosphodiesterase activity [20,21,23–25]. At a concentration of 5 mM all three phosphodiesterase inhibitors almost completely inhibited Na+-dependent Gly–Sar uptake. 3.3 Pentoxifylline decreases the maximal capacity for dipeptide uptake Gly–Sar uptake across the apical membrane was measured at apical pH 6.5 at various concentrations of Gly–Sar (Fig. 4). In the presence of pentoxifylline (5 mM) Gly–Sar uptake was reduced compared to control at all Gly–Sar concentrations. Analysis of the Michaelis–Menten kinetics confirmed that the maximal capacity (Vmax) for Gly–Sar uptake was significantly reduced from 8246 ± 913 to 5369 ± 117 pmol cm− 2 [15 min]− 1 (p < 0.05, n = 6) by pentoxifylline but the affinity (Km) was not affected (1.0 ± 0.2 mM and 1.6 ± 0.6 mM in the absence and presence of pentoxifylline, respectively, p > 0.05). Reduction in the capacity for dipeptide uptake is consistent with a reduction in the driving force (the H+-electrochemical gradient) for uptake through inhibition of NHE3. Previously, we have shown that S1611 and forskolin similarly decrease the Vmax for dipeptide uptake across the apical membrane of Caco-2 cell monolayers [12]. 3.4 Caffeine inhibits pHi recovery (NHE3 function) after dipeptide-induced acidification Previous work demonstrated that NHE3 (but not NHE1 or NHE2) is selectively activated by the intracellular acidification associated with H+/dipeptide symport in Caco-2 cell monolayers [10,12]. S1611, the removal of extracellular Na+, or factors that increase [cAMP]i, such as forskolin and VIP, all inhibit the ability of the cell to regulate pHi after H+/solute-induced intracellular acidification [10,12,18,26]. pHi was measured using the pH-sensitive dye BCECF (Fig. 5). Caco-2 cell monolayers loaded with BCECF were superfused across the apical membrane with Na+-containing, pH 7.4 modified Krebs' solution. The basolateral solution was maintained as Na+-free, pH 7.4 solution throughout the experiment. Cells were exposed to Gly–Sar (20 mM) under conditions where hPepT1 activity would be maximal but where there would be no concurrent NHE3 activity (pH 5.5, Na+-free). After 5 min, Gly–Sar had induced a large intracellular acidification. When Gly–Sar was removed and the apical solution returned to a Na+-containing, pH 7.4 solution, pHi recovered rapidly back to baseline levels. The same monolayers were then incubated with caffeine (5 mM) in both the apical and basolateral solutions for 10 min. Gly–Sar was then reintroduced for 5 min in the continued presence of caffeine. The degree and rate of acidification caused by Gly–Sar was not altered by caffeine suggesting that hPepT1 is not directly regulated by phosphodiesterase inhibitors. However, the ability of the cell to recover pHi after Gly–Sar-induced acidification was attenuated after the 15 min exposure to caffeine (Fig. 5A). The degree of attenuation was quantified by calculating the H+ efflux rate over the first 30 s of pHi recovery (Fig. 5B). In the presence of caffeine H+ efflux was reduced from 57.6 ± 5.8 to 20.5 ± 3.9 μM s− 1 (n = 4, p < 0.01). To confirm that phosphodiesterase inhibitors do not have any direct inhibitory effect on dipeptide transport, hPepT1 was expressed in Xenopus laevis oocytes, as described previously [7]. Incubation of hPepT1-expressing oocytes with phosphodiesterase inhibitors such as caffeine (5 mM) had no effect on dipeptide uptake (data not shown). 3.5 Inhibition of NHE3 by caffeine also inhibits the H+/amino acid transporter hPAT1 (SLC36A1) The indirect nature of the inhibition of hPepT1 by phosphodiesterase inhibitors suggests that any apical solute transporter dependent upon the transmembrane H+-electrochemical gradient will similarly be regulated by these compounds. The H+-coupled amino acid transporter hPAT1 (SLC36A1) has been isolated from Caco-2 cell monolayers [27]. As well as mediating the uptake of a wide variety of amino acids, hPAT1 can also transport orally-active drugs such as the anti-epileptic vigabatrin [28]. Previously we have identified that amino acid uptake into hPAT1-expressing oocytes is Na+-independent but hPAT1-mediated amino acid uptake into Caco-2 cells is partially Na+-dependent [26,29,30]. Intracellular acidification caused by the hPAT1 substrate β-alanine selectively activated Na+/H+ exchange by NHE3 [26]. Like H+-coupled dipeptide uptake, H+-coupled amino acid uptake into Caco-2 cells is inhibited by forskolin, S1611 and VIP in a Na+ and pH-dependent manner via inhibition of NHE3 [26,29,30]. Uptake of the hPAT1 substrate β-alanine [16] was measured across the apical membrane of Caco-2 cell monolayers at apical pH 6.5 for 15 min (Fig. 6). Caffeine (5 mM) reduced β-alanine uptake in the presence (p < 0.001) but not the absence of extracellular Na+ (p > 0.05) suggesting that H+-coupled amino acid uptake via hPAT1 is also modulated indirectly through regulation of NHE3. 4 Discussion The di/tripeptide transporter hPepT1 acts as a high-capacity route for solutes across the first barrier to oral-bioavailability, the brush-border membrane of the small intestine. Many, orally-active peptidomimetics and amino acid-conjugated pro-drugs have been identified as hPepT1 substrates [3,4]. There is an increasing number of examples of physiological regulation (hormonal, neural, paracrine) of hPepT1 and of regulation of hPepT1 in certain disease states and after surgery (reviewed by [14]). Another, less studied, factor which may affect the degree to which drugs are absorbed across the small intestinal epithelium is interaction with co-administered drugs or components of diet. Exposure of Caco-2 cell monolayers to the hPepT1 substrate Gly–Gln for 4 days resulted in a subsequent increase in capacity for dipeptide uptake and in hPepT1 expression [31]. Another study found that an array of flavonoids, which are found ubiquitously in foods of plant origin, either inhibit, have no effect or increase the hPepT1-mediated uptake of the antibiotic cefixime into Caco-2 cell monolayers [32]. In this study we identify that incubation of human intestinal epithelial cells with either dietary or orally-active therapeutic phosphodiesterase inhibitors reduces Gly–Sar uptake through a reduction in hPepT1 capacity. The data presented here show that the inhibition of Gly–Sar uptake by phosphodiesterase inhibitors is both Na+- and pH-dependent (Figs. 1 and 2) suggesting that inhibition is not a direct effect on hPepT1 but rather through NHE3. When NHE3 is inhibited (e.g. by the removal of extracellular Na+ or by addition of S1611) the cells are no longer able to maintain pHi during solute-induced acidification and, therefore, the driving force (the transmembrane H+ electrochemical gradient) for further dipeptide uptake is reduced. Previously, we have shown that hPepT1 can be inhibited by other factors which are known to increase cAMP in intestinal epithelial cells such as the enteric neuropeptides VIP and PACAP [18]. Although caffeine, theophylline and pentoxifylline can elicit effects through pathways other than increasing cAMP, a number of factors suggest that they are acting here as phosphodiesterase inhibitors. Firstly, incubating Caco-2 cell monolayers with all three compounds produced an increase in [cAMP]i. The increase is relatively small compared to that produced by forskolin. However, this could be due to the cAMP signal being compartmentalised (as demonstrated in cardiomyocytes [33]) and thus the local change may be much greater than the measured, global change. Secondly, phosphodiesterases can also regulate the levels of cGMP but none of the compounds tested here produced a significant change in [cGMP]i (data not shown). Lastly, the concentrations of caffeine, theophylline and pentoxifylline required to produce inhibition of hPepT1 (IC50 = 1.0, 0.6 and 0.5 mM, respectively, Fig. 3) are much closer to those for inhibition of phosphodiesterases than for other effects such as antagonism of adenosine receptors (IC50≈ 2–80 μM) and release of intracellular calcium (EC50≈ 5–20 mM) [20–22,34–36]. Theophylline (100 μM) has been shown to abolish increased glucose absorption resulting from activation of A2 receptors by luminal adenosine (5 mM) in mouse small intestine in vivo [37]. However, in addition to the significantly higher affinity for theophylline, A2 receptors are positively coupled to adenylate cyclase so antagonism of the receptor should cause a decrease in [cAMP]i and not the increase noted here [22,37]. Caffeine (and to a lesser extent theophylline) is found in high levels in many beverages either naturally or by addition. For example, the average cup of coffee contains 85 mg of caffeine [38]. Pentoxifylline is given orally as 400 mg tablets to treat intermittent claudication as part of peripheral arterial disease [39]. Although it is difficult to predict the actual concentration at the surface of the small intestinal epithelium it is conceivable that caffeine and pentoxifylline may reach concentrations high enough to elicit an effect on hPepT1 similar to that measured here using Caco-2 cell monolayers. There is evidence that caffeine given orally in doses equivalent to diet can elicit changes in small intestinal function. A study using human volunteers found that ingestion or intraluminal jejunal perfusion of caffeine (75–300 mg) caused a rapid reversal of net fluid absorption to net secretion [40]. Although the cell signalling pathway by which caffeine stimulated fluid secretion was not investigated, this study does indicate that the concentrations of caffeine used here in vitro may be sufficient to cause changes in the small intestinal epithelium in vivo. Since the initial reports of phosphodiesterase activity it has become evident that there are many types of phosphodiesterase. Currently there are 11 different PDE families each of which can contain several isoforms and splice variants [24,41]. Many of these isoforms are tissue or cell-specific and recent studies have shown that they can be expressed in distinct sub-cellular locations where they act to limit site, amplitude and duration of cyclic nucleotide signals [24,33]. Little is known about which PDE isoforms are found in the small intestinal epithelium. Studies reporting tissue distribution of PDE mRNA expression often include a small intestinal sample but generally it is not clear whether the PDE is expressed specifically in enterocytes. A single study of PDE3 isoforms identified that PDE3B but not PDE3A mRNA was highly abundant in the epithelium and underlying smooth muscle layers of developing small intestine of rat [42]. Caffeine, theophylline, IBMX and pentoxifylline generally inhibit all PDE isoforms with roughly equal efficacy [25,43,44]. However, since many PDE isoforms and variants are located in distinctive cell and sub-cellular locations they are now considered to be good targets for specific therapeutic agents. Often the resulting drugs are given orally, for example, the PDE5 inhibitor sildenafil used to treat erectile dysfunction [45]. PDE5 is expressed at the mRNA level in the small intestine [46,47]. Further work is required to identify which PDE isoforms are expressed in small intestinal epithelial cells both along the proximal–distal and crypt–villus axes and which isoforms play a functional role in regulating solute transport at the brush-border membrane. In conclusion, we have shown that non-nutrient components of diet and orally-delivered therapeutic agents both regulate activity of the nutrient and drug transporter hPepT1. This regulation is an indirect effect through inhibition of apical NHE3 function, which acts as a pH homeostatic mechanism during H+-coupled dipeptide transport. Functional cooperativity between hPepT1 and NHE3 results not only in the absorption of di/tripeptides but also stimulation of Na+ absorption [8–10]. The H+-coupled amino acid transporter hPAT1 (SLC36A1) is regulated by caffeine in a similar, indirect manner to hPepT1 (Fig. 6). Thus, through modulation of a single homeostatic mechanism (NHE3), a luminal compound may influence several different absorptive mechanisms. The results presented here suggest that any regulation of NHE3 activity by luminal contents, co-administered drugs or by neurohormonal signals may have significant impact on the general absorptive state of the cell. Any other nutrient transporter dependent on the maintenance of the H+ electrochemical gradient for function may be regulated in a similar manner.

J_Affect_Disord-1-5-1894757

Outcome of prenatal depression and risk factors associated with persistence in the first postnatal year: Prospective study from Rawalpindi, Pakistan

Background Rates of prenatal and postnatal depression in developing countries are high. Prolonged depression during the postnatal period is associated with impaired infant growth and development. Little is known about the factors predicting the persistence of prenatal depression beyond the first few postnatal months. 1 Introduction The rate of postnatal depression in developing countries ranges from 16% to 35% (Ghubash and Abou-Saleh, 1997; Patel et al., 2002; Aydin et al., 2005; Cooper et al., 1999) and is a major contributor to the ‘burden of disease’ in these countries. Depression around childbirth is associated with low birth weight and impaired weight gain in the first year of the infant's life (Rahman et al., 2004). The outcome is worse in infants whose mothers remained persistently depressed from the third trimester throughout the first postnatal year. It is, therefore, important to predict which mothers, depressed prenatally, are likely to remain depressed throughout the next year. Relatively fewer studies have examined the long-term outcome of postnatal depression. A recent review of studies from developed countries concluded that in about 30% women with postnatal depression, symptoms persist for up to a year after giving birth (Goodman, 2004). There are no studies exploring the long-term outcome of postnatal depression in developing countries. Such longitudinal studies identify the course of depressive disorder and the factors that promote persistence thus helping to formulate effective preventive and treatment strategies (Weich and Araya, 2004). They can also help in targeting limited health resources in developing countries towards those at the greatest risk of poor outcome. In an earlier paper (Rahman et al., 2003a,b), we assessed 632 women in rural Pakistan and found that 160 (25%) of the sample had ICD-10 Depressive disorder. In this paper, we present the results of the one-year follow-up of those mothers who were depressed in the third trimester of pregnancy. 2 Method 2.1 Study area, subjects and sampling The study was carried out in a rural sub-district of Rawalpindi, Pakistan. This is a mainly agrarian low-income rural area about 60 km south-east from the city of Rawalpindi. All married women aged 17 to 40 in their third trimester of pregnancy (n = 701) were identified from 10 Union Councils (each consisting of 5–10 villages; total population 150,000) over a period of 4 months (Fig. 1). Subjects were identified by obtaining official lists from 120 government-employed Lady Health Workers (LHWs) working in the area, who routinely collect data on new pregnancies. Six-hundred and seventy out of 701 (95%) agreed to take part. Written informed consent was obtained from all subjects after the procedure had been fully explained. Fourteen (2%) were excluded because of a physical illness or complication of pregnancy, 21 (3%) had anxiety disorder and 3 (0.5%) had learning disability and were excluded. Out of the remaining, 160 were diagnosed with ICD-10 Depressive Episode, giving a prevalence rate of depressive disorder in the prenatal period (T1) of 25%. Out of these 160, four had infants born prematurely and were excluded from this study. Two mothers discontinued due to severity of depression. Ten mothers had stillbirths or neonatal deaths, and 1 newborn had a congenital abnormality and were excluded. Fourteen subjects dropped out of the study. Thus, one hundred and twenty-9 mothers were assessed at 3 months (T2), 6 months (T3) and 12 months (T4) postnatal. 2.2 Data collection Mental state assessments were carried out at all time points by two trained and experienced clinicians using the Schedules for Clinical Assessment in Neuropsychiatry (SCAN), developed by the World Health Organization as an internationally validated semi-structured interview generating ICD-10 diagnoses of Depressive Disorder (World Health Organization, 1994a). All interviews were carried out after translation, back-translation and cultural adaptation of the interview schedule using an established procedure (Rahman et al., 2003a,b). Interrater reliability was established prior to the study when both interviewers independently assessed 20 women (10 had clinical depression) and agreed on the diagnosis of 19 (κ = 0.90). Psychological symptoms were assessed at T1 by the same interviewers using the Self-Reporting Questionnaire (SRQ-20) (World Health Organization, 1994b). This consists of twenty items designed to identify psychological symptoms associated with anxiety and depression. Each item has a yes/no answer. The time span refers to the individual's feelings over the past 30 days. Each item is scored 0 or 1. The maximum score is therefore 20. Disability in mothers was assessed at T1 using the Brief Disability Questionnaire (BDQ) (VonKorff et al., 1996). This is an 8-item questionnaire that rates current problems in carrying out daily activities on a scale of 0 (not at all) to 2 (definitely), with a maximum score of 16. This instrument has been validated in a 15-center cross-national, multilingual study (VonKorff et al., 1996). Socio-demographic variables (age, education, employment, family structure and composition) were assessed at T1 by the same interviewers using a specially designed Personal Information Questionnaire (PIQ). Education was categorised into no education versus at least four years primary education. Four years of schooling was chosen as a cut-off because many Pakistani female children attend primary school for four years, after which many children from low-income families stop attending. Family structure was categorised into nuclear family (parents and children only) or extended family (three generations, or one or both parents with married sons, their wives and children). Socioeconomic status was assessed at T1 by inquiring if the household was in debt and by asking Lady Health Workers, who lived in the same locality and had intimate knowledge of the families being studied, to rate the household on a 5-point Likert scale ranging from 1 (richest) to 5 (poorest). A single dichotomous variable of ‘poverty’ was created by combining these 2 measures, i.e., subjects who were both in debt and rated below 3 on the socioeconomic 5-point Likert scale were classified as being poor. Maternal financial empowerment within the household was measured at T1 by asking the mothers if they were given a lump-sum amount of money for day-to-day household expenses, and whether they could take independent decisions about its use. Mothers who answered ‘yes’ to both questions were classified as financially empowered within the household. Social support was assessed by inquiring if the woman received any support during pregnancy from relatives or friends. A brief list of life events and difficulties was administered at T1. These were derived from the Life Events and Difficulties Schedule (LEDS) (Brown and Harris, 1989), a semi-structured instrument that measures events and difficulties experienced during the previous year. LEDS has been translated and culturally adapted for use in the study area (Husain et al., 2000). Based on the data from this study, 9 types of events or difficulties that accounted for the majority of the severe events and difficulties reported in that population were used in a modified semi-structured interview. We recorded only those events and difficulties which rated as severe according to the Brown and Harris Rating Scale. In order to do this we discussed the context in which they occurred with the local lady health worker (who lived in the same community and had intimate knowledge of the families being studied). 2.3 Statistical analysis All analyses were carried out with STATA, version 7 (StataCorp., 2001) Subjects who were depressed at all 4 time points (persistent cases) were compared with the rest (non-persistent cases). T-test was used to compare psychological symptoms, disability and life event scores. Univariate analyses (relative risk, Fisher's two sided exact p) was performed between potential risk factors and chronic depression. Associations were considered significant at the 5% level. The simultaneous effects of the measured risk and protective factors on persistent depression were analyzed using logistic regression analysis, including all the variables studied in the model (listed in Table 1). The study was approved by the ethics committees of Rawalpindi Medical College, Pakistan and University of Manchester, UK. 3 Results 3.1 Sample characteristics 129 depressed women completed the one-year follow-up. Their average age was 27.5 years (SD = 5.3). All were married, the average age of marriage being 20 years (SD = 2.8). Forty-four percent were uneducated; only 4% were employed outside the home. Eighty-eight percent of the fathers were employed and about 22% of them were absent from home for 6 months or more due to employment in the cities. The average reported monthly family income was 2500 rupees (US$42). Sixty-seven percent of the families were rated below 3 on the 5-point socioeconomic scale by the LHWs and 53% of the households were in debt. Sixty-two (48%) were in debt and rated below 3 by the LHWs and were classified as ‘poor’ in this study. Nine percent were primigravid, 16% already had one child, 21% two children, 17% three children and the remaining 37% had three or more. Thirty-seven percent lived in nuclear families (parents and children only) while the remaining lived in extended families (three generations, or one or both parents with married sons, their wives and children). Fifty-seven percent delivered at home with a traditional birth attendant, 98% without any reported complication. The gender of newborns was equally distributed. Of the 129 women, 121 (94%) were depressed at 3 months, 98 (76%) at 6 months and 80 (62%) at 12 months. Eighty out of 129 (62%) mothers depressed during the third trimester of pregnancy were still depressed at 12 months postnatally but 7 of these had not been depressed at 6 months; thus 73 (57%) were depressed at all time points. A comparison of the 73 women who had persistent depressive disorder with the remainder on scores of the Self-Reporting Questionnaire, Brief Disability Questionnaire and Life Events Checklist is shown in Table 1. Persistent depression was significantly associated with SRQ score and BDQ score, and weakly associated with Life Events score. Unadjusted relative risks with other factors are shown in Tables 2 and 3. Persistent depressive disorder was associated with having an uneducated husband, family size (5 or more children), poverty and lack of a confiding relationship. Independent predictors of persistent depression selected by multiple logistic regression were: high score on the SRQ in the third trimester of pregnancy (odds ratio (OR) 1.3, 95% CI 1.1 to 1.6, p < 0.01) and poverty (OR 3.1, 95% CI 1.2–8.4, p > 0.05) (Table 4). 4 Discussion To our knowledge, this is the first study from the developing world exploring the course of prenatal depression and factors predicting its persistence during the first year of the newborn's life. The study was community based and used standardised and valid instruments to diagnose depression. Depression around childbirth is a serious public health problem in south Asia, affecting about one in four women (Patel et al., 2002; Rahman et al., 2003a,b). Recent studies also provide strong evidence that maternal depression is associated with poor growth in infants living in poor communities in developing countries (Rahman et al., 2004; Patel et al., 2004), and the outcome is worse in infants of mothers with persistent depression. In developed countries, maternal depression is associated with long-term cognitive, emotional and behavioural problems in children, and the impact is worse where the depressive episode is severe or prolonged (Grace et al., 2003; O'Connor et al., 2002). Therefore, prolonged maternal depression has serious consequences not only for the mother but also for infant growth and development. The main findings of this study are that over half of mothers depressed in the third trimester of pregnancy continued to be depressed one year after giving birth. In similar studies from developed countries, depressive levels have been found to decrease steadily over time. In an early prospective study from USA, O'Hara et al. (1984) followed up 99 women from second trimester of pregnancy to 6 months after giving birth, and found that although almost one half of the subjects had depressive scores that would place them in the mildly depressed range during the second trimester, less than 12% of the subjects were in the mildly depressed range at the 9-week and 6-month follow-ups. In a more recent American study, Campbell and Cohn (1997) followed up 70 women meeting the criteria for clinical depression at 2 months postnatal and found that at 4 months postnatal, 48% continued to be depressed; at 6 months 30% and at 12 months, 24% continued to meet the criteria for depression. Beeghly et al. (2002) followed up 106 women with high depression scores at 2 months postnatal and found that 35% and 31% continued to have high scores at 6 and 12 months respectively. Rubertsson et al. (2005) followed up a national cohort of Swedish women and found that out of 333 women with high depression scores in early pregnancy, 79 (24%) continued to have high scores at one year postpartum. A follow-up study of an Australian cohort from early pregnancy till five years after birth suggested that in the majority of women who experienced depressed mood after birth, the symptoms were not severe and did not continue beyond a few weeks (Najman et al., 2000). In our study, prevalence of antenatal depression is high and comparatively fewer women recover in the first year after childbirth. The low rate of recovery could be due to the adverse circumstances experienced by many women in developing countries. Persistent depression was associated with several factors that preceded the birth: poverty, already having 5 or more children, an uneducated husband, and lack of a friend or confidant. Similar factors have been reported in studies from poor communities in developed countries. Horowitz and Goodman (2004) found that women still depressed two years after giving birth were more likely to be poor and have less social support. Bernazzani et al. (1997) found that lower occupational status, lower income, prenatal depression and stressful life events in 12 months prepregnancy were associated with depression persisting at 6 months postnatal. Yonkers et al. (2001) found that persistent postnatal depressive symptoms were linked with the presence of other young children at home. In the current study, multivariate analyses suggested poverty to be the major predictor of persistence of depression after severity of depressive symptoms in the prenatal period was adjusted for. The gender of the newborn did not predict persistence. In south Asia, giving birth to a female infant was found to be associated with postnatal depression (Patel et al., 2002), especially in mothers who already had more than two girl-children (Rahman et al., 2003a,b). The preference for male children is deeply rooted in South Asia. Women are often blamed for the birth of girls. However, this study suggests that the effects of having a female child on the mother's mood may only be transient. Similarly, lack of social support during pregnancy (in terms of assistance provided in daily activities) was not associated with persistence while lack of a confidant or friend was. It may be that there are qualitative differences in the type of social support that predicts a worse outcome of depression, and studies using better measures of social support may be required. While only physically healthy mothers were included in the study, we did not include any physical measures to screen mothers for common physical problems such as anaemia or iodine-deficiency. However, chronic malnutrition in the mothers indicated by a low body-mass index was not significantly associated with persistent depression. Other limitations include a relatively small sample size and the fact that all the women came from one sub-district of Rawalpindi. Cross-sectional epidemiological studies in Pakistan suggest that the prevalence rate of depressive disorder in women is high even in the non-postnatal period (Husain et al., 2000; Mirza and Jenkins, 2004). Other longitudinal studies would be required to try and understand the origins of depression in these women, which may well pre-date the first pregnancy, or even have its onset in adolescence. However, the postnatal period assumes importance because of the impact of maternal depression on the infant's development. Thus, early identification of mothers whose depression is likely to persist and providing extra support could improve outcomes in both mothers and their infants. The strongest predictor of persistent depression in this study was a high score on the SRQ-20 in the third trimester pregnancy. Recent meta-analyses suggest that psychological symptoms (mainly depression and anxiety) during pregnancy, along with past history of psychiatric illness are primary risk factors for postnatal depression (O'Hara and Swain, 1996; Beck, 2001). The few studies that have explored the long-term outcome of postnatal depression suggest that depressive symptoms during pregnancy (Bernazzani et al., 1997; Verkerk et al., 2003) not only predict early postnatal depression but are also associated with persistence of depression. The utility of the SRQ-20 in predicting persistent depression suggests that this questionnaire, which has been specifically developed for use in primary care in developing countries, could also serve as a useful screening instrument during the antenatal period.

Int_Arch_Occup_Environ_Health-4-1-2175018

Quality of life and illness perception in working and sick-listed chronic RSI patients

Objective To study differences between working and sick-listed chronic repetitive strain injury (RSI) patients in the Netherlands with respect to indices of quality of life and illness perception. Introduction The growing global concerns in the 1990s about the effects of work-related upper extremity musculoskeletal disorders (i.e. Repetitive Strain Injuries, or RSI) on the health and well-being of workers, and about the economic and social impact of these conditions, has led to a variety of research on the international level (e.g. Sluiter et al. 2001; Lee et al. 2005; Bongers et al. 2006; Waters et al. 2007), and in the Netherlands governmental actions, new rules, regulations, and professional guidelines were developed (Sluiter et al. 2001; Douwes et al. 2001; NVAB 2003; Blatter et al. 2004). In addition, the epidemiological studies begun in the 1990s that tried to unravel the multifactorial predisposing and precipitating factors of RSI complaints managed to find several significant factors, albeit the perceived relationships were neither very pronounced nor very specific (e.g. Bongers et al. 2006). In contrast to the recently expressed concerns about multidisciplinary curative actions in the US (Feuerstein and Harrington 2006), many preventive ergonomic actions have been introduced in high risk sectors in the Netherlands during the “Covenant” periods. To prevent long-term work disability, occupational physicians have learned to start taking action as early as three weeks after onset of sickness absence for RSI-like complaints. Long-term sickness absence because of RSI-complaints has decreased in the past few years (Bongers et al. 2006). Moreover, when workers’ sickness absence is of very long duration, some employers invest in their workers’ health by paying for intensive multidisciplinary treatments aimed at a return to work (Meijer et al. 2006). Despite all these changes, the number of members in, for example, the Dutch RSI Patients’ Association doubled during the last seven years, and prevalences of over 25% for upper extremity complaints are still found; rates are comparable to other European countries (Bongers et al. 2006). All medical professionals deal with patient groups that have medical disorders that may be cared for but not cured with additional costs (Meerding et al. 1998), and also therefore, increasing interest in the patient’s perspectives has grown in the last decade. Perspectives about what are important points of interest to deal with in treatment and knowledge about their quality of life both reveal interesting information for specific focus in treatment (Gafni et al. 1998). Illness perception may be distorted more in work-related disorders compared to other diseases when the ability to return and to keep on working in the job that may have caused the health complaints is questioned. Up to now, only little or even no information was available on chronic RSI patients’ quality of life and illness representation (Picavet and Hoeymans, 2004; Broadbent et al. 2006). Therefore, in 2005 a study that aimed to describe the working vs. sick-listed RSI patients’ perspectives in the Netherlands with respect to their complaints, quality of life, and illness perception was begun in a large group of chronic RSI patients. Methods Population Patients with RSI were identified by approaching all 3,250 members of the Dutch RSI patients’ association. The Dutch RSI patient association was started 11 years ago. In the Netherlands, patients with the same disease are allowed and administratively helped by Governmental Services to join and start an association. Their goal is to gather and spread relevant information to all patients with the same disease in The Netherlands. In addition, their input in sickness-related policy matters is increasingly asked for. Members that were registered in March 2005 were eligible for inclusion. This group defined the study population in this cross-sectional study. Data collection In April 2005 questionnaires were sent by the secretariat of the RSI patients’ association to the home addresses of all their current members. An accompanying letter was included from the patients’ association. Patients were requested to complete a set of self-report measures. Return envelopes that were addressed directly to the research institute were included as well. One week later, a reminder letter was sent to all members. Data collection was stopped four weeks after sending the reminder. A total of 1,185 (36%) questionnaires were returned by mail. Outcome measurements Quality of life was assessed in three ways: Through seven subscales of the Dutch version of the 36-item Short Form Health Survey (Ware and Sherbourne 1992; Aronson et al. 1998): physical role functioning, emotional role functioning, social functioning, pain, mental health, vitality, and physical functioning. Scores ranged between 0 and 100 (higher scores indicated better functioning). Visual analogue scales (VAS) (Streiner and Norman 2003) were used to rate the level of present general quality of life with respect to health (0 = the worst imaginable; 10 = the best imaginable) and to rate the level, retrospectively, of the estimated general quality of life with respect to health before the RSI complaints started (0 = the worst imaginable; 10 = the best imaginable). The patients were also asked to report on their current work-ability by means of an 11-point scale (ranging from 0 for the least ability imaginable to the participant’s own maximum level of 10, the highest level imaginable) (e.g. Croon et al. 2005). The brief illness perception questionnaire (IPQ-B) (Broadbent et al. 2006) assessed cognitive illness perception through items on illness “consequences”(how much does your illness affect your life?), “identity” (how much do you experience symptoms from your illness?), “timeline” (how long do you think your illness will continue?), “personal control” (how much control do you feel you have over your illness?), “treatment control” (how much do you think your treatment can help your illness?). It assessed emotional illness perception with items on illness “concern” (how concerned are you about your illness?) and “emotions” (how much does your illness affect you emotionally? [e.g. does it make you angry, scared, upset, or depressed?]), and it assessed illness understanding with an item on illness “comprehensibility” (how well do you feel you understand your illness?). Scores ranged from 0 to 10, and averages on the group level were calculated (Broadbent et al. 2006). The last question of the IPQ-B is open ended, centring on the most important factors believed to cause the illness. The demographic, complaint-related, and work activities characteristics were used to describe the two groups. Demographics included gender, age, education, and current work activities. Complaint-related variables were pain and stiffness/tingling intensity (assessed with VAS scales ranging from 0 = no complaints at all to 10 = the worst complaints imaginable) (Streiner and Norman 2003), pain duration (in years), extent of pain complaints (number of upper body regions affected) (Sluiter et al. 2001), ranging between 0 and 16. Work activity characteristics were assessed by four psychosocial work characteristics subscales: work pressure, social support from colleagues or supervisor, and job control (from the “Dutch Questionnaire on the Experience and Assessment of Work”, VBBA) (Van Veldhoven and Meijman 1994; Sluiter et al. 2003). VBBA subscale scores ranged between 0 and 100, with higher scores being more unfavourable. In order to compare the two groups of RSI patients, we classified the “working” group as those patients that had been present at their work for a minimum of 8 h during the previous week, and the “sick-listed” group were RSI patients with a sickness absence certificate following the Dutch social system (i.e. longer than 1 year sick-listed, or those that stated that they did not work according to their contract hours during the previous week before they filled in the questionnaire). Statistical analysis Scale scores were calculated according to the original descriptions of the scales used. VAS scores were calculated in mm, ranging from 0 to 100. Means and standard deviations were calculated for every outcome. We divided the respondents in two groups, one group of patients that indicated they worked in a paid job, and one group of patients that indicated they were sick-listed. After testing assumptions, either multivariate ANOVA controlling for age, gender, and education level or nonparametric tests (Mann–Whitney) were performed to test differences between the two groups of RSI patients; P-values < 0.01 were considered significant. Average differences over 1 on a 10-point scale and over 10 on a 100-point scale were considered clinically important in terms of effect size (Streiner and Norman 2003). Results A total of 1,185 respondents responded, and data from 1,121 questionnaires could be used in the analyses for this study. Table 1 provides a description of demographic and both complaint-related and work-related variables for the groups under study. Table 1Total study population and working vs. sick-listed groups: gender, education, age, pain intensity, pain duration, work status, and work-abilityTotal group (n = 1,121)Working group (n = 745)Sick-listed group (n = 376)Difference between groups: P-valueGender (% females)676278a < 0.001Education (% high )6771 60a < 0.001Age in years (mean (SD))40.8 (8.7)40.0 (8.3)42.3 (9.4)a < 0.001Pain intensity (0–100) (mean (SD))41.3 (25.4)36.2 (24.7)51.6 (23.9)a < 0.001Stiffness/tingling intensity (0–100) (mean (SD))37.8 (26.8)33.7 (25.8)45.7 (27.0)a < 0.001Pain duration in years (mean (SD))5.8 (3.2)5.6 (3.1)6.2 (3.3)0.01Pain extent (mean (SD) number of body regions)6.4 (3.9)5.7 (3.6)7.7 (4.1)a < 0.001Work pressure (0–100) (mean (sd))b48.4 (21.1)48.7 (20.8)45.9 (23.6)0.885Social support from colleagues (0–100) (mean (SD))b27.8 (20.1)27.4 (20.0)31.5 (20.4)0.157Social support from direct supervisor (0–100) (mean (SD))b35.1 (26.2)34.4 (26.1)42.0 (26.1)0.138aKolmogorov–Smirnov test P-values < 0.01bFilled in by n = 84 in the sick-listed group RSI patients in the sick-listed group were two years older on average, and this group contained a relatively higher proportion of women and lower proportion of highly educated persons. In the sick-listed group, 76% was sick-listed longer than one year. In both groups, complaints were present for several years and extended over several upper body regions. In both groups, over 85% received a diagnostic label, provided by a medical professional, that belong to the umbrella term RSI. With respect to complaint-related variables, the patients in the sick-listed group reported significantly more pain in more regions, and they reported more stiffness/tingling complaints. The sick-listed group reported significantly lower ability to work, and no differences between groups were found in the psychosocial work characteristics experienced. With respect to co-morbidity, no other diseases were reported by 46% in the sick-listed group compared to 64% in the working group; depression and burnout were reported by 12 and 11% in the sick-listed group versus 6 and 5% in the working group. In the working patient group, 35% filled some kind of administrative function, 11% worked in IT, 10% worked in a physical job, and 10% in a management function. In both groups, almost 80% reported long- or short-term overexertion during work as the only main cause of the onset of their illness. Experienced quality of life on the studied SF-36 subscales is shown in Tables 2 and 3. In Table 2, a matrix of zero-order correlation coefficients is shown for the complete population between work-ability, pain intensity, pain duration, the SF-36 subscales, and the eight B-IPQ dimensions. The highest correlation coefficients were found between the two SF-36 subscales mental health and vitality (r = 0.75), and the SF-36 subscale pain and VAS scale pain intensity (r = 0.72). Table 2Matrix of zero-order correlation coefficients between 1: work-ability, 2: pain intensity, 3: pain duration, 4–10: SF-36 subscales, and 11–18: the B-IPQ outcomes (n varies between 1,027 and 1,114)1234567891011121314151617181 work-ability–2 pain intensiy−0.48–3 pain duration−0.080.12–4 sf physical role0.54−0.45−0.02–5 sf mental health0.34−0.250.030.32–6 sf emotional role0.26−0.180.050.31–7 sf social functioning−0.010.05−0.01−0.07−0.010.09–8 sf pain0.61−0.72−0.050.590.310.24−0.03–9 sf vitality0.40−0.300.010.400.750.47−0.080.39–10 sf physical functioning0.60−0.52−0.180.480.270.22−0.020.610.38–11 consequences−0.610.510.08−0.58−0.35−0.27−0.001−0.62−0.41−0.49–12 timeline−0.130.180.32−0.10−0.030.010.01−0.15−0.07−0.160.19–13 personal control0.44−0.36−0.0040.350.330.18−0.0010.420.350.34−0.36−0.06–14 treatment control0.29−0.19−0.090.160.100.050.040.210.150.15−0.16−0.160.33–15 identity−0.510.640.07−0.54−0.29−0.230.04−0.67−0.35−0.490.680.23−0.33−0.10–16 concern−0.380.46−0.03−0.46−0.35−0.270.01−0.49−0.37−0.300.570.16−0.36−0.100.58–17 comprehensibility0.20−0.160.040.220.200.14−0.030.190.210.12−0.210.000.310.22−0.16−0.29–18 emotional response−0.380.35−0.07−0.46−0.51−0.390.01−0.41−0.47−0.250.590.08−0.35−0.080.480.64−0.27–Table 3Mean (SD) scores by groups on SF-36 subscales physical role functioning, mental health, emotional role functioning, social functioning, pain, vitality, physical functioning, and work-abilityTotal group (n = 1,121)Working group (n = 745)Sick-listed group (n = 376)Difference between groups: P-valueSF-36 scales (0–100) (mean (SD)) Physical role functioning34.7 (38.2)43.4 (39.2)17.0 (29.0)a < 0.001 Mental health68.5 (16.7)70.6 (15.6)64.3 (18.0)a < 0.001 Emotional role functioning73.9 (38.3)78.7 (34.8)64.1 (43.1)a < 0.001 Social functioning54.5 (9.4)54.6 (8.9)54.2 (10.4)0.208 Pain54.7 (21.6)60.7 (19.2)42.7 (22.0)a < 0.001 Vitality53.6 (17.9)55.7 (17.1)49.3 (18.6)a < 0.001 Physical functioning76.5 (18.0)81.7 (14.6)65.9 (19.6)a < 0.001 Work-ability (0–10) (mean (SD))5.4 (2.7)6.6 (1.8)3.1 (2.6)a < 0.001aKolmogornov–Smirnov test P-values < 0.01 As shown in Table 3, the average scores of all scales but social functioning differed significantly between the two groups to the detriment of the sick-listed RSI patients on the other SF-36 subscales. Clinically relevant differences were found in work-ability and in the SF-36 scales for physical role functioning, emotional role functioning, pain, and physical functioning. General quality of life with respect to present health was rated as 52.7 on average (SD: 24.4) in the total group of RSI patients, and the average in the working group and sick-listed groups were 57.9 and 42.2, respectively. The group difference was significant (P-value Kolmogornov–Smirnov test < 0.001). In contrast, the estimated general quality of life with respect to health before the onset of RSI complaints was rated as 83.1 on average (SD: 15.4) in the total group of RSI patients. There was only a non-significant difference in the average scores between groups (83.1 in the working group and 83 in the sick-listed group (P-value Kolmogornov–Smirnov test = 0.796)). Combining these results, the general quality of life with respect to health was subjectively decreased by 37% in the total group over time. However, the difference in decrease was significant (Kolmogornov–Smirnov test, P-value < 0.001): only 31% in the working group but 49% in the sick-listed group of patients. Scores on the eight illness perception dimensions are shown in Table 4. Table 4Mean (SD) illness perception scores (B-IPQ) by groupsIllness perception dimensionsTotal group (n = 1,121)Working group (n = 745)Sick-listed group (n = 376)Difference between groups: P-valueConsequences6.3 (2.6)5.6 (2.5)7.6 (2.1)a < 0.001Timeline8.3 (2.0)8.2 (2.1)8.5 (1.7)0.557Personal control 6.3 (2.0)6.7 (1.8)5.6 (2.1)a < 0.001Treatment control 5.2 (2.6)5.7 (2.5)4.4 (2.6)a < 0.001Identity6.2 (2.4)5.8 (2.4)7.1 (2.1)a < 0.001Concern5.5 (2.6)5.2 (2.6)6.1 (2.6)a < 0.001Comprehensibility6.9 (2.1)7.1 (2.0)6.6 (2.3)0.014Emotional response5.4 (2.6)5.1 (2.6)6.0 (2.5)a < 0.001aKolmogornov–Smirnov test P-values < 0.01 With the exception of the dimension timeline and comprehensibility, there was a significant difference in the six other illness perception dimensions between the two groups. Clinically relevant differences between the two groups to the detriment of the sick-listed RSI patients were found for the illness perception dimensions of “consequences”, “personal-” and “treatment-control”, and “identity”. Discussion The aim of the present study was to describe working vs. sick-listed RSI patients’ perspectives in the Netherlands in 2005 with respect to their complaints, quality of life, and illness perception. The sick-listed RSI patients reported more severe and extensive complaints in the upper extremity, but the long duration (6 years) of their complaints was comparable to the working RSI patients. The sick-listed patients, however, experienced a considerably more decreased quality of life because of their complaints, as well as more distorted illness perceptions. The quality of life in sick-listed patients was lower than that of the working patients when the SF-36 subscales scores were taken into account. Interestingly, both groups showed comparable but low social functioning scores (averaging around 55). This was also actually quite low compared to a group of patients with rheumatoid arthritis, who averaged over 70 and compared with a small sample of Dutch workers with RSI who averaged 79 (Sprangers et al. 2000; Picavet and Hoeymans 2004). The experienced quality of life decrease over time was found to be substantial in both groups, which might be explained by the severity, duration, and extent of the reported complaints. It is possible to argue that the anchor question that was used to calculate the quality of life decrease was posed post-hoc. But that was one of the proposed solutions among quality of life researchers in the debate about changed perspectives and response shift: by asking for the two rates at a given time in the form of an existing test, there cannot be any recalibration, re-evaluation of re-prioritization, and re-conceptualization (Osborne et al. 2006; Visser et al. 2005). This may have been the first time that illness representation was assessed in a large group of RSI patients using the recently described brief illness perception questionnaire (IPQ-B) (Broadbent et al. 2006). Illness representation covers the dimensions of illness identity, consequences, cause, timeline, and cure or control. These perceptions about the illness influence patient behaviours, and changing perceptions may improve recovery (e.g. Petrie et al. 2002). If an average difference of over 1 on a 0–10 point scale is considered clinically relevant, illness consequences, personal control, treatment control, and identity (i.e. the number of symptoms because of the illness) clearly differed between sick-listed and working RSI patients. We believe that it would be useful to focus future informational treatment strategies in the sick-listed patient group on changing their illness perception. We successfully compared two fairly large groups of chronic RSI patients with respect to their current complaints, quality of life, and illness perception. However, some methodological considerations are in place: (1) the used case-definition to identify sick-listed RSI patients may have influenced the outcomes although it is not clear in what direction this may have caused a bias; (2) because not all variables complied with all assumptions for multivariate testing, that would have allowed for controlling for different possible confounders, non-parametric test outcomes have been presented. Post-hoc multivariate testing did not show any differences in the outcomes as were presented. (3) It is acknowledged that a relatively modest proportion of the total study population responded. Considering the variation in responses and the main purpose of this study, the response should not be considered as a bias. Our population, however, is a selected population in several ways and this has been described in the methods section: participants were members of the RSI patients association, they had long-term complaints, it was not possible to take a random sample in the theoretically existing Dutch group of chronic RSI patients who work or are sick-listed, and the number of responders of the total sample was not as much as would be desirable. However, the diagnostic label that was communicated with over 85% of these chronic patients by different physicians was one of the syndromes that fall under the umbrella diagnosing label of “RSI” (Sluiter et al. 2001). In addition, comparisons between two groups of this specific group of patients were reported upon for this study. Summarizing, besides its possible shortcomings, this study is thought to be unique with respect to studying a large group of chronic RSI patients as most of the published literature on work-related upper extremity musculoskeletal disorders deal with acute and sub-acute populations. It would be interesting to study how the relation between the variables under study developed over time: if the severity of complaints, quality of life level, and illness perception dimensions are relatively stable over time in this subgroup of chronic patients, they could be used as prognostic factors to decide on interventions with respect to future ability to work. It may also be that if activities are increased because of work resumption, they will act as “intervention” in the place-then-train ratio and influence the measures over time. We do know that the differences in the severity of RSI complaints and quality of life indices, to the detriment of the sick-listed group of RSI patients, could be used as an indication for referral to multidisciplinary treatment programmes, because all of these parameters have recently been shown to be influenced positively, even in patients with long-term complaints (Meijer et al. 2006).

Pediatr_Nephrol-3-1-1766479

The management of anemia in pediatric peritoneal dialysis patients

Anemia is common in chronic renal failure. Guidelines for the diagnosis and treatment of anemia in adult patients are available. With respect to the diagnosis and treatment in children on peritoneal dialysis, the European Pediatric Peritoneal Dialysis Working Group (EPPWG) has produced guidelines. After a thorough diagnostic work-up, treatment should aim for a target hemoglobin concentration of at least 11 g/l. This can be accomplished by the administration of erythropoietin and iron preparations. Although there is sufficient evidence to advocate the intraperitoneal administration of erythropoietin, most pediatric nephrologists still apply erythropoietin by the subcutaneous route. Iron should preferably be prescribed as an oral preparation. Sufficient attention has to be paid to the nutritional intake in these children. There is no place for carnitine supplementation in the treatment of anemia in pediatric peritoneal dialysis patients. The treatment of anemia in chronic kidney disease has been summarized in the NKF/DOQI guidelines [1], which have recently been updated [2]. European best practice guidelines have also been recently published [3]. These published guidelines pay no, or very limited, attention to the special situation in children. The European Pediatric Peritoneal Dialysis Working Group (EPPWG) was established in 1999 by pediatric nephrologists with a major interest in peritoneal dialysis and has, among other things, published guidelines on commencing elective chronic peritoneal dialysis [4]. One of the functions of the group is to establish expert guidance in important clinical areas associated with peritoneal dialysis in conjunction with other members of the multidisciplinary team. These guidelines were initiated and discussed at meetings of the group and developed by e-mail discussion to develop a consensus of opinion based upon cumulative clinical experience and reported studies. The present guidelines apply to the management of anemia in pediatric peritoneal dialysis patients. Definition of anemia More than in adult patients, hemoglobin and hematocrit values are age dependent in children [5], (Tables 1, 2, 3, 4). Also, concentrations of ferritin, transferrin, and iron are dependent on age. Early iron deficiency may be diagnosed by an increase in hypochromic red blood cells [6]. The technique for measuring this parameter is not available everywhere. Since children with renal failure generally do not have underlying co-morbidity contributing to anemia, it is obvious that in children on peritoneal dialysis normal hematocrit levels should be aimed for (opinion). The work-up of the anemic child is not different from the work-up that is advocated for adult patients by the DOQI guidelines, although the importance of a stool test for occult blood may be less useful in this age group [2, 7] (opinion):Iron deficiencyClinical historyAssessment of nutritional statusHemoglobin and hematocritRed blood cell indicesReticulocyte countIron parametersSerum ironTable 1. Reference ranges of anemia parameters in children [5]: hemoglobin and hematocrit (M male; F female)Hemoglobin (g/dl)Hematocrit (%)1–3 days14.5–22.545–702 months9.0–14.028–426–12 years11.5–15.535–4512–18 years, M13.0–16.037–49                      F12.0–16.036–46Table 2. Reference ranges of anemia parameters in children [5]: ferritinFerritin (ng/ml)Newborn25–2001 month200–6002–5 months50–2006 months–15 years7–140Table 3. Reference ranges of anemia parameters in children [5]: iron (M male; F female)Iron (μg/dl)0–2 months100–2502–12 months40–1001–12 years50–120Thereafter, M50–160                    F40–150Table 4. Reference ranges of anemia parameters in children [5]: transferrinTransferrin (mg/dl)Newborn130–275Adult200–400 After this work-up has been completed, iron and/or erythropoietin therapy should be initiated to obtain a target hemoglobin concentration of at least 11 g/dl (hematocrit 33%), although there is a tendency to increase this target [8] (opinion). A target hemoglobin concentration of at least 11 g/l should be aimed for in children on peritoneal dialysis (opinion). Erythropoietin Most experience has been obtained with subcutaneous administration of recombinant human erythropoietin [9, 10]. Erythropoietin is available as erythropoietin alpha and beta. Erythropoietin alpha and beta differ noticeably in their formulation excipients. No clear prescription schedules for starting erythropoietin are available for children, and schedules provided in the literature vary greatly. It seems reasonable to start with a subcutaneous dosage of 50–100 U/kg body weight 2 or 3 times per week. In some selected patients (particularly if started in the predialysis period) a lower frequency or dosage can be attempted. In studies in adult patients once weekly subcutaneous administration of erythropoietin was effective for the treatment of renal anemia [11, 12, 13]. It is well known that younger children need relatively more erythropoietin than older ones [14] (evidence). Maintenance recommendations vary from 300 U/kg/week for a child with a weight of <20 kg to 120 U/kg/week for a child with a weight of >30 kg [10] (opinion). Each child will need the dosage titrated to achieve the target hemoglobin concentrations. For the titration of erythropoietin the DOQI guidelines can be used [2]. If the increase in hematocrit after initiation of erythropoietin therapy is less than 2% over a 2- to 4-week period, the dose should be increased by 50%. If the absolute rate of increase in hematocrit after initiation of erythropoietin therapy or after a dose increase exceeds 8% per month, the weekly dose of erythropoietin should be decreased by 25%. Side effects of erythropoietin therapy are rare; increased clotting tendency, hypertension, and seizures are to be considered the consequence of the therapeutic effect rather than an adverse effect of the preparation. Nevertheless, blood pressure should be carefully monitored during therapy. Recently, attention was drawn to a severe side effect of erythropoietin therapy, pure red blood cell aplasia due to the occurrence of neutralizing antierythropoietin antibodies [15, 16]. Subsequent investigations showed that this was in most cases associated with subcutaneously administered erythropoietin alpha from one brand, although some cases have also been described using the other brand available on the market. The application of erythropoietin beta is associated with a much lower incidence of this severe side effect. In many countries the subcutaneous administration of erythropoietin alpha is discouraged at present. The administration of darbepoetin alpha is not associated with pure red blood cell aplasia, but it should be remembered that experience with this drug is still limited. Subcutaneous administration of drugs is psychologically distressing, especially for children (opinion). The use of ultrafine needles and special injection pens may help to alleviate the upset. In adult peritoneal dialysis patients a noncompliance with erythropoietin administration was reported in between 35% and 55% of patients [17, 18]. For children no data on erythropoietin are available, but from clinical practice it is known that subcutaneous administration is frightening and a source of conflict between child and caregivers. Limited data are available on the compliance with another drug which has to be administered subcutaneously, recombinant human growth hormone. Noncompliance was reported to be between 50% and 91% [19, 20, 21]. In one study it was reported that noncompliance increased significantly from 41% at 1 year to 91% at 2 years [21]. Intravenous administration is a more expensive alternative, but will rarely be applied in pediatric peritoneal dialysis patients [22]. Despite some initial discouraging but still frequently cited reports in the literature, erythropoietin can be administered very well by the intraperitoneal route [23, 24]. If erythropoietin is administered in a small volume of dialysis fluid (50-ml bags are commercially available), bioavailability is similar to that after subcutaneous administration of the same amount of hormone [25] (evidence). Mean dosage needed for maintaining the target hematocrit decreased from 279 to 194 U/kg/week if the drug was administered in a 50-ml dialysis bag during the daytime [26, 27]. In a more recent study in 20 patients on nightly intermittent peritoneal dialysis mean dosage was 179 U/kg/week [28]. Dialysis adequacy is a major factor of concern in such a regimen: KT/V urea was ≥2.2 in the group studied. However, in some patients the application of intraperitoneal erythropoietin may be limited by the inability to obtain adequate dialysis (opinion). In patients not achieving adequate dialysis, the daytime period should be used for additional dialysis exchanges. In this category, that period will not be available for intraperitoneal erythropoietin therapy. Peritonitis frequency does not need to increase using intraperitoneal erythropoietin: the two pediatric studies reporting peritonitis rates mention one episode every 14.7 and 11.2 treatment months, respectively [19, 21]. One early study in children was broken off because of a high peritonitis rate [29]; possibly this was due to insufficient training of the caregivers. Recently, an erythropoietin analogue was developed (darbepoetin alpha or NESP = novel erythropoiesis stimulating protein), which is a hyperglycosylated erythropoiesis-stimulating protein with a presumed threefold longer half-life than erythropoietin in man [30, 31, 32, 33, 34] (evidence). In children a two- to fourfold longer half-life was reported [35]. The pharmacokinetics when administered intravenously and subcutaneously appear to be the same in adult and pediatric patients. A randomized comparative study of darbepoetin and erythropoietin in pediatric patients with chronic or end-stage renal disease is just starting. There are no data available with respect to the intraperitoneal administration of darbepoetin. The development of an orally active agonist of the erythropoietin receptor will be an interesting future feature [36]. Erythropoietin resistance may be due to a number of causes:InfectionHyperparathyroidismMalnutritionHemolytic disordersFolate or vitamin B12 deficiencyUnderdialysisVitamin C deficiencyACE inhibitorsAnti-erythropoietin antibodies Since there is an excellent review of these in the adult guidelines [2, 3, 37], and they are not essentially different in children, they are not discussed in detail here. In children responding poorly to erythropoietin therapy, special emphasis should be put on the possible contribution of inflammation or hyperparathyroidism [38]. The possible occurrence of pure red cell aplasia due to the development of neutralizing antierythropoietin antibodies has been discussed before. Erythropoietin should be administered by the subcutaneous or intraperitoneal route in children on peritoneal dialysis (evidence). Although there is sufficient evidence to advocate the intraperitoneal administration of erythropoietin, most pediatric nephrologists still apply erythropoietin by the subcutaneous route in their peritoneal dialysis patients. Iron Iron supplementation is indicated in virtually all pediatric patients with renal anemia who are treated with erythropoietin. According to the DOQI guidelines, transferrin saturation should be maintained above 20% and serum ferritin concentration above 100 ng/ml [2] (evidence). There is no reason to anticipate that these guidelines should be different for children and for adults (opinion). It may be difficult to maintain sufficient iron stores in children on peritoneal dialysis, using oral iron preparations only. Iron supplements should be given to prevent iron deficiency and to maintain adequate iron stores. A dosage of 2–3 mg/kg body weight per day is recommended and administered in two to three divided doses either 1 or 2 h after food. Iron supplements should not be added to the infant formula or nutrition supplements [39] (evidence). If possible, they should be prescribed with vitamin C to enhance absorption (opinion). However, it is important not to oversupplement with vitamin C for risk of increased oxalate formation. Iron supplements should ideally not be taken with cereals and legumes, tannins (tea, cocoa, chocolate) and dairy products as these interfere with absorption. Micronutrient supplements should also be prescribed following individualized dietary assessment and should account for the potential peritoneal dialysis losses of folic acid and vitamins C and B6. Compliance with oral iron preparations for micronutrient supplements may be difficult and must be reinforced by both medical and dietetic staff [39, 40]. Parenteral iron preparations, commonly used in patients on hemodialysis, are more difficult to apply in children on peritoneal dialysis. After the recent approval of iron gluconate, iron sucrose, and iron saccharate, the application of iron dextran should also be abandoned in the United States [41, 42, 43, 44] (evidence). Recommended dosage is 2 mg iron per kg per week for intravenous iron sucrose treatment [7]. Before starting intravenous therapy the administration of a test dosis is recommended. It is clear, however, that the intravenous administration of iron preparations is cumbersome, and oral administration is preferred [39] (opinion). In rare cases, for example with noncompliance with oral medication, intermittent intravenous administration will be indicated. Limited but positive experience has been obtained with the intraperitoneal administration of iron dextran, both in rats and men [45, 46, 47]. Reports on the intraperitoneal administration of iron gluconate or iron sucrose are lacking. Iron should preferably be prescribed as an oral preparation (evidence). In rare cases intermittent intravenous administration will be indicated. Carnitine Several studies in adult patients suggest that intravenous supplementation with l-carnitine reduces requirements for erythropoietin by 38–50% [48, 49, 50] (opinion). Published data on the effect of l-carnitine supplementation on the treatment of anemia in children are very scarce. One study showed an increase in the hematocrit by 34% in two children on hemodialysis with l-carnitine supplementation without modification of erythropoietin dosage [51]. Another study in 16 children on dialysis, of whom five were on peritoneal dialysis, showed no beneficial effect of oral supplementation with l-carnitine on erythropoietin requirement [52]. There is no precise place for carnitine supplementation in the treatment of anemia in pediatric peritoneal dialysis patients (opinion).

Diabetologia-3-1-2039833

Fatty acid-induced mitochondrial uncoupling in adipocytes as a key protective factor against insulin resistance and beta cell dysfunction: a new concept in the pathogenesis of obesity-associated type 2 diabetes mellitus

Type 2 diabetes is associated with excessive food intake and a sedentary lifestyle. Local inflammation of white adipose tissue induces cytokine-mediated insulin resistance of adipocytes. This results in enhanced lipolysis within these cells. The fatty acids that are released into the cytosol can be removed by mitochondrial β-oxidation. The flux through this pathway is normally limited by the rate of ADP supply, which in turn is determined by the metabolic activity of the adipocyte. It is expected that the latter does not adapt to an increased rate of lipolysis. We propose that elevated fatty acid concentrations in the cytosol of adipocytes induce mitochondrial uncoupling and thereby allow mitochondria to remove much larger amounts of fatty acids. By this, release of fatty acids out of adipocytes into the circulation is prevented. When the rate of fatty acid release into the cytosol exceeds the β-oxidation capacity, cytosolic fatty acid concentrations increase and induce mitochondrial toxicity. This results in a decrease in β-oxidation capacity and the entry of fatty acids into the circulation. Unless these released fatty acids are removed by mitochondrial oxidation in active muscles, these fatty acids result in ectopic triacylglycerol deposits, induction of insulin resistance, beta cell damage and diabetes. Thiazolidinediones improve mitochondrial function within adipocytes and may in this way alleviate the burden imposed by the excessive fat accumulation associated with the metabolic syndrome. Thus, the number and activity of mitochondria within adipocytes contribute to the threshold at which fatty acids are released into the circulation, leading to insulin resistance and type 2 diabetes. Introduction Type 2 diabetes mellitus is generally associated with an enhanced energy intake and too little physical exercise. Also genetic factors determine the susceptibility to develop this disease. Whole-body insulin resistance and a state of low grade inflammation are early marker of the disease process [1]. Subsequently, hyperglycaemia develops due to an accelerated decline in beta cell function [2–4]. Type 2 diabetes mellitus is often accompanied by other co-morbidities such as hypertension and dyslipidaemia. Together, they constitute the metabolic syndrome [2]. Here we present the concept that removal of fatty acids within white adipocytes by fatty acid-induced uncoupled mitochondrial β-oxidation protects the organism against fatty acid leakage out of adipocytes, thereby preventing fatty acid-induced insulin resistance in liver and muscle and lipotoxicity in pancreatic beta cells. A consequence of this concept is that mitochondrial dysfunction in adipocytes, either inherited or acquired, makes the organism more prone to develop insulin resistance and type 2 diabetes. Mitochondria and fatty acids One of the main functions of mitochondria, in addition to producing ATP, is to remove fatty acids by β-oxidation. In this way, mitochondria are able to remove NEFA and to protect the organism against fatty acid-induced insulin resistance and pancreatic beta cell lipotoxicity. β-Oxidation can take place in all cell types relevant for glucose homeostasis, including muscle, liver and adipocytes. During this process, fatty acids are oxidised in the mitochondrial matrix by NAD+ and flavin adenine dinucleotide (FAD), yielding acetyl coenzyme A, which is further degraded to CO2 by the citric acid cycle. The resulting NADH and FADH2 need to be recycled into NAD+ and FAD by the respiratory chain so that additional fatty acid molecules can be oxidised. This requires a supply of ADP, which is converted into ATP by respiratory chain activity. The rate of conversion of ATP back into ADP, which is determined by the metabolic activity of the cell, determines the rate at which fatty acids can be removed by mitochondrial activity. If the capacity of the cell to oxidise fatty acids is to be enhanced, ATP needs to be reconverted into ADP at an increased rate. This occurs during exercise in contracting muscle. The resulting enhanced rate of ADP generation provides a major additional sink for the oxidation of fatty acids within the body. Only when mitochondria are in the uncoupled state can regeneration of NAD+ and FAD occur without the conversion of ATP into ADP. The released energy is then converted into heat. Uncoupling is a physiological adaptation process, normally regulated by specific proteins [5]. It is remarkable that, like uncoupling proteins, fatty acids are capable of inducing mitochondrial uncoupling, especially when their concentrations exceed the binding capacity of fatty acid binding proteins [6–8]. This implies that when unbound fatty acids are present in the cytosol of cells, mitochondria become less efficient at ATP production and generate more heat during the oxidation of NADH and FADH2 derived from food. Mitochondrial uncoupling in adipocytes by fatty acids may protect the organism against fatty acid-induced insulin resistance and lipotoxicity White adipocytes contain large amounts of mitochondria in their tiny cytosolic compartment [9, 10]. There is no obvious reason why these cells need such a large capacity to produce ATP. We propose that, when uncoupled by fatty acids, these mitochondria prevent the release of fatty acids out of the adipocytes when the antilipolytic action of insulin is attenuated. Evidence is accumulating that inflammation of adipose tissue is an early step in the pathogenesis of type 2 diabetes. The combination of excessive food intake with a sedentary lifestyle results in the formation of large adipocytes overloaded with triacylglycerols. Through as yet unknown mechanisms the adipose tissue becomes inflamed and infiltrated by leucocytes. This leads to the release of inflammatory cytokines such as TNF-α [11–14], a potent inducer of insulin resistance in adipocytes which also induces lipolysis. As a result, fatty acids are released out of the large triacylglycerol pool [15]. We propose that, initially, a part of these fatty acids are removed inside adipocytes by partially uncoupled mitochondrial β-oxidation. In doing so, these mitochondria create a threshold for fatty acid release into the circulation, which would otherwise trigger the development of whole-body insulin resistance and pancreatic beta cell lipotoxicity, the latter expected to occur when the body cannot utilise these released fatty acids as fuel, for example, in response to high food intake and little physical exercise. Our proposed mechanism is outlined in Fig. 1. When low amounts of fatty acids are released out of the triacylglycerol pool in insulin-resistant adipocytes, these fatty acids can be removed by coupled β-oxidation. The rate of fatty acid removal by this pathway is determined by the rate of ADP regeneration within adipocytes, which is likely to be a more or less constant factor, merely determined by cellular metabolism and not affected by variations in the lipolytic rate. We propose that, when the rate of fatty acid release exceeds the rate of fatty acid removal by coupled β-oxidation, the increase in cytosolic concentrations of fatty acids not bound to binding proteins (unbound) induces a partial uncoupling of the mitochondria. Uncoupling by fatty acids has been shown in multiple in vitro experiments [5–8, 16]. Evidence of uncoupling in vivo is provided by a study in which exposure of heart muscle to fatty acids resulted in a decrease in ATP production and a concomitant increase in oxygen consumption [17]. When uncoupled, the mitochondria are able to remove much larger quantities of fatty acids through generation of heat. The degree of uncoupling increases in with further elevations in fatty acid concentration, thus generating more heat. This represents a dynamic adaptation of the efficiency of the mitochondrion, which is determined by the intracellular concentration of fatty acids. If, however, the rate of fatty acid release exceeds the maximum clearance rate, fatty acids may reach concentrations that are toxic to the mitochondrion [18]. At this point β-oxidation capacity collapses and fatty acids are released out of adipocytes and redistributed over other tissues in the body. These ectopic triacylglycerol deposits are associated with the development of insulin resistance in muscle and liver and lipotoxicity in pancreatic beta cells [19]. At this stage of the disease process physical exercise will, by virtue of the generation of large amounts of ADP in muscle tissue, protect against lipotoxicity and the development of insulin resistance. This is because the muscle switches to β-oxidation to regenerate ATP when fatty acids reach a concentration sufficient to make the muscle resistant to insulin-stimulated glucose uptake [20, 21]. Fig. 1Consequences of fatty acid release from the adipocyte triacylglycerol pool. In insulin-sensitive adipocytes, fatty acid concentrations are kept low by insulin-induced antilipolytic action, re-esterification of fatty acids and mitochondrial β-oxidation. TNF-α induces insulin resistance and lipolysis. At low cytosolic concentrations of unbound fatty acids, the flux through mitochondrial β-oxidation is limited by the rate of ADP generation by cellular metabolism. At intermediate concentrations of unbound fatty acids, uncoupling of mitochondria is induced leading to continuous oxidation of fatty acids, independent of ADP supply. This process generates heat and keeps cytosolic fatty acid concentrations low. When the rate of fatty acid release from the triacylglycerol pool exceeds the rate of fatty acid removal, high cytosolic concentrations of unbound fatty acids develop, which induce mitochondrial damage [18]. This results in a decline in the capacity to remove fatty acids and the release of large amounts of fatty acids into the circulation. Unless these are removed by muscle activity they form ectopic triacylglycerol deposits and induce whole-body insulin resistance and beta cell damage. Dotted arrows indicate consequences; continuous arrows, fluxes Discussion In our model for the development of obesity-induced whole-body insulin resistance and beta cell damage leading to type 2 diabetes, we propose a key role for fatty acid removal by adipose tissue mitochondria. The disease process, as outlined in Fig. 2, is initiated by the development of cytokine-induced adipose tissue insulin resistance [11–15] and results in the release of fatty acids into the cytosol of the adipocytes. Further increases in cytosolic fatty acids induce mitochondrial dysfunction [18], decreasing the rate of β-oxidation, augmenting the fatty acid concentration, leading to their release into the circulation. These fatty acids are responsible for the development of insulin resistance in liver and muscle and lipotoxicity in pancreatic beta cells. Fatty acids do not appear to induce insulin resistance in adipocytes [22], which otherwise would create a positive feedback loop whereby the release of fatty acids induces additional insulin resistance and further promotes the release of fatty acids. Fig. 2Proposed sequence of events leading to the development of hyperglycaemia during the metabolic syndrome. When adipocytes become overloaded with triacylglycerol, low-grade inflammation develops and inflammatory cytokines such as TNF-α induce insulin resistance in the adipocytes. This results in an elevated state of lipolysis. When fatty acids are inadequately removed within adipocytes because of mitochondrial dysfunction (for example, induced by fatty acids or HAART therapy, fatty acids appear in the circulation, where they induce insulin resistance of muscle and liver and malfunction of pancreatic beta cells. The elevated circulating fatty acid concentrations may also uncouple mitochondria in artery wall smooth muscle cells, thereby elevating the risk of hypertension [46]. Thiazolidinediones (TZDs) ameliorate the disease process in two ways: (1) by creating more mitochondria in adipose cells [23, 36, 37], thereby enhancing the capacity for oxidation of fatty acids; and (2) by enhanced re-esterification of fatty acids [39, 40] Is there any evidence to support our model? Several studies have shown that in adipose tissue from obese animal models and humans, mitochondrial (mt) DNA copy number and expression of mitochondrial genes is decreased [23, 24]. In the diabetic state, mitochondrial β-oxidation of fatty acids was found to be attenuated. In addition, mitochondria showed an abnormal morphology [24]. These data support the concept of mitochondrial dysfunction in adipose tissue in states of obesity and type 2 diabetes. Furthermore, intake of high-fat food by healthy persons rapidly increases the basal metabolic rate. This increase is blunted in obese individuals, whereas obese individuals exhibit an increased thermogenesis. These observations are suggestive for a rapid effect of ingested fatty acids on thermogenesis by mitochondrial uncoupling. The data also suggest that, in obese individuals, mitochondria are already uncoupled or damaged and less responsive to further fatty acid-induced uncoupling [25]. In vitro studies have shown that fatty acid-induced uncoupling of the mitochondrial respiratory chain depends on the chemical nature of these fatty acids. Unsaturated fatty acids, such as oleic acid, are better uncouplers than saturated fatty acids [8]. This could imply that the inclusion of saturated fatty acids in the diet makes mitochondria less prone to uncoupling, resulting in the earlier release of fatty acids, i.e. at a lower fatty acid level. A further implication of our model is that a decline in mitochondrial function in adipocytes reduces the ability of these cells to store triacylglycerol. Inadvertently released fatty acids out of the triacylglycerol pool enter the circulation and are redistributed in other tissues. This occurs, for example, in individuals starting on highly active antiretroviral therapy (HAART) [26–29]. The nucleoside analogues included in HAART inhibit mtDNA polymerase and induce a ∼30–50% reduction in mtDNA content in adipocytes and other tissues. HAART is associated with the redistribution of peripheral fat to the central and other compartments and with an elevated risk of developing the metabolic syndrome and type 2 diabetes. We see this clinical phenotype as a result of release of fatty acids by peripheral adipocytes, which in turn is due to a decrease in mitochondrial capacity to remove fatty acids. The fatty acids, after entering the circulation, induce insulin resistance and lipotoxicity to pancreatic beta cells and become stored as ectopic triacylglycerol deposits. Another example of a mitochondrial dysfunction is represented by patients carrying a 3243A>G mutation in mtDNA. This mutation results in an attenuated mitochondrial function. Most of these patients develop the maternally inherited diabetes and deafness (MIDD) syndrome. Remarkably, these patients usually have a BMI of <25 kg/m2. Furthermore, they exhibit ectopic triacylglycerol deposits in multiple tissues [30–33]. This clinical picture suggests the attenuated storage of triacylglycerol in adipose tissue. Around mid-life, these patients show a decrease in insulin secretion, which may result from lipotoxicity to pancreatic beta cells. Congenital lipodystrophy is another clinical example showing that an inadequate storage of triacylglycerol in adipocytes contributes to ectopic triacylglycerol deposits and the development of severe whole-body insulin resistance and diabetes [34]. In contrast, improved mitochondrial function specifically in adipocytes is seen in individuals taking thiazolidinediones. These drugs are widely used to ameliorate whole-body insulin resistance in patients suffering from type 2 diabetes mellitus and the metabolic syndrome. These drugs bind to peroxisome proliferator-activated receptor-γ receptors which are highly expressed in adipocytes and in cells from the immune system, but low in liver and muscle [35]. These drugs increase mitochondrial copy number and mitochondrial gene expression specifically in adipose tissue [23, 36, 37]. Clinically, these drugs improve storage of triacylglycerol in peripheral adipocytes and ameliorate fatty acid-related insulin resistance. Patients taking these drugs tend to store more triacylglycerol in their adipocytes, leading to weight gain [38]. Together, these clinical data suggest that correct mitochondrial function is needed for adequate storage of fatty acids as triacylglycerol in adipocytes. By this means the organism is protected against the fatty acid-induced development of insulin resistance and lipotoxicity to the pancreas. As thiazolidinediones also induce the expression of genes involved in the neogenesis of glycerol, these drugs also enhance the re-esterification of fatty acids [36, 39, 40]. This also contributes to the removal of fatty acids from the circulation and protects adipocyte mitochondria against fatty acid-induced damage. Because part of the glycerol in adipocytes is synthesised through pathways that are dependent upon mitochondrial function [36], this suggests that changes in mitochondrial function also may affect the rate of glycerol production and thereby the rate of fatty acid esterification. To be able to perform, uncoupled β-oxidation mitochondria require the presence of sufficient amounts of oxygen. It has been suggested that a state of hypoxia occurs in expanding adipose tissue during the development of obesity and that this state may contribute to the development of the metabolic syndrome [41, 42]. Our model predicts that hypoxia in adipocytes would lead to a reduced rate of fatty acid removal inside these cells and to more fatty acid release into the circulation. The sensitivity of an individual to the development of type 2 diabetes as a result of lifestyle is determined by genetic factors. Multiple genetic variants that modulate the risk of an individual for the development of diabetes have been identified, including genetic variants related to mitochondrial function. A high penetrance mutation in mtDNA has been found to be associated with the MIDD syndrome. This mutation predominantly affects the activity of complex I of the respiratory chain [43]. In the same biochemical pathway a mutation in the LARS2 gene, which encodes mitochondrial leucyl-tRNA synthetase 2, also modulates the risk for diabetes [44]. Furthermore, genetically determined changes in expression levels of components of the mitochondrial respiratory chain have been found to be associated with an increased risk of developing type 2 diabetes mellitus [45]. These genetic factors are likely to result in a decreased capacity of the mitochondria to remove fatty acids through uncoupled β-oxidation. Thereby, a genetic predisposition can lower the threshold for fatty acid release by adipocytes into the circulation and as a result enhance the risk for type 2 diabetes. Our model of the pathogenesis of type 2 diabetes mellitus requires experimental verification of several points, and it certainly does not exclude the involvement of additional factors such as the coregulation of whole-body insulin action and insulin secretion through adipokines and the involvement of uncoupling proteins in setting the threshold for fatty acid-induced uncoupling of mitochondria. However, we see the way fatty acids interact with mitochondria in the cytosol of adipocytes as a major initiating event in the disease process leading to the metabolic syndrome and type 2 diabetes mellitus.

Sex_Abuse-2-2-1764595

Assessing Sexual Arousal with Adolescent Males Who Have Offended Sexually: Self-Report and Unobtrusively Measured Viewing Time

Sexual arousal was assessed using three approaches: the Affinity (Version. 1.0) computerized assessment of unobtrusively measured viewing time (VT), Affinity self-report ratings of sexual attractiveness, and a self-report sexual arousal graphing procedure. Data were collected from 78 males, aged 12–18 (M=15.09; SD=1.62), who acknowledged their sexual assaults. The pattern of responses to all three assessment techniques was remarkably similar, with maximal sexual interest demonstrated and reported for adolescent and adult females. Both self-report procedures could significantly distinguish those adolescents who assaulted a child from those who assaulted peers or adults. The self-report procedures could also significantly discriminate those adolescents with male child victims. The Affinity VT approach significantly differentiated those adolescents who assaulted male children from those who assaulted other individuals. No assessment technique could accurately identify those adolescents with exclusively female child victims. Overall, the results suggest that structured, self-report data regarding sexual interests can be useful in the assessment of adolescents who have offended sexually. Introduction Much of the treatment that was provided in the 1980's and 1990's to adolescents who offended sexually was predicated on the notion that deviant sexual interests played a critical role in the sexual assaults. For example, adolescents were often asked to track their sexual interests in violence and/or prepubescent children for lengthy time periods by completing deviant fantasy logs and charts and to participate in punishment-based procedures, such as covert sensitization, to reduce the strength and frequency of their presumed deviant sexual interests. In the last several years, however, this heightened focus on the role of deviant sexual arousal has shifted considerably; in part because it is now recognized that most adolescents who commit a sexual offense do not display primarily deviant sexual interests. For example, in their analysis of archival penile plethysmograph (PPG) data, Seto, Lalumière, and Blanchard (2000) reported that 25% (or 10 out of 40) of a small sample of adolescents who had offended sexually demonstrated maximal sexual interest in prepubescent children. With an overlapping yet augmented sample, Seto, Murphy, Page, and Ennis (2003) found that 30% (76 out of 253) of adolescents who had offended sexually responded equally or more to child stimuli during PPG assessments. In a study of 136 adolescents who had committed a sexual offense, Worling (2004) found that 36% of the adolescents were rated by clinicians as having sexual interests in prepubescent children and/or sexual violence. Although deviant sexual interests likely play a role in the etiology and/or maintenance of adolescent sexual offending for some adolescents, there are likely other factors to consider such as intimacy deficits, antisociality, attitudes supportive of sexual offending, and opportunity, for example. Measurement of Sexual Arousal Given the assumption that individuals will be reluctant to reveal their sexual thoughts and feelings during an interview, some believe that it is necessary to utilize a physiological measure of sexual interest. The most popular technique for measuring sexual arousal with adult males is the PPG. Although it has been argued by some that the PPG can provide valuable information for adult male clients who have offended sexually (Lalumière & Harris, 1998; Seto, 2001), there are some concerns regarding the reliability and validity of PPG data collected from adult males (Konopasky & Konopasky, 2000; Marshall & Fernandez, 2000). There are also several ethical and empirical concerns regarding the use of the PPG with adolescents (Becker & Harris, 2004; Hunter & Lexier, 1998; Worling, 1998). For example, adolescence is a period of emerging sexual development and teenagers are continuously developing and refining their sexual scripts, identities, and preferences. The potential for iatrogenic harm from exposing adolescents to visual and/or auditory depictions of deviant sexual activities is considerable, yet this has never been examined. Second, there is little compelling evidence regarding the reliability or validity of a PPG procedure with adolescents. Indeed, in studies with adolescents, PPG data are significantly influenced by such variables as the adolescent's age (Kaemingk, Koselka, Becker, & Kaplan, 1995) and history of physical and sexual abuse (Becker, Hunter, Stein, & Kaplan, 1989; Becker, Kaplan, & Tenke, 1992). Furthermore, although PPG data are predictive of subsequent sexual offending for adults (Hanson & Bussière, 1998; Hanson & Morton-Bourgon, 2004), the available research with adolescents indicates that there is no significant relationship between phallometrically-measured sexual deviance and sexual assault recidivism (Gretton, McBride, Hare, O’Shaughnessy, & Kumka, 2001; Gretton et al., 2005). Becker et al. (1992) also found that a majority of adolescents who denied their sexual offenses provided invalid deviant-arousal data. In a chart-review study by Seto et al. (2000), it was found that PPG data were only moderately discriminative for adolescent males who sexually assaulted male victims and that adolescents who offended against female children could not be differentiated from a nonoffending population. Hunter, Becker, and Goodwin (1994) similarly found that only those adolescents with male victims demonstrated significant deviant arousal using the PPG. In their review of the research, Becker and Harris (2004) concluded that the PPG may be useful only for the following adolescent clients: (i) older teenage males, (ii) individuals who acknowledge their sexual assaults, and (iii) those who offend sexually against males. Given the scientific and ethical concerns regarding phallometric assessment, a number of researchers and clinicians have begun to use an alternative physiological measure: unobtrusively measured viewing time (VT). With this procedure, clients are asked to rate the sexual attractiveness of photographs of a variety of models while the response time to provide the ratings is unobtrusively recorded. The assumption underlying this technique is that people will look longer at stimuli that they find sexually attractive relative to stimuli that they find sexually unattractive. In several VT assessment systems that are commercially available, the models in the photographs are clothed and are not displayed in sexual poses. For many clinicians, therefore, the VT assessment addresses some of the ethical concerns raised by the PPG. In studies with nonoffending adults, it has been demonstrated that VT is significantly correlated with self-reported ratings of sexual arousal (e.g., Harris, Rice, Quinsey, & Chaplin, 1996; Lang, Searles, Lauerman, & Adesso, 1980; Quinsey, Ketsetzis, Earls, & Karamanoukian, 1996) and with sexual arousal as measured by the PPG (Harris et al., 1996; Quinsey et al., 1996). There have also been several investigations of the efficacy of VT with adults who have offended sexually. For example, Harris et al. (1996) found that men who sexually assaulted children viewed slides of children longer than they viewed slides of adults. In several studies conducted with the Abel Assessment for Sexual InterestTM (AASI), a VT assessment methodology using photographs of clothed models, Abel and his colleagues have reported encouraging data with respect to the internal consistency and discriminant validity of the AASI with adult males who acknowledged sexual assaults against children (Abel, Jordan, Hand, Holland, & Phipps, 2001; Abel, Lawry, Karlstrom, Osborn, & Gillespie, 1994; Abel, Huffman, Warberg, & Holland, 1998). In one of the few studies of the utility of the AASI conducted outside of the developer's laboratory, Letourneau (2002) found that PPG and AASI responses from 57 adult males were significantly correlated for most stimulus categories, and that both assessment approaches significantly differentiated those participants with male child victims from those with victims from other groups. Furthermore, the AASI VT procedure could identify those men with adolescent female victims. More recently, Gray and Plaud (2005) compared the AASI and the PPG using data collected from 39 men who had offended sexually against a child less than 11 years of age. They found that both procedures significantly identified sexual interest in children. There have been few published studies of the utility of VT with adolescents. In one paper, Smith and Fisher (1999) used Abel's VT system (AASI) with 81 adolescent males. They concluded that this particular approach yielded little convincing evidence with respect to reliability or validity. However, it is important to note that Abel (2000) later disputed many of the results and interpretations presented by the investigators. More recently, Abel et al. (2004) evaluated the AASI with data collected from 1,704 males aged 11 to 17. The authors reported that VT for images of children was moderately correlated to the number of child victims (r=.18) and the number of acts of child sexual offending (r=.23). It was also noted that VT for child stimuli could moderately differentiate those adolescents who offended sexually against children from those who offended against peers or adults (AUC=.64). Despite the prevailing assumption that self-reported sexual interests from individuals who commit sexual offenses will be necessarily biased and prejudicial, there is growing evidence of the utility of self-reported data. For example, Laws, Hanson, Osborn, and Greenbaum (2000) found that self-reported sexual interests showed more classification accuracy than PPG data with respect to victim gender in a sample of men who offended against children. Laws et al. also found that the self-report methodology used (card sort) had excellent levels of internal consistency. In a similar study, Day, Miner, Sturgeon, and Murphy (1989) found that self-report data from a structured questionnaire regarding sexual thoughts, feelings, and behaviors could accurately classify men according to the gender of their child victims. Turning to research with adolescents, Seto et al. (2000) found that most participants who acknowledged a sexual interest in children were classified as “pedophilic” using the PPG. Furthermore, Seto et al. found that those participants who offended sexually against children but who denied sexual interest in children scored significantly lower on the pedophilic index. Daleiden, Kaufman, Hilliker, and O’Neil (1998) found that adolescents who offended sexually disclosed significantly more deviant sexual behaviours relative to both nonsexual offenders and a nonoffending group. Results such as these suggest that individuals may be quite open regarding deviant sexual thoughts, interests, and/or behaviors. Method Participants After obtaining informed consent (and parental consent when required), data were collected from 78 males aged 12–18 (M=15.09; SD=1.62). Participants were assessed at a residential treatment facility in Minnesota (n=44) or at one of three community-based treatment centers in the Greater Toronto (Ontario, Canada) area (n=34). Females were not specifically excluded; rather the facility in Minnesota provides services only to males, and the 34 consecutive Canadian referrals who chose to participate were male. All participants for this study acknowledged a contact sexual offense. During the course of this investigation, none of the adolescents who completely denied their sexual offenses and who were approached to participate (n=5) volunteered for this study. Sixty-seven percent (52/78) of the adolescents committed a sexual offense against at least one child (defined as under the age of 12 and 4 or more years younger than that adolescent at the time of the offense); the remainder (33% or 26/78) of the adolescents committed sexual assaults against peers or adults exclusively (5 of these 26 offended against male peers). With respect to the participants' ethnic origin, 81% (63/78) were Caucasian, 12% were African American/Canadian, 5% (4/78) were native American/Canadian, and 2% (2/78) were Hispanic. Information regarding intellectual functioning was available for 52 of the 78 adolescents: 19% (10/52) of these adolescents had IQ's less than 80, 6% (3/52) of the adolescents had IQ's above 120, and 75% (39/52) had an IQ's within the average range (i.e., IQ between 80 and 120). Assessors reported that 50% (39/78) of the participants disclosed a childhood sexual victimization history and 42% (33/78) of the adolescents disclosed physical abuse within their families. Self-Report Sexual Arousal Graphs To collect self-reported sexual arousal data, adolescents were asked to complete 2 graphs (1 graph for each gender) on which they rated their sexual arousal for 8 age-based categories: 0–3 years, 4–6 years, 7–9 years, 10–12 years, 13–15 years, 16–18 years, 18–24 years, and over 24 years (see Appendix A for self-report arousal graphs and instructions). Adolescents were first asked to rate their level of sexual arousal to the various age groups if there was no physical force or violence involved in the sexual interaction; they were then asked to use a different color of pen and indicate how their sexual arousal would change for each age group when the thought of forced sexual contact was introduced. All ratings were made on a scale from 0 (low) to 10 (high). For all participants, the graph to address sexual arousal to males was completed first, followed by the graph to capture sexual arousal to females. The graphing procedure was completed with the adolescents by a psychologist or social worker during the completion of a comprehensive assessment. There were 7 different clinicians (1 in Minnesota and 6 in Ontario) at the various agencies who assisted the adolescents to complete the graphs. Affinity Assessment of Sexual Interest The Affinity procedure is explained in detail elsewhere (Glasgow, Osborne, & Croxen, 2003), and assessors followed the assessment protocol outlined in the Affinity 1.0 manual (Glasgow, 2001). Briefly, the assessment involves the computerized presentation of photographs of 28 males and 28 females in 4 age categories: toddlers, preadolescents, adolescents, and adults. All of the photographs are of clothed individuals, and none of the models are depicted in sexual poses. The 56 images are presented in a fixed, random order, and participants are asked to rate the sexual attractiveness of each image using a mouse pointer. The on-screen self-report rating scale accompanying each photograph includes the anchors “Very unattractive,” “Neutral,” and “Very attractive,” and there are 19 different, unnumbered gradients on the scale. The Affinity computer program registers a self-report rating score for each photograph ranging from 0 to 18, and it records the time taken (in seconds; accurate to ±0.02 s) to provide each attractiveness rating. The later measure is referred to as on-task latency (OTL). Immediately prior to the debriefing procedure where the nature of the Affinity program was explained, participants were asked to rate—on a scale from 1 (Not at all upsetting) to 10 (Very upsetting)—how upsetting they found the experience of rating the Affinity photographs. They were also asked to indicate how enjoyable the experience was using a scale from 1 (Not at all enjoyable) to 10 (Very enjoyable), and participants’ comments regarding the assessment process were recorded by the assessor. Average Sexual Arousal Scores As a result of the fact that the Affinity program uses four age groups (toddler, preadolescent, adolescent, and adult) for each gender, the eight age groups examined using the self-report graphing procedure were collapsed to approximate the same four Affinity groups to facilitate raw-score comparisons. Specifically, self-report sexual arousal graph data were collapsed as follows: ages 0–3 and 4–6 (toddler); ages 7–9 and 10–12 (preadolescent); 13–15 and 16–18 (adolescent); 18–24 and 24 and over (adult). To facilitate visual comparisons between the three assessment techniques, raw scores were converted to z scores. The average standardized raw scores for the three assessment approaches are displayed in Fig. 1. The raw data distributions are remarkably similar across approaches, despite the differences in measurement scales and assessment methodology. Specifically, average self-reported sexual arousal ratings and VT scores for female adults and adolescents are significantly higher than the arousal data for all remaining stimulus categories. Furthermore, there was significant overlap in the arousal data provided for males of any age and for female toddlers and preadolescents.Fig. 1Average standardized raw scores for Affinity On-Task Latency (OTL; n=78), Affinity Self-Report (n=78), and Self-Report Sexual Arousal Graph (n=72) procedures across stimulus groups. Bars represent 95% confidence intervals. MTOD: Male Toddler; MPRE: Male Preadolescent; MADO: Male Adolescent; MADU: Male Adult; FTOD: Female Toddler; FPRE: Female Preadolescent; FADO: Female Adolescent; FADU: Female Adult Internal Consistency Internal consistency estimates (Cronbach's α) for the 8 Affinity stimulus categories are displayed in Table 1. With the exception of the α for OTL measure for the 7 photographs of female adolescents, all internal consistency estimates were above .70 for the OTL values across the age/gender categories. Note, however, that higher internal consistency estimates were found for all Affinity stimulus categories based on self-report ratings.Table 1Internal consistency estimates for Affinity on-task latency and Affinity self-report ratingsAffinity on-task latency (measuredAffinity self-report ratings (scaleStimulus categoryin seconds) Cronbach's αfrom 0–18) Cronbach's αFemale toddlers.82.96Female preadolescents.79.95Female adolescents.62.87Female adults.72.94Male toddlers.73.97Male preadolescents.82.97Male adolescents.77.94Male adults.77.94 Internal consistency estimates (Cronbach's α) for the self-report sexual arousal graphs were .83 for females (collapsed across the four age groups) and .89 for males (collapsed across the four age groups). For children 12 and under (for both males and females), the internal consistency estimate was .92; for the 4 age groups of both males and females over the age of 12, internal consistency was .77. Validity The assumption underlying VT is that, when asked to rate the sexual attractiveness of a particular model, individuals look longer at photographs that they find sexually arousing relative to the time that they spend looking at photographs that they find not sexually arousing. The correlations pertaining to this issue are displayed in Table 2, and it can be seen that for 7 of the 8 stimulus categories, the time spent viewing the slides (Affinity OTL) was significantly correlated with self-report ratings for the same slides (Affinity Self-Report). The average correlation between Affinity OTL and Affinity self-report ratings across all slide categories was r=.31, p < .01. There was, however, a significant, negative correlation between OTL and Affinity Self-Report for the slides of the adult females. In other words, there was a marked tendency for those adolescents who rated the photographs of adult females as sexually attractive to provide their ratings more quickly than those adolescents who found the photographs less sexually attractive. This, of course, is counter to the assumption that forms the premise of the VT approach.Table 2Correlations between Affinity on-task latency, Affinity self-report, and self-report sexual arousal graph proceduresSelf-report sexualStimulus groupAffinity self-reportArousal graphsMale Toddler Affinity OTL.38**.35** Affinity self-report.64**Male preadolescent Affinity OTL.67**.55** Affinity self-report.79**Male Adolescent Affinity OTL.61**.46** Affinity self-report.63**Male adult Affinity OTL.52*.21* Affinity self-report.50**Female toddler Affinity OTL.49**.44** Affinity self-report.80**Female preadolescent Affinity OTL.48**.38** Affinity self-report.73**Female Adolescent Affinity OTL.24*.16 Affinity self-report.34**Female Adult Affinity OTL−.26*−.17 Affinity self-report.48**Note. OTL: On-task latency. n=78 for all correlations between Affinity self-report and Affinity OTL. n=72 for all correlations with the Self-report sexual arousal graphs.*p < .05; **p < .01. Although the Affinity self-report is based on photographs of various models, and the self-report sexual arousal graphs simply require a rating in response to a numeric age group, the correlations between these two measures of sexual arousal were significant for all 8 stimulus categories. The correlations between OTL and the self-report graphs were also significant for 6 of the 8 stimulus categories. The validity of the two Affinity procedures (OTL and self-report) and the self-report graphing procedures was examined by comparing subgroups of adolescents formed on the basis of the age and gender of their victims. First, an Affinity OTL deviance index was calculated for each participant by dividing the highest mean OTL (in seconds) for male or female toddlers or preschoolers by the highest mean for male or female adolescents or adults.1 An Affinity Self-Report deviance index was calculated using the same mathematical procedure: the highest mean Affinity self-report rating for male or female toddlers or preschoolers was divided by the highest mean self-report rating for male or female adolescents or adults. Finally, a deviance index was calculated in the same fashion for the self-report graphing procedure by dividing the highest rating for a male or female child aged 0 to 12 by the highest rating for a male or female aged 13 and over. To examine the discriminative validity of the three assessment procedures, the area under the Receiver Operating Characteristic (ROC) curve (AUC) was calculated for several different between-group comparisons (see Table 3). Both of the self-report procedures could significantly identify those adolescents with single child victims, multiple child victims, or male child victims (whether defined as ever a male child or exclusively male child victims), and there were no significant differences between the two self-report methodologies, all ZΔ’s ≤ .03, all p’s ≤ .05. The Affinity OTL deviance index was only able to discriminate those adolescents with a male child victim from those adolescents with victims from all other age/gender groups, and neither self-report measure was superior to the Affinity OTL for this discrimination, all ZΔ’s ≤ .03, all p’s > .05. Interestingly, none of the deviance indices examined correctly identified those adolescents with female child victims: whether this was defined as if ever a female child victim or exclusively female child victims. It should also be pointed out that of the 30 adolescents who had 2 or more child victims, 23/30 (or 77%) had at least one male child victim. Only 7/30, or 23%, had 2 or more child victims that were exclusively female.Table 3Area under the ROC curve (AUC) classification data for three measures of deviant sexual interestSexual assault victim(s)Affinity on-task latency deviance index AUC (n=78)95% CIAffinity self-report deviance index AUC (n=78)95% CISelf-report sexual arousal graph deviance index AUC (n=72)95% CIEver a child victim.61.47–.75.67*.54–.79.66*.53–.792 or more child victims.60.47–.73.73**.61–.86.70**.59–.82Ever a male child victim.69**.57–.81.72**.60–.84.72**.60–.84Only male child victim(s).73**.60–.86.74**.60–.89.76**.64–.88Ever a female child victim.42.30–.55.48.35–.61.45.33–.58Only female child victim(s).43.29–.56.42.28–.56.41.27–.56Note. Affinity On-Task Latency Deviance Index = (highest mean on-task latency for male or female toddlers or preadolescents)/(highest mean on-task latency for male or female adolescents or adults). Affinity Self-Report Deviance Index = (highest mean self-reported ratings for male or female toddlers or preadolescent)/(highest mean self-reported ratings for male or female adolescents or adults). Self-Report Sexual Arousal Graph Deviance Index = (highest rating for male and female children aged 0–3, 4–6, 7–9, and 10–12)/(highest rating for males and females aged 13–15, 16–18, 18–24, and 24 and over).*p < .05; **p < .01.Table 4Mean (and SD) deviance indexes for between-group comparisonsComparisonAffinity on-task latency deviance index (n=78)FAffinity self-report deviance index (n=78)FSelf-report sexual arousal graph deviance index (n=72)FEver a child victim Yes.72 (.20)1.6.17 (.27)2.2.29 (.29)1.79 No.65 (.24).08 (.24).19 (.32)2 or more child victims Yes.74 (.20)1.6.24 (.31)7.4 **.34 (.25)4.1* No.67 (.23).08 (.20).19 (.31)Ever a male child victim Yes.77 (.20)7.1 **.24 (.31)7.4 **.37 (.28)7.1 ** No.65 (.21).08 (.20).18 (.29)Only male child victim(s) Yes.84 (.20)10.9 **.32 (.37)11.7 **.39 (.29)4.2* No.66 (.21).09 (.19).20 (.29)Ever a female child victim Yes.66 (.18)1.9.10 (.16)1.6.24 (.27)0.2 No.73 (.24).18 (.31).27 (.32)Only female child victim(s) Yes.64 (.19)0.1.08 (.13)0.2.17 (.26)0.2 No.72 (.22).17 (.29).28 (.31)Participant a victim of sexual abuse Yes.68 (.22)0.4.20 (.03)0.9.24 (.29)0.7 No.72 (.21).31 (.05).27 (.30)Participant a victim of physical abuse Yes.67 (.23)0.7.13 (.22).07.20 (.25)1.7 No.71 (.21).15 (.28).29 (.32)Location of participant Minnesota.70 (.22)0.0.23 (.03)0.1.26 (.31)0.2 Greater Toronto area.70 (.22).33 (.07).23 (.28)Note. Affinity On-Task Latency Deviance Index = (highest mean on-task latency for male or female toddlers or preadolescents) / (highest mean on-task latency for male or female adolescents or adults). Affinity Self-Report Deviance Index = (highest mean self-reported ratings for male or female toddlers or preadolescent)/(highest mean self-reported ratings for male or female adolescents or adults). Self-Report Sexual Arousal Graph Deviance Index = (highest rating for male and female children aged 0–3, 4–6, 7–9, and 10–12)/(highest rating for males and females aged 13–15, 16–18, 18–24, and 24 and over).*p < .05; **p < .01. Given that there are few published data using the Affinity assessment system, one-way ANOVAs were also computed for the 3 deviance measures for each of the comparisons described above. The means and standard deviations for these comparisons are presented in Table 4 to facilitate the calculation of effect sizes and comparisons with future investigations. The pattern of group differences is similar to that found with the ROC analyses. Also note in Table 4 that there were no significant differences on any of the deviance measures as a function of sexual or physical victimization history or location of data collection. It should also be pointed out that there was no significant correlation between participant age and the deviance index calculated for Affinity OTL (r=.02), Affinity Self-Report (r=.09), or the self-report sexual arousal graphs (r=−1.1), all p’s > .05. Correlations between the three deviance measures and the number of known child victims are displayed in Table 5.Table 5Correlations between deviance indices and number of known child victimsVariableAffinity on-task latency deviance index (n=78)Affinity self-report deviance index (n=78)Self-report sexual arousal graphs deviance index (n=72)Number of child victims.07.23*.14Number of male child victims.33**.50**.34**Number of female child victims−.13−.04−.05Note. Affinity On-Task Latency Deviance Index = (highest mean on-task latency for male or female toddlers or preadolescents)/(highest mean on-task latency for male or female adolescents or adults). Affinity Self-Report Deviance Index = (highest mean self-reported ratings for male or female toddlers or preadolescent)/(highest mean self-reported ratings for male or female adolescents or adults). Self-Report Sexual Arousal Graph Deviance Index = (highest rating for male and female children aged 0–3, 4–6, 7–9, and 10–12)/(highest rating for males and females aged 13–15, 16–18, 18–24, and 24 and over).*p < .05; **p < .01. Combinatory deviance scores were also computed to determine whether the results from multiple assessment methods could enhance discriminatory power. Specifically, the following combined deviance scores were examined: Affinity OTL + Affinity self-report; Affinity OTL + self-report arousal graph; Affinity self-report + self-report arousal graph; and Affinity OTL + Affinity self-report + self-report arousal graph. For the six between-groups comparisons outlined in Table 3, there was no significant increase in predictive efficiency for any of the combinatory deviance scores, all ZΔ’s ≤ .06, all p’s > .05. The highest AUC value obtained for any comparison was 0.79 (95% CI; .66–.91), and this represented the ability to detect multiple child victims using the combination of Affinity self-report plus the self-report sexual arousal graphing procedure. Given the relative novelty of the self-report sexual arousal graphs, a final examination of the validity of this technique involved an analysis of the frequency with which adolescents ever rated their sexual arousal over the “0” line for male or female children aged 0 to 12 years. The data related to this analysis are presented in Table 6, and it can be seen that if an adolescent ever provided a rating over “0” for children aged 0 through 12 years, this was significantly related to the choice of a single child victim, multiple child victims, and male child victims. As with all previous analyses, however, this particular procedure could not discriminate those adolescents with female child victims.Table 6Frequency of self-report sexual arousal graph ratings over “0” for children aged 0 to 12Ever more than “0” sexual arousalNoYesTotalEver a child victim No16925 Yes153247 Total314172χ2 = 6.9*Ever 2 or more child victims No261945 Yes52227 Total314172χ2 = 10.6**Ever a male child victim No261844 Yes52328 Total314172χ2 = 11.87**Only male child victim(s) No282755 Yes31417 Total314172χ2 = 5.86*Ever a female child victim No192342 Yes121830 Total314172χ2 = 0.20Only female child victim(s) No213253 Yes10919 Total314172χ2 = 0.97Note. Ratings were provided on a scale from 0 (low) to 10 (high).*p < .05; **p < .01. Finally, participants were asked to rate how upset they were after completing the Affinity rating task. On the scale from 1 (Not at all upsetting) to 10 (Very upsetting), the mean rating was 2.38 (SD=1.88). Only 6 adolescents provided a rating of 5 or higher, and 66% of the participants provided a rating of 1 or 2. Anecdotally, those adolescents who rated the Affinity procedure as somewhat upsetting most often commented that there were too many photographs of young children or that there were too many photographs of males. Participants were also asked to rate how enjoyable they found the Affinity computerized assessment using a scale from 1 (Not at all enjoyable) to 10 (Very enjoyable). The mean rating for this variable was 3.70 (SD=2.10). Those adolescents who found the procedure enjoyable most often commented that it was “interesting”; those who rated the Affinity assessment as Not at all enjoyable most often mentioned that it was “boring.” Discussion Adolescent sexual interest was assessed using three different assessment procedures: the Affinity (version 1.0) VT procedure, the Affinity self-report procedure, and a self-report sexual arousal graphing procedure. Overall, the internal consistency estimates for all three measures were acceptable for most age and gender groups, and all three assessment approaches significantly differentiated those adolescents with male child victims from those adolescents who never offended sexually against a male child. On the other hand, none of the assessment techniques could successfully differentiate adolescents with female child victims from those who offended sexually against other groups. The finding that those adolescent males with male child victims demonstrated the highest interest in prepubescent children—either through VT or self-report—is consistent with results from several studies in which the PPG was used (Becker et al., 1989; Hunter et al., 1994; Murphy, DiLillo, Haynes, & Steere, 2001; Seto et al., 2000). As has been noted by many of these researchers, it is quite likely, therefore, that many adolescents who commit a sexual offense against a male child are motivated, at least in part, by a sexual interest in prepubescent children. Of course, not all adolescents who target male children demonstrate deviant sexual arousal, and other factors, such as opportunity or antisocial attitudes, for example, need to be considered in the etiology and/or maintenance of their sexual offenses. Similarly, the fact that many participants who offended sexually against female children did not demonstrate or report significant sexual interest in prepubescent children is also consistent with PPG data collected from adolescents (e.g., Seto et al., 2000) and with both PPG and VT data collected from adults (e.g., Letourneau, 2002). As has been noted by others, this may be a result of the fact that the various assessment procedures are not yet able to discern such a sexual interest. Alternatively, given the fact that this finding has been observed with both adults and adolescents—using PPG, VT, and self-report—it is possible that there are many males who commit sexual offenses against prepubescent girls for reasons other than deviant sexual interest. As noted with respect to some of those who offend against boys, it could be that factors such as opportunity or abuse-supportive attitudes, for example, are critical in the formation and/or continuation of offending behaviors. The AUC reported by Abel et al. (2004) for the AASI to differentiate those adolescents who ever assaulted a child from those who sexually assaulted a peer or adult was .64. The AUC for the Affinity VT in the present study (AUC=.61) was remarkably similar for the same comparison. As noted clearly by Abel et al. (2004), this suggests that VT alone should not be used to classify adolescents who have offended sexually with respect to victim age. Of course, despite the fact that both of the self-report procedures significantly differentiated adolescents with versus without child victims, no one assessment methodology produced an AUC over .74 regardless of the discrimination examined, and no combinatory deviance score produced an AUC over .79. This indicates that there is certainly much room for improvement, and that neither self-report procedure should be used on its own, or in combination, to determine whether or not an adolescent has committed a sexual offense against a person from a specific age/gender category. Of course, it is also possible that researchers will never find that they can get better discrimination between subgroups on the basis of victim gender and/or age because of the fact that sexual interests and attitudes are still being formed during adolescence. Although victim gender and/or age may be proxy markers of sexual interest for some adults who offend sexually, particularly those who offend against boys, it is likely that the choice of victim is less often related to fixed sexual interests for adolescents. Despite the encouraging results reported regarding both self-report procedures, it should be stressed that there were no adolescents who completely denied their sexual crimes included in this study. As such, these results should not be generalized to that subgroup. It would be interesting to investigate the utility of both VT and self-report with those who deny their sexual offenses as Becker et al. (1992) found that the PPG was not useful with this group as most of these adolescents were classified as “nonresponders.” There were also no female participants included in this study. Given the relative ease with which both Affinity and the self-report graphing procedures can be carried out with both male and female adolescents, it would be beneficial to learn about the psychometric properties of these assessment techniques with female participants. It would also have been ideal to have included a group of adolescents without a history of sexual offending. This would have permitted analyses regarding normative responses for both VT and self-report, and there may have been greater discrimination accuracy for specific subgroups if nonoffending youth were included. Finally, it should also be pointed out that there were only 78 participants in this investigation; therefore, the probability of a Type 1 error should be considered given the number of statistical tests herein. Although VT was significantly correlated with ratings of sexual interest for the same photographs within most of the stimulus categories, there was a significant negative correlation between VT and sexual attractiveness ratings for photographs of adult females. It is difficult to comment on the uniqueness of this finding, however, as few researchers have published correlations between self-reported ratings and VT scores for separate stimulus categories. Rather, it is more common for researchers to report an average correlation across the various age/gender categories (r=.31 in the present study). One can quickly see in Fig. 1 that, on average, adolescents viewed photographs of adolescent and adult females for approximately the same amount of time, but that the sexual attractiveness ratings provided for the photographs of the adult females were significantly higher than those ratings given to the photographs of adolescent females. Given that the relationship between VT and self-reported attractiveness ratings were in the expected direction for most stimulus categories, it may be that this result is spurious and sample specific. On the other hand, there may be something about the seven photographs of adult women used in the Affinity program that contributed to this unexpected result. Alternatively, it is possible that some participants provided exaggerated ratings to the slides depicting adult females in an effort to appear more socially desirable. Finally, it is possible that there is an inverted-U-shaped distribution of VT responses such that respondents provide the fastest ratings to categories that they find both highly attractive and highly unattractive. Given that the majority of adolescents in this study found only 2 of the 8 categories of photographs as highly sexually attractive (adolescent and adult females), the fast VT responses to the 6 remaining categories would account for an overall positive correlation between low VT and low attractiveness ratings. In other words, participants provided low average attractiveness ratings after low average viewing times for 75% (42/56) of the slides. It would be ideal for researchers to examine the possibility of such an inverted-U-shaped distribution in future research with VT technology; particularly given the assumption underlying the technique that there is a linear relationship between viewing times and attractiveness ratings for all age/gender categories. Most of the adolescents found that participating in the Affinity VT assessment was not very upsetting, and many even found the process to be at least somewhat enjoyable. Given the potential intrusiveness of the PPG procedure, it is interesting that researchers have never collected information from adolescents regarding their experience of providing PPG data. Given the current scientific limitations of the PPG with adolescents, and the ethical concerns involved in conducting such an assessment, it is encouraging that at least one VT approach (Affinity 1.0) can produce some useful information regarding sexual interests without significant negative impact. Furthermore, it is important to point out that age at time of assessment, childhood sexual victimization history, and physical abuse history were not significantly related to deviance indices calculated for Affinity VT, Affinity Self-Report, or the self-report graphing procedure. Recall that some researchers have found that PPG data are correlated with these factors. The results of the present study also lend support to the growing evidence that structured, self-report methodologies can provide the assessor with valuable information regarding an adolescent's sexual interests. For example, adolescents with a history of sexual offenses who rated their sexual arousal to children 12 years of age and under over the “0” mark on the self-report arousal graphs were significantly more likely to have offended against a single child victim, multiple child victims, or male child victims. Of course, regardless of the particular assessment methodology chosen, it is always prudent to supplement any sexual-interest data with information from official documentation and with reports from parents and other adults who may be familiar with the adolescent. It is also essential to be mindful of the fact that sexual interests are not necessarily crystallized for many adolescents and, furthermore, that victim choice may not necessarily be a proxy for sexual preferences. Even when deviant sexual interests are identified, however, there is very little research regarding how best to help adolescents in this regard. Although covert sensitization is by far the most popular behavioral treatment for deviant-arousal reduction in programs in the United States for adolescents who have offended sexually (McGrath, Cumming, & Burchard, 2003), there are no published data regarding the efficacy of this approach for adolescents. In addition to the continued refinement of measurement strategies, therefore, it is critical that we examine the impact of treatment and management techniques for those adolescents who evidence deviant sexual interests.

Eur_Child_Adolesc_Psychiatry-3-1-1914287

Problem behavior in a community sample of 14- and 19-month-old children

Few studies have examined the presence, structure, and stability of behavior problems in a community sample of 14- and 19-month-old infants. A questionnaire with items on emotional, attentional, and impulsive behavior and social communication was completed by the parents of 6,491 infants aged 14 months and 1,803 infants aged 19 months. Particularly externalizing behavior problems were reported to present sometimes or often for 50% of more of the sample and could be considered as common. In contrast, social communication problems were reported to be present in less than 10% of the sample. Overall, boys showed more problem behaviors than girls. Principal component analysis at the 14 months data revealed seven factors, which could all be replicated in the 19 months data. Pearson correlations between scores at 14 months and 19 months were highest for oppositional and attention factors (0.68 and 0.63) and lowest for the inhibiton factor (0.38). More than 50% of those scoring in the top 10% for total problem score at 14 months were in the top 10% at 19 months. These results will facilitate the recognition of psychopathology at very early age and the study of its development over time. Introduction Psychopathological research has only recently started to focus on the expression of problematic development at a very young age, i.e., younger than 18 months, and few data are available on the expression of problem behavior in somewhat older children between 1- and 3-year-old [5, 9, 23]. Indeed, the lack of normative data for normal and deviant development in children younger than 18 months may lead to inadequate early recognition of serious behavior problems [8, 33, 34]. The differentiation of normal behavior from problem behavior is particularly challenging in young children [19]. Many symptoms of psychiatric disorders in childhood and adolescence are considered to be relatively normal in preschool-age children. For example, during the preschool years, aggressive, hyperactive, and non-compliance behaviors and separation problems are relatively common and can be viewed as relatively normal within a developmental perspective [6]. Still, these behaviors can be quite stressing for parents and are generally referred to as problem behaviors. Children younger than 18 months are learning to control themselves, to become attached to people and objects, and to explore, and are only beginning to form stable behavior patterns. Earlier studies reported that problem behavior affected between 7.3% and 24% of 2- and 3-year-old children [5, 12, 14, 18, 20, 22, 24, 26, 27, 31, 32], but there are few data for children younger than this age. One developmental screening study reported that about 10% of infants had emotional and behavioral problems in the first year of life [15], and another study reported that about 6% of the 1-year-old children had high scores on the Difficult Child domain of the Parenting Stress Index [5]. An important reason to study behavior at a young age is that it may predict problem behavior at a later age. For example, indices of temperament at 3 years were found to predict the presence and the type of psychiatric disorder in young adulthood [10], and behavior problems at 2 years, such as aggression, non-compliance, and difficult temperament, were valuable predictors of lasting problems at a later stage of development [17]. Approximately 50% of the 2- and 3-year-old children with externalizing psychiatric disorders continued to have these problems later in life [6, 21]. Such problems appear to be stable from an age of 18 months onward [23], but to our knowledge stability has not been evaluated in younger children. More knowledge on the expression and stability of problem behaviors before the age of 18 months would be advantageous, especially for doctors and nurses working in well baby clinics. In the Netherlands, infants visit well baby clinics according to a fixed scheme. At 14 months there is a relatively extensive check up. Since this is a moment for vaccination the attendance rate for this check up is high. Therefore, more knowledge on problem behaviors at especially 14-month-old was desirable. The current study was designed to investigate behaviors at this young age and had the following aims. The first aim was to examine the occurrence of problem behaviors or lack of competences at younger age that are considered to be psychopathological in later childhood. A second aim was to examine whether the constructs of problem behaviors and competences usually found in older children are already present at 14 and 19 months of age. A third aim was to examine the short-term stability of parent-reported problem behavior and lack of competences between age 14 and 19 months. To achieve these aims, we used a questionnaire with 55 items on mainly externalizing and social communication behaviors. Method Sampling The sample was a total population birth cohort of children born between August 2000 and August 2001 (N = 12,297) in the province of Utrecht in the Netherlands. The province has a population of about 1.117.997 and includes rural and highly urbanized areas. Parents of 14-month-old children received a request from the provincial inoculation center to complete our questionnaire concerning infant problem behavior. This request was accompanied by an information brochure that explained the purpose of this study (at time t1). Parents of 6,491 infants (i.e., 53% of the 12,297 infants eligible) returned the questionnaire and consented to participate. About equal numbers of boys and girls participated. Unfortunately, we were not allowed access to information on non-responders, and so we investigated potential selection bias by comparing the data of responders to demographic data for the province of Utrecht [25]. Responders lived somewhat less often in urban areas (responders 46.6%, province of Utrecht 48.1%) and had a higher educational level, which was defined as parents with at least a high school or college degree (responders sample: mothers 43.4%; fathers 44.2%; province of Utrecht: mothers 38.9%; and fathers 36.0%). Chi-square test for differences confirmed significant differences with respect to urbanization grade (χ2 = 86.437, df 4, P < 0.000) and educational level of the mother (χ2 = 328.56, df 2, P < 0.000) and of the father (χ2 = 313.18, df 2, P < 0.000). Further, our sample contained significantly fewer children (responder sample: 5.3%, province of Utrecht: 17.94%) of non-Dutch descent (χ2 = 695.785, df 5, P < 0.000). To parents of children born in June, July, and August 2001 (N = 4,051) who had also received the questionnaire at 14 months and had returned the questionnaire the first time, the questionnaire was sent again (at time t2) when the children were 19 months of age. This time 1,803 subjects (44%, 919 boys, 889 girls) participated. There were no significant differences in demographic characteristics between the 19-month-old group and the 14-month-old group. For further analyses of the 14- and 19-month-old data, we used the 14-month-old data (N = 6,491), and the 19-month-old sample (N = 1,803), respectively. When both samples were compared we used the combined sample (N = 1,803). Description of participants Approximately 88% of the parents were Dutch, 1.3% were Turkish, 1.8% were Moroccan, 1.3% were from the Caribbean (Surinam or the Dutch Antilles), 0.4% were from other Western countries, and 0.5% were of non-Western origin. Questionnaires were completed by mothers (80.5%), fathers (10.7%), and by both (0.6%). Mean child age was 14.80 months (SD = 1.84). The mean number of children in the family was 1.80 (SD = 1.03). Mean birth weight was 3,537.5 g (SD = 823.9). Of all infants, 1% had a physical or mental disability, 4.1% had a physical disease, and 6.3% used medication. Parental education level varied: 43.8% had a high level of education, 38.1% had a medium level of education, and 15.5% had a low level of education. Measures Behavior problem questionnaire We used a combination of 55 items on a 3-point Likert scale (0 ‘‘never,’’ 1 ‘‘sometimes,’’ 2 ‘‘often’’) focusing primarily on externalizing behavior, social communication problems, and internalizing items. In selecting items, we first composed a large pool of potentially interesting items concerning children younger than 18 months by reviewing instruments for older children, namely The Infant-Toddler Social and Emotional Assessment; The Child Behavior Checklist for 1.5–5; and the Vineland Social-Emotional Early Childhood Scales [1, 7, 29, 30] and based on our clinical experience with very young children. Then a multidisciplinary panel of experts with clinical and research experience with infants and toddlers made a final selection of items based on two criteria, namely, the item had to be specific for externalizing, social communication or internalizing problem behaviors and to be usable for infants younger than 18 months. The final selection included items that covered attentional problems, hyperactivity, oppositional behavior, aggressive behavior, verbal and non-verbal aspects of social communication (such as initiating and responding) and internalizing behaviors such as problems with eating or sleeping. Data analyses Items were recoded in such way that a score of two represented the complete presence of an unwanted behavior or the complete absence of a wanted behavior. Common and uncommon problems were identified using the approach of Koot and Verhulst [18]. Common problems were defined as all items that scored 1 (present sometimes) or 2 (present often) for 50% or more of the sample. Uncommon problems were defined as all items that scored 1 or 2 for less than 10% of the sample. The structure of problem behavior was examined by entering all items in a principal component analysis (PCA) with varimax rotation. The number of factors was based on the scree test [13]. Items with a minimum factor loading of ≥0.30 were included [2, 4]. To validate the findings, PCA was performed on two split half samples Pearson correlations were calculated to examine the interrelationship between the factor sumscores. Finally, in order to examine possible two higher-order factors, PCA was performed requiring all items to load on two factors and Pearson correlations were calculated to examine their interrelationship. Stability was examined parametrically by calculating Pearson correlations between scores at 14 and 19 months. Further, stability was examined non-parametrically by examining how many children with high scores (above the 75th or 90th percentile) on Total Problem scores and single factor scores at 14 months still had high scores at 19 months. ANOVA was performed to assess the effect of sex on the scores of each problem item and for the factor sum scores. Cohen’s d value is reported as an index of effect size (small, d = 0.2; medium, d = 0.5; or large, d > 0.8) [11]. All analyses were two-tailed and a 0.05 level of significance was used. These analyses were performed using the Statistical Package for Social Sciences (SPSS for Windows, version 11.5). Results Frequencies Common and uncommon problem behaviors and their frequency of occurrence are shown in Table 1. Almost all common problem behaviors were externalizing items, whereas uncommon problems were social communication and internalizing items. Seven of 13 common behavior problems and two of eight uncommon behavior problems were more common among boys than among girls. Table 1Common and uncommon problem behavior at 14 months% With score 1 (sometimes)% With score 2 (often)% With score 1 or 2Common problem behavior  Quickly shifts from one thing to othera61.926.087.9  Angry moodsa63.59.673.1  Demands must be met56.814.170.9  Can’t stand waiting, wants everything now51.116.267.3  Uses toys for fantasy playa35.229.865.0  Wants a lot of attention50.314.464.7  Accident prone56.46.462.8  Stays close to parent54.04.558.5  Easily upset46.37.553.8  Doesn’t sit still during storya39.414.053.4  Won’t share toys or other thingsa46.15.952.0  Doesn’t keep trying46.75.151.8  Can’t sit still, restless, hyperactivea40.011.051.0Uncommon problem behavior  Fails to follow with eyes8.01.49.4  Doesn’t show interest in new objects7.10.67.7  Doesn’t make happy noises6.70.87.5  Doesn’t make noises spontaneously5.81.16.9  Has less fun than others4.51.56.0  Doesn’t react at calling namea5.30.65.9  Seems unhappy without good reason4.90.75.6  Doesn’t imitate simple gesturesa4.30.95.2Note: Common is defined as >50% with score 1 or 2. Uncommon is defined as <10% with score 1 or 2. Please note further that items covering lack of competences were recoded in such way that a score of 2 represented the complete presence of an unwanted behavior or the complete absence of a wanted behavior. This explains why all items have the same-sign factor loadingsa Significantly more boys (P < 0.001) Factor structure of problem behavior The principal component analysis with varimax rotation with eigenvalues >1.5 for two equally sized split half samples resulted in two very similar solutions. The scree test suggested a seven-factor PCA solution (eigenvalue >1.5) with a percentage explained variance of 34.6% (see Appendix to this paper for details). The PCA for the 19-month-old children showed the same factors as the analysis at 14 months, with 37.7% of explained variance. Five of the 55 items had higher loadings at 19 months on other factors than at 14 months. Pearson correlations between the factor scores (Table 2a) suggested a higher-order division into two problem domains, one reflecting total problem behaviors and one reflecting competences. Oppositional problem behavior at 14 months of age was mostly related to the factors Attention, Inhibition, and Dysregulation, with r values varying between 0.37 and 0.52 and the factor Language Development was mostly related to the factors Explorative Behavior and Communicative Intent (r values of 0.41 and 0.42, respectively). Indeed, principal component analysis confirmed the underlying structure of the higher-order factors when all items were forced to load on two factors and were named Total Problem Behavior (eigenvalue 6.9, explained variance 11.5%) and Competences (eigenvalue 4.5, explained variance 7.5%). Pearson correlation of these higher order constructs revealed a low correlation (0.12). Table 2Pearson correlation matrix of factor sumscores of (a) 14-month-old and (b) 14- and 19-month-oldQuestionnaire domainsOppositional behaviorLanguage developmentAttentionExplorative behaviorCommunicative intentDysregulationInhibition14 monthsaOppositional behavior1Language development−0.0011Attention0.522**0.106**1Explorative behavior0.0050.412**0.167**1Communicative intent0.102**0.422**0.142**0.351**1Dysregulation0.436**0.0220.3150.060**0.124**1Inhibition0.379**0.098**0.235**0.076**0.308**0.308**114 months19 monthsbOppositional behavior0.678Language development0.0290.476**Attention0.498**0.112**0.625**Explorative behavior0.0890.264**0.188**0.452**Communicative intent0.057**0.255**0.131**0.206**0.466**Dysregulation0.353**0.069**0.300**0.080**0.127**0.445**Inhibition0.306**0.100**0.247**0.086**0.120**0.226**0.380Note. *Correlation is significant at the 0.05 level (2-tailed)**Correlation is significant at the 0.01 level (2-tailed) A significant sex difference (boys had higher scores than girls) was found on several factors, namely Oppositional Behavior (F = 15.20, df 1, P < 0.000), Language Development (F = 108.88, df 1, P < 0.000), Attention (F = 22.56, df 1, P < 0.000), Explorative Behavior (F = 18.05, df 1, P < 0.000), and Communicative Intent (F = 15.76, df 1, P < 0.000). The effect sizes were small (Oppositional Behavior, d = 0.09; Language Development, d = 0.26; Attention, d = 0.11; Communicative Intent, d = 0.10). No significant sex differences were found for Dysregulation and Inhibition. At 19 months, the same sex differences were found, except for Communicative Intent. Stability over 5 months Pearson correlations between the factor sum scores at 14 and 19 months varied between 0.38 and 0.68, with the highest correlation being found for the Oppositional (0.68) and the Attention factors (0.63) and the lowest for the Inhibition factor (0.38). This indicates moderate stability of the constructs of problem behavior over the 5-month period (See Table 2b). The most stable problems were problems of Communicative Intent, Oppositional Behavior, and Attention (see Table 3). At least half of the group with scores at or above the 75th percentile range at 14-month-old had high scores on problem behaviors at 19 months. With regard to the Total Problem sum scores, 75.5% of the group with scores at and above the 90th percentile of the Total Problem score at 14 months still had problem behaviors at 19 months within the75th percentile range. Also, 53.1% of the children at or above the 90th percentile still had high scores within the highest range 5 months later. On the other hand, about 88.1% of the children with scores below the 75th percentile at 14 months remained below the 75th percentile at 19 months. However, 3.4% of the infants with the least problems on the Total Problem score at 14 months scored at or above the 90th percentile at 19 months (see Fig. 1). Table 3Mean factor scores at t1 (14 months) and t2 (19 months) and percentage of children with scores above 90th and 75th percentile on both 14 months t1 and 19 months t2 (N = 1,803)Behavioral dimensionMean factor scores at t1 and t2% Of children scoring >90th percentile at both t1 and t2% Of children scoring >75th percentile at both t1 and t2Oppositional behavior0.370.4244.860.3Language development0.470.4431.355.8Explorative behavior0.420.4434.250.3Attention0.420.4245.056.7Inhibition0.640.6529.043.7Communicative intent0.500.9246.851.3Dysregulation0.730.3527.551.1Note. The columns represent the percentage, of those children scoring more than 90th/75th percentile at t1, who were also above 90th/75th percentile at t2Fig. 1Transitions (expressed as percentages of children) between 14 months and 19 months for the 90th and 75th percentile cutoff of the total problem behavior score Discussion Recently, more interest has been paid to the manifestation of early problem behavior in infants and toddlers. As expected, we found externalizing behaviors to be among the most common problem behaviors in 14-month-old infants. Indeed, more than 70% of the parents reported that their children often or very often showed behavioral items such as “quickly shifts activities,” “angry moods,” and “demands must be met immediately.” These rates are fairly comparable with the prevalence rates reported earlier for somewhat older children [18]. Although these and other common oppositional behaviors are symptoms of DSM-IV axis-I disorders such as Attention Deficit Hyperactivity Disorder, Oppositional Defiant Disorder, and Conduct Disorder at older ages, these high scores for 14-month-old infants are not necessarily psychopathological [6, 18]. Instead, the high occurrence of parent-reported problem behaviors at this early age can best be interpreted as “relatively normal” [18, 23]. While the symptoms and frequency of externalizing problem behaviors change with age, it is of interest to categorize these changes by their severity and to study their developmental trajectories in order to classify expected abnormal outcomes [3]. Further work should therefore focus on longitudinal studies to decide where the cut-off point is and then start to index a risk for psychopathology. Lack of competences in social communication and internalizing problem behaviors were relatively rare, occurring in about 5–9% of the community sample. These results confirm that internalizing symptoms with possible underlying constructs of withdrawal and depression are less prevalent at this early age [4]. This is not to say that these behaviors are unimportant or should be neglected. For example, other social communication items such as “failure to follow with the eyes,” “does not react to calling name,” and “does not imitate simple gestures” may indicate the presence of more serious psychopathology, such as autistic spectrum disorder. It would be interesting to determine whether the risk of later psychopathology is significantly higher for a combination of problem behaviors and for which combination this accounts. Also, the frequency or severity of these social communication problems is associated with a serious risk of problems later in development. Principal component analysis of the data for the 14-month-old children indicated that seven factors could identify important behavioral dimensions. The observation that PCA revealed the same factor solutions for the original as well as for both split half samples is strong evidence for the presence of these distinct behavioral dimensions at 14 months. While the CBCL 1.5–5 [1] can detect externalizing problem behavior in 18-month-old children, our study indicates that these behaviors are present at an even younger age, 14 months. The two behavioral dimensions with the highest explained variance were dimensions describing problems in oppositional behaviors and language development. The dimension Language Development was correlated with the behavioral dimension Communicative Intent, which covers items of more nonverbal-related behaviors, and with Explorative Behavior. Ever since the work of Piaget, explorative behavior has been considered crucial for cognitive development. Explorative behavior is also an important construct in the field of attachment theory, where the infant is thought to use the attachment figure as a secure base for exploring its environment. Interestingly, the finding of a separate attentional factor refers to the emergence of attentional systems of higher-level control at 12–36 months, which contribute to the development of more goal-oriented behaviors and the ability to inhibit actions [28]. The behavioral dimension termed Inhibition consisted of items consistent with the construct of passive inhibition, namely items focusing on inhibition to novelty, separation distress, and affect-relating internalizing behaviors [16]. Eating and sleeping problems as well as more anxiety-related problem behavior formed a Dysregulation dimension, as found in other studies [4]. Our findings suggest that even at this young age the structures that underlie behavior problems are becoming clearer. These behavioral dimensions are in line with dimensions described in other questionnaires. For example, our Oppositional Behavior dimension includes many items found in the Aggressive behavior factor of the CBCL 1.5–5 [1]. Items of the Communication scale of the Vineland are found in our dimension Language Development and Communicative Intent in our analysis. The latter dimension also includes some items of the Socialization scale of the Vineland [29, 30]. Next, our dimension measuring Attention consists of items from Attention dimensions of other scales (the Attention Problems of the CBCL 1.5–5 and the Attention scale of the ITSEA) [7]. Also, our dimension Explorative behavior is partly in line with ITSEA’s Mastery motivation and partly with the Socialization scale of the Vineland. Furthermore, our dimension Dysregulation consists mainly of items of the Internalizing dimension of the CBCL 1.5–5. Finally, most of the items seen in our dimension Inhibition are found in the Other Problem dimension of the CBCL 1.5–5 and some are from the Internalizing and Competence domain of the ITSEA. Inspection of the factorsumscores correlation matrix suggested two higher-order dimensions, one consisting of problems behaviors (Oppositional behavior, Attention, Dysregulation, and Inhibition) and a second one covering competences (Language Development, Explorative Behavior, and Communicative Intent). This higher order division was confirmed in a factor analysis forcing a two-factor solution. The two higher-order factors Total Problem Behavior and Competences correlated only very weakly with each other (0.12), suggesting a fairly amount of independence. Apparently a relative high level of problem behaviors is not necessarily accompanied by a relative low level of competences and vice versa. This finding is in accordance with studies using the ITSEA in older children. In these studies too, competences emerged as a separate factor from factors on problem behaviors [4, 5, 9]. For the clinical practice our result indicates that even at a very young age, a thorough evaluation of a child’s functioning should include both an investigation of problem behaviors and an evaluation of competences. With regard to sex differences, boys were reported to have significantly more problem behaviors than girls on all factors, except for dysregulation and inhibition, consistent with the finding that boys show more externalizing behaviors [18]. We found no sex differences in dysregulation and inhibition, whereas other studies report girls to be more fearful and to have more sleeping problems than boys [14, 27]. Boys were reported to have more problems in social communication behaviors, which suggests that boys acquire these competencies later than girls. This could play a role in the emergence of problem behavior among boys, whereas the early acquisition of competencies could be a protective factor for girls [4]. Contradictory reports on the presence of sex differences in problem behaviors may also be due to cultural or social factors, such as social role expectancy issues, and these may well complicate the interpretation of results. Problems of communicative intent, oppositional problems, and attentional problems had the highest correlations and remained moderately stable over the 5-month period between 14 and 19 months of age. The least stable were inhibition and dysregulation problems, perhaps reflecting a more temporary nature of these behaviors, as reported earlier by Briggs-Gowan and Carter [4]. At least 53% of the infants with scores within the 90th percentile of each factor sum score at 14 months still exhibited these problem behaviors at 19 months of age. These results are in accordance with other studies that have also shown substantial stability in problem behavior [21, 23]. These studies investigated children older than 19 months of age, whereas our results suggest that these problem behaviors are already moderately stable at an even younger age, emphasizing the need for early assessment. We can only speculate on the determinants of this stability, such as the child’s temperament, family environment, or parental variables. It would be interesting to investigate these determinants in relation to the frequency, severity, and pervasiveness of behavior problems over a longer period of time, for example, at 14, 19, and 36 months of age, in order to gather information on how problem behavior evolves. Limitations One should note that we did not aim to present a new instrument to measure problem behaviors at this young age. Instead, we wanted to report on common and uncommon problem behaviors in this young age group. Although the study population was large and community based, it was not fully representative of the general population. Our sample contained more parents with a higher socioeconomic status than expected compared to the general population. Thus scores for problem behaviors might have been higher in the non-responder group. A second limitation is that we only used parent reports. Information from other sources can be of additional value to evaluate the level of problem behavior [8, 18]. Future work should include measures of impairment of functioning and should focus on more in-depth assessments of psychopathology and level of cognitive development. Finally, it may be informative to compare a population-based sample with a clinically referred group. Clinical implications This study underlines the growing need to identify problem behavior in early development. A better understanding of what is normal and abnormal in early infant-toddler development is of great value for the recognition of early psychopathology. Normative data for the prevalence and stability of behavior problems of 1- and 2-year-old children in a normal population may make it possible to provide early effective preventive intervention programs, in addition to background information for clinicians and researchers in order to develop such programs. In summary, externalizing behaviors were the most common, and social communication the least common, problems reported in a population sample of 14-month-old children. Boys were reported to show significantly more problems of externalizing behavior and competences. These behaviors were moderately stable over a 5-month period, between the ages of 14 and 19 months.

J_Med_Internet_Res-9-2-1913940

Language Preferences on Websites and in Google Searches for Human Health and Food Information

Background While it is known that the majority of pages on the World Wide Web are in English, little is known about the preferred language of users searching for health information online. Introduction The World Wide Web has more than 15 billion pages [1] and has become an important source of health-related information [2-5]. In the United States, for example, it has been reported that 82% of female users and 77% of male users used the Internet to obtain medical information on a routine basis [6]. Google searches have even been shown to assist physicians in the correct diagnosis of medical ailments [7]. However, two thirds of the pages on the Web are published in English [8,9], even though the world has over 5 billion non-English speakers [10], including approximately 700 million non-English-speaking Internet users [8]. In fact, the vast majority of the world’s 6900 living linguistic groups [10] have little Web content available in their language [8,9]. Adding to this problem, search engines such as Google do not translate search terms into other languages [9]—perhaps a surprise to many users. To overcome linguistic barriers, the World Health Organization (WHO) and the Food and Agricultural Organization (FAO) now publish their websites in six and four major languages, respectively. However, other globally authoritative organizations, such as the Centers for Disease Control (CDC) in the United States, primarily publish online information in only one or two languages of domestic importance. In spite of the significant challenges created by linguistic differences in effectively communicating health information to the world’s peoples online, we could find little quantitative data on this issue. Do the world’s online users, presumably wealthier and more educated than the general population, primarily search online for health information in their local language, or do they employ Web-prevalent languages such as English? Are current online translation efforts by the world’s health and food agencies beneficial, and should these agencies be spending more resources on these efforts? In a world of human migration, which language(s) should domestic governments and international agencies use in order to communicate online health information to target populations? To transmit information to front-line health professionals in developing nations, a group that can include international aid workers from wealthy nations, which language(s) should be employed to target a particular nation? Do indigenous peoples search for online health-related information using search terms belonging to their own language or the colonial language? Real-time, accurate communication of health information might be especially critical during a pandemic infectious disease outbreak or famine. To begin to answer these questions, we have used a case-study approach that examines linguistic preferences in Internet search engine queries. Specifically, we measured search patterns on Google for four health- and food-related terms in seven languages in 227 nations. The four search terms we chose for our study were “avian flu,” “tuberculosis,” “schizophrenia,” and “maize” (corn). We chose “avian flu” because it is an ideal model for searching for online information concerning an emerging infectious disease pandemic; as of August 2006, avian influenza (virus subtype H5N1) had killed 141 people in 10 countries [11,12] in addition to prompting the slaughter of millions of animals [13]. We chose “tuberculosis” because it is a good model for searching an established global infectious disease, as it is a major cause of death for HIV-positive patients [14] and currently afflicts about 15 million people in 207 nations [15]. We chose “schizophrenia” because psychiatric and neurological diseases affect more than 450 million people globally [16,17] and because online mental health information has the potential to help nations that have few mental health specialists [17]. Schizophrenia afflicts over 24 million people worldwide [18]. Finally, because there are currently around 850 million chronically undernourished people in the world [19], under frequent threat of famine, and because malnutrition is a major underlying contributor to infectious disease susceptibility [20], we chose “maize” (corn) because it is an important search term for global food security agencies. Maize supplies one third of all human calories in Latin America and Sub-Saharan Africa [19] and, combined with its genetic relatives rice and wheat, supplies approximately 50% of all human calories globally, either directly or via animal feed [19]. In this paper, we quantitatively demonstrate the need for health and food website translation that not only targets the world’s major languages, but also linguistic minorities within nations, including immigrants, foreign aid workers, and indigenous groups. Methods Measuring the Extent to Which Health and Food Agencies Translate Websites When an online publisher translates its Web site across languages, the same source domain name can be found in the different result lists of a search engine when entering different translated search terms. In an URL, the domain name is the identifier before the first slash (eg, for the WHO, the domain name is www.who.int). Therefore, to estimate the percentage of institutions, news agencies, and other sources that translate their health and food websites, we measured the frequency that domain names overlap (institutional sources overlap) in different results lists when Google is queried using different language translations of the same search term (eg, English “tuberculosis” versus Bahasa Indonesia “tuberkulosa”). The language translations used in this study are shown in Multimedia Appendix 1. We first extracted URLs from Google.com listed as “the most relevant” (about 500-1000 URLs) based on the Google page rank algorithm [21] on July 10, 2006. We then excluded nonunique domain names from within each query group; the remaining unique subset was then compared to the URLs retrieved using the comparison language. Measurements of Online Search Rates and Language Choice For each of the four terms and their translations, we measured the rate at which users from 227 different countries searched Google (Multimedia Appendix 2). The country of origin for each search was identified using the geographic locations of Internet Protocol (IP) addresses. The search rates were based on proprietary Google Inc. (Mountain View, CA) data using an algorithm that measured 20% of all first-page Web search queries from January 2004 to April 2006. The search rate, M, for a term, T, originating from a specific country was then estimated from this large sample: the algorithm calculated the ratio of searches using a specific term (T) divided by all searches (A) from that nation during the time period, such that M = T/A. Multimedia Appendix 2 contains the standard error (SE) for each search rate. To calculate the standard error, a normal distribution for the statistic was assumed. To convert standard error to the 95% confidence interval of the estimate, the following formula was used: 95% CI = M ± (1.96 × SE). Only the search rates for selected languages are shown for each nation, not the total search rate across languages or the total number of searches, which are proprietary. For ease of database searching and validation, we chose languages that share a Latin-based script similar to English. For detailed analysis, the languages chosen were English, Bahasa Indonesia, Spanish, Portuguese, German, and French, as they represent the primary or secondary language of about 100-500 million people worldwide and/or are significant post-colonial languages [10]. Because Turkey has reported cases of avian influenza [11,12], Turkish was added as a language of interest. A search term may be common to multiple languages, not all of which are noted here. Multimedia Appendix 1 contains a complete list of the language translations used to retrieve search rate data from the Google database. Health and Food Security Indicators For the number of human avian influenza cases (subtype H5N1), we employed the WHO Epidemic and Pandemic Alert and Response Database [12], updated on August 9, 2006. For the number of poultry outbreaks of type H5 since 2003, we used the August 16, 2006 Avian Flu Update from the Organisation Mondiale de la Santé Animale (OiE) [World Organisation for Animal Health] [13]. For tuberculosis (TB), we used Millennium Development Goal (MDG) Indicator 23, the estimated TB prevalence in 2004 from the Global Tuberculosis Database of the WHO Global Health Atlas [15]. The number of psychiatrists per 100000 people in 2005 was obtained from Project Atlas: Resources for Mental Health and Neurological Disorders, from the WHO Global Health Atlas Database [17]. The most recent maize consumption data (kcal/person/day in 2004) was from the Supply Utilization Accounts (SUA) Database of the Food and Agricultural Organization Statistical Division (FAOSTAT) [19]. Results Measuring the Extent to Which Institutions Currently Translate Online Information We measured the extent to which the world’s institutions, including health and food agencies, news organizations, and other sources, currently publish Web content in multiple languages. To quantify this, we measured the institutional source (host-domain) overlap between search results (first 500-1000 URLs) retrieved using English-language queries versus select comparison languages (Table 1). For example, as shown in Table 1, “avian flu” resulted in 906 hits, which included 539 unique hosts. The same search in French resulted in 801 hits, which included 375 hosts. Only 7.1% of the 539 hosts found in the English search were also found in the French search, while 10.1% of the 375 hosts found with the French search were also found in the English search. This means that approximately 7.1%-10.1% of all hosts have an English/French translation. For “avian flu,” which has afflicted 56 people in Indonesia [12], only 6.8% and 2.4% of institutional domain names overlapped between Bahasa Indonesia searches and English-language searches, respectively (Table 1). For “tuberculosis,” we discovered only about 9% overlap in the institutional domain names retrieved for English and French/Portuguese translations: French and Portuguese are widely spoken by TB-afflicted nations in Africa and Brazil [15]. The host domain overlap was typically less than 10% between European languages; we were surprised by this low rate given the presence of common governmental institutions in Europe (eg, European Union). We extrapolate that 80%-90% of health- and food-related institutions do not translate their websites into multiple languages, even when the information concerns pandemic disease. This does not exclude the possibility that other agencies, such as domestic health agencies, might be translating this information. We then specifically screened for Web pages belonging to the WHO and the CDC, both authoritative agencies for infectious disease information. Because the first page of search results is the most viewed [22], the rank order in which search results appear is critical. When we searched for “avian flu” in English on Google.com, the CDC website was the first hit, followed second by the WHO website. In the French translation, however, no CDC-affiliated Web pages appeared in the first 100 hits. When we searched Google Indonesia using the Bahasa Indonesia translation of “avian flu,” a WHO-affiliated Web page did not appear until page three of the search results (rank 21), and the first CDC-affiliated page appeared on page six (rank 63). Most significant, by searching for “avian flu” in Turkish using Google Turkey, we were unable to retrieve the websites of either the WHO or CDC in the first 500 search results. As a cautionary note, we found that a single accent or special character in the search term sometimes changed the rank order of health search results significantly, consistent with more detailed analysis conducted for nonmedical terms [9,23]: for example, for avian flu, we found that “gripe aviária” (accent, Portuguese) yielded 49/100 top search results of Brazilian origin (ie, “br” in domain name), whereas “gripe aviaria” (no accent, Portuguese) yielded 40/100 top search results from Spanish-speaking nations and only 4/100 search results from Brazil. This is significant, because we found that Brazil, Portugal, and other Portuguese-speaking nations (Angola, Mozambique) searched Google for this term with and without the accent at nearly equal rates (Multimedia Appendix 2). Table 1 Overlap in the institutional domain names retrieved by Google to measure the extent to which institutions translate websites across languages English Search Term Comparison Language English Language Search Comparative Language Search URLs Unique Hosts Host Overlap with Comparison Languages URLs Unique Hosts Host Overlap with English Avian flu French 906 539 7.1% 801 375 10.1% Avian flu Indonesian 906 539 2.4% 472 190 6.8% Tuberculosis FrenchPortugueseDutch 834 473 8.5% 790 426 9.4% Tuberculosis GermanDanishAfrikaans 834 473 0.0% 809 443 0.0% Tuberculosis Bahasa Indonesia 834 473 0.8% 462 273 1.5% Schizophrenia SpanishPortuguese 813 444 10.8% 764 518 9.3% Schizophrenia Bahasa IndonesiaMalay (Bahasa Malayu) 813 444 17.1% 748 463 16.4% Maize Spanish 828 421 7.8% 830 543 6.1% Language-Specific Searching of Infectious Disease Information Though we did not retrieve many WHO- and CDC-affiliated Web pages when we searched across different languages, one possibility is that Internet users, many of whom are well-educated, are supplementing their Google searches for online health information by searching in English. If true, then there would be less of a need for the WHO and other global agencies to translate online information into diverse languages. In our case study, we found the actual results to be variable (Table 2; Multimedia Appendix 2): for “avian flu,” we often found that only 1% of searches in non-English-speaking nations were in English, whereas for “tuberculosis” or “schizophrenia,” about 4%-40% of searches in non-English countries employed English. Brazil, which had an estimated 141000 cases of TB in 2004 [15], had an 18-fold higher query rate for “tuberculose” (Portuguese) than “tuberculosis” (English). However, the comparison is more important for languages that have much less Web content, such as Bahasa Indonesia and Turkish. Indonesia, which had the highest number of reported cases of human H5N1 viral infections in 2006 [12], had a 15-fold higher query rate for “avian flu” in Bahasa Indonesia (“flu burung”) than in English. We found that Turks used Turkish to search Google for health terms at 3- to over 1000-fold higher rates than English, French, German, or Spanish (Table 2; Multimedia Appendix 2). We did, however, find sites in Turkish that had translated information from the WHO. Whether or not the WHO, CDC, and FAO wish to leave it up to others to translate their information accurately and rapidly must be decided based on their confidence level of the eHealth capabilities of each target nation. Given the Google language-based search patterns, we conclude that during times of infectious disease outbreaks, though English may be useful, global agency–affiliated Web pages translated into local languages would likely be highly accessed and would have the benefits of being viewed as authoritative and accurate and of being transmitted in real time. Language of Online Mental Health Information Searches In terms of mental health, many developing nations have 10- to 100-fold fewer psychiatrists per capita than many developed nations [17] (Table 2). For this reason, global accessibility of online mental health information has the potential to be very beneficial to physicians and families of patients in developing nations [24]. Given ethnic taboos [24], we first asked whether or not Internet users from developing nations are searching online for mental health information—potentially useful information for global mental health experts. Indeed, we found that the search rate for “schizophrenia” was similar between developed and developing nations in the local language, demonstrating an active need for online mental health information in poor countries. We have made the full mental health dataset available (Multimedia Appendix 2). Given that Google estimates 3-fold more search results concerning schizophrenia in English versus the next 10 languages combined (data not shown), we asked whether middle- and low-income nations searched for this topic in English. We found that people from Brazil, a nation of about 180 million people, searched for “schizophrenia” in Portuguese at a 28-fold higher rate than in English (Table 2). Table 2 Google search rates for selected health terms in local languages relative to English* Search Term Country Public Health Comparison Metric Local Language English Searches (% of Local Language Searches) Avian flu Human Cases† (Poultry Outbreaks)‡ United States 0 (0) English 100.0 France 0 (1) French 1.6 Germany 0 (1) German 0.8 Turkey 12 (176) Turkish 0.9 Democratic Republic of Congo 0 (0) French 1.1 Cote d’Ivoire 0 (3) French 1.4 Burkina Faso 0 (4) French 100.0 Mozambique 0 (0) Portuguese 143.8 Mexico 0 (0) Spanish 3.3 Brazil 0 (0) Portuguese 1.0 Indonesia 56 (211) Bahasa Indonesia 6.5 Tuberculosis TB Cases § United States 10510 English 100.0 France 5901 French 9.8 Germany 5243 German 14.3 Turkey 32371 Turkish 29.1 Democratic Republic of Congo 307554 French 23.6 Cote d’Ivoire 116349 French 8.6 Burkina Faso 46815 French 15.9 Mozambique 123360 Portuguese 28.4 Mexico 45710 Spanish same term Brazil 141115 Portuguese 5.6 Indonesia 605759 Bahasa Indonesia 1779.3 Schizophrenia Psychiatrists per 100000|| People United States 13.70 English 100.0 France 22.00 French 6.6 Turkey 1.00 Turkish 10.5 Democratic Republic of Congo 0.04 French 43.0 Cote d’Ivoire 0.20 French 7.1 Burkina Faso 0.05 French 9.5 Mozambique 0.04 Portuguese 24.4 Mexico 2.70 Spanish 4.1 Brazil 4.80 Portuguese 3.6 Indonesia 0.21 Bahasa Indonesia 115.4 *Based on sampling 20% of all searches on Google.com from January 2004 to April 2006. †Subtype H5N1, from the WHO Epidemic and Pandemic Alert and Response Database [12], updated August 9, 2006. ‡Type H5 outbreaks since 2003, from the OiE [13], updated August 16, 2006. §Estimated TB prevalence in 2004 from the WHO Global Tuberculosis Database [15]. ||2005 data from Project Atlas: Resources for Mental Health and Neurological Disorders, from the WHO Global Health Atlas Database [17]. In contrast, people from Indonesia, a nation of 220 million people, searched for “schizophrenia” at similar rates in Bahasa Indonesia as in English. As with infectious disease searching, our data would suggest that users from most nations tend to search Google for “schizophrenia” in their official language at up to 10-fold higher rates than other languages (Multimedia Appendix 2). Many of the world’s people might therefore benefit if the world’s most authoritative mental health agencies (eg, US National Institute of Mental Health) translated information into other languages, even though this is not part of their domestic mandate. The Online Search Rates of Immigrant Minorities In some developed nations, there is concern that immigrant groups might spread infectious diseases. In Europe, the TB prevalence is 43/100000 people in Turkey but lower in wealthier nations such as Germany (6/100000); in Asia, the TB prevalence is 273/100000 people in Indonesia, compared to 48/100000 in Singapore [15] (Table 2). Unlike many of their neighbors, both Turkey and Indonesia have reported human cases of avian influenza [12]. Governments may be interested to know whether their immigrant communities consult infectious disease–related websites originating from their adopted country or their native country. As a case study, we examined the search rates of Turks and Indonesians after they migrated to other nations in Europe or Asia, respectively, as determined by searches in Turkish and Bahasa Indonesia originating outside Turkey and Indonesia. The Turkish and Indonesian languages are distinct relative to many of their surrounding nations, permitting us to discern the search behavior of these populations after they have emigrated, provided they search in their native language. As shown in Table 3, we detected searches in Turkish for “avian flu” and “tuberculosis” throughout Europe, including high rates in Belgium and Austria, respectively. In Asia, we could detect searches in Bahasa Indonesia for “tuberculosis” and very high rates for “avian flu” throughout Asia and the Pacific region, including Hong Kong, Singapore, and Australia. We cannot exclude that these (presumed) Turkish and Indonesian immigrants also searched online in their adopted language(s), nor could we measure the fraction of their searches in their native language versus adopted language(s). However, based on the fraction of a country’s population that belongs to a particular immigrant group versus the fraction of searches conducted in the immigrant language compared to the adopted language, then if every person in a country searched the Internet for the same term at the same rate, but in different languages, we could then extrapolate that Turkish immigrants in Belgium searched for “avian flu,” “schizophrenia,” and “tuberculosis” 100%, 40%, and 15% of the time, respectively, in Turkish rather than French (Table 3). Using the same simplistic assumptions, in Austria, 39% of searches for “tuberculosis” by Turkish immigrants were in Turkish rather than German. In reality, immigrant groups are likely searching for a term such as “avian flu” at a higher rate than the general population when the corresponding disease affects their homeland and is in the news. However, we also found high search rates for “tuberculosis” and even “schizophrenia” in Turkish in these nations, which are less featured in the news. We conclude that it is important for health officials to be aware that if they wish to disseminate health information to susceptible immigrant groups, they should not rely on websites published in the majority language(s) of the nation. High priority domestic health-related websites should be multilingual, particularly those that concern infectious disease. Our analysis also revealed surprises: for example, in Bangladesh, one of the world’s poorest and most populated nations, with 136 million residents, we found that the search rate for “avian flu” in Bahasa Indonesia (“flu burung”) was equivalent to the search rate in English (Table 3); the Bahasa Indonesia translation of “schizophrenia” was also high in Bangladesh. The high Bahasa Indonesia search rate may reflect the fact that many Bangladeshi citizens work in Singapore and Malaysia, nations that speak a similar language, Malay (Bahasa Malayu) [25]; it would appear that when they return to Bangladesh, Bangladeshis continue to use the terminology they learned while away, but the reason is unclear. We suggest that international health organizations aiding Bangladesh should publish or meta tag health information in Malay (Bahasa Malayu) to reach health practitioners in that nation. This result also highlights the practical value of analyzing linguistic preferences during online searching. Finally, we were surprised to find that online searches for “avian flu,” “schizophrenia,” and “maize/corn” in Spanish in the United States occurred at less than 1% of the English search rate (Table 3). The US online Hispanic population (legal and illegal) is estimated to be 12% of the total US online population, of which nearly half use Spanish for some (28%) or all (21%) of their Internet usage [8,26]. Therefore, our data suggest that Latin American immigrant groups in the United States search for health information to a lesser degree in Spanish than might be predicted, although these data could also be a result of the digital divide between the groups. Table 3 Online search rates of immigrant minorities Search Term Country Immigrant Language Comparative Major Language* of Adopted Country Immigrant Language Searches per 10000 Major Language Searches Relevant Immigrants per 10000 Total Population† Turkish immigrants in Europe Avian flu Belgium Turkish French 56.1 39.7‡ Switzerland Turkish German 9.2 N/A United Kingdom Turkish English 31.2 N/A Tuberculosis Austria Turkish German 60.2 155.4§ Belgium Turkish French 5.8 39.7‡ Germany Turkish German 11.8 212.2|| Switzerland Turkish German 10.7 N/A United Kingdom Turkish English 2.8 N/A Schizophrenia Belgium Turkish French 15.9 39.7‡ Switzerland Turkish French 62.2 N/A United Kingdom Turkish English 2.9 N/A Indonesian/Malaysian immigrants in Asia/Pacific Avian flu Australia Bahasa Indonesia English 234.7 625.2¶ Bangladesh Bahasa Indonesia English 10474.0 N/A Hong Kong Bahasa Indonesia English 3424.4 N/A India Bahasa Indonesia English 154.4 N/A Singapore Bahasa Indonesia English 2268.0 N/A Tuberculosis Hong Kong Bahasa Indonesia English 23.3 N/A Singapore Bahasa Indonesia English 17.0 N/A Schizophrenia Australia Bahasa Indonesia English 6.0 625.2¶ Bangladesh Bahasa Indonesia English 687.4 N/A Hong Kong Bahasa Indonesia English 295.1 N/A India Bahasa Indonesia English 16.0 N/A Spanish immigrants in English-Speaking North America# Avian flu Unites States Spanish English 102.8 > 503.0** Schizophrenia Unites States SpanishPortuguese English 91.8 > 503.0** Corn (maize)‡‡ Unites States SpanishIndigenous (“choclo”/”elote”) English 37.9 > 503.0** *a major language for which Google search rates were accessible, not necessarily the largest linguistic group of the nation. †Total population data are from United Nations Population Division, 2005 data. URL: http://www.un.org/esa/population/unpop.htm. ‡Data are from Institut National De Statistique, 2003 data. Population et Ménages: Mouvement de la population et migrations, "Immigrations extérieures par nationalité et groupe d'âges – Belgique." URL: http://www.statbel.fgov.be. §Data are from Statistics Austria, Volkszählung. Hauptergebnisse I – Österreich, 2001 census data. URL: statistik.at/neuerscheinungen/vzaustria.shtml. ||Data are from Statistisches Bundesamt (Federal Statistical Office, Germany), 2006 data. URL: http://www.destatis.de/themen/e/thm_bevoelk.htm. ¶Data are from Australia Bureau of Statistics, Cultural and Language Diversity, 2001 Census data. URL: http://www.abs.gov.au/. #“Tuberculosis” is not included as it is the same term in both English and Spanish. **Data are from Department of Homeland Security Yearbook of Immigration Statistics. Legal immigrants, 2005 data. URL: uscis.gov/graphics/shared/aboutus/statistics/ybpage.htm. ‡‡The search term used was “corn.” Table 4 Search rates in European languages in Sub-Saharan Africa Country Colonial Language* Minority Language Minority Language Searches per 10000 Colonial Language Searches Search Term: Avian Flu Angola Portuguese English 6629 Cameroon French German 323 Ghana English French 2024 Ghana English German 1744 Ghana English Dutch 498 Kenya English French 2941 Mozambique Portuguese English 14286 Mozambique Portuguese French 4444 Nigeria English French 4235 Nigeria English German 897 Rwanda French English 1481 Senegal French English 732 Search Term: Tuberculosis Angola Portuguese French EnglishSpanish 1504 Cameroon French Portuguese English 2405 Democratic Republic of Congo French Portuguese EnglishSpanish 2364 Ghana English FrenchPortuguese 836 Kenya English FrenchPortuguese 1128 Mozambique Portuguese French EnglishSpanish 2842 Nigeria English FrenchPortuguese 506 Rwanda French Portuguese EnglishSpanish 2808 South Africa English Afrikaans 1547 Search Term: Schizophrenia Angola Portuguese Spanish English 1547 Cameroon French English 1922 Democratic Republic of Congo French English 4304 Ghana English French 488 Mozambique Portuguese Spanish English 2441 Rwanda French English 18453 Senegal French English 3287 *When two languages are noted, the term is the same in both languages. Search Rates in European Languages in Sub-Saharan Africa Sub-Saharan Africa suffers from high rates of infectious disease, including TB and HIV [18], and high rates of malnutrition [19]. As demonstrated in Table 4, our search results suggest that the dissemination of online health or food security information to this region by international agencies should not be limited to the colonial language(s) of the target nations. For example, in English-speaking Ghana, 20% of searches for avian flu were in French (“grippe aviaire”), 17% in German (“vogelgrippe”), and 5% in Dutch (“vogelgriep”) relative to English (Table 4). In Mozambique, a former Portuguese colony, we found 1.4 times more searches for “avian flu” in English relative to Portuguese, and a high rate in French. In the Democratic Republic of Congo (DRC), a former French colony with more than 300000 TB infections in 2004 [15], 24% of the searches for tuberculosis were in English relative to French. Because Internet use is only 1.9% is Ghana, 0.7% in Mozambique, and 0.2% in DRC [8], it is plausible that these high rates of non-colonial language searches using Google may reflect searches by health professionals trained in other nations, including workers from international agencies who would be expected to have better Internet access than the general population. The Effect of Region-Specific Cultural and Indigenous Terminology Finally, we measured the effects of cultural bias within the same linguistic group, using a term important for human nutrition and food security, “maize.” Maize is eaten directly, but it is also a major source of animal feed worldwide. Maize is known as “corn” in the United States and the United Kingdom, but as “maize” in many other English-speaking nations. We found that the US searches for “corn” were at a 28-fold higher rate than for “maize,” while other English-speaking nations such as Nigeria and Zimbabwe queried “maize” at a 1.5- to 4-fold higher rate than “corn,” respectively (Table 5); in the latter nations, “corn” may also refer to any large cereal grain (eg, wheat). This search behavior has consequences, as we found only three domain names that overlapped between searches for “maize” versus “corn” out of the first 50 unique Google search results. Therefore, when African nations attempt to retrieve information about growing maize in English, they may unknowingly be excluding authoritative information from the United States and other Western English-speaking nations, such as practical information from the US National Corn Growers’ Association, whose website does not appear in the first 100 Google hits for “maize,” but ranks sixth when “corn” is searched. Similarly, when international organizations such as the FAO wish to transmit knowledge, for example, using Spanish or Portuguese in Latin America, indigenous terminology usage may make some of this information inaccessible. Maize originated from Southern Mexico and fed indigenous Latin American civilizations [27]. In Mexico and Guatemala, we found that an indigenous synonym for maize, “elote,” was searched at a similar rate to the Spanish term, “maíz” (Table 5). In Peru, however, we found that a different indigenous term, “choclo,” was searched at a 2-fold higher rate than “maíz.” We conclude that cultural and indigenous linguistic divisions may be preventing large numbers of food security and nutrition websites from reaching those people aiding 800 malnourished people or 1.2 billion agricultural workers that live in developing countries [19]. Cultural and indigenous bias may be particularly prevalent for terms related to crops, diseases, or pathogens that have pre-colonial origins. Table 5 The effect of region-specific cultural and indigenous terminology Country Calories from Maize (kcal/person/day)* Search Term Comparison Search Rate Ratio Canada N/A† corn:maize 36:1 United States 512 corn:maize 28:1 United Kingdom 115 corn:maize 7:1 India 38 corn:maize 2:1 Nigeria 179 corn:maize 1:2 Kenya 775 corn:maize 1:3 Tanzania 646 corn:maize 1:4 Zimbabwe 720 corn:maize 1:4 Spain N/A† maíz:elote:choclo 10:1:4 Venezuela 467 maíz:elote:choclo 16:1:3 Colombia 312 maíz:elote:choclo 20:1:5 Mexico 1081 maíz:elote:choclo 14:16:1 Guatemala 869 maíz:elote:choclo 12:10:1 Peru 145 maíz:elote:choclo 11:1:25 Argentina 132 maíz:elote:choclo 41:1:51 *Data from FAOSTAT [19]. †N/A: Data not available. Discussion Principal Results In a world where infectious disease pandemics and threats of famine are always present, and in spite of the fact that the World Wide Web offers great hope for rapid and accurate sharing of information between peoples, we have demonstrated that one linguistic group does not or cannot access the health and food security websites of a different linguistic group. Our data suggest at least three reasons for this. The first reason is that the websites of most institutions are not published in more than one or two languages. When we sampled Web pages found by the English-language queries “avian flu,” “tuberculosis,” “schizophrenia,” and “maize/corn” and their counterpart queries in other languages, the Google search results typically overlapped by only less than 10% in terms of the domain names retrieved, indicating that 90% of the relevant Web pages had not been translated into at least two languages (see Table 1). For example, when Turkish or Bahasa Indonesia was used as the search language, Web pages from very authoritative sources, such as the CDC or WHO, were not retrieved by Google. We also found that a single linguistic accent or special character in the search query could significantly alter the number and content of health-related search results retrieved by Google. Therefore, one reason for the linguistic digital divide is that the majority of health and food Web pages are not translated into multiple languages and/or that their cross-language retrieval by search engines is poor. The first problem would not be important if the world’s online community, better educated than the general public, searched in English, since the majority of the world’s Web pages are published in English [8,9]. However, we found that there was a 2- to 100-fold higher Google search rate for health and food terms in the native language of a country compared to English (see Table 2). Finally, within a nation, it might be assumed that language would not be a problem if a domestic agency only published their health websites in the majority language of their own people. However, we found that in Asia and Western Europe, a subset of immigrants from Indonesia and Turkey, respectively, searched Google for health and food information in their native language, not the language(s) of their adopted countries (see Table 3). In Sub-Saharan Africa, we detected unexpectedly high search rates for health information in non-colonial European languages (see Table 4), perhaps reflecting the presence of international aid workers. Finally, in Latin America, we found that indigenous words were used to search Google for information about food, rather than the colonial language of Spanish (see Table 5). Therefore, domestic agencies, in addition to global agencies, face a linguistic challenge when publishing information online: their target audiences still require information to be published in different languages, even though Internet users are presumably more educated and thus more multilingual than the general population. Recommendations Given our observation that the world’s peoples appear to be searching for health and food terms in their local language or mother tongue, not in English, previous online language translation efforts by the WHO and FAO have no doubt been worthwhile. This is also revealed by the high page rank (first page results) of WHO-affiliated search results when Google is searched for infectious disease information in one of the WHO’s six online languages. We recommend that these efforts be continued and further expanded to include more languages; this recommendation applies to global agencies, but also to domestic agencies, in order to meet immigrant or indigenous needs and/or to make information accessible to other nations. To achieve such translation goals, better health-specific translation software must be developed and more translators are needed who specialize in human health and food security terminology. For example, improved cross-language search retrieval [23] of health information by online search engines would be beneficial. These investments should then be used to initially target health and food security information to the world’s most important linguistic groups, which include speakers of Chinese (1080 million people), Hindi (about 500 million), English (350-500 million), Spanish (390 million), Arabic (255 million), Portuguese (190 million), Bengali (215 million), Russian (255 million), Bahasa Indonesia (200 million), Japanese (127 million), Punjabi (104 million), German (123 million), and French (119 million) [10]. In a world that is primarily non-English speaking, such attempts will help to reduce the linguistic digital divide in health and food information on the World Wide Web. Furthermore, as we did in this study (see Table 2 to Table 5), we recommend that when global or domestic health and food security organizations wish to use the Internet to disseminate information to other nations [28] or to their own immigrant or indigenous communities, they should first consult search engine query rates for different translations of possible search terms in order to determine which online languages are most needed. Multimedia Appendix 2 contains extensive linguistic online search pattern data to help health and food agencies better select languages for targeted website publishing. In order to measure search rates for other subjects of interest, we note that free online tools exist, such as Google Trends. In some situations, such as targeting indigenous groups, who often speak the majority language of a nation, all that may be needed is to imbed translated keywords into a majority-language website (eg, Spanish) so that search engines such as Google can cross-retrieve relevant information. Future Studies Though this study examined the extent to which agencies such as the WHO are publishing information in multiple languages, we did not systematically address the quality of health and food information available in different languages. Quality analysis would especially be important for minor languages that have little content available on the Web: for example, we estimate that there are more than 4000-fold fewer search results in Bahasa Indonesia than in English for “tuberculosis,” more than 200-fold fewer search results for “avian flu” in Arabic or Japanese than in English, and about 500-fold fewer search results for “schizophrenia” in Arabic than in English. Though these precise numbers are considered to be unreliable [29], they do illustrate the point that most of the world’s linguistic groups, even major ones, have much less available online health information relative to English. As to the quality of the websites that are available, this will require systematic analysis, which poses significant methodological problems [30]. One could, however, perform a subjective survey-based evaluation by multilingual physicians, as has been performed to evaluate disease-specific websites published in English [31]. Such a survey should include quantifying to what extent online information from the WHO, CDC, and FAO is being translated into the world’s minor languages in order to help these agencies determine where they need to target their online translation efforts.

J_Neurol-2-2-1705504

Genetics of human hydrocephalus

Human hydrocephalus is a common medical condition that is characterized by abnormalities in the flow or resorption of cerebrospinal fluid (CSF), resulting in ventricular dilatation. Human hydrocephalus can be classified into two clinical forms, congenital and acquired. Hydrocephalus is one of the complex and multifactorial neurological disorders. Introduction Human hydrocephalus is a significant medical condition with an estimated incidence of 1 in 1500 births [1]. Hydrocephalus is characterized by abnormalities in the flow or resorption of cerebrospinal fluid (CSF), resulting in ventricular dilatation. However, hydrocephalus is far more complicated than a simple disorder of CSF circulation [2]. Although commonly considered a single disorder, human hydrocephalus is a collection of a heterogeneous complex and multifactorial disorders [3]. Genetic factors are involved in the pathogenesis of hydrocephalus [4–6]. For the purposes of this review, we categorize hydrocephalus as congenital, which is present at birth and often associated with developmental defects; and acquired, which occurs after development of the brain and ventricles [7–12]. The development and progression of congenital hydrocephalus is a dynamic process that is not yet well understood. It is thought that it may develop at an important and specific embryonic time period of neural stem cell proliferation and differentiation in the brain [13, 14]. Congenital hydrocephalus may occur alone (non-syndromic) or as part of a syndrome with other anomalies (syndromic) [15, 16]. In syndromic forms, it is hard to define the defective gene because of the association with other anomalies. We will mainly focus on isolated forms of hydrocephalus. In genetic terms, the isolated (non-syndromic) form of hydrocephalus is a primary and major phenotype caused by a specific faulty gene. It is estimated that about 40% of hydrocephalus cases have a possible genetic etiology [10]. In humans, X-linked hydrocephalus (HSAS1, OMIM) comprises approximately 5–15% of the congenital cases with a genetic cause [10, 17–20]. Although there is strong evidence for genetic causes, only one hydrocephalus gene (X-linked) has been identified in humans. Besides genetic factors, many other factors influence the development of congenital hydrocephalus, such as congenital malformations, intracerebral hemorrhage, maternal alcohol use [21, 22], infection [6, 23–25], and X-ray radiation during pregnancy [26, 27]. Genetics in hydrocephalus Congenital hydrocephalus is the more common of the two forms of hydrocephalus, and is probably the consequence of abnormal brain development and perturbed cellular function, which emphasizes the important roles that congenital hydrocephalus genes play during brain development. In general, the recurrence risk for congenital hydrocephalus excluding X-linked hydrocephalus is low. Empiric risk rates range from <1% to 4% [28–30], indicating the rarity of autosomal recessive congenital hydrocephalus [10, 20, 31, 32]. However, multiple human kindreds with congenital hydrocephalus have been reported [10, 15, 20, 32–43]. The loci or genes for human autosomal recessive congenital hydrocephalus have not yet been identified, but there is at least one locus for this trait. Furthermore, like animal models, since there is heterogeneity among clinical phenotypes, there may be more genetic loci in human autosomal recessive congenital hydrocephalus. One kindred was reported in which congenital hydrocephalus was transmitted in an autosomal dominant fashion. This condition was associated with aqueductal stenosis but was not associated with mental retardation or pyramidal tract dysfunction. The lack of mental retardation and pyramidal tract dysfunction was in contrast to X-linked or recessive congenital hydrocephalus with stenosis of the aqueduct of Sylvius (HSAS), in which these abnormalities are commonly seen [44]. Another study identified a kindred with a microdeletion of 8q12.2-q21.2 which subsequently developed hydrocephalus. This trait was also transmitted in an autosomal dominant fashion [45]. Molecular genetic studies have revealed that the responsible gene for X-linked human congenital hydrocephalus is at Xq28 encoding for L1CAM (L1 protein) [46]. The mutations are distributed over the functional protein domains. The exact mechanisms by which these mutations cause a loss of L1 protein function are still under investigation. Another form of this disorder, acquired or adult-onset hydrocephalus is mostly sporadic and characterized by ventricular enlargement in the absence of significant elevations of intracranial pressure; therefore this form is termed normal pressure hydrocephalus (NPH). Definite changes in CSF flow, resorption, and associated dynamics have been found in NPH patients, and these changes may represent a pathogenic mechanism or a secondary phenomenon [47]. Adult-onset hydrocephalus may develop either as a result of decompensation of a “compensated” congenital hydrocephalus, or it may arise de novo in adult life secondary to an acquired disturbance of normal CSF dynamics. The latter may be due to late-onset aqueductal stenosis or disruption of normal CSF absorptive pathways [11, 48]. Acquired (adult-onset, or NPH) form of inherited hydrocephalus is very rare. Recently, an X-linked adult-onset NPH [49] and a form of familial NPH that is transmitted in autosomal dominant fashion [50] have been reported, but detailed genetic linkage studies have not been carried out yet. The genetic etiology of this form is therefore totally unknown. Hydrocephalus has been observed in many mammals [51–59]. Animal hydrocephalus models have many histopathological similarities to humans and can be used to understand the genetics and pathogenesis of brain damage [59–64]. It has been well documented in the animal models that in the majority of cases, congenital hydrocephalus is a genetic disease. Furthermore, many congenital hydrocephalus loci have been mapped and identified in the animal models. Hydrocephalic Texas strain (HTX) rat model of inherited congenital hydrocephalus is characterized by onset in late gestation, a complex mode of inheritance, and ventricular dilatation associated with abnormalities in the cerebral aqueduct and subcommissural organ (SCO), a structure that is important for the patency of the aqueduct of Sylvius and normal CSF flow in the brain. Quantitative trait locus (QTL) genetic mapping has been performed from the progeny of a backcross of HTX rat with the non-hydrocephalic Fischer F344 strain. The disease has been linked with loci on chromosome (Chr) 9 (peak markers D9Rat2), 10 (between markers D10Rat136 and D10Rat135), 11 (peak markers D11Arb2 and D11Rat46) and 17 (peak markers D17mit4 and D17Rat154) respectively. The severity of hydrocephalus in HTX rat seems to be influenced by different genetic loci [65–68]. Another study suggested that the HTX strain is homozygous carrier of an autosomal recessive hydrocephalus gene with incomplete penetrance [69]. The genetics of another hydrocephalus inbred strain, Wistar-Lewis rats (LEW/Jms) which demonstrate inherited congenital hydrocephalus, is less clear with possible traits as an autosomal recessive [70] or semidominant or multigenic (possible QTL) with a possible locus on sex chromosomes [71], but none of the loci has been localized. In mouse models, three QTL loci associated with congenital hydrocephalus have been identified and labeled as Vent8a, Vent4b, and Vent7c. As a major QTL controlling variance in ventricular size, Vent8a is located on Chr 8 (near the markers D8Mit94 and D8Mit189). The Other two loci, Vent4b and Vent7c, show strong epistatic interactions affecting ventricular size in the developing embryo. Vent4b is located on Chr 4 (near D4Mit237 and D4Mit214), and Vent7c is located on Chr 7 (between D7Mit178 and D7Mit191) [72]. The autosomal recessive congenital hydrocephalus-1 (hy1) mouse has been characterized phenotypically by a dome-shaped head that is sometimes seen at birth or develops during the first 2 weeks. Internally, dilatation of the entire ventricular system is observed [73, 74]. A more severe phenotypic form, hydrocephalus-2 (hy2) mouse [75, 76], and an obstructive hydrocephalus (oh) mouse with communicating hydrocephalus and secondary aqueductal stenosis have also been described [77, 78]. Unfortunately subsequent efforts to identify genetic loci have not been done on these non-inbred mouse strains. In mouse targeted insertional mutagenesis, the accidental insertion of a transgene into a crucial genomic locus could yield important information, which has happened twice in hydrocephalus genetic studies. The transgenic mouse line OVE459 demonstrates autosomal recessive congenital hydrocephalus. This is caused by a Bdnf transgene-induced insertional mutation on a single locus on mouse Chromosome 8 (near marker D8Mit152). The OVE459 insertion locus is overlapped with that of autosomal recessive hydrocephalus-3 (hy3) mouse that phenotypically shows lethal communicating hydrocephalus with perinatal onset [79, 80]. The transgene insertion resulted as a rearrangement of Hydin exons in OVE459 mice. Subsequently, a single CG base-pair deletion in exon 15 of Hydin was also discovered in hy3 mice carrying the spontaneous hy3 mutant allele [81, 82]. In another targeted insertional mutagenesis resulting in congenital hydrocephalus, the CYP2J2 transgene interferes with the expression of a brain-specific isoform of the regulatory factor X4 (RFX4), which belongs to the winged helix transcription factor family. This brain specific isoform is called variant transcript 3 or RFX4_v3 and is crucial for normal brain development as well as for the genesis of the SCO. Loss of a single allele prevents formation of the SCO and leads to an autosomal dominant congenital hydrocephalus. This obstructive hydrocephalus appeared to be secondary to failure of development of the SCO [83]. The autosomal recessive congenital hydrocephalus (ch) mouse was reported decades ago [79]. Recently this mouse has been shown to have a mutation on another winged helix/forkhead transcription factor gene, Foxc1 (Mf1) on mouse Chromosome 13 [84, 85]. There is a recent report of 6 children with hydrocephalus from 3 different families with subtelomeric deletions from chromosome 6p. Three forkhead genes within this region (FOXF1 and FOXQ1) or proximal to it (FOXC1) were evaluated as potential candidate disease genes but no disease causing mutations were identified [86]. The autosomal recessive hydrocephalus with hop gait (hyh) mouse exhibits dramatic dilation of the ventricles at birth and invariably develops hopping gait. The hyh mouse shows a markedly small cerebral cortex at birth and dies postnatally from progressive enlargement of the ventricular system. The small cortex in hyh mouse reflects altered development of the neuronal cells. In this mouse, it is postulated that neural progenitor cells withdraw prematurely from the cell cycle, producing more early-born, deep-layer cerebral cortical neurons but depleting the cortical progenitor pool, and creating a small cortex. Genetic linkage analysis localized the hyh locus between markers D7Mit75 and D7Mit56 on mouse Chr 7. Later, the hyh gene was identified as α-SNAP (soluble NSF attachment protein α) [87]. Homozygous mutant mice harbor a missense mutation M105I in a conserved residue in one of the α-helical domains. The hyh mutant was not a null allele and is expressed; however, the mutant protein is 40% less abundant in hyh mice. The autosomal recessive hemorrhagic hydrocephalus (hhy) homozygous mutant mouse has dilated lateral ventricles and a patent aqueduct, with no histological abnormalities either in the subarachnoid space or in the choroid plexus. Multiple hemorrhages in the meninges and throughout the brain parenchyma can be observed in the advanced stages of hydrocephalus. The hhy locus has been localized on mouse Chr 12 [88]. Recently, several new congenital hydrocephalus models have emerged in zebra fish mutagenesis screening. These models have been shown to have the defects in embryogenesis and early development leading to enlarged brain ventricles. However, genetic loci for these models have not been identified yet [89, 90]. Genetic studies in animal models have started to open the way for understanding the underlying pathology of hydrocephalus. In contrast to research with animal models, human hydrocephalus genetic research has lagged far behind. To date, at least 43 mutants of hydrocephalus have been described, and 10 congenital hydrocephalus genes have been identified. Among them, only one hydrocephalus gene has been identified in humans (see Table 1). Table 1Summary of current known loci (or mutants) of hydrocephalus in vertebratesSpeciesStrainClinical formTrait*LocusChromosomeHuman syntenic regionHuman GeneReferencesHumanCARUnknownunknown10,15,27,32–43HumanCADUnknown8q12.2–21.2 or unknown 44,45HumanAOADNPHunknown50HumanCX-linkedL1camXXL1CAM46HumanAOX-linkedUnknownXX49RatHTXCQTLD9Rat2 9q38 5q21.1, 18p11.22–31 65–68RatHTXCQTLD10Rat136, D10Rat13510q32.1–10q32.317q21.3–q25.365–68RatHTXCQTLD11Arb2, D11Rat4611q233q27–28, 22q11.21,10p12.2 65–68RatHTXCQTLD17Mit4, D17Rat15417q12.1 1q43, 10p11.21–p13 65–68RatLEW/JmsCAR, (QTL)unknownunknown70,71MouseC57BL/6JCQTLVent8a88p11–23, 13q11–3472MouseC57BL/6JCQTLVent4b46p, 972MouseC57BL/6JCQTLVent7c719q10–1372Mousehy1CARunknownunknown73,74Mousehy2CARunknownunknown75,76Mousehy3CARHydin816q22.2HYDIN79–82MouseC57BL/10JCARhyh719q13.3a-SNAP87, 100MouseC57BL6/JCADRfx41012q24RFX483MouseBALB/cHeACARhhy1214q3288MousechCARMf1136p25FREAC-384MouseSTOCK tbCARohunknown77,78MouseC57BL/6*CBA/JCARMdnah5155p15.2DNAH5106MouseC57BL/CBACADOtx21414q21-q22OTX2122Mouse129P2/OlaHsdCARMsx154p16.3-p16.1MSX1107,108MouseC57BL/6CARSocs71117q12SOCS7118MouseC57BL/6JCARNmhc-b1117q13MYH10121Zebrafishm404/m491CARapounknown89,90Zebrafishm409/m432CARcudakunknown89,90Zebrafishm691CAReagleunknown89,90Zebrafishm591CARendeavorunknown89,90Zebrafishm584CARenterpriseunknown89,90Zebrafishm492/m510CARgalileounknown89,90Zebrafishm445/m585/m700CARgumowyunknown89,90Zebrafishm727CARhubbleunknown89,90Zebrafishm221/m470/m680CARinterrailunknown89,90Zebrafishm733CARkeplerunknown89,90Zebrafishm728CARneilunknown89,90Zebrafishm481CARpan twardowskiunknown89,90Zebrafishm172/m476CARuchu hikoushiunknown89,90Zebrafishm766CARvoyagerunknown89,90Zebrafishm331CARvikingunknown89,90Zebrafishm479/m627CARyuraunknown89,90Zebrafishm111/m307/m512/m97CARzezemunknown89,90* Genetic trait, AR: autosomal recessive, AD:autosomal dominant, QTL: Quantitative trait locus, ** clinicalform, C: congenital, AO: Adut-Onset Developmental, physiological and anatomical pathology of hydrocephalus The neuropathology of hydrocephalus has been adequately elucidated. Cerebral ventricle dilatation secondary to disturbed CSF flow has been observed as an inheritable trait in a variety of laboratory animals (as well as in humans). In most cases, defective development of the cerebral aqueduct or the subarachnoid space has been observed [61]. Affected individuals may have severe developmental delay and radiographic findings of hydrocephalus [91]. The morphological and developmental changes in the ventricular system have been well studied in three major rat models of congenital hydrocephalus: 6-aminonicotinamide (6-AN)-induced, LEW/Jms and HTX mutant rats. Comparative morphological studies revealed that 6-AN-induced hydrocephalus was comparable to the Dandy-Walker syndrome. The LEW/Jms and HTX mutant models were identical with regard to the form of presentation and progression of hydrocephalus in the postnatal period; but the pathogenesis of these two conditions in the fetal period was different. The LEW/Jms rats showed primary congenital aqueductal stenosis in early prenatal life and the hydrocephalic state appeared before pulmonary maturation was completed. However, although the model has been considered to be of congenital communicating hydrocephalus [64], the HTX fetuses demonstrated secondary closure of the aqueduct in the perinatal period. This secondary closure of the aqueduct in HTX rats is believed to be due to retrograde degeneration of the thalamus caused by apoptotic cell death [92, 93] and failure in cell proliferation [94, 95]. The HTX rat also shows a reduction in the secretory cells of the SCO. Regarding the role of the SCO in hydrocephalus pathogenesis, serial brain sections through aqueduct regions containing the SCO from HTX rats, in comparison with normal Fischer F344 strain, have been studied and found that reduced SCO glycoprotein immunoreactivity precedes both aqueduct closure and expansion of the lateral ventricles in the HTX rate (as it’s redundant) [96, 97]. Although some studies have addressed the activation of macrophages and microglia (the resident mononuclear phagocytes of the brain) within the brain in animal hydrocephalus models, little is known of their state of activation or regional distribution in human congenital hydrocephalus. In one experiment, brain tissue samples of 10 human fetal cases with hydrocephalus and 10 non-hydrocephalic controls were stained immunohistochemically with antibodies directed against MHC class II and CD68 antigens, and lectin histochemistry was done with tomato lectin. Hydrocephalus cases showed focal collections of CD68 and tomato lectin-positive macrophages along the ependymal lining of the lateral ventricles, particularly within the occipital horn. By comparison, brain tissue samples from controls showed few or no ependymal or supraependymal macrophages and the few macrophages that were present were not as intensely immunoreactive as in the hydrocephalus cases. The macrophage response detected at the ependymal lining of the ventricles and within the periventricular area in hydrocephalus may be related both to the severity of hydrocephalus and the age of the fetus [98]. Microglia that are normally interspersed throughout the intermediate zone and circumscribing the basal ganglia were within normal confines in all cases examined. Unexpectedly, hydrocephalic cases also showed focal regions of hypovascularization or alterations in the structure and orientation of capillaries within periventricular areas, compared with controls [98]. In summary, the pathological studies of hydrocephalus clearly indicate that impaired and abnormal brain development in the early development stage caused by altered neural cell fate and perturbed regulation of cellular proliferation and apoptosis. The abnormal brain development subsequently leads to the accumulation of the CSF in cerebral cavities. All these cellular and developmental events eventually lead to the congenital hydrocephalus accompanied by possible secondary inflammatory reaction and neurovascular pathogenesis. The molecular and cellular etiology of hydrocephalus One of the possible mechanisms leading to the pathogenesis of hydrocephalus is the disruption of neural cell membrane proteins that play an important function during brain development. The L1 protein coded by human X-linked hydrocephalus gene is a member of the immunoglobulin superfamily of neural cell adhesion molecules that is expressed in neurons and Schwann cells, and seems to be essential for the brain development and function. Hirschsprung’s disease (HSCR) is characterized by the absence of ganglion cells and the presence of hypertrophic nerve trunks in the distal bowel. There have been several reports of patients with X-linked hydrocephalus and HSCR with a mutation in the L1CAM gene. Therefore, decreased L1CAM may also be a modifying factor in the development of HSCR [99]. The gene carrying the mutation for autosomal recessive hydrocephalus in the hyh mouse codes α-SNAP protein. α-SNAP is essential for apical protein localization and cell fate determination in neuroepithelial cells [100]. α-SNAP plays a key role in a wide variety of membrane fusion events in eukaryotic cells. Membrane fusion is required for two main cellular functions: 1) the transport of molecules to distinct inter- and intracellular compartments and thereby maintenance of the functional and structural organization of eukaryotic cells; 2) the intercellular communication such as the regulated exocytosis (secretion) of neurotransmitters by neuronal cell, which occurs temporally and spatially as the precise sequential regulation events at the plasma membrane during the early brain development [87]. In the hyh mouse model, altered neural cell fate is also accompanied by abnormal localization of many apical proteins implicated in regulation of neural cell fate, including E-cadherin, beta-catenin, atypical protein kinase C (aPKC), inactivation-no-afterpotential D-like (INADL), SNAP receptor (SNARE), and vesicle-associated membrane protein-7 (VAMP-7) [100]. Furthermore, disturbed astrocyte metabolism in the early brain development in the kaolin-induced rat model of hydrocephalus has also been reported [101]. Hydrocephalus may also be caused by a malfunction of the ependymal cells [102–105]. Within the neonatal brain of the mouse, the hy3 protein (hydin) is confined to the ciliated ependymal cell layer lining the lateral, third and fourth ventricles. Hydin is not closely related to any previously known protein, with the exception of a 314 amino acid domain with homology to caldesmon, an actin-binding protein, suggesting that hydin interacts with the cytoskeleton [81, 82]. The protein of axonemal heavy chain 5 gene (Mdnah5), dynein is also specifically expressed in ependymal cells, and is essential for ultra structural and functional integrity of ependymal cilia. In Mdnah5-mutant mice, lack of ependymal flow causes closure of the aqueduct and subsequent formation of triventricular hydrocephalus during early postnatal brain development. The higher incidence of aqueductal stenosis and hydrocephalus formation in patients with ciliary defects proves the relevance of this novel mechanism in humans [106]. Hydrocephalus may be caused by malfunction of mesenchymal cells. In mice, Msx1 is a regulatory gene involved in epithelio-mesenchymal interactions in limb formation and organogenesis. In the embryonic brain, the Msx1 gene is expressed along the dorsal midline. The most important features observed in homozygous Msx1 mutants were the absence or malformation of the posterior commissure (PC) and of the SCO, the collapse of the cerebral aqueduct, and the development of hydrocephalus. The heterozygous mutants developed an abnormal PC and smaller SCO, as revealed by specific antibodies against SCO secretory glycoproteins. About one third of the heterozygous mutants also developed hydrocephalus; therefore the phenotype may be determined by the Msx1 gene dosage during a specific developmental period [107, 108]. In the autosomal recessive congenital hydrocephalus (ch) mouse model, a truncated protein lacking the DNA-binding domain of the forkhead/winged helix gene, Mf1, was generated. Mesenchymal cells from Mf1lacZ embryos differentiate poorly into cartilage in micromass culture and do not respond to added bone morphogenetic protein 2 (BMP2) and transforming growth factor-beta 1 (TGFB1). The differentiation of arachnoid cells in meninges of the mutant mice is also abnormal. The levels of developmental growth factors such as TGFB1 and BMP2 are dramatically increased in the ch mouse, and it is possible that phenotypic hydrocephalus of Mf1 mouse is due to the secondary effect of these elevated growth factors. Corresponding to studies in the ch mouse, human patients with deletions in the region containing human Mf1 homolog FREAC3 were found to develop multiple developmental disorders, including hydrocephalus [84]. Another winged helix transcription factor causing congenital hydrocephalus when mutated, RFX4_v3 transcript, is dynamically expressed in the developing brain from the neural plate stages. The RFX proteins belong to the winged-helix subfamily of helix-turn-helix transcription factors, and bind to ‘X-boxes’ in target DNA sequences and regulate expression of the downstream target genes [83]. Disruption of both RFX4_v3 alleles by insertional mutagenesis severely alters early brain morphogenesis, reduces Msx2 expression, and causes a deficiency in WNT signaling [83]. This may suggest that RFX4_v3 is probably upstream of Mf1 in the signaling pathway during early brain development. Hydrocephalus may be caused by perturbation of growth factor signaling [109, 110]. Developmental abnormalities in congenital hydrocephalus provide the clues for the perturbation of major signaling pathways in the development [111]. TGFB is an important cytokine and growth-signaling molecule in the brain. In mouse models, severe hydrocephalus has been observed in transgenic mouse overexpression of TGFB1 in astrocytes [112, 113]. In the HTX rat, increased level of TGFB3 may contribute to the development of hydrocephalus [114]. In mouse models, fibroblast growth factor-2 (FGF-2) seems to play a predominant role in the proliferation of neuronal precursors and in neuronal differentiation in the developing cerebral cortex even at relatively late stages of brain neurogenesis. Administration of FGF-2 to embryonic brain induces hydrocephalic brain morphology and aberrant differentiation of neurons in the postnatal cerebral cortex [110]. IGF binding protein-1 (IGFBP-1) modulates the cellular action of insulin-like growth factors (IGFs), some of which are expressed in the fetal brain. Hydrocephalus has been observed in mouse models that overexpress liver-specific IGFBP-1 during fetal life. The hepatic over-expression of IGFBP-1 may have endocrine effects on brain development and induction of congenital hydrocephalus [115]. Other studies have shown that up-regulation of certain growth factors in the brain could lead to altered brain fluid dynamics [116, 117]. SOCS7 is a member of the suppressor of cytokine signaling (SOCS) protein family. SOCS proteins have a similar structure: an N-terminal domain of variable length, a central Src homology-2 domain, and a C-terminal SOCS box. Biochemical and genetic studies have revealed that SOCS family members play an important role in the termination of cytokine and growth factor signaling. Homozygous Socs7 mutant mice were born in expected numbers, were fertile, and did not exhibit defects in hematopoiesis or circulating glucose or insulin concentrations. Strikingly however, these homozygous Socs7 mice were 7–10% smaller than their wild-type littermates, and within 15 weeks of age approximately 50% of the homozygous Socs7 mice died as a result of hydrocephalus. In situ hybridization studies in normal mice have revealed that Socs7 is prominently expressed in the brain, suggesting that SOCS7 plays an important functional role in early brain development [118]. We can therefore hypothesize that loss of SOCS7 function will lead to increased expression of cytokines resulting in developmental abnormalities and congenital hydrocephalus due to its inhibitory role in cytokine signaling. Hydrocephalus may also be caused by the disruption of extracellular matrix (ECM). In the TGFB1 over-expression mouse model, the changing expressions of a remodeling protein - matrix metalloproteinase-9 (MMP-9) and its specific inhibitor- tissue inhibitor of metalloproteinases-1 (TIMP-1) were also found to be important factors in the spontaneous development of hydrocephalus by altering the ECM environment [119]. Furthermore, increased expression of cytokines such as TGFB1 might also reciprocally play an important role by disrupting the vascular ECM remodeling, promoting hemorrhages, and altering the re-absorption of CSF [120]. In another mouse model, ablation of the nonmuscle myosin heavy chain II-B (NMHC-B) results in severe hydrocephalus with enlargement of the lateral and third ventricles. These defects may be caused by abnormalities in the cell adhesive properties of neuroepithelial cells and suggest that NMHC-B is essential for both early and late developmental processes in the mammalian brain [121]. Hydrocephalus may also be caused by the disruption of major early brain developmental patterning molecules. Autosomal dominant hydrocephalus in Otx2 mutation mice is characterized by eminent dilatation of lateral ventricles and a ballooned cerebrum. Histopathology shows edematous change of the periventricular white matter, suggesting that Otx2 functions as a brain developmental organizer, and a disruption of this gene is a likely cause of hydrocephalus [122]. In conclusion, many genetic loci of hydrocephalus have been defined in animal models, which is building a foundation for better understanding of molecular etiology of hydrocephalus; however, genetic research of hydrocephalus in humans is limited. The histopathological similarities of animal models can be used to understand the genetics and pathogenesis of human hydrocephalus. For example, the histopathological and morphological appearance of hydrocephalic HTX rats is demonstrated in Fig. 1, in which hydrophilic HTX rat exhibits large cerebral ventricles, a progressively-thinned cortical mantle, and stretched internal capsule fibers[123]. Review of the molecular etiologies shows a very diverse set of pathogenetic mechanisms. Perturbation of almost any molecule that plays a crucial role in early brain development, and sequentially regulation of dynamics of cerebrospinal fluid, could lead to the pathogenesis of congenital hydrocephalus. The 10 known hydrocephalus genes mostly code for important cytokines, growth factors, or related molecules in the cellular signal pathways during early brain development. Fig. 1Comparison of rat brain morphology by MRI and histology at 4, 11 and 21 days of age in non-hydrocephalic and hydrocephalic HTX rats. T2-weighted MRI scans of coronal sections from a non-hydrocephalic HTX rat at the level of the thalamus at 4, 11 and 21 days (A, D, G) shows small ventricular and subarachnoid spaces compared to an age-matched HTX littermate (B, E, H) that exhibits large cerebral ventricles, a progressively-thinned cortical mantle, and stretched internal capsule fibers (*). Congenital closure of the cerebral aqueduct becomes life-threatening by 21 days of age. Histological sections (C, F, I) at the level of the midbrain at the same ages demonstrate extreme thinning of the cortical mantle. MRI images are modified from Jones et al (2000) [123] with permission by Maney Publishing; histological samples are from the doctoral thesis of Janet M. Miller, PhD Future prospects It is essential to recognize that molecular genetics is the only current scientific approach that can be used to study hydrocephalus in which the usual concern about whether an observed phenomenon is a consequence or a cause is completely addressed. Despite our knowledge of the genetics of hydrocephalus in animal models, we have very limited knowledge about the genetic and molecular mechanisms that cause human hydrocephalus. Without this knowledge, it is impossible to say whether the pathogenesis of human hydrocephalus is comparable to that seen in animal models, and impractical to extrapolate data gained from animal models to humans. In order to better understand human hydrocephalus and to develop more appropriate translational research, it will be necessary to conduct large-scale genetic studies of human hydrocephalus. If, and when, more heritable forms of human hydrocephalus are identified, and underlying genes and their functions are characterized, then this knowledge could be used to improve patient care in a variety of different ways such as prenatal diagnosis and new potential therapeutic approaches. Possible new mechanisms other than altered CSF circulation and resorption, if uncovered via the genetic research, may also help explain why patients with hydrocephalus may experience symptomatic progression despite functioning shunts. Efforts to identify genetic variants associated with susceptibility to genetic diseases rely on three major approaches: pedigree and sib-pair linkage analysis and population association studies. The differences among these study designs reflect their derivation from biological versus epidemiological traits. Like most common diseases, it would be very difficult to identify and recruit large pedigrees in hydrocephalus that show hereditary transmission of the condition. Therefore, the last two approaches, sib-pair linkage analysis and population association studies, are the best options for the genetic study on this disease. For any study, but particularly in the case of genetic mapping for a common disease, a large sample size is crucial in achieving statistical significance. Recent advances in the genomics and statistical methodology in genetic mapping will certainly help in making well-powered studies more feasible, by reducing the number of genetic markers or workload required for these studies. For example, since many genes and loci response for hydrocephalus in animal models have been mapped, candidate genes approach will certainly be the very first choice to test the collected human hydrocephalus population for linkage and association analysis. In collaboration with the Hydrocephalus Association (HA), our group has initiated a genetic study of human hydrocephalus. As part of a prospective study that has been approved by the Johns Hopkins Institutional Review Board, we are collecting blood samples from both congenital and acquired NPH hydrocephalus patients. The objectives of this study are to identify the genetic loci responsible for the development of hydrocephalus, to examine the relationship between genotype and phenotype and to define the functions of these genes during early development. This is the first large-scale research study of its kind and information gained from this study will undoubtedly provide invaluable information concerning the developmental mechanisms of this disease in humans. Such knowledge will hopefully lead to the more reasonable treatment schemes, the better diagnostic tools, and the more effective therapeutic modalities.

Pediatr_Nephrol-4-1-2259254

Muscle wasting in chronic kidney disease: the role of the ubiquitin proteasome system and its clinical impact

Muscle wasting in chronic kidney disease (CKD) and other catabolic diseases (e.g. sepsis, diabetes, cancer) can occur despite adequate nutritional intake. It is now known that complications of these various disorders, including acidosis, insulin resistance, inflammation, and increased glucocorticoid and angiotensin II production, all activate the ubiquitin–proteasome system (UPS) to degrade muscle proteins. The initial step in this process is activation of caspase-3 to cleave the myofibril into its components (actin, myosin, troponin, and tropomyosin). Caspase-3 is required because the UPS minimally degrades the myofibril but rapidly degrades its component proteins. Caspase-3 activity is easily detected because it leaves a characteristic 14kD actin fragment in muscle samples. Preliminary evidence from several experimental models of catabolic diseases, as well as from studies in patients, indicates that this fragment could be a useful biomarker because it correlates well with the degree of muscle degradation in dialysis patients and in other catabolic conditions. Maintenance of protein stores in chronic kidney disease In uremic patients, protein stores are frequently depressed when assessed by a low serum prealbumin and weight loss, which includes loss of muscle mass [1, 2]. In pediatric patients with chronic kidney disease (CKD), linear growth is impaired and muscle mass is reduced. Although these findings have been attributed to “malnutrition”, many uremic patients with muscle wasting have not developed the problems because of inadequate diet; instead, they have complications that induce a complex series of physiological and biochemical adaptations, resulting in protein catabolism [3, 4]. In children and adults with CKD, these complications include metabolic acidosis, insulin resistance, increased glucocorticoid production, high levels of angiotensin II (Ang II), and inflammation [5–8]. Many observational studies and mechanistic investigations have attempted to explain this loss of protein stores, and especially the loss of muscle mass. There are at least three conclusions from these studies. First, rates of protein synthesis in muscle are generally unchanged, whereas rates of protein degradation tend to increase [5, 9]. Second, the daily rates of protein turnover in cells are so high (3.5–4.5 g protein/kg per day) that even a small increase in protein degradation (and/or a decrease in protein synthesis) will cause marked protein depletion over time [10]. Third, the increase in muscle protein degradation in uremia and most other catabolic disease states is mostly due to programmed activation of the ubiquitin–proteasome system (UPS) [5, 11]. Therefore, to understand muscle wasting, one must understand the UPS. The ATP-dependent, ubiquitin–proteasome system (UPS) Over the past two decades, progress in understanding the action and regulation of the UPS has been at the center of attempts to understand the control of protein turnover. The UPS includes concerted actions of enzymes that link ubiquitin (Ub), a member of the heat-shock protein family, to protein substrates that are destined for degradation (Fig. 1). When a chain of at least four to five ubiquitins are linked to a protein, it is marked for degradation in a second step mediated by the proteasome [12, 13]. Specifically, the tagged protein will be recognized by the 26S proteasome, a very large multicatalytic protease complex that not only recognizes Ub-conjugated proteins but also removes Ub, unwinds the protein, and injects it into the central core of the 26S proteasome. Once inside this central, tube-like structure, the protein substrate is degraded into small peptides [14]. Fig. 1The ubiquitin–proteasome pathway of protein degradation. Ubiquitin (Ub) is conjugated to proteins destined for degradation by an ATP-dependent process that involves three enzymes (E1–E3). A chain of five Ub molecules attached to the protein substrate is recognized by the 26S proteasome, which removes Ub and digests the protein into peptides. The peptides are degraded to amino acids by peptidases in the cytoplasm or used in antigen presentation. (Reproduced with permission from [15]) Three enzymatic components are required to link Ub to proteins that are destined for degradation. There appears to be only one E1 (Ub-activating) enzyme and around 40 E2 (Ub-carrier or conjugating) proteins. The key enzyme, however, is the E3 enzyme, which accounts for the exquisite specificity of proteins to be degraded. There are at least a thousand E3 enzymes (Ub ligase), and each can recognize a specific protein substrate and catalyze the transfer of an activated Ub from the E2 carrier protein to the substrate protein [10, 15]. Since the initial reports that the UPS recognizes specific proteins and tags them for destruction, knowledge about proteolytic processes in the proteasome has exploded. Thousands of proteins have been recognized as being degraded by the UPS, and novel cellular functions are now known to be regulated by Ub conjugation. In terms of protein breakdown, the major functions of the pathway are: Rapid removal of proteins Protein degradation is irreversible, and hence, destruction of a protein generally leads to a complete termination of cellular process mediated by the protein. Consequently, protein degradation is critical for the regulation of metabolism and cell turnover. The rapid degradation of specific proteins also permits cells (as well as the organism) to rapidly adapt to a change in physiological conditions (e.g. requiring a switch to glucose as an energy source involves converting protein stores into amino acids that can be used in gluconeogenesis). Regulation of gene transcription Ub conjugation affects gene transcription because many transcription factors become conjugated to Ub, and transcription activators are degraded by the proteasome [16]. This process regulates transcriptional activity by removing “spent” activators and resetting a promoter for additional rounds of transcription [17]. Second, the ability of transcription factors to function varies with their location within the cell. For example, nuclear factor (NF)-κB, a proinflammatory transcriptional factor, is kept outside the nucleus because movement into the nucleus is blocked by its association with an inhibitory protein, IκB. Destruction of IκB, initiated by the IKK kinase and carried out by the UPS, frees NF-κB, which then translocates to the nucleus to stimulate gene transcription [18]. A clinical application of this function of the UPS has developed in oncology. Bortezomib (Velcade, PS-341), a proteasome inhibitor, has proven to be beneficial in patients with multiple myeloma and is currently in clinical trials for the treatment of other cancers [19, 20]. The proposed mechanism of action involves the ability of bortezomib to prevent the UPS-induced destruction of IκB, thereby blocking the activation of NF-κB (an antiapoptosis transcription factor), leading to an increase in apoptosis [21]. Inhibition of the proteasome, therefore, will induce apoptosis of the neoplastic cells [22]. In addition, myeloma cells are also particularly dependent upon NF-κB to produce essential growth factors [especially inerleukin 6 (Il-6)]; when NF-κB is inactive, the growth of myeloma cells is depressed. Quality-control mechanism The UPS selectively eliminates abnormally folded or damaged proteins that have arisen because of missense or nonsense mutations, biosynthetic errors, proteins damaged by oxygen radicals, or by denaturation. For example, in cystic fibrosis, the mutant form of the transmembrane conductance regulator protein (CFTR) is selectively degraded before it reaches the cell surface [23]. The UPS catalyzes destruction of this mutant CFTR because its tertiary structure is abnormal. Another example is the degradation of misfolded proteins within the endoplasmic reticulum. Endoplasmic-reticulum-associated degradation (ERAD) of proteins removes misfolded proteins by targeting them for destruction by proteasomes in the cytoplasm [24]. Influencing the function of the immune system The UPS is responsible for creating antigens from the degradation of foreign proteins (e.g. viral particles). The antigens are presented on the major histocompatibility complex as class I molecules. In this way, the 26S proteasome exerts dual roles of removing foreign proteins and creating a stimulus of the immune system [15, 25]. As a source of amino acids When carbohydrate calories are rapidly needed or when cells must respond to catabolic diseases/conditions, there is breakdown of cell proteins, especially skeletal muscle proteins. The UPS degrades muscle proteins to provide amino acids that can be converted to glucose (i.e. gluconeogenesis). An undesired consequence of this activity could be an inappropriate loss of muscle protein. Functions of Ub not associated with proteolysis Ub can also be conjugated to proteins as a monomer (rather than as the typical Ub chain). When this occurs on cell-surface proteins, the protein is internalized into the endocytic pathway to be degraded in lysosomes [26, 27]. Uremia-activated mechanisms that accelerate loss of muscle protein Results from rodent models of CKD have established that accelerated muscle protein catabolism involves many of the same cellular mechanisms that cause muscle wasting in other catabolic conditions, such as cancer, starvation, insulin deficiency/resistance, or sepsis [10, 28]. The principal mechanism causing muscle atrophy in CKD involves activation of the UPS. Evidence for this includes the presence of higher levels of mRNAs encoding certain components of the UPS, as well as a similar pattern of increases and decreases in the expression of about 100 atrophy-related genes (also called atrogenes) [5, 28]. Changes in atrogenes include decreased expression of various growth-related genes and increased expression of components of the UPS. Patients with different clinical conditions associated with muscle atrophy exhibit similar increases in mRNAs encoding components of the UPS (e.g. an increase in mRNAs encoding Ub and proteasome subunits) [11, 29–31]. In these cases, changes in gene expression are most likely due to transcriptional regulation because we have shown that uremia or abnormal insulin responses increase the transcription of Ub and subunits of the proteasome [5, 6, 32]. Additional evidence linking the UPS to protein degradation in catabolism is the finding that the increase in protein degradation in the muscle of rats with CKD (and other muscle-wasting conditions) can be blocked by inhibitors of the proteasome [5, 6, 33]. Considered together, these results indicate that muscle wasting is a specific and carefully orchestrated program. Other questions are why are proteins degraded, and how is the complex program triggered in widely varied pathological conditions (e.g. acidosis in renal failure, low insulin levels in fasting and diabetes, inactivity, or glucocorticoids and cytokines in sepsis and other inflammatory conditions)? In fasting and in other disease states, acceleration of muscle-protein breakdown mobilizes amino acids, which are used for protein synthesis in tissues and for conversion to glucose in the liver [10]. However, if excessive protein degradation persists, the protein loss will have deleterious effects. In CKD, the breakdown of tissue protein produces nitrogenous waste products, which must be excreted to prevent the accumulation of uremic toxins [3, 10]. Finally, in muscle wasting conditions, contractile proteins are lost differentially, whereas in conditions causing atrophy (e.g. aging), all components of muscle cells seem to be affected. What accounts for muscle-specific response? The answer to this question lies in the involvement of the UPS. The UPS degrades a specific protein depending on which E3 ubiquitin ligase is activated. For example, two Ub ligases, atrogin-1 (also known as MAFbx) and MuRF-1, are found specifically in muscle, and their expression increases dramatically (8- to 20-fold) in catabolic states, causing loss of muscle protein [28, 34, 35]. In mice lacking the genes for either atrogin-1 or MuRF-1, muscles grow normally, but in response to muscle denervation, the ensuing atrophy is 30–50% slower [34]. In addition, the muscle’s content of atrogin-1 mRNA can be considered a biomarker for the rate of proteolysis in muscles responding to a catabolic condition [36–38]. In uremia, the initial cleavage of myofibrillar proteins is mediated by caspase-3 Myofibrillar proteins (including actomyosin) comprise about two thirds of the protein in muscle, the major store of amino acids for new protein synthesis and for gluconeogenesis. The UPS readily degrades the main proteins in myofibrils (i.e. actin, myosin, troponin, or tropomyosin), but it does not readily break up the myofibril into its main component proteins [39]. This means that another proteolytic system must initially digest myofibrils to create substrates that can be degraded by the UPS [40]. Many catabolic conditions are associated with inflammation and/or cell injury, and these conditions activate a cysteine protease called caspase-3. We tested caspase-3 in an in vitro system using purified actomyosin and found that caspase-3 cleaves actomyosin and leaves a characteristic 14kD actin fragment [40]. When we activated caspase-3 in cultured muscle cells, we found that UPS rapidly degraded myofibrillar component proteins, and again, the 14kD C-terminal fragment of actin was left in the muscle cells [40]. The protein-cleaving action of caspase-3 is important because blocking caspase-3 will reduce overall protein degradation in muscle [40]. In addition, we have found the same cleavage processes are present in muscles of rodent models of uremia, type-1 or type-2 diabetes, or in Ang-II-induced hypertension [7, 35, 40, 41]. Moreover, we found that the 14kD actin fragment also accumulates in muscles of patients with loss of muscle mass due to painful osteodystrophy, uremia, or burn injury [42]. In the latter study, we found that the rate of protein degradation in muscle (measured from the turnover of labeled amino acids) directly correlated (r = 0.78) with the level of the 14kD actin fragment in the same muscle. In addition, there was a lower level of the 14kD actin fragment in muscle of hemodialysis patients who participated in 18 weeks of an endurance exercise training program. In summary, the level of the 14kD actin fragment is directly associated with measured protein degradation in muscle, and the accumulation of the fragment responds to a beneficial therapeutic intervention. If these properties hold up in other trials, the level of the 14kD actin fragment could be used as a “biomarker” of increased muscle protein degradation [42]. Signals triggering muscle atrophy in kidney disease and other catabolic states Complications of CKD, as well as the complex syndrome of uremia, can trigger muscle protein breakdown. The triggering complications include metabolic acidosis, decreased insulin action (including insulin resistance), increased glucocorticoid production, high levels of Ang II, and inflammation [5–8]. Metabolic acidosis is known to cause accelerated protein degradation in infants, children, adults, the elderly, and patients with CKD (Table 1). The mechanism by which metabolic acidosis causes muscle wasting involves activation of the UPS and caspase-3 [40, 43]. In addition, acidosis changes hormone actions, such as insulin resistance and increased glucocorticoid production, which are involved in activating protein degradation [32, 41, 43–45] (Table 2). It is important to emphasize that the correction of metabolic acidosis has been shown to decrease protein breakdown in various clinical trials, indicating why maintaining normal serum bicarbonate levels should be part of standard clinical care (Table 1). Table 1Evidence that metabolic acidosis induces catabolism of protein and amino acids in normal infants, children, and adults, as well as in patients with chronic kidney disease (CKD)Subjects investigatedOutcome measurementsTrial outcomeInfants [68]Low-birth-weight, acidotic infants were given NaHCO3 or NaClNaHCO3 supplement improved growthChildren with CKD [69]Measured rates of protein degradation in children with CKDProtein loss was ∼ 2-fold higher when HCO3 was < 16 mM compared with > 22.6 mMNormal adults [70]Acidosis induced and then measured amino acid and protein metabolismAcidosis increased amino acid and protein degradationNormal adults [71]Induced acidosis and then measured nitrogen balance and albumin synthesisAcidosis induced negative nitrogen balance and suppressed albumin synthesisChronic renal failure [72]Nitrogen balance before and after treatment of acidosisNaHCO3 improved nitrogen balanceChronic renal failure [73]Essential amino acid and protein degradation before and after treatment of acidosisNaHCO3 suppressed amino acid and protein degradationChronic renal failure [74]Muscle protein degradation and degree of acidosisProteolysis was proportional to acidosis and blood cortisolChronic renal failure [75]Nitrogen balance before and after treatment of acidosisNaHCO3 reduced urea production and improved nitrogen balanceHemodialysis [76]Protein degradation before and after treatment of acidosisNaHCO3 decreased protein degradationHemodialysis [77]Serum albumin before and after treatment of acidosisNaHCO3 increased serum albuminCAPD [78]Protein degradation before and after treatment of acidosisNaHCO3 decreased protein degradationCAPD [79]Weight and muscle gain before and after treatment of acidosisRaising dialysis buffer increased weight and muscle massCAPD continuous ambulatory peritoneal dialysisTable 2Metabolic acidosis in otherwise normal humans changed hormonal levels or responses to hormonesHormoneAcidosis-induced responseGrowth hormone (GH) [80–84]Suppressed GH secretionLower IGF-1 responseInsulin [44, 85, 86]Suppressed insulin-stimulated glucose metabolismInsulin-like growth factor (IGF)-1 [81, 84, 87]Decreased IGF-1 in plasma, and kidney and liver (but not in muscle)Thyroid hormone [82, 88]Decreased plasma T3 and T4 levels plus a higher plasma thyroid-stimulating hormoneGlucocorticoids [89]Increased glucocorticoid productionParathyroid hormone (PTH) [90, 91]Decreased sensitivity of PTH secretion to changes in plasma calciumVitamin D [91]Suppressed activation to 1,25 (OH)2 cholecalciferol The finding that various diseases with muscle wasting are caused by activation of the UPS, plus the fact that coordinated changes in the expression of genes in muscle occur in different catabolic states, suggest that catabolic states activate a common cellular signaling pathway [28]. One signaling pathway is a decrease in phosphatidylinositol 3-kinase (PI3K) activity (Fig. 2). The involvement of this signaling pathway follows from the finding that several catabolic illnesses, including sepsis, acidosis, uremia, and diabetes, are characterized by insulin resistance [41, 44, 46, 47]. In normal muscle, binding of insulin or insulin-like growth factor (IGF)-1 to their receptors increases the activities of PI3K and its downstream target, Akt. In insulin-resistant conditions or with deficiency of IGF-1, the activity of this signaling pathway is depressed [35, 48, 49]. When PI3K activity falls, there is decreased production of phosphatidylinositol-3,4,5 phosphate (PIP3), leading to decreased phosphorylation and activity of the downstream serine/threonine kinase, Akt. This is a key step, because activated Akt is a major stimulator of growth-related processes via phosphorylation of the downstream kinases, GSK1 and mTOR/S6kinase, stimulating protein synthesis. On the other hand, reduced PI3K-Akt signaling (as occurs in insulin resistance) enhances protein breakdown in muscle [41]. The rise in muscle protein losses is associated with two catalytic processes: first, caspase-3 is activated to break down the complex structure of muscle; second, there is increased expression of the E3 ubiquitin ligases, atrogin-1 and MuRF-1, to degrade the proteins made available by caspase-3 [35–37] (Fig. 2). The result is muscle wasting. Fig. 2The balance between muscle hypertrophy and atrophy depends on whether protein synthesis is more active than degradation or vice versa. In protein synthesis, insulin-like growth factor (IGF)-1 and insulin signaling leads to increased phosphatidylinositol 3-kinase (PI3K), which promotes Akt phosphorylation (Akt-P). Akt-P promotes phosphorylation of GSK1 and mTOR/S6 kinases, leading to increased protein synthesis. Akt-P also phosphorylates the forkhead (FoxO) transcription factor, preventing it from entering the nucleus to promote expression of atrogin-1, MuRF-1, and other atrogenes, thereby blocking protein degradation. In protein degradation, the opposite pathway happens, but additionally, decreased Akt-P leads to increased caspase-3 activity, further promoting degradation. In inflammation, it is thought that tumor necrosis factor (TNF)-α and other inflammatory cytokines phosphorylate the inhibitor of nuclear factor (NF)-κB (IκB), to free NF-B to enter the nucleus and promote MuRF-1 expression, and ultimately, muscle protein degradation. (Reproduced with permission from [15]) How are the two mechanisms activated? In studies of muscles from insulin-deficient rats or db/db mice (a model of insulin resistance), we found that accelerated muscle protein degradation increases the level of the proapoptotic factor, Bax [35, 41]. This is relevant because activated Bax causes the release of cytochrome C from mitochondria. Cytochrome C release, in turn, activates caspase-3 to trigger actomyosin/myofibril cleavage, leaving behind the 14kD actin fragment marker [35]. The activation of atrogin-1, and hence the UPS, involves another mechanism: changes in activity of the fork-head transcription factors (FoxO 1, 3 and 4). When these transcription factors are phosphorylated by Akt, they cannot enter the nucleus to stimulate transcription of atrogin-1. However, when PI3K/Akt activities are low, the FoxOs are not phosphorylated, so they can enter the nucleus to increase the transcription of atrogin-1, resulting in an increase in muscle protein degradation [35–37]. The influence of Akt on the other E3 ubiquitin ligase involved in muscle proteolysis, MuRF-1, is not as clearly established, and it may be linked to inflammation, as activation of NF-κB will cause overproduction of MuRF1 and muscle atrophy [37]. Besides acidosis and depressed insulin/IGF-1 action, another complication of CKD that participates in muscle wasting is increased glucocorticoid production. The complexity of these interactions is great because increased glucocorticoids can cause insulin insensitivity [32, 41, 50], and both insulin deficiency and insulin resistance increase glucocorticoid production. Glucocorticoids exert a permissive effect on protein degradation in muscle caused by several catabolic conditions. For example, activation of muscle protein breakdown does not occur in adrenalectomized animals with metabolic acidosis or with acute diabetes unless the animals are also given a physiological dose of glucocorticoids [32, 51–53]. Similarly, the increase in muscle proteolysis induced by Ang II or sepsis is largely eliminated by inhibiting the glucocorticoid receptor [7, 54]. This response to glucocorticoids is “permissive” because the same physiological level of glucocorticoids does not stimulate muscle protein degradation unless the animals are either acidotic or insulin deficient. These complex interactions actually make “physiological sense” because glucocorticoids evolved to integrate stress responses in different tissues. When glucose is needed, glucocorticoids mobilize amino acids from muscle protein. At the same time, glucocorticoids induce gluconeogenic enzymes in liver to catalyze the conversion of the amino acids to glucose and urea. In children, accelerated loss of protein stores associated with glucocorticoid therapy frequently results in impaired linear growth. One glucocorticoid-dependent mechanism causing these defects is impaired response to insulin (insulin insensitivity). An allied mechanism is impaired response to IGF-1 arising from decreased or impaired action of growth hormone. Indeed, administration of growth hormone has been shown to improve linear growth of children being treated chronically with glucocorticoids [55]. Moreover, growth hormone has been shown to improve growth of children with chronic catabolic diseases such as CKD; whereas growth hormone improves growth, it does not improve growth in children to (or close to) a normal height for age [56–58]. The impaired response is likely due to the multiple complications of CKD, resulting in retarded growth. Interestingly, there is a small uncontrolled study of five children who were being treated with chronic intermittent hemodialysis and growth hormone. The children were then placed on an intensified daily hemodialysis regimen (3 h/day, five to six times a week) for a median time of 1.5 years [59]. The new regimen led to significant catchup growth into the range of a normal height for age. Intensified dialysis plus growth hormone could correct acidosis, improve insulin and IGF-1 signaling and responses to other hormones, and remove unidentified uremic toxins. These responses emphasize the complexity of sorting out mechanisms for muscle wasting in CKD. Another catabolic factor associated with CKD is Ang II. Infusion of Ang II into rodents causes both anorexia and muscle protein loss by mechanisms that depend on glucocorticoids [7]. Also, there is the knotty problem of understanding the influence of inflammation. In this case, the mechanism(s) causing muscle wasting in inflammation is not clear. Suggested responses include the ability of certain inflammatory mediators to cause insulin resistance, as well as a more direct influence of inflammatory mediators on muscle protein metabolism (Fig. 2). The problem is difficult because the link between an increase in inflammatory markers in uremic patients (e.g. C-reactive protein) to loss of muscle mass has not been established [60, 61]. Finally, there is an intriguing protein—myostatin—a member of the transforming growth factor (TGF)-β family of cytokines. It is produced in skeletal and cardiac muscle and regulates muscle growth by limiting it [62]. Overexpression of myostatin in muscle leads to loss of protein mass via inhibition of Akt phosphorylation with an increase in active FoxO1; this increases the expression of atrophy-related genes [63, 64]. Underexpression of myostatin results in skeletal muscle hypertrophy [65]. An increase in myostatin expression is found in several cachexia-associated disease states. However, there is limited information about the influence of kidney disease on myostatin expression and function beyond changes in myostatin mRNA [66, 67]. Conclusion In this brief review of mechanisms causing muscle protein losses, we discussed how a complex series of biochemical reactions are coordinated to create a genetic program that degrades muscle proteins. We also identified an initial step in muscle proteolysis that leaves behind a biomarker in muscle, the 14kD actin fragment, resulting from caspase-3. We emphasized how the UPS causes muscle wasting in uremia, as well as the role UPS plays in the regulation of cellular functions, ranging from the control of the cell cycle to activities that promote cancer. Indeed, inhibitors of proteasome activity have emerged as novel chemotherapeutic agents. Involvement of the UPS in such a wide range of functions explains why the 2004 Nobel Prize in Chemistry was awarded to Avram Hershko, Aaron Ciechanover, and Irwin Rose for their discovery of Ub and its role in orchestrating cellular protein turnover (http://nobelprize.org/chemistry/laureates/2004/).

J_Fluoresc-3-1-2064943

Single-Pair FRET Microscopy Reveals Mononucleosome Dynamics

We applied spFRET microscopy for direct observation of intranucleosomal DNA dynamics. Mononucleosomes, reconstituted with DNA containing a FRET pair at the dyad axis and exit of the nucleosome core particle, were immobilized through a 30 bp DNA tether on a polyethyleneglycol functionalized slide and visualized using Total Internal Reflection Fluorescence microscopy. FRET efficiency time-traces revealed two types of dynamics: acceptor blinking and intramolecular rearrangements. Both Cy5 and ATTO647N acceptor dyes showed severe blinking in a deoxygenated buffer in the presence of 2% βME. Replacing the triplet quencher βME with 1 mM Trolox eliminated most blinking effects. After suppression of blinking three subpopulations were observed: 90% appeared as dissociated complexes; the remaining 10% featured an average FRET efficiency in agreement with intact nucleosomes. In 97% of these intact nucleosomes no significant changes in FRET efficiency were observed in the experimentally accessible time window ranging from 10 ms to 10’s of seconds. However, 3% of the intact nucleosomes showed intervals with reduced FRET efficiency, clearly distinct from blinking, with a lifetime of 120 ms. These fluctuations can unambiguously be attributed to DNA breathing. Our findings illustrate not only the merits but also typical caveats encountered in single-molecule FRET studies on complex biological systems. Introduction Fluorescence (or Förster) Resonance Energy Transfer (FRET) is a process in which the energy of an excited donor fluorophore is transferred non-radiatively to an acceptor molecule [1]. The efficiency of energy transfer E is given by: where R is the distance between donor and acceptor and R0 is the Förster radius, at which 50% energy transfer occurs (typically 5 nm for Cy3–Cy5, a commonly used FRET pair). FRET is a powerful tool to study the structure and function of biological molecules, such as DNA. When extended to the single-molecule level, single pair FRET (spFRET) can potentially be applied to determine the conformational distribution of an ensemble of molecules and the dynamics of individual molecules [2–4]. We exploited spFRET to study the structure and dynamics of single nucleosomes, the fundamental units of compaction and organization of eukaryotic DNA. The nucleosome core particle consists of ∼ 50 nm DNA wrapped nearly twice around a histone-octamer protein-core [5]. Nucleosomal DNA has to unwrap from the nucleosome core to sterically allow processes such as transcription, replication and repair. Accessibility to nucleosomal DNA is facilitated by ATP-dependent remodeling enzymes in vivo [6]. However, it is known that spontaneous conformational changes of the nucleosome expose occluded sites in the DNA as well [7]. DNA breathing, the transient unwrapping and rewrapping of a stretch of DNA from the nucleosome core, has recently been studied in detail with a variety of fluorescence techniques. The equilibrium constant of this process was determined with bulk FRET measurements [8]. Unwrapping lifetimes of 10–50 ms were obtained with stopped-flow FRET measurements and Fluorescence Cross Correlation Spectroscopy (FCCS) [9]. Interestingly, based on their single pair FRET work, Tomschik et al. concluded that unwrapping of nucleosomal DNA occurs to a much larger extent than was previously anticipated [10]: they suggested that 30–60% of the nucleosomal DNA was unwrapped with a lifetime on the order of ∼150 ms before rewrapping. Although the conceptual beauty of FRET studies is undisputed, there are a number of important caveats in single-molecule FRET studies of biomolecules, such as fluorophore blinking, photobleaching and sample immobilization. Here, we addressed these issues. spFRET microscopy on mononucleosomes revealed two dominant types of dynamics: acceptor blinking and intramolecular rearrangements that we attribute to DNA breathing, which only became apparent after suppression of blinking. Upon immobilization, we observed three different populations: 90% of the nucleosomes dissociated or represented donor-only species, and 10% remained intact. Of these fully wrapped nucleosomes, 97% showed stable FRET on timescales between 0.01–10 s, while 3% showed dynamics with a dwell time of 120 ms that we attribute to conformational changes in the nucleosome. Material and methods DNA preparation A 177 base pair (bp) DNA was constructed by PCR using the 601 nucleosome positioning element [11] as template. PCR primers were as follows. Forward primer: Biotin-TTTGAATTCC CAGGGAATTG GGCGGCCGCC CTGGAGAATC CCGGTGCCGA GGCCGC (acceptor labeled nucleotide is underlined). Reverse primer: ACAGGATGTA TATATCTGAC ACGTGCCTGG AGACTAGGGA GTAATCCCCT TGGCGGTTAA AACGCGGGGG ACAGCGCGTA CG (donor labeled nucleotide is underlined). We used either Cy3–Cy5 or ATTO550-ATTO647N as donor-acceptor FRET pair. PCR products were purified with a GFX PCR DNA & Gel Band Purification Kit (GE Healthcare). The position of the labels was chosen such that after reconstitution the acceptor was located at the nucleosome exit, and the donor near the dyad axis, as illustrated in Fig. 1. Donor and acceptor were predicted to be ∼4 nm apart, as deduced from the nucleosome crystal structure [12], resulting in a FRET efficiency E of approximately 0.8 for the Cy3–Cy5 pair (R0 ∼5 nm), and of approximately 0.9 for the ATTO550-ATTO647N pair (R0 ∼6 nm). Fig. 1FRET system for the study of mononucleosome dynamics. a The 177 bp DNA construct, indicating the position of the labels 80 bp apart in a fragment containing the 601 nucleosome positioning sequence. A biotin label allowed for immobilization of the construct. b, c Illustrations of the mononucleosome structure, indicating the position of donor and acceptor upon reconstitution. The distance between the labels was ∼4 nm, at which efficient FRET takes place. Unwrapping of the DNA from the nucleosome core will be accompanied by a decrease in FRET due to increasing separation between donor and acceptor Nucleosome reconstitution Recombinant histone octamers were mixed with the DNA construct at a 1:1 ratio, in TE (1 mM EDTA, 10 mM TRIS pH 8.0) and 2 M NaCl. Mononucleosomes were reconstituted by salt dialysis against 0.85, 0.65, 0.5 and finally against 0.1 M NaCl, all buffered with TE. Bulk fluorescence measurements Bulk fluorescence experiments were carried out on a Luminescence Spectrometer (LS55, Perkins Elmer). All experiments were performed at room temperature (22°C). The nucleosome concentration was 10–50 nM. The donor dye was excited at 515 nm and the emission was recorded from 535 to 700 nm. The acceptor dye was excited at 615 nm and the emission was recorded from 635 to 700 nm, to obtain acceptor-only emission spectra. The FRET efficiency was determined from the enhanced fluorescence of the acceptor using the (ratio)A method [13]: where and are the acceptor and donor extinction coefficient respectively at wavelength λ, is the fluorescence intensity of the acceptor when excited at wavelength λ, and d+ is the fractional labeling coefficient of the donor. The fluorescence intensity of the acceptor was determined at its maximum value. d+ was determined from DNA and fluorophore absorption peaks in an absorption spectrum of the labeled DNA, measured from 230 to 700 nm with a spectrophotometer (Pharmaspec UV-1700, Shimadzu). Single-molecule FRET measurements Cleaned glass slides were amino functionalized with 10 μg/ml poly-d-lysine, and subsequently incubated for 4 h with an amine reactive polyethylene glycol (PEG) mixture: 20% mPEG-succinimidyl propionate 5,000 molecular weight (Nektar Therapeutics) and 0.2% biotin-PEG-n-hydroxysuccinimide 3,400 molecular weight (Nektar Therapeutics) in 0.1 M sodium carbonate buffer (pH 8.2). A flow cell was assembled by sealing a poly-dimethylsiloxane channel with a PEG functionalized slide. A 0.1 mg/ml streptavidin (Roche) solution was incubated for 5 min, and subsequently washed away. A sample, which typically consists of 10–50 pM of labeled mononucleosomes in 50–200 mM NaCl, 10 mM TRIS.HCl pH 8.0, 0.03% NP-40, and 10–50 nM unlabeled mononucleosomes, was injected in the channel and immobilized. An enzymatic oxygen scavenger system (1% glucose, 2% β-mercaptoethanol (βME) or 1 to 2 mM Trolox (Sigma), 0.2 mg/ml glucose oxidase, and 0.04 mg/ml catalase), was added to the buffer to extend the lifetime of the fluorophores before photobleaching. The buffer was degassed prior to use to further reduce the oxygen concentration. The flow cell was mounted on a microscope equipped with a 100X oil-immersion TIRF microscope objective (NA = 1.45, NIKON) and temperature-stabilized at 22°C using a water circulating bath connected to all parts of the setup in contact with the sample. The 514 nm line of an Ar+ laser (Coherent) was used to illuminate an area of ∼ 600 μm2 with a power of 0.9 mW. In the case of alternating excitation, a 636 nm diode laser (Power Technology) was used to illuminate an area of ∼900 μm2 with a power of 0.3 mW. Both beams were circularly polarized and were displaced parallel to the optical axis of the objective, so that an evanescent excitation field was generated by total internal reflection of the light at the glass-water interface. The excitation intensity at the interface in the evanescent field is ∼4 times higher than the incident beam intensity at the critical angle [14]. We therefore estimated that the resulting excitation intensities at the interface were ∼0.6 kW/cm2 for the 514 nm excitation and ∼0.13 kW/cm2 for the 636 nm excitation respectively. The fluorescence was collected by the objective and filtered through a custom-made dual color band pass filter (Chroma), that rejects scattered laser light, and a long pass filter (OG530, Schott). The fluorescence was further split into a donor and an acceptor channel by a custom-made dichroic wedge mirror (0.5° angle, center wavelength of 630 nm, Chroma) placed in the infinity path of the microscope [15]. A +150 mm achromatic lens (Thorlabs) projected the separate images on a multiplication gain CCD camera (Cascade 512B, Roper Scientific) operating at a frame rate of 20 to 100 Hz. Data analysis The simultaneously acquired donor and acceptor images (typically 80 by 80 pixels) were aligned with respect to one another through their cross correlation. The first 50 donor and acceptor frames were overlaid, and their intensities averaged. Low frequency background signal was filtered out with a high-pass FFT filter. The location of the fluorophores was then determined by applying a threshold of 2 times the background noise level. A time-trace of donor and acceptor intensities was then calculated by integrating the pixel intensities 1.5 pixel around the fluorophore center for each frame and each image. In the case of alternating excitation, the acceptor intensity upon direct excitation was retrieved by deinterleaving the acceptor time-trace. The FRET efficiency was calculated from [2]: where IA and ID are acceptor and donor intensity respectively, and is a parameter to correct for photophysical properties of the dyes. ΦA and ΦD are acceptor and donor quantum yield, and ηA and ηD are acceptor and donor detector efficiency respectively. As a first approximation γ was set to unity. A more accurate estimate for γ was obtained from experimental intensity time traces where donor bleaching took place after acceptor bleaching. In these cases the FRET efficiency could also be calculated from donor quenching: where ID0 is the donor intensity after bleaching of the acceptor. Combining Eqs. 3 and 4 results in: Experimental results Bulk fluorescence spectra reveal proper reconstitution of mononucleosomes The results of bulk fluorescence and absorption experiments on reconstituted mononucleosomes are shown in Fig. 2. The reconstituted sample showed efficient FRET, indicated by a distinct peak of fluorescence at the acceptor maximum emission wavelength (670 nm for ATTO647N, see Fig. 2a). This peak was not present in the labeled DNA-only sample, confirming that the donor and the acceptor were in close proximity due to mononucleosome reconstitution. As a control, we diluted the mononucleosome sample in 2 M NaCl, as this high ionic strength disrupts nucleosome structure [16]. As predicted, over 90% of the energy transfer signal was lost. Fig. 2Bulk fluorescence emission and absorption spectra revealed proper reconstitution of mononucleosomes. a Bulk fluorescence emission spectra. A distinct peak of fluorescence at the acceptor emission wavelength was seen after reconstitution, which was not present for the labeled DNA only. The peak disappeared upon dilution of the reconstituted material in 2M NaCl, an ionic strength at which nucleosome structure is disrupted. b Bulk absorption spectrum of the fluorescently labeled DNA construct. The stoichiometry of the labels was obtained by comparison with the absorption spectra of ATTO550 and ATTO647N (as provided by the manufacturer), which are plotted with dotted lines The observed average FRET efficiency in the reconstituted mononucleosomes was 0.75 ± 0.1, which was in good agreement with FRET values predicted by the position of the FRET pair in the nucleosome. From the bulk FRET experiments, and the predicted maximum FRET efficiency of ∼0.9 for a mononucleosome with the ATTO550-ATTO647N FRET-pair, we estimated the reconstitution yield to be at least 85%. Some residual donor emission can be accounted for by incomplete acceptor labeling. With absorption measurements on the DNA construct (Fig. 2b) we determined that the acceptor:donor:DNA stoichiometry was ∼0.7:0.9:1. Together, these bulk data show that the labeled DNA construct and the histone proteins properly formed mononucleosomes upon reconstitution (see Fig. 1). spFRET microscopy reveals individual nucleosomes together with a large population of dissociated nucleosomes To investigate mononucleosome subpopulations and dynamics, spFRET measurements were performed in a wide field microscope. Figure 3 shows an example of typical single-molecule fluorescence images of immobilized mononucleosomes. In Fig. 3a, an acceptor channel image was superimposed on a donor channel image. 10% of the immobilized fluorophores showed efficient FRET, as indicated by colocalized fluorescent spots in the acceptor channel upon donor excitation, and thus represent fully reconstituted mononucleosomes. Fig. 3Single molecule fluorescence image of immobilized mononucleosomes. a False color representation of averaged donor and acceptor channel images, excited at 514 nm. The arrows point at molecules that featured efficient FRET from donor to acceptor. The majority of the molecules however did not show FRET and appears in red. b The same field of view excited at 636 nm, allowing for unambiguous identification of acceptor fluorophores In contrast, 90% of the fluorophores did not show FRET at all. This conflicts with the bulk experiments, where after correction for incomplete labeling an average FRET efficiency of 0.75 was found. As mentioned before, there was a fraction of donor only labeled species (∼30%), but this alone could not explain the observed discrepancy between bulk and single-molecule measurement. When we directly excited the acceptor fluorophores (see Fig. 3b), we found that most of the donors were colocalized with an acceptor. Therefore we conclude that FRET signal was lost during the single-molecule measurement, due to disassembly of a large fraction of the nucleosomes. It is known that nucleosomes become unstable and dissociate when they are diluted to low concentrations [17, 18]. For our wide field spFRET measurements we diluted to pM fluorophore concentrations to resolve individual fluorophores. We ensured that the nucleosome concentration was always above 10–50 nM. By adding an excess of unlabeled mononucleosomes and 0.03% non-ionic detergent (NP-40) to our buffer. Thastrom et al. [19] reported that under these conditions nucleosomes do not dissociate in bulk solutions. We found that even 50 nM of unlabeled nucleosomes, far above the dilution-driven dissociation threshold, did not retain proper nucleosome folding, excluding dilution effects to be the cause. It is known that H2A–H2B histone dimers can spontaneously be exchanged from the protein core [20], which in our case would result in a transient loss of FRET. However, we found the same amount of disassembled nucleosomes upon immobilization when the octamer protein core was crosslinked by dialysis against 0.05% glutaraldehyde in 1 mM EDTA. We confirmed that the crosslinking itself did not dissociate nucleosomes with bulk fluorescence experiments. This suggests that not the histone protein core dissociates, but rather that the wrapped DNA loosens or significantly rearranges itself around the protein core. We confirmed that mononucleosomes in free solution (in the same buffer used for single-molecule experiments) remain stable for hours at room temperature using bulk fluorescence measurements. Therefore we consider the dissociation of the nucleosomes described here to be associated with their immobilization to the functionalized cover glass. As an alternative immobilization strategy we performed experiments with biotinylated BSA-functionalized cover glasses instead of PEGs. Biotinylated BSA is often used for single-molecule studies involving nucleic acids, whereas PEGs are often used for studies involving DNA-protein complexes [21]. Biotinylated BSA yielded even less intact nucleosomes. The exact nature of the interactions of the nucleosomes with the modified cover slides remains unclear, but the destabilizing effect of the surface forms a hurdle for obtaining large datasets of spFRET measurements. On the 10% immobilized mononucleosomes showing FRET, irreversible loss of FRET was only found after photobleaching, implying that their nucleosomal structure remained intact after immobilization. Single-molecule fluorescence footprint of individual nucleosomes Example intensity time-traces of intact single nucleosomes are shown in Fig. 4a and b. Donor and acceptor intensity were clearly anti-correlated, indicative of their FRET interaction. The intensity of a single donor (Cy3) was 1.4 ± 0.3 × 103 counts/10 ms at a signal to noise ratio (SNR) of 5. When the donor was quenched by FRET, both the intensity and SNR decreased. The intensity of an acceptor (Cy5) excited via FRET was 0.9 ± 0.2 × 103 counts/10 ms at a SNR of 3. After ∼4 s of continuous illumination at ∼0.6 kW/cm2 either donor or acceptor photobleached, limiting the observational window to a few seconds. The total number of emitted photons until bleaching from a FRET pair was ∼105 (calculated with gain G = 33 counts/photon, detection efficiency ηD and ηA ∼15%). The observed average FRET efficiency of the high FRET level was ∼0.5 ± 0.13, slightly lower than the values measured in the bulk. From the traces where donor bleaching takes place after acceptor bleaching, we estimated the correction factor ã for photophysical parameters of Cy3/Cy5 to be ∼0.7 ± 0.3 (Eq. 4). The corrected FRET efficiency was then ∼0.6 ± 0.3, in good agreement with values obtained from bulk measurements. Fig. 4Single molecule FRET traces from individual mononucleosomes. The top panels in a and b show the intensity time traces of donor and acceptor for green excitation; the middle panel shows the intensity time traces of donor and acceptor for red excitation, which were acquired in alternation with the green excitations. The bottom panels show the calculated FRET efficiency. The fluctuations between high and low FRET states featured perfect correlation with the corresponding acceptor intensity traces excited at 636 nm. c Histogram of the FRET efficiencies of multiple single molecule traces. d Histogram of the FRET efficiencies of a single trace. The spread in FRET efficiency was larger between different traces than within a single trace We found that the standard deviation in FRET efficiency of the high FRET state of the entire population (0.13) was larger than the standard deviation within individual traces (0.06), as shown in the histograms in Fig. 4c and d. This observation can be accounted for either by different nucleosome populations with slight variations in FRET efficiency, or by local differences in rotational freedom of the dyes due to immobilization. Acceptor blinking is the dominant source of spFRET dynamics The FRET traces shown in Fig. 4a and b are highly dynamic and fluctuate between a high FRET state (E ∼0.6, lifetime 2.5 s) and a low FRET state (E ∼0.1, lifetime 0.13 s). Interestingly, the characteristics of these fluctuations, i.e. both on and off time, and the low FRET level, are remarkably similar to those observed by Tomschik et al. [10] who performed analogous experiments. This similarity however, is remarkable in view of the completely different FRET-label location. Tomschik et al. labeled the nucleosome opposite to the dyad axis and probed the DNA at the most internal position of the nucleosome, whereas our substrate has labels at the most exterior position. Widom et al. have previously shown that the enzymatic accessibility of the DNA inside a nucleosome strongly reduces as the DNA is more internal in the nucleosome [7], suggesting a higher frequency of unwrapping events in our experiments. Because of the nearly complete absence of acceptor emission, we investigated the nature of these fluctuations in order to exclude reversible transitions of the acceptor to an inactive state (acceptor blinking, resulting in a Förster radius of effectively zero [22]) as the origin of these events. By alternating donor excitation with direct acceptor excitation we could directly monitor the acceptor condition as shown in Fig. 3b. After deinterleaving the data into two time-traces, one for green excitation and one for red excitation, it became obvious that the fluorescence intensity of the acceptor upon direct excitation correlated perfectly with the enhanced emission of the acceptor due to FRET. Thus, the low FRET state must be attributed to blinking, due to a dark-state level of the acceptor. Further evidence that these fluctuations were caused by acceptor blinking was provided by experiments with alternative acceptor dye (ATTO647N, emission spectrum similar to Cy5). Alternating excitation of the acceptor dye revealed a strong positive correlation between sensitized emission of the acceptor and direct excitation of the acceptor. In this case the low FRET state was also present, but with a much shorter lifetime of 0.046 s. In conclusion, our data confirm that the fluctuations between a high and a low FRET state reflect photophysical processes in the acceptor dye rather than nucleosome conformational changes. We further analyzed the single-molecule FRET traces for dynamics other than blinking. Therefore we filtered out blinking events by the application of a threshold on low FRET efficiencies (≤ ∼0.1–0.2, dependent on the noise in the measurement). Although careful inspection did occasionally reveal anticorrelated features of donor and acceptor channel, these features had a lifetime below the time resolution of our measurements. To confirm that we did not overlook any dynamics, we analyzed the fluorophore intensity noise in the high FRET state, which in the absence of dynamics should be limited by shot noise. The theoretical noise σtot in the measurement was estimated by [23]: where G is the multiplication gain factor, F is the excess noise factor due to the multiplication gain register, S is the number of photons that reach the camera, Φ is the camera quantum yield, D is the dark count, and σR is the readout noise. The first contribution represents photon shot noise after multiplication, the second contribution represents the camera dark noise after multiplication, and the third the ADC converter electronic noise. Readout noise and dark noise were calculated from the standard deviation of an area of the chip that was not illuminated by fluorescence to be 130 counts/10 ms. The actual noise σ in the single-molecule fluorescence traces was estimated by the standard deviation of the measured fluorophore intensity. The measured and calculated noise were tested for equality with an F test: where α is the significance level at which the test was performed (0.05), and ν1, ν2 are the degrees of freedom used to calculate σ and σtot respectively. We found that the total measured noise was significantly (typically 1.5 times) higher than that predicted by photon statistics and camera noise only. This implied that the traces contained dynamic events that cannot be fully resolved, originating from either photophysical processes (short blinking events, or intersystem crossing), or fast nucleosome dynamics. Hence, to accurately capture these events, blinking had to be further suppressed, and the sampling frequency had to be increased. Suppression of blinking In order to suppress blinking, we first tested a different acceptor dye (ATTO647N), which was reported to have superior photochemical stability compared to Cy5 [24]. As mentioned before, this acceptor dye showed blinking as well, as seen in the example traces and histograms of Fig. 5a and b. Although a small fraction of molecules did not show any dynamics in FRET, the majority significantly blinked. In the case of Cy5 93% of all acceptors excited via FRET showed significant blinking, with a lifetime of ∼0.13 ± 0.05 s, and lifetime of the high state of ∼0.8 ± 0.1 s. In the case of ATTO647N, 94% of all acceptors excited via FRET showed blinking, with a lifetime of ∼0.046 ± 0.02 s, and lifetime of the high state of ∼1.2 ± 0.2 s. In conclusion, the use of a different dye did not suppress blinking to the required level, but just yielded different blinking statistics. Fig. 5Fluorophore blinking in spFRET traces obtained from mononucleosomes. In the presence of βME, nucleosomes labeled with Cy5 a or ATTO647N b both show severe blinking in 95% of the traces. Example traces (top) and blinking lifetime histograms (bottom) are shown. c,d In the presence of an alternative triplet quencher, Trolox, blinking of both dyes is significantly suppressed. The example traces (top) show the absence of blinking in 90% of the traces, while the blinking lifetime histograms (bottom) show a small but finite amount of fast blinking still present in ∼10% of the traces Recently a different approach to reduce blinking was described by Rasnik et al. [25]. They replaced the triplet quencher βME in the oxygen scavenger system by a water-soluble analog of vitamin E, Trolox. Using this approach, Cy5 blinking in single-molecule FRET measurements on DNA constructs was eliminated. We tested the effect of Trolox in the imaging buffer on blinking of FRET pair labeled mononucleosomes. Results are shown in example traces and histograms in Fig. 5c and d. Both for Cy5 and ATTO647N blinking was dramatically suppressed in the presence of 1 to 2 mM Trolox: over 90% of the traces showed no observable blinking. Noise analyses of the intensity fluctuations in most of these traces were fully accounted for by camera noise and photon statistics (shot noise) only. Thus, within our time resolution (10 ms), no effect of short time scale blinking, or inter system crossing, was detected. Surprisingly, less than 10% of the observed FRET pairs still showed some extent of blinking indicated by fast excursions into a FRET state below 0.2, with a typical off time of 14 ± 1 and 13 ± 1 ms respectively, as shown in the histograms in Fig. 5c and d. Because the lifetime of these blinking events was on the order of the smallest sampling time used, blinking events were not identified by alternating excitation of the acceptor dye, but only by FRET efficiencies below the noise threshold. Direct excitation of the acceptor did only reveal some occasional blinking in the acceptor traces, with the same lifetime of 13–14 ms. To confirm that a small but finite amount of fast blinking still occurred in the presence of Trolox, we performed spFRET measurements on a FRET pair that was separated by 11 basepair duplex DNA. This construct does not exhibit structural changes that affect the FRET intensity. In this case we also observed a small, but finite amount of blinking in a number of traces (data not shown), with a lifetime similar to that measured on mononucleosomes. A fraction of the immobilized nucleosomes shows dynamics clearly distinct from blinking The suppression of blinking finally allowed us to unambiguously identify non-blinking dynamic events in the FRET traces. From a sample of 236 mononucleosomes that showed FRET, we found that over 95% of the traces essentially show stable FRET efficiency, as illustrated in Fig. 6a and b; all anti-correlated features in the FRET efficiency were short-lived and fall within the noise of the measurement. Thus, the upper limit for dynamic events that could have been missed in this population was 10 ms (the sampling time used). Interestingly, 3% of the traces showed dynamic events clearly distinct from blinking (examples shown in Fig. 6c and d), as judged by the following criteria: (1) the acceptor signal of a low FRET event was significantly higher than zero. (2) No correlated change in acceptor intensity was detected using alternating excitation. (3) Events persisted for at least two data-points. We found 14 events with an average FRET change ΔE of −0.23 and an average dwell time of 120 ± 5 ms, as summarized in the histogram and cumulative distribution function in Fig. 6e and f respectively. The lifetime was determined by fitting a cumulative exponential distribution to the data, independent of binning and therefore a more accurate way of determining the lifetime when using small datasets (Fig. 6f) than fitting a distribution to binned data. The lifetime of the high FRET state could not be determined accurately, due to the short time window that was available due to photobleaching. Since we explicitly checked the vitality of both fluorophores, we ruled out photodynamics and we could unambiguously attribute the observed features to DNA breathing dynamics. Fig. 6A fraction of the immobilized nucleosomes showed dynamics clearly distinct from blinking. a,b After suppression of blinking with Trolox, over 95% of the FRET traces do not show FRET dynamics. The theoretical photon and instrument noise is approximately indicated by the grey bars. c,d ∼3% of the intensity traces (top panels) showed FRET fluctuations (bottom panels) clearly distinct from blinking: the acceptor intensity was significantly higher than zero, and events persisted multiple data points (see insets). These fluctuations clearly exceeded the noise. e Histogram and cumulative distribution plot f of the lifetime of the dynamic events. An exponential fit to the data gave an average lifetime of 120 ms Discussion and conclusion Time-traces of spFRET microscopy on single reconstituted mononucleosomes revealed two types of dynamics: acceptor blinking and intramolecular rearrangements. Intramolecular rearrangements became only apparent after suppression of blinking. Both Cy5 and ATTO647N showed severe blinking in a deoxygenated buffer in the presence of 2% βME. Replacing the triplet quencher βME with Trolox effectively eliminated most blinking effects. The lifetime of DNA unwrapping that we obtained after rigorous elimination of blinking events (∼120 ms) was comparable to the 150–180 ms obtained by Tomschik et al. [10], despite the very different location of the labels in the nucleosome, and probably less important, the different DNA sequence and origin of the histones. However, we observed a very similar lifetime (∼130 ms) for Cy5 blinking under comparable buffer conditions (2% βME). The FRET efficiency of the open states in our experiments was significantly above the detection threshold, so we can explicitly exclude photophysics as the origin of the observed changes in FRET efficiency. Our single-molecule measurements revealed at least three subpopulations in the reconstituted and immobilized nucleosome sample: 90% of the fluorophores represented dissociated nucleosomes or donor only species, 10% represented intact nucleosomes. Of these, 97% remained stable on time-scales ranging from 10 ms to 10 s of seconds, while 3% showed intervals with reduced FRET efficiency and a lifetime of 120 ms clearly distinct from blinking. Why most nucleosomes dissociate upon immobilization to the cover slip remains unknown. Immobilization of the molecules is necessary for extension of the available observation time. The time limit is given by photobleaching, one of the key advantages of this method with respect to, for example, Fluorescence Correlation Spectroscopy. However, the close proximity to the surface provides ample opportunity for interactions with it. Surface induced nucleosome dissociation has been reported before in Atomic Force Microscopy (AFM) studies. Using AFM in liquid, Nikova et al. observed an unwrapping of ∼25 nm of DNA from nucleosomes absorbed to a mica surface [26]. This unwrapping was attributed to a depletion of H2A–H2B histone dimers induced by the high surface charge of the mica, resulting in unwrapping of DNA. Although PEGs are neutral polymers that are commonly used to reduce non-specific surface binding of proteins, they may affect nucleosomes in different ways: PEG molecules have been reported to interact strongly with unfolded proteins [21], and could therefore possibly interact with histone tails. Furthermore, histone proteins are known to be adhesive to glass or plastic [27]. The large fraction of dissociated nucleosomes we report here was not observed by Tomschik et al. [10]. Because of the internal position of the labels they used, at least 50 bp of DNA had to be detached from the histone core before FRET was completely lost. We labeled the DNA at the very end of the histone bound part, and accordingly a detachment of 10–20 bp of DNA would already result in complete loss of FRET. Furthermore, the exterior part of the DNA is largely constrained by the mobile H2A–H2B dimer, whereas the labeled part of the DNA in the nucleosomes used by Tomschik et al. is mostly constrained by the more stable H3–H4 tetramer. A labeling strategy by Li et al. [8, 9], who end-labeled a 601 nucleosome positioning element together with either histone H3 or H2A, provides a more comparable construct. Based on stopped-flow FRET and FCCS experiments they deduced an unwrapping rate of 4 s−1 an unwrapping lifetime of 10–50 ms. The 3% of our traces that showed dynamics typically featured multiple unwrapping events before photobleaching. Though photobleaching obstructs quantification of the unwrapping rate, it is of the same order of magnitude as observed by Li et al. The lifetime of the unwrapped state we observed is five to ten times larger. This discrepancy may in part be explained by differences in experimental conditions and nucleosome constructs; we can however not exclude the possibility that we overlook short-lived unwrapped states, biasing our data to a longer lifetime. The absence of observations of DNA unwrapping in the majority of the intact nucleosomes reported in this study is in strong contrast with the extent of DNA breathing dynamics found by Li et al. [8, 9]. Two possible explanations could account for this difference: (1) The most frequently occurring DNA unwrapping occurs at a rate that exceeds the time resolution of our experiment. The rare dynamics (3%) that we observe would reflect the release of multiple histone-DNA contacts, a process that would occur less often and on longer time scales than unwrapping of only the first DNA-octamer. However, unwrapping of 10–20 bp of DNA would induce a more dramatic reduction in FRET efficiency than the reduction we observed, which is consistent with unwrapping of 10 bp or less. (2) The immobilized nucleosomes did not undergo breathing dynamics. It should be kept in mind that because of the disruption of 90% of the nucleosome upon immobilization, we only probed a subset of nucleosomes that do not dissociate upon immobilization. These nucleosomes could either be resistant to unwrapping of the DNA, or immobilized in such a way that DNA dynamics are inhibited due to interactions with the surface, while still retaining proper folding. In either case, immobilization is expected to have major impact on nucleosome dynamics, emphasizing the need for a more inert immobilization than point attachment to a PEG coated surface. Our findings demonstrate that experimental conditions can have a profound impact on the data obtained when probing nucleosome structure and conformational dynamics. Immobilization effects and blinking dynamics have to be accounted for, and where possible suppressed in order to extract biologically relevant data from spFRET experiments. We have shown that DNA breathing kinetics obtained from carefully optimized spFRET experiments approaches values obtained from bulk experiments, opening the way to more complex single-molecule studies of chromatin dynamics.

Cardiovasc_Intervent_Radiol-3-1-2039794

Sexuality and Body Image After Uterine Artery Embolization and Hysterectomy in the Treatment of Uterine Fibroids: A Randomized Comparison

In this paper the effect of uterine artery embolization (UAE) on sexual functioning and body image is investigated in a randomized comparison to hysterectomy for symptomatic uterine fibroids. The EMbolization versus hysterectoMY (EMMY) trial is a randomized controlled study, conducted at 28 Dutch hospitals. Patients were allocated hysterectomy (n = 89) or UAE (n = 88). Two validated questionnaires (the Sexual Activity Questionnaire [SAQ] and the Body Image Scale [BIS]) were completed by all patients at baseline, 6 weeks, and 6, 12, 18, and 24 months after treatment. Repeated measurements on SAQ scores revealed no differences between the groups. There was a trend toward improved sexual function in both groups at 2 years, although this failed to reach statistical significance except for the dimensions discomfort and habit in the UAE arm. Overall quality of sexual life deteriorated in a minority of cases at all time points, with no significant differences between the groups (at 24 months: UAE, 29.3%, versus hysterectomy, 23.5%; p = 0.32). At 24 months the BIS score had improved in both groups compared to baseline, but the change was only significant in the UAE group (p = 0.009). In conclusion, at 24 months no differences in sexuality and body image were observed between the UAE and the hysterectomy group. On average, both after UAE and hysterectomy sexual functioning and body image scores improved, but significantly so only after UAE. Introduction When hysterectomy looms as a definite treatment for fibroid related menorrhagia, women and their sexual partners often express their concern about negative effects of the operation on their sexual wellbeing [1–4]. Furthermore, the uterus is considered important for a woman’s self-image and sexual image and some women fear that they will be “less of a woman” to their sexual partners [2, 4, 5]. Therefore a hysterectomy may also interfere with the woman’s body image. These concerns may cause some women to choose alternative treatment options [5]. During the last decade, uterine artery embolization (UAE) has emerged as an alternative treatment for hysterectomy [6]. UAE has been investigated in several case series and three randomized controlled trials [7–13], but the advantage of UAE in terms of sexual wellbeing and body image in comparison to hysterectomy remains unknown, since randomized controlled trials on this subject are not available as yet. Randomized controlled data are preferential to nonrandomized data, since nonrandomized data tend to overestimate treatment effect and are prone to selection bias [14]. We initiated a randomized trial comparing UAE and hysterectomy and have published several short-term results previosly [15–17]. In the present paper, we compare changes in sexuality and body image in patients with symptomatic uterine fibroids who were randomly assigned to either UAE or hysterectomy. Materials and Methods Study Design The details of our study design have been described elsewhere and are discussed here briefly [15]. The EMMY trial (EMbolization versus hysterectoMY) is a prospective randomized multicenter clinical trial with 28 participating hospitals in the Netherlands. Eligible patients met the following inclusion criteria: they were clinically diagnosed with uterine fibroids (confirmed by ultrasonography), had menorrhagia as the major complaint, were premenopausal, and had no wish to conceive. All patients were candidates for undergoing a hysterectomy: other treatment options either had failed, were undesired, or had provided unsatisfactory results. Eligible patients were informed verbally and in writing about possible risks and benefits of both procedures, and were invited to participate in the trial. After written informed consent had been obtained, patients were randomly assigned (1:1) to UAE or hysterectomy by computerized randomization, stratified for hospital. The study was approved by the Dutch Central Committee Involving Human Subjects (www.ccmo.nl) and the local ethics committees of all participating hospitals. Pre-assessment All patients were assessed by the attending gynaecologist. Current symptoms and a complete medical and gynaecological history were recorded. All patients underwent a pelvic ultrasound to determine the number of fibroids and the size of the largest fibroid. Sociodemographic characteristics were assessed by means of a questionnaire. Procedures Uterine artery embolization UAE was performed by an interventional radiologist. A catheter was introduced into the femoral artery and advanced over the aortic bifurcation to the contralateral internal iliac artery, and digital subtraction angiography was performed to identify the origin of the uterine artery. When catheters were placed correctly, the UAE was carried out. Polyvinyl alcohol (PVA) particles of 355–500 μm were used in all procedures. UAE patients were advised to refrain from sexual intercourse for at least 2 weeks and thereafter, depending on their complaints. Hysterectomy The type of hysterectomy and the route of access were determined by the gynecologist. The following procedures were allowed: abdominal hysterectomy, vaginal hysterectomy, laparoscopically assisted vaginal hysterectomy, and laparoscopic hysterectomy. Both supravaginal and total hysterectomies were allowed. After discharge hysterectomy patients were advised not to have sex until the first outpatient visit at 6 weeks. Questionnaires Sexual functioning and body image were assessed by means of patient questionnaires. At baseline, questionnaires were completed by all patients before randomization. Follow-up questionnaires were completed at 6 weeks and 6, 12, 18, and 24 months after treatment. The questionnaire consisted of the Sexual Activity Questionnaire (SAQ), the Body Image Scale (BIS), and the Mental Component Summary (MCS; as part of the Medical Outcome Study Short Form [MOS SF-36]). The SAQ, originally designed to investigate sexual functioning in women treated with tamoxifen because of a positive family history of breast cancer [18], is also useful in women with nonmalignant gynecological disorders [19]. In Table 1 the SAQ is displayed. The questionnaire consists of three dimensions, and each dimension yields its own score: pleasure from sexual intercourse (desire, enjoyment, and satisfaction; score range, 0–18), discomfort during intercourse (score range, 0–6), and habit (frequency; score range, 0–3). Pleasure and habit are good with higher scores, while discomfort is worse with higher scores. The SAQ was only filled out by patients who were sexually active during the month before receiving the questionnaire. Table 1The Sexual Activity Questionnaire (SAQ) and the Body Image Scale (BIS)SAQ Dimension pleasure  1. Was having sex an important part of your life in the last month?  2. Did you enjoy sexual activity in the last month?  3. Did you desire to have sex with your partner in the last month?  4. In general were you satisfied after sexual activity in the last month?  5. How often did you engage in sexual activity in the last month?  6. Were you satisfied with the frequency of sex in the last month? Dimension discomfort  1. Did you notice dryness of your vagina in the last month?  2. Did you feel pain or discomfort in the last month? Dimension habit  1. How did the frequency of sexual behavior in the last month compare with what is usual for you?BIS In the last month  1. Have you been feeling self-conscious about your appearance?  2. Have you felt less physically attractive as a result of your menstrual bleeding problem?  3. Have you been dissatisfied with your appearance when dressed?  4. Have you been feeling less feminine as a result of your menstrual bleeding problem?  5. Did you find it difficult to look at yourself naked?  6. Have you been feeling less sexually attractive as a result of your menstrual bleeding problem?  7. Did you avoid people because of the way you felt about your appearance?  8. Have you been feeling the treatment has left your body less whole?  9. Have you felt dissatisfied with your body?  10. Have you been dissatisfied with the appearance of your scar? Questions on the BIS shown in italics have been omitted Furthermore, participating patients were asked to judge their current sexual life in two questions: “How would you describe the quality of your current sexual life?” and “How well can you live with your current sexual life?” One of the following responses could be ticked: “very good,” “good,” “moderately good,” “neither good nor bad,” “moderately bad,” “bad,” and “very bad.” The BIS was designed in order to asses changes in body image among cancer patients [20]. The questionnaire was also found reliable and valid in women with benign gynecological conditions [21]. Table 1 displays the questions of which the BIS consists. Unfortunately, questions 2, 4, and 6 were not represented correctly in the questionnaire due to a translation error. Therefore we decided to omit these questions in order to attain a comparable outcome between groups. The BIS score ranges from 0 to 30. A higher BIS score indicates a worse body image. In order to assess mental health before treatment as a predictor of worse sexual function after treatment, the MCS score of the 36-item MOS SF-36 was used [22]. This was done because bad preoperative mental health is associated with bad sexual life outcome [23]. Sample Size and Endpoints of the Present Study The sample size was based on the primary endpoint of the clinical study, published elsewhere [24]: the elimination of menorrhagia in such way that hysterectomy could be avoided in the UAE group in at least 75% of patients within 2 years after the primary interventions. The objective of the present study was to compare the following endpoints between both interventions: the number of sexually active patients, the scores yielded by the SAQ and the BIS, the subjective quality of sexual life. For this analysis, no separate power calculation was made. Statistical Analysis Data entry was performed using SPSS data entry for Windows 3.0. A random sample of 10% of the questionnaires was visually double checked by an independent second investigator, revealing a false entry level of 0.3%. All false data entries were corrected. All analyses were performed using SPSS (release 11.5.1) statistical software. Missing items in the questionnaire were treated as follows. For the SAQ no advice is available on scoring missing items. If one item was missing, we regarded the dimension score (which the item was part of) as being missing completely, according to earlier research [25]. For the BIS no missing values were allowed since three questions were omitted already. In these cases, the entire score was omitted. For the three omitted questions, the individual mean score of the remaining seven questions was imputed. First, we determined how many patients in both treatment groups were sexually active before and after treatment. An overall percentage per treatment group was calculated, after which sexual activity in subgroups was assessed: those being sexually active and those not being sexually active before treatment. Second, the mean of the dimension scores were plotted for the hysterectomy and UAE group. Third, the mean differences between all follow-up SAQ dimension scores and baseline were compared between treatment groups. This was only possible for patients who were sexually active both before and after treatment. Within group changes were analyzed as well. Similar analyses were performed for the BIS. Differences in pleasure, discomfort, habit (SAQ), and body image (BIS) between groups over time were tested with repeated measurement analysis, excluding the 6-week measurement for the SAQ, since a high proportion of patients was expected not to be sexually active at that time and therefore not to yield a score. Differences in continuous variables were tested with Student’s t test. Differences in data with skewed distributions were tested with the nonparametric Mann-Whitney U test. The two added questions on the quality of current sexual life were compared between the groups using the chi-square test (or Fishers’ exact test when appropriate). Furthermore, the quality of sex life and the additional questions at the various follow-up moments were compared to baseline, yielding two different options: “worse,” “the same,” or “better” compared to baseline. Differences between the groups were compared with the chi-square test (or Fishers’ exact test when appropriate). Logistic regression analysis was performed to investigate variables associated with a worse sexual life quality at 24 months after treatment compared to baseline versus unchanged or improved quality of sexual life. First, univariate analysis was performed with the following baseline characteristics: intended treatment, age, BMI, parity, ethnicity, having a partner, employment status, smoking status, number of fibroids, uterine volume measured by ultrasound, dominant fibroid volume measured by ultrasound, existence of any comorbid disease, and mental health status (MCS-MOS-SF36) at baseline. Second, covariates with p values <0.1 in the univariate analysis were selected for multivariate analysis. All analyses were two-tailed and carried out based on the intention-to-treat principle. A p value <0.05 (two-tailed) was considered statistically significant. Results Patients were enrolled between March 2002 and February 2004. Of 349 eligible patients, 177 were randomized: 89 were allocated hysterectomy and 88 were allocated UAE. In the hysterectomy group 14 patients refused the allocated treatment, compared to 7 patients in the UAE group, and these patients withdrew from participation. In 4 UAE patients embolization failed bilaterally. These patients subsequently had a hysterectomy but were analyzed in the UAE group according to the intention-to-treat principle (Fig. 1) [15]. The baseline characteristics of all randomized patients are reported in Table 2. No differences between the groups are apparent, as expected considering the randomized design. Fig. 1FlowchartTable 2Baseline characteristicsUAE (N = 88)Hysterectomy (N = 89)Age (yr), mean (SD)44.6 (4.8)45.4 (4.2)Body mass index (weight [kg]/length [m]2), mean (SD)26.7 (5.6)25.4 (4.0)Parity (n)  030 (34.1%)20 (22.5%)  ≥158 (65.9%)69 (77.5%)Ethnicity (n)  Black24 (27.3%)20 (22.5%)  White54 (61.4%)57 (64.0%)  Other10 (11.4%)12 (13.5%)Marital statusa (n)  Single16 (18.2%)13 (14.8%)  Married55 (62.5%)54 (61.4%)  Living apart together5 (5.7%)4 (4.5%)  Divorced12 (13.6%)15 (17.0%)  Widow0 (0%)2 (2.3%)Partner relationshipa (n)  No partner13 (15.3%)19 (22.4%)  Partner72 (84.7%)66 (77.6%)Employment statusa (n)  Employed68 (77.3%)69 (78.4%)  Unemployed20 (22.7%)19 (21.6%)Smoking status (n)  Current smoker21 (23.9%)23 (25.8%)  Former smoker11 (12.5%)14 (15.7%)  Nonsmoker56 (63.6%)52 (58.4%)Comorbid diseaseb  Any comorbid disease24 (27.3%)22 (24.7%)Number of fibroids, median (range)2 (1–20)2 (1–9)Uterine volume (cm3), median (range)321 (31–3005)313 (58–3617)Fibroid volume (dominant fibroid; cm3), median (range)59 (1–673)87 (4–1641)Mental Component Summary (SF-36), mean (SD)40.9 (10.7)41.5 (11.0)aSome values are missingbAny of the following: hypertension, diabetes, astma, clotting disease, system disease, or other The proportion of questionnaires available for analysis ranged from 96.0% (baseline) to 98.7% (6 weeks, 24 months). Before treatment 54 of 81 (67%) and 46 of 75 (61%) of the participating patients were sexually active in the UAE and hysterectomy groups, respectively (Fig. 2A). Six weeks after treatment, sexual activity had decreased in both groups, with significantly more patients showing sexual activity in the UAE group compared to hysterectomy (53% versus 29%; p = 0.004). Hereafter sexual activity was restored in both groups, with no significant differences between the groups at 24 months (p = 0.07). Fig. 2Proportion of sexually active women, by treatment strategy, over time. (A) The weighted average of women who were not sexually active at baseline (B) and those who were sexually active at baseline (C) Patients who were not sexually active before treatment gradually resumed sexual activity after both UAE and hysterectomy (Fig. 2B). After 24 months 31% (hysterectomy) and 52% (UAE) were sexually active (p = 0.118). No significant differences were observed between the groups. Among those patients who were sexually active before treatment, there was a major drop in activity at 6 weeks after treatment, which was partially restored. After 2 years, 86% (UAE) and 90% (hysterectomy) of these patients had resumed their sexual activities (Fig. 2C). Only at 6 weeks had significantly more UAE patients resumed sexual activity (p = 0.01). Figure 3 shows the course of the SAQ dimensions pleasure, discomfort, and habit and the BIS scores over time. There were no differences between the groups at baseline (pleasure, p = 0.76; discomfort, p = 0.44; habit, p = 0.77; body image,: p = 0.60). Repeated measurements revealed no differences between the groups for the SAQ dimensions pleasure (p = 0.343), discomfort (p = 0.246), and habit (p = 0.453) or for body image scores (p = 0.359). Table 3 shows the changes in dimension scores per treatment group for patients who were sexually active at baseline. Positive numbers for pleasure and habit and negative numbers for discomfort and body image are indicative for improvement compared to baseline. No significant differences were found between the two treatment groups at all time points, except for body image at 6 months after treatment: body image had improved significantly more in UAE patients than in hysterectomy patients (UAE, –1.34, versus hysterectomy, no change; difference, –1.34; 95% CI, –2.50 to –0.18; p = 0.02). Within the groups, however, a significant improvement of various dimension scores was noted (indicated in boldface): at 24 months UAE patients reported significantly less discomfort, a higher frequency of intercourse, and a better body image compared to baseline (p = 0.022, p = 0.022, and p = 0.009 respectively). At 24 months hysterectomy patients improved on all dimensions compared to baseline, but differences were not statistically significant. Fig. 3Pleasure, discomfort, habit, and body image, by treatment strategy, over timeTable 3Mean differences in pleasure, discomfort, habit, and body image compared to baseline, by treatment strategy, over timeUAEHysterectomyp valueSAQ: pleasure (0–18)a  6 wk0.29−0.500.47  6 mo1.631.230.61  12 mo0.951.490.50  18 mo1.860.680.13  24 mo0.891.180.74SAQ: discomfort (0–6)b  6 wk−0.25−0.210.96  6 mo−0.58−0.320.41  12 mo−0.47−0.470.98  18 mo−0.51−0.290.51  24 mo−0.43−0.490.88SAQ: habit (0–3)a  6 wk−0.030.000.92  6 mo0.480.280.30  12 mo0.180.420.24  18 mo0.270.190.70  24 mo0.280.220.74BIS (0–30)b  6 wk−1.27−0.280.10  6 mo−1.340.000.02  12 mo−0.240.080.64  18 mo−1.24–0.280.15  24 mo−1.06–0.500.36Note. SAQ, Sexual Activity Questionnaire; BIS, Body Image Scale. Boldface numbers indicate a significant difference from baseline within group (p < 0.05)aA higher score represents more favorable sexual functioning (pleasure, habit)bA lower score represents more favorable sexual functioning (discomfort) or body image Table 4 reports the patients’ experienced quality of their sexual life at 6 weeks and 12 and 24 months. No differences between groups were observed. Table 5 presents the proportion of patients who judged the quality of sexual life as better, comparable to, or worse compared to baseline based on the question: “How is the quality of your current sexual life?” A minority of patients reported worsened sexual life quality compared to baseline at all time points. At 24 months the proportion of patients reporting a worse sexual life compared to baseline was slightly higher in the UAE than in the hysterectomy group, but the difference was not statistically significant (29.3% versus 23.5%; p = 0.43). Table 4Satisfaction with sexual life and ability to cope with sexual life, by treatment strategy, over timeBaseline6 weeks12 months24 monthsUAE (N = 81)Hyst. (N = 75)p valueUAE (N = 81)Hyst. (N = 75)p valueUAE (N = 81)Hyst. (N = 75)p valueUAE (N = 81)Hyst. (N = 75)p valueHow is the quality of your current sex life?  Very good1580.84710.487110.358110.54  Good2227252231182921  Somewhat good15131213911812  Neither good nor bad1212231915211615  Somewhat bad66445388  Bad43583444  Very bad33214251How well can you live with your current sex life?  Very good930.771160.3318170.9320170.78  Good2021373035303431  Somewhat good1012151611131011  Neither good nor bad201710564106  Somewhat bad89251325  Bad44252222  Very bad65011100Table 5Changes in sexual wellbeing compared to baseline, by treatment strategy, over timeUAE (n = 81)Hysterectomy (n = 75)p value6 months  Worse16130.68  The same2527  Improved322512 months  Worse17140.81  The same2926  Improved252724 months  Worse22160.32  The same2720  Improved2632 Univariate logistic regression analysis revealed BMI, number of fibroids, and presence of comorbid disease to be associated with a worsened quality of sexual life at 24 months compared to baseline (p < 0.1). Allocated treatment and mental health at baseline were not associated with worse outcome. BMI was not significant in the multivariate analysis (p = 0.163). The other two parameters remained significantly associated with worse outcome: a higher number of fibroids (OR, 0.69; 95% CI, 0.51–0.94; p = 0.018) predicted a decreased risk of a worsening sexual life at 24 months compared to baseline, while the presence of comorbid disease (OR, 3.20; 95% CI, 1.38–7.41; p = 0.007) was associated with an increased risk of a worse sexual life quality at 24 months after treatment compared to baseline. Discussion In our trial no differences, on average, in sexual function at 24 months were observed between UAE and hysterectomy. After both treatments the dimensions “pleasure” and “habit” improved, while the dimension “discomfort” decreased at 24 months compared to baseline, although only the UAE group showed significant improvement in discomfort and frequency of intercourse. Sexuality and body image after UAE have never been evaluated in a randomized controlled trial before. However, some nonrandomized series have addressed this subject. The improvement of sexual functioning after UAE was in accordance with the limited number of earlier studies on this subject [26–28], which reported a significant improvement in sexual functioning as well. Sexual functioning after hysterectomy has received much attention in the literature. Improvement in sexual functioning after hysterectomy might indirectly be associated with the consequences of surgery, such as less worry about unwanted pregnancy, absence of vaginal bleeding, and more time for sexual activities by the cessation of monthly periods [29]. Most reports on sexual functioning after hysterectomy found significant benefits in various aspects of sexual functioning [25, 30–35]. In contrast, the improvement in sexual functioning within our hysterectomy group at 24 months of follow-up was not statistically significant in our trial. This might be explained by insufficient power, which is partly due to the relatively small sample size for the SAQ scores in our population: change scores were only available from women who were sexually active at baseline and at follow-up. Furthermore, the studies on sexuality after hysterectomy used varying follow-up intervals, ranging from 6 months to 2 years. In our series, hysterectomy patients experienced significantly improved pleasure and habit scores at 12 months of follow-up. These results might indicate that improvement in sexual functioning after hysterectomy could be temporary. On the other hand, a review article recently suggested that there is no scientific proof for either improvement or deterioration of sexual functioning after hysterectomy, unless hysterectomy was performed based on a sound clinical indication [29]. In our patient group, however, there was no doubt about the indication: complaints were serious and lengthy enough to warrant a hysterectomy. Although, on average, SAQ scores for the group as a whole improved, there was a small group of women who reported a deterioration in self-reported quality of sexual life, consistent with other reports [30–32, 34, 36]. For obvious reasons, the prospects of deterioration of sexual life might be of greater concern to patients than the improvement after treatment. Regression analysis of worsened sexual life revealed that women with a low number of fibroids (possibly representing a group of women with less severe complaints) and women with other concomitant chronic diseases might not benefit from fibroid treatment in terms of sexual life improvement. The improved body image after UAE in our series is in accordance with another study that reported higher self-consciousness after UAE [26]. After hysterectomy a period of temporary worsened body image following the operation has been described previously [25, 36, 37]. This was not confirmed by our results, but in accordance with these studies, a normal or even improved body image was found 1 year after hysterectomy. There are several limitations of our study that need to be addressed. First, the values of the SAQ dimensions were based on all sexual active women at the various points in time. Reported values may reflect an ever-changing group of sexually active women. At any stage, some women may cease while others may initiate sexual activity. Of all non-sexually active women before treatment, 50% became active after treatment. When comparing pre- and posttreatment scores, these women cannot be included, since their baseline scores were missing. Second, the significant difference between resumption of sexual activity between the groups at 6 weeks is probably explained by differences in counseling between both treatments as mentioned earlier: hysterectomy patients in our trial were told not to have sex until the first outpatient visit at 6 weeks, while UAE patients were advised to refrain from intercourse for at least 2 weeks and thereafter, depending on their complaints. Third, we allowed all kinds of hysterectomies, thereby possibly biasing the results in the hysterectomy group. However, various reports found no long-term differences in sexual functioning after various surgical routes of hysterectomy [25, 32, 35]. Finally, as described above, three questions on the BIS were omitted because of wrongly posed questions, thereby creating asymmetry in the treatment arms. Since the remaining questions in both groups were identical, the comparison between the groups is still valid but reduces the comparability with other series. In conclusion, at 24 months no differences in sexuality and body image were observed between the UAE and the hysterectomy groups. After both UAE and hysterectomy, on average, sexual functioning and body image scores improved, but significantly so only after UAE.

Pediatr_Nephrol-3-1-1805047

Normative data on the Bonn Risk Index for calcium oxalate crystallization in healthy children

Bonn Risk Index (BRI) is being used for the assessment of urinary calcium oxalate (CaOx) crystallization. There are no published data regarding BRI during growth. The objective of this study was to establish age- and sex-dependent BRI values in healthy children and adolescents. A total of 1,050 Caucasian subjects aged 3–18 years (525 males, 525 females) without a history of kidney stone disease were enrolled in the cross-sectional study. The study group was divided into 15 ranges according to age, each comprising 70 subjects. Urinary ionized calcium [Ca2+] was measured using a selective electrode while the onset of spontaneous crystallization was determined using a photometer and titrating with 40 mmol/L ammonium oxalate (Ox2−). The calculation of BRI value was based on the ratio of [Ca2+] to the required amount of ammonium oxalate added to 200 ml of urine to induce crystallization. The median BRI was 0.26 1/L and the values of the 5th and 95th percentiles were 0.06 1/L and 1.93 1/L, respectively. BRI correlated positively with body-area-related BRI (1/L × 1.73 m2) (R = 0.18; P < 0.05), whereas a negative correlation was found between BRI and body weight (1/L × kg) (R = −0.85; P < 0.05). Neither sex nor age differences were detected in BRI across studied children and adolescents. The values of Bonn Risk Index were constant during growth and there was a limited influence of age and sex on BRI in children over 3 years of age. The BRI may be valuable in the evaluation of pediatric patients at risk for kidney stones, particularly if the BRI from stone formers is demonstrated to be higher than in normal children. Introduction Urolithiasis is a frequently reported condition in children and is diagnosed even in neonates and infants [1, 2]. The disease may be the first sign of congenital and acquired metabolic disturbances, or the consequence of anatomic or genetic abnormalities [2]. The majority of kidney stones are composed of calcium oxalate and calcium phosphate [1]. Insight into this pathology is increasing and is focused on the pathogenesis of deposit formation in the urinary tract [3–5]. Investigation is being conducted into more effective methods which would enable the detection of risk factors for urolithiasis. One of the risk factors is an increase in the crystallization of calcium oxalate (CaOx) in urine. Kavanagh and Laube recently published a review of methods used to assess the crystallization of CaOx in urine [6]. During the last few years, activity products (AP) of crystallization have usually been assessed by means of the APCaOx index or as the relative supersaturation (RS) of urinary calcium oxalate (RSCaOx) using the computer program EQUIL [7, 8]. Laube et al. showed that the ratio of calcium ions [Ca2+] to the amount of ammonium oxalate added to 200 ml of urine to induce crystallization [designated here as (Ox2−)] may be, at the moment of spontaneous crystallization of CaOx, an indicator of the risk of CaOx crystal formation [9]. The authors made this determination with a direct urine collection, without an initial processing. This ratio is known as the Bonn Risk Index (BRI): BRI=[Ca2+]/(Ox2−) 1/L [9, 10]. Due to its characteristics, this index is an accurate indicator of the individual state of balance between the quantities of the most important promoters and inhibitors of the crystallization process within urine [9, 10]. Measuring BRI is simple, cost-effective, and the results are repeatable. In patients with calculi formation, CaOx assessed using BRI is significantly higher when compared to healthy subjects [9]. A lack of published studies concerning BRI in pediatric patients led us to conduct the study in children and adolescents. The purpose of this study was to define the BRI value in healthy subjects aged 3–18 years, in relation to age and sex. Materials and methods The study was performed on a group of 1,050 healthy Caucasian children and adolescents (group I) aged 3–18 years (mean ± SD: 10.51 ± 4.33), comprised of 525 boys and 525 girls. The study population was divided into 15 age groups, consisting of 70 children in each 1-year group (35 boys and 35 girls). The children reported no history of dietary restrictions. All participants met the criteria of the standard dietary energy and nutrient intakes recommended in Poland [11]. These children were free of infection at the time of the examination (serum C-reactive protein CRP <0.4 mg/dL, blood leukocyte count <10 × 109/L). Prior to inclusion, all participants were screened regarding serum levels of protein, albumin, calcium, phosphate, potassium, uric acid, creatinine and alkaline phosphatase as well as urine concentrations of citrate, oxalate, potassium, calcium and phosphate. Urinary dipstick testing (Bayer Diagnostics, Bridgend, UK) detecting nine parameters, including leukocytes, protein and blood, did not reveal any abnormalities. Children with a family history of kidney stones were excluded from the study. Subjects with diseases known to affect oxalate, calcium and phosphate metabolism and children treated with antibiotics were excluded. All children were screened using renal ultrasound examination to exclude urolithiasis (Toshiba SSH-140A apparatus; probe Convex 3.75 MHz). Participants and their legal guardians gave informed consent, and the study was approved by the Ethical Committee of the Medical University of Bialystok. Bonn Risk Index The Bonn Risk Index was assessed using the method of Laube [12]. Each child had a 24-h urine collection into sterile containers, without additional preserving substances, which was stored at 4°C. The testing was always performed twice using the same urine collection from each subject. Two consecutive urine samples (each 200 mL) were incubated immediately after collection, at a temperature of 37°C and the calcium ion concentration was measured using calcium ion-selective electrodes of type Rapilab 855 (Bayer, Germany) and titrated with ammonium oxalate solution (40 mmol/L) at a rate of 0.75 mL/min. The onset of spontaneous crystallization was detected using an Eppendorff photometer (filter 585 nm) with a decrease in light transmission to 98% of the initial value. Each analysis was repeated twice. The BRI is presented as [Ca2+] mmol/L /(Ox2−) mmol = 1/L. Calibration and quality assurance procedures, based on the calibration curves, were conducted every day. Statistical analysis was performed using the program Statistica 6.0 PL. Mann-Whitney test was used for the analysis of two non-parametric independent variables, with P < 0.05 considered statistically significant. Assessment of the rank of two independent variables was conducted using Spearman correlation, with P < 0.05 considered statistically significant. For the purpose of plotting the curve of spontaneous crystallization (an association between the number of calcium ions and the amount of added ammonium oxalate leading to the spontaneous crystallization), we used the computer program with the range of values as a scatterplot with the option of adding curves. Results The characteristics of the study group are presented in Fig. 1. The anthropometric traits of participants, based on weight and height measurements and body mass index (BMI), were within the normal range in each subgroup, relative to the Polish references described by other authors [13]. Fig. 1Body mass index (BMI) in studied girls and boys aged 3–18 years compared to age and sex-matched reference range Figure 2 shows the detailed results for the whole study group for BRI, defined as the ratio of [Ca2+] concentration in urine to the amount of added ammonium oxalate (Ox2−) necessary for the spontaneous crystallization of CaOx. BRI in healthy children ranged between 0.06 and 1.93 1/L. The values between the 5th and 95th percentiles are found between the two borderline arrows. This diagram also presents the minimum and maximum concentrations of [Ca2+] and the required amount of added oxalate (Ox2−). The concentration of [Ca2+] in urine ranged from 0.25 mmol/L (5th percentile) to 0.89 mmol/L (95th percentile) with a median of 0.42 mmol/L, and an amount of added (Ox2−) ranging from 0.46 mmol (5th percentile) to 3.53 mmol (95th percentile) with a median of 1.65 mmol. Fig. 2The spontaneous crystallization of CaOx in urine of 1,050 children aged 3.00–17.99 years. The x-axis shows amount of ammonium oxalate (Ox2−) in mmol necessary for the onset of spontaneous crystallization. The y-axis shows the concentration of calcium ions [Ca2+] before adding Ox2−. The extremes of the hyperbola represent minimum and maximum values. The minimum hyperbola is defined by the equation [Ca2+]=0.27/(Ox2−), and the maximum is [Ca2+]=1.67/(Ox2−). The BRI values for the 5th to the 95th percentile are found between the hyperbolas [Ca2+]=0.39/(Ox2−) and [Ca2+]=0.94/(Ox2−), median [Ca2+]=0.65/(Ox2−) Figure 3 presents the BRI values in children representing various age groups. We considering the subjects in two main subgroups, younger children aged 3–9 years, and older children and adolescents aged 10–18 years. In children 3–9 years of age, the lowest values of BRI were found in the youngest children, i.e. those 3 years old (median 0.18 1/L), whilst the highest values were found in 9-year-old children (median 0.34 1/L). However, statistical analyses did not show a difference between the values across the various age groups. The lowest median BRI value was 0.03 1/L in 9-year old children, whilst the highest median value was 2.48 1/L in 4-year-old children. The BRI values for younger children corresponding to the 5th percentile ranged from 0.03–0.08 1/L, and the 95th percentile from 1.38–2.27 1/L. Fig. 3Bonn Risk Index (BRI) in the whole studied group (left box) and in separate 1-year age groups for children aged 3–18 years In older children and adolescents aged 10–18 years, the lowest BRI values were found in the participants who were 12 years old (median 0.17 1/L), whilst the highest were in 17-year-old adolescents (median 0.31 1/L) (Fig. 3). However, statistical analysis did not show significant differences between the age subgroups. The minimum value of BRI was 0.02 1/L, and was found in 11-year-old children, whereas the maximum value was 3.1 1/L in children aged 10 and 17 years. The BRI values for older children corresponding to the 5th percentile ranged between 0.05 and 0.13 1/L, whilst the 95th percentile was from 1.69–2.66 1/L. The crystallization values of CaOx, based on BRI in healthy children aged 3–18, did not exceed 2.66 1/L. No significant differences were found in the BRI values between boys and girls in either age group. Table 1 presents the BRI and BRI related to the 1.73 m2 body surface area and body mass (kg). The median BRI in relation to body surface area (1/L × 1.73 m2) was 0.39 1/L × 1.73 m2, with 5th and 95th percentile medians of 0.09 and 3.01 1/L × 1.73 m2, respectively. However, the median BRI in relation to 1 kg of weight was 0.008 1/L × kg with corresponding 5th and 95th percentiles of 0.0015 and 0.06 1/L × kg, respectively. A weak positive correlation was found between BRI and BRI/1.73 m2 (R = 0.18, P < 0.05), and a negative correlation was found between BRI/1.73 m2 and BRI/kg (R = −0.86, P < 0.05). No differences were found between the values of BRI/1.73 m2 and BRI/kg in the various age groups or in relation to sex. Table 1Bonn Risk Index values in children aged 3–18 years in relation to body surface area and body weight MedianMinimumMaximum5th percentile95th percentileBRI (1/L)0.260.023.100.061.93BRI/1.73 m2(1/L × 1.73 m2)0.390.036.650.093.01BRI/kg (1/L × kg)0.0080.00060.150.00150.06 Discussion Urinary stones are considered a major health problem in society, both in adults and in children, due to their recurrent nature and the cost of treatment [14]. The pathogenic pathways leading to stone formation in kidneys have not been fully explained. During processes of calculus formation, a number of phenomena have been reported such as excess of crystallizing substances, nucleation, crystallization, aggregation and stone formation [15, 16]. However, it has not been explained why calculi do not form in all subjects, despite a large amount of urinary crystallization products. Over the past few years, there have been many attempts to define the risk factors leading to urinary stone formation on the basis of the ability to form oxalate crystals, a main component of calculi [7, 8, 17]. Tiselius et al. described the practical importance of assessing the activity of calcium oxalate ions in urine using the APCaOx index [7]. Stone formation has also been evaluated in terms of excess urinary calcium oxalate, using the computer program EQUIL [18, 19]. Laube et al. showed a strong correlation between the concentration of free calcium ions [Ca2+] in urine and the quantity of ammonium oxalate (Ox2−) added to invoke spontaneous urinary crystallization of CaOx [9]. The authors suggested that the BRI index in healthy subjects was significantly lower than in those forming urinary calculi. The study, based on a group of 72 healthy subjects, provided mean BRI values of 1.05 ± 1.038 1/L with a range of 0.06–4.89 1/L. In other studies involving 85 healthy adult subjects, the mean BRI values were similar, 0.89 ± 0.91 1/L. Higher BRI values have been described by Lewandowski et al. whose study group of 15 Caucasian adults demonstrated a mean BRI of 4.90 1/L, and 15 African adults had a mean BRI of 2.04 1/L [10]. The literature review did not reveal any larger or more comprehensive studies regarding BRI in children. Our findings in the large group of healthy children showed that the normal BRI value ranged from 0.06 1/L (5th percentile) to 1.93 1/L (95th percentile). The maximum values obtained in children were lower than those described by other authors in adult populations [6, 9, 20]. This may be due to the fact that the mechanism which inhibits crystal formation in children’s urine is more effective when compared to adults. Teller et al. observed that urinary glycosaminoglycans in children reduced the aggregation of crystals more effectively than in adults [21]. A similar age-related effect regarding stone formation has been observed by Bergsland et al., who showed that the urine of subjects aged less than 20 years produced a strong inhibitory effect on crystal formation [22]. Ricchiuti et al. investigated the process in boys aged less than 10 years and found increased activity of plasma protease inhibitor (inter-alpha inhibitor), blocking urinary crystallization during various stages [23]. Our results did not show a significant difference in BRI values between boys and girls. A minor difference was found in relation to age, but this was not significant. The BRI/1.73 m2 and BRI/kg behaved similarly. High correlations between BRI, BRI and body weight, and BRI and body surface area of studied subjects enable practical, more objective use of BRI and allow for comparisons between individuals. For this reason, these results are presented as median values with ranges for the 5th to 95th percentiles for the whole study group. The study has several limitations. So far, the use of our normative data appears to be limited to the Polish population. It is difficult to relate the results to other geographical regions or to different ethnic groups as there are no published data regarding BRI in children. Further comparative and prospective investigations are needed because a cross-sectional study is not able to reveal whether some children with a normal BRI value will be at risk for stone disease in the future. Moreover, some studied children may have had urolithiasis despite normal results of renal ultrasound. However, this study provided consistent normative data on BRI, due to both the large age representation and the stringent selection criteria. In summary, the Bonn Risk Index, expressed as the ratio of ionized calcium to the amount of ammonium oxalate necessary to initiate spontaneous CaOx urinary crystal formation, is lower in healthy children and adolescents aged 3–18 years than in studies of adults by other authors. The BRI during growth appears to be independent of age and sex. Thus, our results may contribute to the effective screening of kidney stone disease in pediatric subjects. We conclude that the BRI may be valuable in the evaluation of pediatric patients at risk for kidney stones, particularly if the BRI from stone formers is demonstrated to be higher than in normal children.

Crit_Care-6-3-137445

Clinical review: Complications and risk factors of peripheral arterial catheters used for haemodynamic monitoring in anaesthesia and intensive care medicine

In order to evaluate the complications and risk factors associated with peripheral arterial catheters used for haemodynamic monitoring, we reviewed the literature published from 1978 to 2001. We closely examined the three most commonly used arterial cannulation sites. The reviewed papers included a total of 19,617 radial, 3899 femoral and 1989 axillary artery catheterizations. Factors that contribute to higher complication rates were investigated. Major complications occurred in fewer than 1% of the cases, and rates were similar for the radial, femoral and axillary arteries. We conclude that arterial cannulation is a safe procedure. Introduction Indwelling arterial catheters are used routinely for continuous haemodynamic monitoring in the operating room during major surgery and in critically ill patients. Arterial cannulation provides easy and convenient access, allowing multiple blood samplings and blood gas analysis. Arterial cannulation has generally been found to be a safe procedure, with few serious complications [1,2,3,4,5,6]. Approximately 8 million and 2.5 million arterial catheters are placed yearly in the USA and Europe, respectively [7]. Despite the frequency with which arterial catheters are employed, there is little information on the impact of cannulation site on risk for complications [8]. We reviewed reports concerning arterial cannulation sites such as the radial, femoral, axillary, brachial, ulnar, dorsal pedis, tibial posterior and temporal arteries with regard to thrombotic, infectious and vascular complications. Material and methods The PubMed, Medline Express and Winspirs databases (publication years 1978-2001) were searched to identify pertinent articles. The keywords 'artery catheter', 'radial artery catheter', 'femoral artery catheter', 'axillary artery catheter', 'catheter complication', 'hemodynamic monitoring' and 'peripheral methods' were used. We reviewed the papers and identified further articles from the references of the papers found in the initial search. We reviewed all studies concerning the use of radial, femoral, axillary, brachial, ulnar, dorsal pedis, tibial posterior and temporal artery catheters for haemodynamic monitoring that dealt with complications and risk factors in adults. We excluded studies conducted in the paediatric patient population. We opted not to perform statistical analyses because the data selection of the published reports do not follow the same criteria, and most of them give no specific information regarding patient population, catheter material, cannulation technique and times. Therefore, comparison between studies and statistical analyses are unlikely to be valid. Results We identified 78 studies that met our criteria. Radial artery The radial artery is the most common site for arterial cannulation, and the studies that reported complication rates are listed in Table 1 [9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32]. The most common complication was temporary occlusion of the artery, the incidence of which ranged from 1.5% [17] to 35% [10] (mean 19.7%), the variability being due to the different methods employed and the degree of effort invested in detecting this complication. Table 1 Complications following radial artery cannulation Reference Cases Permanent ischaemic damage (n) Temporary occlusion (n) Sepsis (n) Local infection (n) Pseudoaneurysm (n) Haematoma (n) Bleeding (n) [9]* 80 1 9 0 0 - - - [9]* 34 1 10 0 2 - - - [10] 100 0 35 - - - - - [11] 148 0 27 - - - - - [12] 29 0 5 - - - - - [13] 62 0 16 0 0 - - - [14] 333 0 100 0 16 - 29 - [15] 40 0 2 - - - - - [16] 100 2 14 0 1 - 31 - [17] 197 0 3 3 4 2 - 1 [18] 178 0 7 1 3 - - 1 [5] 1699 0 360 0 1 - 206 - [19] 26 0 6 0 1 1 1 - [20] 100 0 14 - - - - - [21] 100 0 33 0 1 - 13 - [22] 118 0 18 0 0 - 21 - [23] 200 0 15 - - - 61 - [24] 25 0 3 0 0 - - - [25] 200 0 60 0 0 - 32 - [26] 193 0 50 0 4 - - - [27] 88 0 15 - - - - - [28] 2900 - - 4 12 5 - - [29] 38 0 3 - - - 5 - [30]* 45 0 4 - - - 10 - [30]* 44 0 11 - - - 9 - [31] 12,500 - - - - 6 - - [32] 40 0 11 - - - - - Mean incidence (%) 0.09 (4/4217) 19.7 (831/4217) 0.13 (8/6245) 0.72 (45/6245) 0.09 (14/15,623) 14.40 (418/2903) 0.53 (2/375) The total number of cannulations was 19,617. *One study examining two different populations. -, Not investigated. Generally, temporary occlusion of the artery has no serious sequelae. Permanent occlusion of the radial artery appears to be rare because it was reported in only four patients (mean incidence 0.09%). Another serious complication is pseudoaneurysm, which was reported in 14 patients (mean incidence 0.09%). Pseudoaneurysm poses a risk for infection, sepsis, rupture [33,34,35] and formation of an extracorporeal pseudoaneurysm [36]. Radial catheterization was associated with sepsis in eight patients (mean incidence 0.13%), whereas local infection at the cannulation site was identified in 45 patients (mean incidence 0.72%). Minor complications such as haematoma formation or bleeding at the puncture site are also shown in Table 1. Other complications include abscess, cellulitis, paralysis of the median nerve [37,38,39], suppurative thrombarteritis [40], air embolism [41], compartment syndrome and carpal tunnel syndrome [42,43,44]. Other rarely reported complications include catheter failure as result of manufacturing defect or incorrect use with resultant catheter replacement [45,46,47,48]. Femoral artery We identified 11 studies that used the femoral artery for haemodynamic monitoring, which are listed in Table 2 [17,18,19,49,50,51,52,53,54,55,56]. Temporary occlusion was reported in 10 patients (mean incidence 1.45%), and serious ischaemic complications requiring extremity amputation was reported in three patients (mean incidence 0.18%) [56]. Table 2 Complications following femoral artery cannulation Reference Cases Permanent ischaemic damage (n) Temporary occlusion (n) Sepsis (n) Local infection (n) Pseudoaneurysm (n) Haematoma (n) Bleeding (n) [49] 46 - - 0 - - - - [50] 50 0 1 0 - - 5 - [51] 85 0 0 0 - - 10 - [17] 113 0 0 0 0 - - 4 [52] 89 - - 0 0 - - 1 [53] 2100 - - 6 - 6 - - [18] 114 0 4 2 0 - - 0 [54] 42 0 1 0 1 - 3 - [55] 220 0 1 4 4 - 8 - [19] 64 0 3 1 0 - 2 - [56] 976 3 - - - - - - Mean incidence (%) 0.18 (3/1664) 1.45 (10/688) 0.44 (13/2923) 0.78 (5/642) 0.3 (6/2100) 6.1 (28/461) 1.58 (5/316) The total number of cannulations was 3899. -, Not investigated. Pseudoaneurysm formation occurred in six patients (mean incidence 0.3%), sepsis was observed in 13 patients (mean incidence 0.44%) and local infection was reported in five patients (mean incidence 0.78%). Bleeding (generally minor) was observed in five patients (mean incidence 1.58%), and haematoma formation was reported in 28 (mean incidence 6.1%). One patient developed an infected haematoma and needed blood transfusion [17] and another patient eventually died from massive retroperitoneal bleeding [57]. Axillary artery The axillary artery was cannulated in a total of 1989 reported cases. The complication rates at this site are summarized in Table 3 [54,55,58,59,60,61]. Serious complications included permanent ischaemic damage in two patients (mean incidence 0.20%), pseudoaneurysm formation in one patient (mean incidence 0.1%) and sepsis in five patients (mean incidence 0.51%). Paresthesia of the hand due to pressure on the brachial nerve plexus was also described [54,59]. Table 3 Complications following axillary artery cannulation Reference Cases Permanent ischaemic damage (n) Temporary occlusion (n) Sepsis (n) Local infection (n) Pseudoaneurysm (n) Haematoma (n) Bleeding (n) [58]* 1000 - - - - 1 - - [58]* 245 0 1 0 - - - 6 [59] 120 0 1 0 3 - 4 - [60] 435 2 9 3 7 - 4 3 [54] 28 0 - 0 1 - 3 - [61] 31 0 - 0 - - 1 1 [55] 130 0 0 2 5 - 5 - Mean incidence (%) 0.20 (2/989) 1.18 (11/930) 0.51 (5/989) 2.24 (16/713) 0.1 (1/1000) 2.28 (17/744) 1.41 (10/711) The total number of cannulations was 1989. * One study examining two different populations. -, Not investigated. Other sites for arterial cannulation Other less frequently reported cannulation sites are the brachial, dorsal pedis, ulnar, tibial posterior and temporal arteries. Only one serious complication was found in a study of 1000 patients [62] in which the brachial artery was used for invasive monitoring in ambulatory patients. This complication was an infected haematoma arising from a pseudoaneurysm. Another study that employed the brachial artery for arterial blood sampling in 6185 patients [63] also showed a small number of complications (incidence 0.2%), mainly paresthesias. The dorsal pedis artery has also been cannulated without any major complications [24,54]. However, because of its distance from the central circulation and associated difficulties in hypotensive patients it is generally avoided [64]. The ulnar artery has been used without serious complications, and in a direct comparison [29] this site exhibited similar complication rates to those encountered with the radial artery. The tibial posterior artery has been used in paediatric patients, without major complications [65,66]. However, there also exists a case report of limb amputation secondary to a tibial artery catheterization [67]. The temporal artery is an end artery of the external carotid artery, and is usually avoided for fear of serious complications. There is a report of three cases of cerebral embolization in paediatric patients [68]. Further studies Other studies that examined arterial catheter complications but that did not specifically attribute their occurrence to one specific site are listed in Table 4 [49,69,70,71]. One study [69] used mainly the radial (n = 210), ulnar (n = 158) and dorsalis pedis arteries (n = 82), whereas another [49] used mainly the femoral (n = 46), brachial (n = 9), axillary (n = 3) and radial arteries (n = 1). One serious complication was found at the femoral site, with compromised circulation to the leg [70]; eventually, thrombectomy was required. Another reported serious complication is air embolism caused by improper use of the flushing device [70]. Table 4 Studies examining arterial catheter complications without specifically attributing them to any specific site Reference Cases Air embolism (%) Permanent occlusion (%) Temporary occlusion (%) Haematoma (%) Bleeding (%) Sepsis (%) Abscess (%) [69] 450 0 0 18.4 - 8.7 0 - [49] 59 0 1.7 6.8 5.1 1.7 0 - [70] 506 0.2 0 - - 0 0 - [71]* 1556 0 0 3.4 0.2 1.5 0.06 0.3 [71]� 565 0 0 4.6 0.3 2.3 0 0.5 *Medical intensive care unit (radial artery 52%, femoral artery 45%). �Surgical intensive care unit (radial artery 78%, femoral artery 11%). -, Not investigated. A recent study [71] examined complication rates, comparing medical with surgical intensive care units. It also directly compared radial and femoral artery catheters. There were no significant differences in terms of sepsis and local infection rates between the femoral and radial arteries, and no significant difference between the medical and surgical intensive care units. Another nine studies specifically looked for septic complications associated with arterial catheter placement, and included a total number of 1681 patients (Table 5) [52,54,72,73,74,75,76,77,78]. No septic complications were identified in seven of those studies [52,72,73,74,75,76,77], and nine cases of sepsis were reported in the remaining two studies [54,78]. All of the cases of sepsis were identified after the line had been in place for longer than 4 days. The studies in which sepsis was reported predominately used the radial artery. Table 5 Incidence of sepsis following arterial cannulation Reference Cases Sepsis (%) [54] 130 4 [72] 172 0 [52] 186 0 [73] 19 0 [74] 155 0 [75] 643 0 [76] 230 0 [77] 75 0 [78] 71 5.5 Risk factors A further objective of the present review was to consider risk factors that might have impacted on complication rates. Thrombotic complications Thrombus formation and occlusion of the artery appear to be caused by changes in the integrity of the vessel wall induced by the presence of the catheter [79]. Recannulation of the occluded artery generally occurs, but this may take up to 75 days [80]. The incidence of thrombus appears to be related to the degree to which the catheter fills the arterial lumen [3]. It has been shown that the incidence of radial obstruction increases linearly with the ratio of outer diameter to vessel lumen diameter [81]. This might explain the higher incidence of radial occlusion in female patients, who generally have vessels of smaller diameter [5,10,14,22,25]. This has led to a preference for 20-gauge catheters for radial artery cannulation [11,14]. Teflon catheters also appear to be associated with a lower incidence of occlusion [11], but in another study [5] this could not be confirmed. Patients with a low cardiac output have a higher incidence of radial artery occlusion [5,14], but no difference was found among patients with episodes of hypotension [21]. Multiple punctures for catheter insertion was shown to be a risk factor for complications in one study [14], but this could not be confirmed in another [5]. A higher incidence of occlusion is found in the presence of a haematoma [5,14,21]. Patients who are treated with aspirin or low-dose heparin have fewer vessel occlusions [13,14]. The age of adult patients does not appear to be a risk factor [14]. An important factor with regard to arterial occlusion is the duration of cannulation. A higher incidence of occlusion was reported when the cannula was left in place longer than 48 or 72 hours [5,9,14,16,25]. Infectious complications Important factors in decreasing the incidence of catheter-related infections are aseptic insertion technique and adequate disinfection of the insertion site [82]. The risk for infectious complications and sepsis increases with the duration of cannulation, especially if the catheter is left in place for longer than 96 hours [78,83]. The catheter material also appears to be of importance, because Teflon has a greater resistance to Escherichia coli and Pseudomonas aeruginosa [84] and a lower affinity for stapylococci [85]. Prophylactic treatment with antibiotics does not appear to reduce the risk for catheter-related infections [78,83], but antibiotic bonding of the catheter may confer a degree of protection [86,87]. Arterial catheter placement via the cut-down technique is associated with a higher risk for infection [83]. Furthermore, sepsis occurs more frequently in the presence of local site inflammation [83]. Infection may also occur through contamination by caregivers of the monitoring system, the flushing device and the infusion system [88,89,90]. Discussion The use of peripheral arterial catheters for haemodynamic monitoring is widespread. The most frequently used site is the radial artery because of its well documented low complication rates and easy access [9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32]. In the present review we report on 19,617 radial artery cannulations, and the main complication was temporary occlusion of the artery (incidence 19.70%). This complication rate is close to the finding of Wilkins [91], who reviewed reports of catheterization of the radial artery with regard to ischaemic complications published between 1969 and1983. He reported a temporary occlusion rate of 23%. In an extensive study of 1699 patients, Slogoff and coworkers [5] reported a 21.2% rate of temporary occlusion. Although temporary occlusion is reported quite frequently, serious ischaemic damage was reported in only two studies [9,16], with a mean complication rate of 0.09%. Nevertheless, there have been reports of serious ischaemic damage after radial artery cannulation that led to necrosis, and amputation of fingers or the whole hand [92,93,94,95,96,97,98,99]. Other major complications such as pseudoaneurysm and sepsis were reported to occur in a mean of 0.09% and 0.13% of cases, respectively. The second most cannulated artery for haemodynamic monitoring is the femoral artery, and we reviewed 3899 cases. Temporary occlusion was reported in only 1.18% of the reviewed cases. The incidence of this complication in the femoral artery is much lower than that in the radial artery. We attributed this to the larger vessel diameter of the femoral artery. Serious ischaemic complications were reported in only one study [56], and the mean complication rate was 0.18%. The incidence rates for other major complications such as pseudoaneurysm and sepsis (0.3% and 0.44%, respectively) were similar to our findings in the radial artery. Some authors caution against use of the femoral artery for cannulation for fear of higher sepsis rates because of the close proximity to the perianal region [52]. We are unable to corroborate this on the basis of the reviewed literature, and recent studies that directly compared the radial and femoral arteries with regard to septic complications identified similar rates of sepsis [53,71]. In particular, in an extensive study of 4932 patients, Frezza and Mezghebe [71] were unable to identify a difference in complication rates between the radial and femoral arteries. In that study the femoral artery was actually the preferred site in the medical intensive care unit. Other authors prefer the femoral artery over the radial artery because the femoral artery is usually palpable even in hypotensive patients and may be the only accessible route for haemodynamic monitoring [64,100,101,102,103]. The blood pressure curve that is obtained from this larger artery is generally more accurate and gives a closer estimation of the aortic blood pressure [64,100,101,102,103]. The third most cannulated artery in the present review of the literature was the axillary artery, with almost 2000 reported cases. Some authors prefer not to cannulate the axillary artery because of its close location to the carotid artery and because of fear of embolism to the brain [101]. On the basis of the data summarized here we cannot confirm this, and no case of embolism to the brain was reported. Sometimes the axillary artery is avoided because of the more difficult approach required, although (particularly in anaesthesia) it is a well known route because it requires the same approach as that for axillary nerve block [3]. The major complications encountered with the axillary artery were similar to those for the radial and femoral arteries, and we conclude that it is a safe route for arterial cannulation. Other reviewed arteries employed for catheterization, such as the brachial, dorsal pedis, ulnar, tibial and temporal arteries, have been used without serious complications, but published reports of their use are limited. Conclusion Incidence rates for major complications such as permanent ischaemic damage, sepsis and pseudoaneurysm formation are low and similar for the radial, femoral and axillary arteries. They occur in fewer than 1% of cases. On the basis of the present systematic review we can conclude that serious complications of the radial, femoral and axillary artery are rare, and that arterial cannulation is a relatively safe procedure. Nevertheless, one must be aware of the possible serious complications that might arise. Therefore, it is important to use an arterial catheter only when it is indicated and to select carefully the best cannulation site individually for each patient, because the different cannulation sites have their specific advantages and attendant risks. Competing interests UJP is President of Pulsion Medical Systems AG, and AP is on the medical advisory board of Pulsion Medical Systems AG, whose PiCCO product uses a femoral arterial line.

J_Gastrointest_Surg-3-1-1852380

Management of Massive Arterial Hemorrhage After Pancreatobiliary Surgery: Does Embolotherapy Contribute to Successful Outcome?

Massive arterial hemorrhage is, although unusual, a life-threatening complication of major pancreatobiliary surgery. Records of 351 patients who underwent major surgery for malignant pancreatobiliary disease were reviewed in this series. Thirteen patients (3.7%) experienced massive hemorrhage after surgery. Complete hemostasis by transcatheter arterial embolization (TAE) or re-laparotomy was achieved in five patients and one patient, respectively. However, 7 of 13 cases ended in fatality, which is a 54% mortality rate. Among six survivors, one underwent selective TAE for a pseudoaneurysm of the right hepatic artery (RHA). Three patients underwent TAE proximal to the proper hepatic artery (PHA): hepatic inflow was maintained by successful TAE of the gastroduodenal artery in two and via a well-developed subphrenic artery in one. One patient had TAE of the celiac axis for a pseudoaneurysm of the splenic artery (SPA), and hepatic inflow was maintained by the arcades around the pancreatic head. One patient who experienced a pseudoaneurysm of the RHA after left hemihepatectomy successfully underwent re-laparotomy, ligation of RHA, and creation of an ileocolic arterioportal shunt. In contrast, four of seven patients with fatal outcomes experienced hepatic infarction following TAE proximal to the PHA or injury of the common hepatic artery during angiography. One patient who underwent a major hepatectomy for hilar bile duct cancer had a recurrent hemorrhage after TAE of the gastroduodenal artery and experienced hepatic failure. In the two patients with a pseudoaneurysm of the SPA or the superior mesenteric artery, an emergency re-laparotomy was required to obtain hemostasis because of worsening clinical status. Selective TAE distal to PHA or in the SPA is usually successful. TAE proximal to PHA must be restricted to cases where collateral hepatic blood flow exists. Otherwise or for a pseudoaneurysm of the superior mesenteric artery, endovascular stenting, temporary creation of an ileocolic arterioportal shunt, or vascular reconstruction by re-laparotomy is an alternative. Recent advances in surgical technique and postoperative management of major pancreatobiliary surgery have reduced the morbidity and mortality. However, the problem of life-threatening postoperative arterial hemorrhage caused by pseudoaneurysm rupture, although uncommon, has received little attention.1–6 Immediate arteriography to identify the site of bleeding and subsequent treatment by radiological intervention have been the first approach for the last decade.7–9 The efficacy of transcatheter arterial embolization (TAE) is well established.8–18 Nevertheless, occlusion by TAE sometimes causes distal end-organ damage, even though bleeding has been controlled.9–11 Indeed, complete interruption of hepatic inflow may lead to fatal hepatic necrosis, but risk factors for whole liver infarction have yet to be identified. Knowledge of which sites are associated with a high risk of hepatic necrosis would enable surgeons to choose an alternative to TAE, such as vascular reconstruction. Recent reports have documented an advantage of endovascular stenting over TAE.19–22 We reviewed the records of patients who experienced massive arterial hemorrhage after pancreatobiliary surgery to help generate guidelines for the management of postoperative pseudoaneurysm rupture. Patients and Methods From January 1993 to December 2005, 351 patients underwent major surgery for malignant pancreatobiliary disease in the Department of Gastroenterological Surgery, Yokohama City University. The cases were pancreatic cancer (n = 139), distal bile duct cancer (n = 58), ampullary cancer (n = 46), hilar bile duct cancer (n = 47), advanced gallbladder cancer (n = 42), and intrahepatic cholangiocarcinoma (n = 19). Procedures included pylorus-preserving pancreatoduodenectomy (PPPD) in 113, conventional pancreatoduodenectomy (PD) in 84, distal pancreatectomy (DP) in 35, segmental resection of the pancreas in 4, total pancreatectomy in 6, bile duct resection with partial hepatectomy in 98, and PD combined with partial hepatectomy (HPD) in 11. Patients undergoing isolated bile duct resection or partial hepatectomy without bile duct resection were excluded. Except for the patients who underwent DP, all patients received biliary tract reconstruction and skeletonization of the hepatic arteries to complete lymphadenectomy within the hepatoduodenal ligament. Pancreatojejunostomy was created as an end-to-side, duct-to-mucosa anastomosis with a stenting tube, or via the pancreatic duct insertion technique with total tube drainage when the pancreatic duct was smaller than 3 mm. Hepaticojejunostomy was created end-to-side in a single layer. Massive postoperative arterial hemorrhage was defined as bleeding requiring a transfusion of 2 or more units of packed red blood cells, an invasive intervention such as laparotomy or TAE and monitoring in the surgical intensive care unit within 24 h of the onset of hemorrhage. The medical records of these patients were analyzed retrospectively. Results Of the 351 patients who underwent a major procedure, 13 (3.7%) presented with massive postoperative arterial hemorrhage (10 men and 3 women; average age, 66 ± 10 years). The demographic and clinical characteristics are summarized in Table 1. Table 1Baseline Characteristics of PatientsCaseDiseaseOrigin of bleedSurgeryInterval (days)aSite of TAERe-laparotomyCause of bleedOutcome1Ampullary cancerRHAPPPD10RHANoPancreatic leakAlive2Gallbladder cancer RHA and RHAHPD8 and 12RHA and PHANoMinor injuryHepatic failure died3Hilar bile duct cancerRHAExtended left hepatectomy9NoneNoUnsuccessful reconstructionCancer recurrence, died4Gallbladder cancerGDAHPD10GDANoPancreatic leakAlive5Distal bile duct cancerGDAPD11GDAYesMinor injuryAlive6Distal bile duct cancerCHAPPPD24CHANoPancreatic leakAlive7Distal bile duct cancerPHAPD7NoneNoPancreatic leakSepsis, died8Hilar bile duct cancerGDARight hepatectomy7CHAYesMinor injuryHepatic failure died9Gallbladder cancerMHA and GDARight hepatectomy13 and 27MHA and CHANoPancreatic leakHepatic failure died10Gallbladder cancerGDARight hepatectomy7GDAYesPancreatic leakHepatic failure died11Pancreatic cancerSPADP17CAYesPancreatic leakAlive12Intrahepatic cholangiocarcinomaSPAHPD9NoneYesMinor injuryHepatic failure died13Pancreatic cancerSMAPPPD34NoneYesPancreatic leakMOF, diedTAE Transcatheter arterial embolization, RHA right hepatic artery, PHA proper hepatic artery, GDA gastroduodenal artery, CHA common hepatic artery, MHA middle hepatic artery, SPA splenic artery, CA celiac axis, SMA superior mesenteric artery, PPPD pylorus preserving pancreatoduodenectomy, HPD pancreatoduodenectomy combined with partial hepatectomy, DP distal pancreatectomy, PD pancreatoduodenectomy, MOF multiple organ failureaDays from surgery to hemorrhage After restoration of hemodynamic stability by volume loading, 10 of 13 patients underwent emergency pan-abdominal angiography visualizing the celiac axis (CA) and superior mesenteric artery (SMA) by standard Seldinger technique. The other three patients required emergency laparotomy without angiography because their clinical status was deteriorating. Of the 10 patients who underwent angiography, an arterial pseudoaneurysm was detected in 7: right hepatic artery (RHA) in two, gastroduodenal artery (GDA) in three, common hepatic artery (CHA) in one, and splenic artery (SPA) in one. Three patients had extravasation in the area of the middle hepatic artery (MHA), proper hepatic artery (PHA), or GDA without a clear source of bleeding (see Table 1). TAE was attempted in nine patients using various coil occlusion devices. Table 1 details the origin of bleeding and the sites of TAE. Complete hemostasis was achieved by TAE in five patients, but hemostasis was only temporary in four: two required a second TAE and two required laparotomy to control rebleeding. Of the three patients with a pseudoaneurysm distal to the PHA, one received selective TAE of the RHA with a successful outcome (patient 1). A second patient had recurrent hemorrhage after TAE of the RHA and second TAE of the PHA was required. Unfortunately, the cluster of coils compressed the portal vein and portal flow was disrupted after the second TAE. This patient died of hepatic failure secondary to hepatic hypoperfusion (patient 2). The third patient had undergone extended left hepatectomy with resection of the RHA and had bleeding from the site of the RHA reconstruction (patient 3). If TAE had been performed at the RHA, inflow to the small remnant liver would be completely interrupted, and fatal hepatic failure would have been the most likely outcome. Instead, we re-operated, ligated the RHA, and created an ileocolic arterioportal shunt to supply the hepatic remnant. The patient survived without hepatic failure. However, portal hypertension developed because this shunt remained patent 6 months after reoperation, and the patient died of spinal metastasis 11 months after the initial operation. In seven patients, the pseudoaneurysm was proximal to the PHA and involved the GDA or CHA. Two patients successfully underwent selective TAE of the GDA for a pseudoaneurysm that originated from the ligated GDA. One patient had a replaced RHA from the CA, which contributed to a favorable outcome (patient 4, Fig. 1). One patient underwent TAE of the CHA uneventfully because the hepatic inflow was narrowly maintained by the left subphrenic artery (patient 6, Fig. 2). In contrast, the patient who had extravasation of the PHA without a discrete source had intimal injury and occlusion of the CHA during angiography. The patient experienced a liver abscess that was difficult to treat and died of sepsis and cancer recurrence during a long hospital stay (patient 7). Of the three patients who underwent right hepatectomy, two had complete disruption of the inflow to the hepatic remnant by TAE of the CHA, leading to fatal hepatic failure (patients 8 and 9), and one underwent unsuccessful TAE of the GDA and required a laparotomy. Vascular ligation at the pancreatic head was successful in achieving hemostasis, but hepatic failure secondary to hemorrhagic shock proved fatal 59 days after the initial surgery (patient 10). Figure 1Case 3. A 74-year-old woman with advanced gallbladder cancer presented with massive hemorrhage 10 days after pancreatoduodenectomy combined with partial hepatectomy of the segments (Couinaud segments) IVb and V. Angiogram of the common hepatic artery [the right hepatic artery (RHA) replaced from the celiac axis (CA)] showed a pseudoaneurysm (arrow) originating from the stump of the gastroduodenal artery (GDA). (A) Complete hemostasis was obtained using transcatheter arterial embolization of this stump. The patient survived without hepatic failure because the hepatic inflow was maintained by the replaced RHA (B). LHA Left hepatic artery, SPA splenic artery.Figure 2Case 4. A 69-year-old man who had undergone a pylorus preserving pancreatoduodenectomy for distal bile duct cancer had massive hemorrhage 24 days after surgery. Angiogram of the celiac axis (CA) showed a pseudoaneurysm (arrow) originating from the common hepatic artery (CHA) (A) Complete hemostasis was obtained using transcatheter arterial embolization proximally and distally to the origin of the pseudoaneurysm, but the proper hepatic artery was occluded. The hepatic arterial inflow was narrowly maintained via the left subphrenic artery (B). The patient had an uneventful course. RHA Right hepatic artery, LHA left hepatic artery, SPA splenic artery. Of the two patients with a pseudoaneurysm originating from the SPA, one underwent TAE of the CA with a favorable outcome because the pancreatoduodenal arcades around the pancreatic head maintained hepatic inflow after DP (patient 11, Fig. 3). The other patient underwent an emergency laparotomy. Complete hemostasis was achieved by ligation, but hemorrhagic shock resulted in subsequent hepatic failure (patient 12). Figure 3Case 8. A 77-year-old man who had undergone a distal pancreatectomy for pancreatic tail cancer had massive hemorrhage 17 days after surgery. Angiogram of the celiac axis (CA) showed a pseudoaneurysm (arrow) originating from the splenic artery (A). Hemostasis was obtained using transcatheter arterial embolization (TAE) of the CA. The common hepatic artery (CHA) was occluded by TAE but the hepatic inflow was maintained via the arcades of the pancreatic head from the superior mesenteric artery (B). The patient had a favorable course. RHA Right hepatic artery, LHA left hepatic artery, GDA gastroduodenal artery. One patient with a pseudoaneurysm of the SMA underwent surgical resection with vascular reconstruction. However, the patient died of recurrent pancreatic cancer and multiple organ failure during the hospital stay (patient 13). Seven of 13 patients died in the hospital; thus, the mortality rate for massive arterial hemorrhage was 54%, and the overall mortality rate was 2.0%. Case reports of some patients are illustrated in Figs. 1, 2, and 3. Discussion Rupture of a pseudoaneurysm, although uncommon, can cause life-threatening hemorrhage even comparatively late after pancreatobiliary surgery.1–6 This morbidity rate for pancreatectomy was reported to range from 2.0% to 4.6%.4–6,11,12 A pseudoaneurysm is a pulsatile hematoma surrounded by fibrous tissue that communicates with the artery via a disruption of the arterial wall. It can rupture into the peritoneal cavity, the gastrointestinal tract, or biliopancreatic ducts through a point of weakness, most commonly the anastomotic site. The Japanese Multi-institutional Study of 1,066 patients who underwent PPPD reported that the incidence of intra-abdominal hemorrhage was 3.5% and that of upper gastrointestinal hemorrhage was 3.2%.23 The etiology of pseudoaneurysm formation has yet to be clearly delineated. It is believed to be most commonly due to pancreatic fistula or anastomotic dehiscence.1–4,10–12 However, pseudoaneurysm can develop far from the pancreatic cut surface, and there is no evidence of a pancreatic leak in some cases. It has been suggested that skeletonization of the visceral arteries may result in iatrogenic vascular injury (e.g., secondary to diathermy).10,14 In three patients in our series, the etiology was thought to be a minor vascular injury that occurred during dissection (see Table 1). Previous reports found that the patients with massive arterial bleeding frequently had septic complications.3,4,17 Therefore, both arterial injury and infection can contribute to massive arterial hemorrhage. Whether or not preoperative obstructive jaundice is an etiologic fact remains controversial.3,24 Some clinical studies found that a preliminary warning bleeding (sentinel bleeding) precedes major hemorrhage.1,2,9–13 This sentinel bleeding probably indicates local infection and an anastomotic leak.2 So, recognition of a sentinel bleeding and prompt intervention can be life-saving. Angiography is necessary to identify the site of bleeding, and TAE is the treatment of choice to control massive bleeding and achieve hemodynamic stabilization. When performing TAE for a pseudoaneurysm, microcoils must be placed both proximally and distally to the origin, not within the pseudoaneurysm itself.11,25 When the origin of the pseudoaneurysm is distal to the PHA, left hepatic artery, RHA, or MHA, TAE should be highly selective to preserve the other branch to the liver. Selective TAE without devascularization of the other lobe provides optimal protection of the liver parenchyma (patient 1).15 Nevertheless, superselective TAE frequently is not possible. Inadvertent occlusion of PHA, even when bleeding is controlled, may cause fatal hepatic infarction (patient 2).10 Therefore, the success of superselective TAE distal to the PHA depends on the expertise of the radiologist. Superselective TAE is difficult or even impossible for a pseudoaneurysm proximal to the PHA.24 When the GDA is ligated close to its divergence from the CHA, you must embolize the CHA to obtain hemostasis. The proximal and distal control of a bleeding pseudoaneurysm of the PHA or CHA usually results in complete occlusion of hepatic arterial flow. Fortunately, we successfully performed TAE of the GDA in patient 4 because the patient had the replaced RHA from CA, and it was unaffected by TAE. Even if the CHA is occluded, the prognosis is good when hepatic inflow can be maintained by a replaced hepatic artery,6,13,14 but success is not uniform. In another case, hepatic inflow was narrowly maintained by the left subphrenic artery after TAE of the CHA (patient 6). A similar case has been described previously.16 The liver can tolerate considerable arterial embolization because of its multiple collateral pathways, mainly via the subphrenic arteries. Unless the surgical procedure includes mobilization of both lobes of the liver, as long as the subphrenic arteries are well developed, TAE proximal to the PHA should lead to a successful outcome. However, TAE proximal to the PHA usually risks occlusion of the CHA with the attendant risk of necrosis and liver failure.9,17 The authors believe that the presence of a few collateral pathways might make liver abscesses difficult to treat and may be associated with high morbidity and mortality rates (patient 7).10 Complete interruption of arterial inflow to the remaining liver after major hepatectomy usually causes imminent hepatic failure (patients 8 and 9). Emergency laparotomy and vascular reconstruction are the most certain treatments for this type of bleeding, but preoperative angiography is highly recommended to identify the bleeding site, although surgery should not be excessively delayed. Endovascular stenting is another option in reestablishing the continuity of the bleeding artery, such as GDA, PHA, or CHA.19–22 A covered stent makes it possible to arrest the bleeding, while preserving patency. Potential disadvantages include longer duration to obtain hemostasis than TAE, technical difficulties in negotiating tortuous arteries, and the risk of arterial rupture due to low flexibility and fragile vascular walls. In the case of a stented hepatic artery, the progressive occlusion of the stent due to intimal hyperplasia would not influence outcome because this process is gradual and allows for the formation of collateral pathways.21 Therefore, a stent-graft delivered into a bleeding artery proximal to PHA may be the treatment of choice. When this procedure is technically difficult, vascular reconstruction should be performed after temporary occlusion by interventional radiology. When vascular reconstruction is difficult and ligation is required to obtain hemostasis, creation of an ileocolic arterioportal shunt is another option,26 which we performed successfully in patient 3. Thus, the indications for TAE proximal to the PHA are limited to cases where the replaced hepatic artery exists or the subphrenic arteries are well developed. Pseudoaneurysm originating from the SPA is well described as a complication of pancreatitis,27,28 but this is an unusual source of bleeding after pancreatobiliary surgery. Selective TAE is indicated because ischemia of the spleen is rare. However, when the origin of the pseudoaneurysm is close to the CA, selective TAE proximal to the SPA is as difficult as the case of bleeding from the GDA. Therefore, when the SPA is divided in surgery, it may be better to leave the proximal part of the ligation site in some degree. TAE of the CA is contraindicated as it will completely interrupt hepatic arterial flow in most cases (patient 11). When the origin of a pseudoaneurysm is the SMA, it may be extremely difficult or even impossible to preserve mesenteric arterial flow. Hence, resection of the pseudoaneurysm with vascular reconstruction is indicated (patient 11). Endovascular stenting may be an alternative deserving further study. Hemorrhagic shock is a potentially fatal complication that may result in hepatic failure (patient 10). Rebleeding after TAE is a poor prognostic factor. As rebleeding is often due to a pancreatic leak or intraperitoneal septic condition, pancreatic drainage and loculated fluid collections are mandatory.6 The management of anastomotic dehiscence after pancreatectomy, whether a completion pancreatectomy is necessary or not, is controversial.1,6,29–33 When management of the dehiscence is difficult or when severe organ failure is present, completion pancreatectomy is probably necessary.1,6,29–31 Analysis of our experience and literature review suggests that management of a pseudoaneurysm must be individualized according to the site of bleeding (Fig. 4). We have developed a protocol for managing massive postoperative arterial hemorrhage and are evaluating it postoperatively in a prospective fashion. We hope to report the value of this approach in the future. Figure 4Scheme for an approach to the management of a pseudoaneurysm according to the site of bleeding. PHA Proper hepatic artery, RHA right hepatic artery, MHA middle hepatic artery, LHA left hepatic artery, SPA splenic artery, GDA gastroduodenal artery, CHA common hepatic artery, CA celiac axis, SMA superior mesenteric artery, TAE transcatheter arterial embolization.

Eur_Spine_J-2-2-1602189

Giant cervicothoracic extradural arachnoid cyst: case report

The pathogenesis, etiology, and treatment of the spinal arachnoid cyst have not been well established because of its rarity. A 57-year-old male was presented with spastic quadriparesis predominantly on the left side. His radiological examination showed widening of the cervical spinal canal and left neural foramina due to a cerebrospinal fluid - filled extradural cyst that extended from C2 to T2 level. The cyst was located left anterolaterally, compressing the spinal cord. Through a C4–T2 laminotomy, the cyst was excised totally and the dural defect was repaired. Several features of the reported case, such as cyst size, location, and clinical features make it extremely unusual. The case is discussed in light of the relevant literature. Introduction Various types of cystic lesions are confronted in the spinal canal and are classified in respect of their relationship to the adjacent structures and nature of the cyst content. Amongst them, cerebrospinal fluid (CSF)-filled meningeal cysts constitute for the majority of the cases [11, 18] and are called as “arachnoid cysts,” “diverticula,” [5, 9] or “pouches” [9]. They may be located in the intradural, extradural, or perineural spaces. Depending on the location, size, and mechanism of origin, the clinical course varies between asymptomatic, incidentally-diagnosed cases [17] to severe myelopathy [11]. When they communicate freely with subarachnoid space, fluctuating symptoms related to changes in CSF pressure may occur [8, 9]. Although arachnoid cysts can be observed anywhere along the length of the spinal canal, middle, and lower thoracic regions constitute for the most frequently involved areas. When progressive neurological findings exist, surgical treatment is warranted. Surgical techniques include excision, fenestration, or placement of a cysto - subarachnoid shunt [6, 11], as well as some less invasive techniques that have been proposed recently [16]. We present a case with a giant cervicothoracic extradural arachnoid cyst and its surgical management. Several features of the cyst, such as its size, location, and clinical features make the case extremely unusual. Case Report A 57-year-old male was referred with symptoms of difficulty in walking, weakness, numbness, and thinning of arms and legs, and impotency, which had all begun two years before. The patient had suffered a serious traffic accident when he was 10 years old, and has had a contracture of his right hand since then. Neurological examination revealed spastic quadriparesis predominantly on the left side (left upper extremity proximally had 2/5 muscle strength, distally had 1/5, and left lower extremity globally had 4/5), left-sided atrophy, hypoesthesia and hypoalgesia in C5 - C8 dermatomes, hypoactivity of deep tendon reflexes at the upper extremity, hyperactivity in patellar reflex, and Babinski sign. Muscle tonus of bilateral lower extremities was increased. The anal sphincter tonus was weak, and the patient reported an intermittent sphincter control defect. Cervical radiographs showed widening of the left neural foramina and the spinal canal. Magnetic resonance imaging (MRI) revealed an extradural cystic mass extending from C2 to T2 vertebral bodies, located left anterolaterally, compressing the spinal cord (Figs. 1a, b, c and 2). It had the same CSF intensity on T1 and T2-weighted images and showed no contrast enhancement. The lesion had caused a concavity in the related vertebral bodies and widening of the neural foramina on the left side. Electrophysiological tests reported chronic severe neurogenic involvement and a delay in the somato - sensorial evoked potential responses in C6 - T1 innervated muscles. Fig. 1T2-weighted median (a), slightly left paramedian (b), and left paramedian (c) sagittal magnetic resonance images of the cervical region shows a cystic mass extending from C2 to T2 vertebral bodies, located left anterolaterally, enlarging left neural foramina (arrows). (Arrowheads spinal cord)Fig. 2Axial magnetic resonance image taken from T1 level shows the CSF-filled extradural cyst, located left anterolaterally, compressing the spinal cord (SC spinal cord) A C4 – T2 laminotomy was performed. An extradural cystic lesion, full of CSF and having a thin and transparent membrane, was observed compressing the dural sac. The lesion had caused widening of C5 – C7 foramina and there was an active flow of CSF to the cyst through the defects located at the exiting points of these three roots. The cyst wall was excised and the dural defects were repaired using muscle pieces and sutures. Despite some degree of the spinal cord and dural sac expansion, the epidural and subarachnoid spaces left enlarged, since the canal was widened. The epidural space was left unfilled for the fear of compression. The samples taken from the cyst wall were reported as fibrohyalinized meningeal tissues. Postoperative course was uneventful. Although the spasticity gradually improved during two years of follow-up, muscle strength and other neurological findings did not change. Discussion Spinal arachnoid cyst is a rare entity. There are various classifications based on etiology, histopathology, or localization of the cysts. Many authors classify spinal meningeal cysts as both intradural and extradural. Extradural cysts include arachnoid cysts, synovial cysts, ganglias, cysts of ligamentum flavum, and cysts originating from the intervertebral discs. Intradural cysts include arachnoid cysts, enterogen (endodermal, neuroenteric) cysts, and ependymal cysts [3, 6, 7, 10, 12 – 15, 18]. The pathogenesis of spinal arachnoid cyst is not clear and probably not homogenous. Congenital [13], traumatic [1], or inflammatory [15] mechanisms were proposed. It was documented that congenital asymptomatic cysts could enlarge due to trauma and become symptomatic [3]. The size of arachnoid cysts may vary in relation to underlying mechanisms. For example, the arachnoid cysts related to arachnoiditis are usually localized and small [6, 18]. Pulsatile CSF dynamics [9], osmotic gradient between the subarachnoid space and cyst [7], and the valve - like mechanism between the cyst and subarachnoid space [10] may play an important role in the enlargement of spinal extradural arachnoid cysts. Doita et al. [4] demonstrated that increased intra abdominal and intrathoracic pressure may lead to size changes of the spinal arachnoid cyst. A review demonstrated that arachnoid cysts involving the cervical region are extremely rare, that they never exceed three vertebral bodies in length, and that none extended into the thoracic region [6], except the case reported by Safriel et al. [14]. That case was an intradural arachnoid cyst situated anteriorly to the spinal canal extending between C5 and T3 levels. The presented cyst was extradural, located at the cervicothoracic region, anterolateral to the spinal cord, and involved eight vertebral bodies extending from C2 to T2 level. To the authors’ knowledge, there is no extradural arachnoid cyst reported in the literature matching that size. The patient had no disposing factor other than trauma. He had suffered a serious traffic accident when he was a child and consequently had a contracture in his left upper extremity. Although there was an excessively long duration (45 years) between trauma and appearance of new symptoms attributable to the cyst, the concordance of the former neurological deficit and location of the cyst convinced us that the etiology of the cyst might have been the childhood trauma. During the surgery, we did not observe any laminar signs that indicated a sustained compression fracture on the posterior elements. Thus, stretching and partial avulsion of the roots may have caused arachnoid and dural tearing during the childhood trauma, causing subsequent collection of the CSF at the epidural space, communicating freely both subarachnoid space and dural sleeves of the roots. Indeed, the occurrence of multiple dural defects, through which the CSF exits, is very rare. This point makes an extra implication on the case and further supports the avulsion theory as the cause of multiple dural tearings, since the neurological deficit due to childhood trauma is concordant with multiple root involvement. Long-standing progressive enlargement of the cyst explains radiological signs such as eroded vertebral bodies and enlarged intervertebral foramina. Conclusion The pathogenesis, etiology, and treatment of the spinal arachnoid cyst have not been well established because of its rarity. Symptomatic cases require surgical intervention. For long segment involvements, hemilaminectomy or laminotomy, as was performed in the current case is preferable to laminectomy in order to prevent postoperative deformity. In surgery, dural defect should be closed after excision. Neurological recovery depends on the size of the cyst and degree and duration of the spinal cord compression. A long-standing spastic myelopathy is unlikely to have a significant improvement [2], as we observed in the current case. Except spasticity, the patient had no improvement on his neurological findings. Thus, in delayed cases, surgery may be taken and should be offered as a prophylactic intervention to prevent further impairment, rather than a curative one. In conclusion, the case presented here is a very unusual occurrence of this rare entity. Although probably traumatic, it had extremely delayed clinical presentation, its location was unusual, and as an extradural cervicothoracic arachnoid cyst, it was the largest amongst those reported.

Virchows_Arch-3-1-2082069

Pseudoneoplastic lesions of the testis and paratesticular structures

Pseudotumors or tumor-like proliferations (non-neoplastic masses) and benign mimickers (non-neoplastic cellular proliferations) are rare in the testis and paratesticular structures. Clinically, these lesions (cysts, ectopic tissues, and vascular, inflammatory, or hyperplastic lesions) are of great interest for the reason that, because of the topography, they may be relevant as differential diagnoses. The purpose of this paper is to present an overview of the pseudoneoplasic entities arising in the testis and paratesticular structures; emphasis is placed on how the practicing pathologist may distinguish benign mimickers and pseudotumors from true neoplasia. These lesions can be classified as macroscopic or microscopic mimickers of neoplasia. Introduction There are many lesions that can simulate a neoplasm in the testis or paratesticular structures. Their incidence among tumors arising within the scrotal sac varies according to different series from 6 to 30% [16, 32]. These pseudoneoplastic lesions can be divided into those that only macroscopically imitate neoplasia (Table 1) and those that microscopically imitate neoplasia, regardless of whether they form a macroscopic mass (Table 2). The latter group causes more problems to the practicing pathologist in terms of the correct classification of a giving lesion. Hereby, we summarize main tumor-like lesions and benign mimickers that may be seen in the testis and paratesticular coverages, with emphasis on morphologic criteria for the differential diagnosis from true neoplasia. Table 1Macroscopic mimickers (pseudotumors) of testicular and paratesticular neoplasiaVascular lesions  Intratesticular hemorrhage  Segmental testicular infarction  Organizad testicular hematocele    Cholesterol granuloma of the tunica vaginalisInflammatory lesions  Nonspecific infectious inflammatory lesions  Specific infectious inflammatory lesions  Non-infectious inflammatory lesions    Idiopathic inflammatory lesions     Idiopathic granulomatous orchitis     Testicular malakoplakia     Testicular sarcoidosis  Meconium periorchitis  Sperm granulomaCysts  Testicular cysts    Albuginea cysts    Parenchymal cysts (Epidermoid cysts)    Rete-testis cysts- Cystic displasia of the rete testis  Epididymal cysts and Spermatoceles  Spermatic cord cystsEctopic tissues  Adrenal cortical rests  Spleno-gonadal fusionTesticular appendagesMiscellaneous other lesionsFibrous pseudotumors; (Fibromatous periorchitis-Nodular periorchitis)  Amyloidosis  Polyorchididm  Sclerosing lipogranulomaTable 2Microscopic mimickers of testicular and paratesticular neoplasiaTesticular  Inflammatory-reactive lesions   Lymphocitic orchitis (Testicular pseudolymphoma)   Rosai-Dorfman disease  Sertoli cell hyperplasia    Pick adenoma    Hamartomatous proliferation testicular feminization syndrome  Interstitial cell hyperplasia    Leydig cell hyperplasia    Testicular “tumor” of the adrenogenital syndrome  Hyperplasia of the rete testisEpididymis  Adenomatoid hyperplasiaTunicas albuginea-vaginalis  Mesothelial hyperplasiaSpermatic cord  Vasitis nodosa  Inflammatory pseudotumor (Funiculitis proliferans)Miscellaneous other lesions Macroscopic mimickers (pseudotumors) of testicular and paratesticular neoplasia This group of lesions refers to those that macroscopically imitate a neoplasia. Frequently, these lesions affect both compartments and are therefore difficult to establish if one or both are affected. For this reason, we classify the tumor-like lesions according to their etiology. Vascular lesions The majority of the vascular disorders of the male gonad are not confused with a tumor because, in addition to causing acute symptoms, they usually affect the entire gonad. However, in the rare situation in which the vascular lesion is segmental, it can simulate neoplasia, as follows: Intratesticular hemorrhage, which occurs spontaneously [24] or in the context of anticoagulation treatment [15], may simulate a tumor at ultrasound examination.Segmental testicular infarction, a lesion clinically characterized by slight local pain unrelated to any acute episode [20], may be related to isolated or systemic vasculitis [43, 103] with morphology of polyarteritis nodosa [21], giant cell vasculitis [94], or Wegener granulomatosis [45]. Up to year 2000, 81 cases of systemic vasculitis with testicular tumor-like lesion have been recorded [45]. Other cases are seen in the context of a hematological disease (sickle cell anemia) [53] or associated with nonspecific perivascular fibrosis. Any of the phases of an infarction can be observed from acute (with hemorrhage) to healing stage (Fig. 1). Currently, the clinical diagnosis can be suspected with Doppler sonography [81], avoiding orchiectomy.Organized testicular hematocele and other hemorrhages in the tunica vaginalis are rarely confused with a neoplasia [89], but this can happen occasionally in long-standing cases because of fibrous thickening with cholesterol granuloma formation in the tunica vaginalis (Fig. 2) [61] Exceptionally, a true testicular neoplasia presents clinically with a hematocele [86].Fig. 1Segmental testicular infarction. a Acute hemorrhagic with reinforcement of the peripheral area (arrows). b Healing fibrosisFig. 2Cholesterol granuloma in the tunica vaginalis (H&E) Inflammatory lesions Similar to what was stated for vascular tumor-like lesions, inflammation that simulates a neoplasia usually has atypical clinical features; these entities can be grouped as follows: Nonspecific infectious inflammatory lesions with a tumor-like presentation are frequently chronic processes causing progressive fibrosis, which may clinically [38] or sonographicaly [23] simulate neoplasia. Rarely a testicular, epididymal, or vas deferens abscess can look like a neoplasia [84]. Very occasionally, testicular neoplasia can clinically imitate an acute inflammatory process [44].Specific infectious inflammatory lesions. The entities of this group that most often have been confused with neoplasias are granulomatous inflammation in tuberculosis [87], brucellosis [49], syphillis [5], fungal infections [42], and parasite diseases [9]. These lesions are usually not problematic for the pathologist.Noninfectious inflammatory lesions. This group of tumor-like lesions include different entities among which can be highlighted:Idiopathic granulomatous orchitis, probably of autoimmune aetiology, of which around 230 cases have been published [4], is characterized by tubular granulomas (tubular orchitis; Fig. 3) or interstitial granulomas (interstitial orchitis). The presence of intratubular giant cells differentiate this entity from infectious granulomatous orchitis [70]. Diffuse testicular hypoechoic involvement with only peripheral low-resistance flow on color Doppler sonography is a typical but not pathognomonic pattern [72].Malakoplakia. This lesion is secondary to a mononuclear decrease in cyclin–guanine monophosphate that impairs the killing of bacteria. Fusion of the phagolysosomes with bacterial rests produces the characteristic Michaelis–Gutman bodies in the cytoplasm of the macrophages (von Hansemann cells) [1] (Fig. 4). Giant cells are occasional or absent. Testicular involvement represents only 12% of genital malakoplakia with around 388 cases in the literature [19, 56]; exceptionally, it may affect the epididymis only [31]. It has been related to idiopathic granulomatous orchitis [56, 57] and chronic xantogranulomatous inflammation of testis [85, 109, 110]; this last also reported in the spermatic cord [64, 100].Sarcoidosis. Testicular involvement in a systemic sarcoidosis [39] is exceedingly rare, and its presentation as the primary form is even rarer; in these cases, the epididymis is affected more with the testicle being involved by contiguity [83].Meconium periorchitis. This is an infrequent lesion (around 30 cases being reported) that typically presents in the first months of life; most times there is a clinical history with obstetric problem that has caused the passing of meconium toward the testicular surrounding structures [29]. The macroscopic appearance is a myxoid material with calcified pearls resulting from the calcification of the remains of squamous cells or lanugo hairs [107] (Fig. 5). Suspicion of neoplasia, although very uncommon in this period of life, may be caused by a scrotal mass or sonographically detectable calcifications [6]. Clinically, the peritesticular and spermatic cord enlargement can simulate a paratesticular rhabdomyosarcoma.Sperm granuloma. As its name indicates, this granulomatous lesion with few giant cells is the consequence of extravasation of spermatozoa generally postvasectomy (40% of vasectomized men and 2.5% of general population) [108]. When it produces a tumor-like lesion, it is usually located in the deferent duct or the epididymis [22] with firm nodules of 0.7 to 4 cm with occasional cysts formation.Fig. 3Idiopathic granulomatous orchitis characterized by tubular granulomas (tubular orchitis)Fig. 4Malakoplakia. a Macroscopic appearance with a homogeneous aspect. b The characteristic Michaelis–Gutman bodies in the cytoplasm of the macrophages (von Hansemann cells) (PAS)Fig. 5Meconium periorchitis. a calcified pearls resulting from the calcification of the remains of squamous cells or lanugo hairs. b Keratin rest in one of the calcifications (H&E) Cysts The majority of the cysts with a tumor-like appearance are paratesticular structures, but the testicle may occasionally have some cystic lesions that can be confused with a neoplasia. Testicular cysts occur in approximately 8–10% of patients with a lump in the testis, including those of the tunica albuginea or the parenchyma [33].Tunica albuginea cysts do not usually cause any diagnostic problem, unless if they are complex cysts [74].Parenchyma testicular cysts can be more difficult to distinguish from a neoplasia and if there is the slightest suspicion of an intracystic content, one must suspect a malignant neoplasia [107]. Special consideration deserves the testicular Epidermoid cyst, which must only be lined with squamous epithelium (Fig. 6). It is recommended that the specimen is examined “in toto” to avoid underdiagnosis of any area of teratoma (especially among postpuberal patients) or intratubular germ cell neoplasia, as sonographically it is not possible to distinguish between these lesions [54, 106]. Epidermoid cysts represent 1% of the masses of the testes. Recent genetic studies have shown that there is no chromosome 12p abnormality [14], thus supporting its distinction from teratoma.Tubular ectasia of the rete testis secondary to obstruction and generally located in the mediastinum area of the testes (Fig. 7) [69], is usually bilateral and very different from cystic dysplasia of the rete testis, a congenital lesion with complete testicular parenchyma substitution [68].Epididymal cysts and spermatoceles are relatively frequent and the majority are in relation to the inflammatory processes. The differential diagnosis with other entities are related to its size, and similar to the other cystic formations a true neoplasia must be considered in case of observing any content in its interior [108].Spermatic cord cysts. The majority of these cysts do not cause any diagnostic doubts, only the occasional Epidermoid cysts [46, 105] can simulate a neoplasia. These cysts can be unilocular or multilocular depending on their origin [65], and the multilocular must be distinguished from the exceptional cystadenomas of probably Müllerian origin [55].Fig. 6Epidermoid cyst, which must only be lined with squamous cell epithelium (H&E)Fig. 7a Tubular ectasia of the rete testis located in the mediastinum area of the testes (arrows). b Microscopic aspect (H&E) Ectopic tissues Ectopic adrenocortical tissue is relatively frequent in the tunica albuginea, rete testis, epididymis, and spermatic cord and occasionally reaches the size to be symptomatic. Its incidence ranges from 2.5 to 15% [17, 102]. It is made up of adrenal cortical nodes surrounded by a connective tissue band and of about 5 mm in diameter on average (Fig. 8); for which reason, they are not clinically palpable. Only in cases of congenital adrenal hyperplasia or Cushing syndrome, ectopic adrenocortical tissue can be prominent and appear as a tumor-like lesion [90]. Splenic–gonadal fusion, as its name indicates, is the fusion of spleen and gonad. It is more frequent on the left side, with about 148 published cases [41]. Morphologically, the ectopic splenic tissue can be in close relation to the head of the epididymis or the upper pole of the testis (Fig. 9) or being separated from it; likewise, there may or may not be a structural continuity between the normal spleen and the ectopic tissue [30]. The same alteration has been described in women, but it is much less frequent that in men, probably because it is much easier to find it clinically in men. In about 30% of the cases, it is associated with complex malformations such as micrognathia, peromelia, or phocomelia (absence of upper portion of a limb) [30, 60, 95]. In three of the reported cases, the fusion was associated with a germ cell tumor of the testis [41]. An exceptional hepato-gonadal fusion is reported [26]. Fig. 8 Ectopic adrenocortical tissue. Adrenal cortical nodes surrounded by a connective tissue band in continuity with epididymis. (H&E) Fig. 9 Splenic–gonadal fusion. The ectopic splenic tissue in close relation with the upper pole of the testis (H&E) Testicular appendages There are five testicular appendages, but for the surgical pathologist, only three can be of interest (appendix testis or hydatid of Morgagni, appendix epididymis, paradidymis or organ of Giraldes; Fig. 10). These structures are not usually the origin of a tumor-like lesion, but in rare cases of large-sized cysts, it may presents as a paratesticular mass [93]. A case of ectopic epididymal tissue in an appendix testis [101] has recently been reported. Fig. 10Normal testicular appendages Miscellaneous other lesions Fibrous pseudotumors. This name refers to a fibrosis phenomenon with paucicellular hyalinized collagen (Fig. 11) presenting as nodular (single or multiple) or diffuse lesion of the testicular tunics [96, 71]. Sometimes, a node can be free (scrotal mouse) [108]. This broadly considered spectrum of lesions has received diverse names: chronic periorchitis, fibromatous periorchitis, nodular periorchitis, fibrous proliferation of the tunica, nonspecific paratesticular fibrosis, granulomatous periorchitis, nodular fibrous pseudotumor, fibrous pseudotumor, inflammatory pseudotumor, fibroma, reactive periorchitis [108], indicating its controversial pathogenesis. Some cases have been reported preceded by trauma or infection, and on occasion, an inflammatory component can be observed and granulation tissue suggesting the possibility that there might be the healing of an inflammatory pseudotumor, (which will be described later on) [10]. Although radiologically, it is not difficult to recognize, upon occasion, an intraoperative frozen section becomes necessary. Amyloidosis. It is usually bilateral and present in a patient with a prior history of amyloidosis [34]; more rarely, it is a primary form that by being a cryptorquidic patient simulates a testicular tumor [13]. Polyorchidism or supernumerary testes is a rare condition, which is easy to recognize. However, the sonography can occasionally be different from that of the normal testis, and the condition may then be tumor suspicious [79]. Fig. 11 Fibrous pseudotumor. a Well outlined fibrous-like nodule. b Paucicellular hyalinized collagen with calcification Microscopic mimickers of testicular and paratesticular neoplasia Lesions or cellular changes that microscopically imitate a neoplasia are included under this category, whether or not they make a clinical mass. These changes are closely related to the structure of the organ in which they arises, as follows. Testicular Inflammatory-reactive lesions Some lesion, already described above such as xantogranulomatous orchitis, idiopathic granulomatous orchitis and malakoplakia could be considered under this category; their microscopic confusion with a neoplasia (seminoma for example) is not currently a usual event; therefore, we preferred to include them in the macroscopic mimickers (pseudotumors). The situation is different with lymphocitic orchitis or testicular pseudolymphoma [2, 3], which is characterized by a lymphocytic and plasmocellular reaction that may be confused with a lymphoma, but immunohistochemistry will show that the cellular infiltrate is polyclonal (Fig. 12). Among these idiopathic lesions, we can include Rosai–Dorfman disease; histological examination of the testicular mass reveals an inflammatory lesion comprising lymphocytes, plasma cells, and sheets of pale staining histiocytes, some containing lymphocytes within their ample cytoplasm, suggestive of emperipolesis. The histiocytes stained positive for CD68, S100 by immunohistochemistry and negative for CD1a, while ultrastructural examination confirmed emperipolesis [25]. Fig. 12Lymphocitic orchitis. Lymphocitic cellular infiltration of polyclonal type. a CD45RA (lymphocytes B). b UCHL-1 (lymphocytes T) Sertoli cell hyperplasia In a series of situations, nonencapsulated nodules of Sertoli cells can be found, especially known in cryptorchid testes as Pick’s adenomas (Fig. 13a) [77]. Because of their appearance, these should be distinguished from the actual Sertoli cell tumors that generally are larger and sometimes there are areas that mimic Call–Exner bodies (Fig. 13b) A differential diagnosis with a yolk-sac tumor is not usually in the scope, but the immunohistochemistry study with AFP, calretinin, α-inhibin, and CD 99 can help [40]. Fig. 13Sertoli cell hyperplasia. a Nonencapsulated nodules of Sertoli cells (Pick’s adenoma). b Sometimes there are areas that mimic Call–Exner bodiesA special consideration merit the androgen insensitivity syndrome or testicular feminization (male pseudohermaphrodism, caused by a failure of androgen receptor binding) that in 63% of cases can have tubular hamartomas (tubules lined by immature Sertoli cells) [82] that must be differentiated from the Sertoli cell adenomas and sex cord tumor with annular tubules [76]. Interstitial cell hyperplasia In testicles with marked tubular atrophy, such as in the Klinefelter’s syndrome, it is possible to see Leydig cell nodules that must be distinguished from Leydig cell tumors. An interstitial growth without expansive pattern favors hyperplasia (Fig. 14) [62]. Fig. 14Leydig cell hyperplasiaNodules of eosinophilic cells appearing to be Leydig cells are found in the patients with adrenogenital syndrome [18, 48] and Nelson’s syndrome (adrenocorticotropic hormone-secreting pituitary adenoma after bilateral adrenalectomy), some of this last syndrome with excessive testosterone production [91]; proof that these interstitials cells are not only morphologically similar to Leydig cells but also have the functional property of these cells. These nodules are usually bilateral and of a large size with cellular pleomorphism and pigmentation. The clinical history and a complete endocrinological profile avoid an unnecessary orchiectomy [78] because only one case of aggressive behavior is published [18]. Hyperplasia of the rete testis The normal rete testis epithelium is flat, but in some hyperestrogenic situations (treatment or hepatic dysfunction), the epithelium may become columnar and rarely a micropapillary growth of bland cells can be observed. The diagnosis of rete testis hyperplasia is subjective, and adenomatous lesions are rarely seen [35, 63]. In some cases, there are intracytoplasmic hyaline eosinophilic globules resembling a yolk sac tumor, but the negative stains of α-fetoprotein or placental alkaline phosphatase help to rule this differential diagnosis [99].Pseudohyperplasia of the rete testis and epithelial reaction in case of germ cell invasion and cryptorchidism must be differentiated from real hyperplasia of the rete testis [62]. Epididymis Benign microscopic mimickers of cancer in the epididymis are very rare. Cysts may occur but do not resemble tumors microscopically. However, some cases of adenomatoid hyperplasia of the rete testis can involve the epididymis, and occasionally, they may become macroscopically apparent [93]. Tunica albuginea and vaginalis Non-neoplastic mesothelial lesions involving the paratesticular region include mesothelial cysts and reactive mesothelial hyperplasia [73]. Mesothelial hyperplasia is the most important benign mimicker of the testicular tunics. It is present as a reactive lesion in hydrocele or hernia but may also be found microscopically in older men [80]. The mesothelial proliferation has an epithelial appearance, and rarely, a spindle cell proliferation can be present. In the differential diagnosis with mesothelioma, the bland nucleus, no true invasion, and associated inflammatory elements can be useful (Fig. 15) [11, 98]. Recently, a case was published with “atypical” mesothelial hyperplasia on one side and “well differentiated” mesothelioma on the contalateral [97]. This case is an example of the subjective interpretation of some mesothelial proliferative lesions because some malignant mesotheliomas lack cellular atypia. In these cases, an extensive confluence or prominent infiltration favors a malignant diagnosis. Unfortunately, the immunohistochemical expression of benign and malignant mesothelial proliferations are similar (low- and high-molecular-weight cytokeratins and vimentin) [11, 28], and only the metastatic neoplasias can be differentiated by the CEA, Ber-EP4, and B72.3 expression [27]. Only deoxyribonucleic acid ploidy can distinguish some borderline lesions [28]. Fig. 15Mesothelial hyperplasia. The bland nucleus, no true invasion and associated inflammatory elements, can be useful to distinguish from malignant mesothelioma Spermatic cord The vas deferens and the soft tissues of spermatic cord can have benign mimickers. Vasitis nodosa is a ductular proliferation, generally after vasectomy [36], although it can be seen following other trauma on that area, [75]. It has a microglandular morphology (Fig. 16) with mild nuclear atipia and perineural growth [7] or benign vascular invasion [8] that may be mistaken for malignancy [104]. The frequent hyperplasia of nerve fibers in the adventitia can explain the painful symptoms in some patients [36]. The coincidence with microscopic sperm granulomas and inflammatory reaction can help in the correct diagnosis. An analogous epididymal lesion also occur [88].Proliferative funiculitis is the inflammation of the spermatic cord usually the result of an extension of vasitis. The soft tissues of the spermatic cord are also the most common site of an inflammatory pseudotumor in the male genital tract [37, 50, 58]. The lesion is ill defined, myxoid with white-gray color, and a moderate cellular proliferation with loose collagen fibers and irregular infiltration of inflammatory cells. An exceptional case has been reported that was largely infiltrated by mast cells [92]. In some cases, a prominent spindle cell proliferation mimics a sarcoma (pseudosarcomatous myofibroblastic proliferation), but low mitotic index, a capillary pattern, inflammatory cells, and absence of atypical mitoses speak against a diagnosis of sarcoma. Unfortunately, immunohistochemistry is only partly helpful because the cells express actin and vimentin, less strongly desmin and exceptionally cytokeratin. Two cases of this lesion in epididymis [12, 51] and one in the rete testis [47] have been reported.Fig. 16Vasitis nodosa. Ductular proliferation with a microglandular morphology Embryonic remnants Although ectopic tissue usually is more problematic for constituting a mass, some of them can cause microscopic diagnostic doubts for which reason they can be included in this group of benign mimics. The presence of seminiferous tubules within the tunica albuginea [67], Leydig cells in rete testis, albuginea, spermatic cord, or within sclerotic tubules [59, 66], prostate gland in the epididymis [52], and special circumstances with muscular rete testis hypertrophy [27] can mimic a neoplasia. The lesions described constitute a large heterogeneous group, without etiological or pathogenic relations among them or with true neoplasias. However, in spite of that, the patient with one or more of these tumor-like and/or benign mimickers can have a concomitant or ulterior true neoplasm.

Graefes_Arch_Clin_Exp_Ophthalmol-4-1-2292474

The effect of acute hyperglycemia on retinal thickness and ocular refraction in healthy subjects

Purpose To quantify the retinal thickness and the refractive error of the healthy human eye during hyperglycemia by means of optical coherence tomography (OCT) and Hartmann–Shack aberrometry. Patients with diabetes mellitus (DM) often experience subjective symptoms of blurred vision associated to hyperglycemia. The nature and origin of this phenomenon are still unclear. Blurred vision during hyperglycemia could be a result of transient refractive alterations due to changes in the lens [5, 12, 15, 25, 27, 36, 37], but it could also be caused by changes in the retina. Macular edema, or retinal thickening due to abnormal fluid accumulation within the macula, is a common cause of visual loss [1, 14, 22]. The degree of retinal thickening has been found to be significantly correlated with visual acuity [24]. Furthermore, a change in retinal thickness, resulting in a change in axial eye length, could also induce a change in ocular refractive error. For instance, it can be calculated that with a 50 μm increase in retinal thickness, the ocular refractive error becomes 0.15 D more hyperopic [30]. Several studies have demonstrated that retinal thickness is affected by DM [2, 7, 8, 18, 21, 23, 26, 32, 33, 35, 38]. In general, an increase in retinal thickness has been reported in patients with long-term DM and advanced stages of retinopathy [7, 8, 18, 32, 33, 35]. However, in diabetic patients with and without minimal diabetic retinopathy, a decrease in retinal thickness has been observed [2, 23]. In healthy subjects, it has been shown that the average retinal thickness did not change during normo-insulinaemic hyperglycemia [13]. It is unclear whether the thickness of the different retinal areas, such as the foveal area, the pericentral foveal area, and the peripheral foveal area, changes during acute hyperglycemia and suppression of insulin. A change in retinal thickness and/or ocular refractive error could explain the subjective symptoms of blurred vision in patients with DM and hyperglycemia. Therefore, in the present study the effect of hyperglycemia on retinal thickness and ocular refractive error was investigated in healthy subjects during suppression of endogenous insulin. Retinal thickness was measured by means of optical coherence tomography (OCT), which is a non-invasive technique that provides cross-sectional retinal images, and produces an objective measurement of the retinal thickness, independent of the refractive status of the eye [10, 11, 29]. Furthermore, aberrometry was used to measure the ocular refractive error. This technique makes it possible to detect small changes in ocular refraction [19]. Subjects and methods Five healthy subjects (two males and three females) participated in the study. The mean age of the subjects was 24.8 years (range 21.2–32.6), and their mean body mass index (BMI) was 24.2 kg/m2 (range 21.4–29.7). The subjects were screened during a first visit, which included medical history-taking, a physical examination (measurement of visual acuity, weight, height and blood pressure) and collecting a fasting blood sample. Exclusion criteria were a history of DM (or a fasting plasma glucose >5.5 mmol/l), a BMI of >30 kg/m2, elevated blood pressure (>140 / 85 mmHg), a visual acuity of <0.5 (Snellen) or a history of ocular pathology. The investigators of the ocular parameters (NW and MD) were not informed about the blood glucose levels. Furthermore, the investigators who induced hyperglycemia (EE and SS) were not informed about the results of the ocular measurements. The study protocol was approved by the Medical Ethics Committee of the VU University Medical Center in Amsterdam, and written informed consent was obtained from all subjects after the purpose and nature of the study had been explained to them. Procedure to induce hyperglycemia After a 10-hour overnight fast, the subjects were given a subcutaneous injection of a low dose (100 μg) of synthetic somatostatin (Sandostatin, Novartis, Basel, Switzerland) in order to suppress endogenous insulin secretion. Each subject underwent an oral glucose tolerance test (OGTT) (75 g glucose) 30 minutes after the somatostatin injection, and blood glucose levels were measured with a blood glucose analyzer (HemoCue Diagnostics BV, Oisterwijk, the Netherlands). Endogenous insulin levels were measured by means of immunometric assays (Luminescence, Bayer Diagnostics, Mijdrecht, the Netherlands) in the Endocrinology Laboratory at the Department of Clinical Chemistry of the VU University Medical Center. The subjects remained in fasting state during the entire procedure. Ocular measurements Retinal thickness was measured with the Stratus OCT (Model 3000, Carl Zeiss Meditec, Dublin, CA, USA), which combines a low-coherence scanning interferometer (wavelength 820 nm) with a video camera to visualize the fundus of the eye. The fast macular thickness OCT scan protocol was used to obtain six cross-sectional macular scans, 6 mm in length, which were positioned at equally spaced angular orientations (30°) centred on the fovea. The cross-sectional images were analyzed with OCT3 mapping software that uses an edge-detection technique to locate the vitreoretinal interface and the anterior surface of the retinal pigment epithelium. Retinal thickness was defined as the distance between these two surfaces. Two OCT scans were made of each subject before, and every 30 minutes during the period of hyperglycemia. In order to quantify the retinal thickness, the foveal map constructed by the software was divided into nine Early Treatment Diabetic Retinopathy Study (ETDRS) areas [6]: the central fovea (central circle with a diameter of 1 mm), and the pericentral area (donut-shaped ring with an inner diameter of 1 mm and an outer diameter of 3 mm) and peripheral area (donut-shaped ring with an inner diameter of 3 mm and an outer diameter of 6 mm), both of which were divided into four quadrants. Retinal thickness was calculated for all separate areas, and for the average pericentral and peripheral regions. Ocular refractive error was determined with an IRX3 aberrometer (Imagine Eye Optics, Paris, France), which performs wavefront analysis of the eye according to the Hartmann-Shack principle [19]. After pupil dilation and paralysis of accommodation with 1.0% cyclopentolate and 5.0% phenylephrine eye-drops, a series of three aberrometry measurements was made before, and every 30 minutes during the hyperglycemic condition. From these measurements, the equivalent refractive error was calculated as: equivalent refractive error (ERE) = sphere + (cylinder / 2). The measurements at baseline and during maximal hyperglycemia were analyzed, and any change was considered to be meaningful if the difference between the measurements was greater than the threshold of 50 μm for retinal thickness and 0.2 diopters (D) for ERE. The threshold of 50 μm exceeded the 95% confidence interval for the detection of a change in retinal thickness, which has been reported to be approximately 40 μm [4, 20, 28]. A refractive change of more than 0.2 D also surpasses the precision (defined as 95% confidence interval) of the aberrometer for measuring sphere, cylinder, and consequently ERE [3, 31]. In each subject, the significance of a change was obtained from the precision of the measurement instruments and the difference in the ocular parameters at baseline and during hyperglycemia. In the whole group, the significance of a change could be determined by means of Wilcoxon matched pairs signed rank sum tests. P-values ≤0.05 were considered to be statistically significant. Results The changes in blood glucose after the administration of somatostatin and glucose are shown in Fig. 1. Mean blood glucose levels rose from 4.0 mmol/l (range 3.4–4.5 mmol/l) to 18.4 mmol/l (range 16.1–22.0 mmol/l) after the OGTT. Endogenous insulin was suppressed by the subcutaneous injection of somatostatin during the glucose load to a mean value of 2.1 pmol/l (range 0.4–4.5 pmol/l), and remained below basal secretion level (<110.0 pmol/l) for 147 minutes (range 75–270 minutes). Subject 01 had a delayed elevation of blood glucose level, compared to the other subjects. This person received a second 75 g oral glucose load after 30 minutes in order to induce a rise in the blood glucose level. In all subjects, the blood glucose and endogenous insulin levels normalized within 6 hours after the OGTT. Fig. 1Graph of normalized blood glucose levels in the five subjects after the administration of somatostatin and glucose. Data are normalized by subtracting the value at baseline from the measured value in each subject. The oral glucose load (75 g) was administered at T 0. Subject 01 received an extra 75 g oral glucose load at T 30 Figure 2 shows the normalized ERE of the five subjects during hyperglycemia. Mean ERE at baseline was 0.6 D (SD 0.6) and 0.7 D (SD 0.6) during maximal hyperglycemia; no significant change was found in the group as a whole. A small, but significant hyperopic shift of 0.4 D (SD 0.2) in ERE was measured in subject 01 during maximal hyperglycemia, compared to the start of the procedure (p < 0.001). No significant change in ERE was found in any of the other subjects. Normalized retinal thickness parameters are shown in Fig. 3a (central foveal area), Fig. 3b (average pericentral foveal area) and Fig. 3c (average peripheral foveal area). Average central foveal thickness, average pericentral foveal thickness, and average peripheral foveal thickness at baseline were 202 μm (SD 8), 277 μm (SD 5), and 243 μm (SD 8). During maximal hyperglycemia, average central foveal thickness, average pericentral foveal thickness, and average peripheral foveal thickness were 203 μm (SD 7), 275 μm (SD 3), and 242 μm (SD 9). No significant differences were found in the group as a whole. Furthermore, none of the subjects had any significant changes in the thickness of the central fovea, the pericentral fovea, or the peripheral fovea during maximal hyperglycemia, compared to baseline. The nine separate ETDRS areas were not affected by hyperglycemia. At baseline and during hyperglycemia, any change in retinal thickness that occurred in the various areas was less than 15 μm, which was within the previously defined threshold of 50 μm. Fig. 2Graph of the normalized equivalent refractive error (ERE) in diopters (D) of the five subjects. Data are presented as mean±SD; three measurements were made of each subject every 30 minutes during the procedure. Data are normalized by subtracting the value at baseline from the measured value in each subject. The oral glucose load was administered at T 0. * Significant difference between ERE at T 0 and T 210 (maximal hyperglycemia), p < 0.001Fig. 3Graphs and maps of normalized retinal thickness parameters in the five subjects during hyperglycemia: (a) central fovea, (b) pericentral fovea, (c) peripheral fovea. Data are normalized by subtracting the value at baseline from the measured value in each subject. Each measured area has been indicated by a dark grey area on the retinal maps. No significant changes in retinal parameters were found in any of the subjects. The oral glucose load was administered at T 0 Discussion Blurred vision is a symptom that occurs frequently in patients with DM and hyperglycemia. The underlying mechanism is still unclear, and therefore the present study was carried out in an attempt to identify a possible cause of this symptom. The effect of reproducible hyperglycemia on retinal thickness and refractive error was studied in healthy young subjects who did not suffer from the systemic effects of DM. No changes in the thickness of the central, pericentral or peripheral foveal areas were found in any of the subjects during hyperglycemia. In addition, no significant change was measured in any of the nine different ETDRS areas of the macula. In their study, Jeppesen et al. [13] also found no significant difference in retinal thickness in healthy subjects during normo-insulinaemic hyperglycemia. Before and 180 minutes after the start of a hyperglycemic clamp they measured the average thickness of the retina, and found that retinal thickness was not affected by hyperglycemia. Although in the present study retinal thickness was measured under different circumstances than in the study of Jeppesen et al. (hypo-insulinaemic hyperglycemia instead of normo-insulinaemic hyperglycemia), the results confirm those of Jeppesen et al. Retinal thickness has been reported to change in patients with long-term DM and retinopathy. A morphological change in the retina may even occur in the early stages of diabetic retinopathy [2, 7, 8, 18, 21, 23, 26, 32, 33, 35, 38]. These changes in retinal thickness are usually due to abnormal fluid accumulation resulting from a breakdown of the blood-retinal barrier [34]. Goebel et al. [8] measured retinal thickness by means of OCT in 136 patients with different stages of diabetic retinopathy and with a mean DM duration of 16 years. Mean foveal thickness was 307 ± 136 μm in the diabetic subjects, compared to 153 ± 15 μm in healthy subjects. It seems that only long-term hyperglycemia and/or long-term fluctuations in blood glucose levels have any significant influence on the blood–retina barrier and retinal thickness. From the findings of the present study, it appears that the blood–retina barrier does not seem to be affected by a single episode of acute hyperglycemia. Nevertheless, the fact that no change in retinal thickness could be determined does not exclude the possibility that there could be early dysfunction of the blood retina barrier. Other means of examination could evidence such a dysfunction of the blood retina barrier following acute hyperglycemia. A factor that could have biased the results of this study is the administration of a synthetic somatostatin analogue to the subjects. Somatostatin is a peptide hormone that inhibits several hormones, including IGF-1 and insulin. IGF-1 is a growth factor that is produced by the hypoxic retina to mediate angiogenesis, resulting in neovascularisation. Somatostatin analogues not only inhibit neovascularisation in patients with advanced diabetic retinopathy, but also stabilize the blood–retinal barrier in patients with diabetic macular edema [16, 17]. It could have been possible that in the present study an increase in retinal thickness during hyperglycemia was prevented by somatostatin. Nevertheless, the efficacy of synthetic somatostatin in the treatment of advanced diabetic retinopathy was investigated by Grant et al. [9]. With maximally tolerated doses of somatostatin (ranging from 200 to 5000 μg/day), after a period of 15 months one out of 22 eyes required panretinal photocoagulation, compared to nine of 24 eyes that were not treated with somatostatin. From the results of the Grant et al. study, it seems that only frequent, large doses of somatostatin over a long period of time have any significant effect on the progression of diabetic retinopathy. Although the effect of somatostatin on the healthy retina has not been investigated yet, it seems to be unlikely that the results of the present study were biased by the administration of one single, low dose (100 μg) of somatostatin. In conclusion, the results of this study indicate that in healthy subjects, hyperglycemia does not cause any change in retinal thickness. Furthermore, ocular refraction in general was not affected by hyperglycemia. However, there were interindividual variations, as illustrated by subject 01, who had a hyperopic shift of refraction during hyperglycemia. Therefore, it seems that a refractive change during hyperglycemia cannot be explained by a change in retinal thickness. It could well be that other refractive components, such as the lens, are involved in causing blurred vision and refractive alterations during hyperglycemia.

Eur_J_Pediatr-4-1-2413085

Rab proteins and Rab-associated proteins: major actors in the mechanism of protein-trafficking disorders

Ras-associated binding (Rab) proteins and Rab-associated proteins are key regulators of vesicle transport, which is essential for the delivery of proteins to specific intracellular locations. More than 60 human Rab proteins have been identified, and their function has been shown to depend on their interaction with different Rab-associated proteins regulating Rab activation, post-translational modification and intracellular localization. The number of known inherited disorders of vesicle trafficking due to Rab cycle defects has increased substantially during the past decade. This review describes the important role played by Rab proteins in a number of rare monogenic diseases as well as common multifactorial human ones. Although the clinical phenotype in these monogenic inherited diseases is highly variable and dependent on the type of tissue in which the defective Rab or its associated protein is expressed, frequent features are hypopigmentation (Griscelli syndrome), eye defects (Choroideremia, Warburg Micro syndrome and Martsolf syndrome), disturbed immune function (Griscelli syndrome and Charcot–Marie–Tooth disease) and neurological dysfunction (X-linked non-specific mental retardation, Charcot–Marie–Tooth disease, Warburg Micro syndrome and Martsolf syndrome). There is also evidence that alterations in Rab function play an important role in the progression of multifactorial human diseases, such as infectious diseases and type 2 diabetes. Rab proteins must not only be bound to GTP, but they need also to be ‘prenylated’—i.e. bound to the cell membranes by isoprenes, which are intermediaries in the synthesis of cholesterol (e.g. geranyl geranyl or farnesyl compounds). This means that isoprenylation can be influenced by drugs such as statins, which inhibit isoprenylation, or biphosphonates, which inhibit that farnesyl pyrophosphate synthase necessary for Rab GTPase activity. Conclusion: Although protein-trafficking disorders are clinically heterogeneous and represented in almost every subspeciality of pediatrics, the identification of common pathogenic mechanisms may provide a better diagnosis and management of patients with still unknown Rab cycle defects and stimulate the development of therapeutic agents. Introduction Germline mutations in the genes involved in the Ras–mitogen-activated protein kinase (MAPK) pathway explain a number of neuro–cardio–facio–cutaneous syndromes, as reported in a paper recently published in this journal [10]. Ras-associated binding (Rab) proteins are small GTPases of the Ras superfamily that continuously cycle between the cytosol and different membranes. In 2002, Pfeffer described 20 different Rab GTPases that have been identified as prenylated proteins localized in distinct membrane-bound compartments [25]. Prenylated proteins are proteins bound to isoprenes, which are intermediaries in the cholesterol synthesis. At the time of the writing this article, more than 60 human Rabs had been identified [26]. This high number of different Rab proteins highlights their importance in the regulation of vesicle trafficking processes, including vesicle formation, motility, tethering and fusion to the acceptor membrane and signaling to other organelles. The ability of Rabs to perform these different tasks in a co-ordinated and regulated manner originates in their highly dynamic conformation and mobility and from their interaction with different Rab-associated proteins, as illustrated in the Rab GTPase cycle (Fig. 1). Rabs can cycle between the active GTP-bound and the non-active GDP-bound forms, assisted by different Rab-associated proteins, such as the GEFs (guanine nucleotide-exchange factors) and GAPs (GTPase-activating proteins), respectively [24, 26]. This conformational cycle is the main driving force for the on/off ‘switch’ mechanism of Rabs and their ability to both regulate binding to downstream effectors and perform the desired function. Rabs are initially synthesized as soluble proteins in the cytosol where they are first recognized by a soluble chaperone-like protein named REP (Rab escort protein) [3]. The REP brings the Rab to the RabGGT (Rab geranylgeranyl transferase) for the addition of—generally—two geranylgeranyl groups. These prenyl groups are added, by covalent thioether bonds, to cysteine residue(s) located at the C-terminus of the Rab protein [7]. This post-translational modification of Rabs is needed to allow for the attachment of the Rab proteins into the lipid bilayer of the organelle; consequently, Rabs are considered to be peripheral membrane proteins (Fig. 1). After performing their function, Rab proteins are extracted from the membranes by RabGDI (GDP dissociation inhibitor), and they remain in the cytosol until they are needed again [7, 24, 37]. Statins are known to inhibit protein isoprenylation. As such, they are able to reduce the production of amyloid-beta and inhibit preamyloid catabolism in the lysosomes, which suggests their possible application in the treatment of Alzheimer [20]. However, the therapeutic role of statins in Rab-related disorders has not yet been established. The potent nitrogen-containing biphosphonates, such as pamidronate, alendronate, ibandronate and zoledronate, inhibit a key enzyme, farnesyl pyrophosphate synthase, which is necessary for Rab activity, hence their inhibitory action on osteoclast activity [29]. Fig. 1The Ras-associated binding (Rab) protein–GTPase cycles. Rab proteins are intrinsically soluble and require a post-translational modification for membrane association. They first associate with a Rab escort protein (REP) and form a stable complex that is the substrate for the subsequent dual prenylation of C-terminal cysteine motifs via Rab geranylgeranyl transferase (RabGGT). RabGGT consists of two different functional subunits (RabGGTα and RabGGTß). After lipid tranfer, REP delivers the prenylated Rab to the donor membrane (the REP cycle is shown with blue arrows). In the absence of REP or RabGGT, Rab proteins remain in the cytosol in an inactive state. The transfer of Rab proteins between acceptor and donor membranes is facilitated by the GDP dissociation inhibitor (GDI) (the GDI cycle is shown with red arrows). Both REP and GDI bind the GDP-bound inactive form of Rab. After REP or GDI dissociate from Rabs at the donor membrane, Rabs can cycle between the inactive (GDP-bound) and active (GTP-bound) states. Rab proteins are activated by guanine nucleotide exchange factors (RabGEFs) and deactivated by GTPase activating proteins (RabGAPs), which accelerate the slow intrinsic rates of nucleotide exchange and GTP hydrolysis. In the active state, Rabs interact with structurally and functionally diverse effectors, including cargo sorting complexes on donor membranes, motor proteins involved in vesicular transport and tethering complexes that regulate vesicle fusion with acceptor membranes Evolutionarily conserved Rabs are usually expressed in all cell and tissue types and regulate more fundamental vesicle transport pathways whereas the less conserved Rab family members function in more specific pathways, which are often organ-dependent. Defects in intracellular vesicle trafficking underlie a large variety of human diseases, including pathologies associated with defects in Rabs or Rab-associated proteins [14, 32]. In this review, we deal with the role of different Rab and rab-associated proteins in inherited (Table 1) as well as certain multifactorial human diseases in different pediatric disciplines. In addition, the review will also focus briefly on recent advances in understanding human disease through the study of Rab proteins from in vitro and mouse knockout studies. Table 1Human monogenic diseases caused by a Ras-associated binding (Rab) protein or Rab-associated protein defectDiseaseOMIMGeneRabRab-associatedDescriptionGriscelli syndrome type I214450MYO5AxAutosomal recessive, albinism, neurological impairmentGriscelli syndrome type II607624RAB27AxAutosomal recessive, albinism, hemophagocytic syndrome, sometimes secondary neurological impairmentGriscelli syndrome type III609227MLPHxAutosomal recessive, albinismChoroideremia303100REP1xX-linked, progressive loss of vision beginning at an early age, and the choroid and retina undergo complete atrophyNon-specific mental retardation300104GDI1xX-linked, affected males show moderate to severe mental retardation, carrier females may also be mildly affectedCharcot-Marie-Tooth disease type IIb608882RAB7xAutosomal dominant, peripheral sensory neuropathy with late onset muscle weakness, foot ulcers and infectionsWarburg Micro syndrome600118RAB3GAP1xAutosomal recessive, microcephaly, microcornea, congenital cataract, mental retardation, optic atrophy, and hypogenitalismMartsolf syndrome212720RAB3GAP2xAutosomal recessive, congenital cataracts, hypogonadism, and mild mental retardation Rab proteins in skin and hear pigmentation Griscelli syndrome (GS) is an autosomal recessive disorder that causes partial albinism [15]. There are three variants of this disease: one is a purely hypopigmentation disorder (GS type III), and two, in addition to the pigmentation defect, are also associated with immunological defects (GS type II) or with primary neurological dysfunctions (GS type I). Griscelli syndrome type II with immunological defects is caused by missense mutations in the gene encoding Rab27a [18]. This Rab regulates the movement of melanosomes to the cell periphery of melanocytes and also regulates the secretion of lytic granules in cytotoxic T lymphocytes (CTL) [35]. The lack of Rab27a thus causes pigment anomalies and dysfunctional T lymphocytes, which is in agreement with the defects observed in the patients that present with albinism and hemophagocytic syndrome. The GS type I with neurological symptoms is caused by mutations in the MYO5A gene encoding the motor protein myosin Va [22], a putative Rab27a effector that drives the peripheral distribution of melanosomes along actin filaments [15]. As myosin Va does not participate in the exocytosis of lytic granules of CTL, the inactivation of this protein does not lead to immunological symptoms. Finally, GS type III in patients with typically pigment abnormalities without additional features is caused by a mutation in the MLPH gene encoding melanophilin. A direct interaction exists between the SHD domain of melapholin and Rab27a and between the C-terminal domain of melanophilin and myosin Va [36]. Therefore, although the Rab27a/Mlph/MyoVa complex is crucial for the trafficking of melanosomes, neurodevelopmental abnormalities are specific for the Myo5a deficiency, and Rab27a uses a different effector for its function in the CTL. Rab proteins in immunology It has already been mentioned that GS type II patients with Rab27a mutations develop the hemophagocytic syndrome, which is characterized by episodes of life-threatening uncontrolled T lymphocyte and macrophage activation [17]. During these episodes, activated T cells and macrophages infiltrate various organs (including the brain), leading to tissue damage, organ failure and death in the absence of immunosuppressive therapy. In practice, only bone-marrow or cord blood stem cell transplantation can be curative for this condition [33]. The important function of Rab proteins in intracellular traffic, endocytosis and exocytosis explains their role in counteracting bacterial and viral infections. Phagocytosis by macrophages leads to the destruction of bacteria. Endocytosis leads to the formation of phagosomes, but phagosomes fuse with lysosomes (phagolysosomes), resulting in the release of toxic products that kill the bacterium as well as the release of the degradation products by exocytosis [19]. However, some bacteria survive this process and avoid bactericide by escaping from the phagosomes into the cytoplasm [19]. Listeria monocytogenes is a paramount example. It has been shown that the stimulation of macrophages with interferon (IFN)-γ, a cytokine that is secreted by natural killer and T helper cells, results in the expression of at least 200 proteins, including Rab5a [11]. In L. monocytogenes-infected macrophages, overexpression of Rab5a stimulates the intracellular degradation of the pathogen [6]. It has been suggested that Rab5a participates in the active recruitment of Rac2 to phagolysosomes, which is important for the destruction of the bacterium [27]. Other microbes develop antiphagocytosis (e.g. Yersinia enterocolitica) or inhibit the formation of phagosomes (e.g. Salmonella typhimurium). In Salmonella infections, there is evidence that Rab5a is stabilized on the phagosome by the Salmonella type III secretory protein SopE [23]. SopE has GEF activity for Rab5a, stimulating activated GTP-bound Rab5a and preventing its membrane extraction by RabGDI. A general conclusion that can be drawn is that parasites recruit endocytic Rabs to prevent or delay the formation of the degradative phagolysosome in the host organism. Viruses enter host cells by receptor-mediated endocytosis, but they usually escape from endosomal vesicles to enter either the cytosol or the nucleus where they replicate. The binding of hepatitis C virus (HCV) non-structural protein (NS) with Rab1 GTPase-activating protein suggests that viruses are able to subvert host cell machinery in this way because this binding is necessary for HCV replication. Rab1 depletion decreases HCV-RNA levels, which could lead to therapeutic applications [34]. Rab proteins in vision Choroideremia is an X-linked disease that involves the degeneration of the retinal pigment epithelium and the adjacent choroid and retinal photoreceptor cell layers, leading to blindness. The mutated gene in the choroideremia is one of the two REP isoforms, REP1 [3, 30]. Although the other isoform, REP-2, seems to be sufficient for the geranylgeranylation of all Rab GTPases in all tissues, the retinal pigment epithelium seems to be the exception. A role for REP-2 in human disease has not yet been discovered. REP-1 is essential for the efficient geranylgeranylation of Rab27a in the retinal pigment epithelium. Thus, a lack of REP1 leads to a lack of functional Rab27a specifically in the retinal pigment epithelium [30, 31]. The degeneration of this epithelium and its adjacent layers may be due to deficient melanosome transport and, consequently, a lack of protection against harmful light exposure. Rab proteins in neurology A subgroup of patients with X-linked non-specific mental retardation have mutations in the GDI1 gene for one of the GDI isoforms, GDI-α [9]. This isoform is particularly abundant in the brain, and dysfunctional membrane recycling of one or more Rab GTPases in brain synapses, leading to aberrant neurotransmission, is likely to underlie the symptoms in this disease. Hereditary sensory and autonomic neuropathies (HSAN), known for many years as familial dysautonomia in Jewish patients, are characterized by the loss of sensation for pain and temperature, alacrima, excessive sweating and the absence of fungiform tongue papillae. The onset is congenital, and the transmission is autosomal recessive. The clinical symptoms of HSAN have been classified into five types [21]. HSAN type II encompasses patients with Charcot–Marie–Tooth disease type II. This is an axonal neuropathy with a median motor conduction velocity of >38 m/s. This peripheral neuropathy starts in the second or third decade of life and presents as mixed motor and sensory involvement associated with ulcers. Four mutations in the Rab7 gene have been identified in four families with Charcot–Marie–Tooth disease type 2B (CMT2B), with symptoms leading to ulcerations and amputations [38, 39]. However, a hallmark feature of patients with HSAN type 1 is lancinating pain, which is not commonly observed in CMT2B. Moreover, HSAN type 1 is associated with a mutation in serine palmitoyl transferase (SPT), a rate-limiting step in the sphingo-lipid synthesis. This overlapping of the two syndromes is possibly explained by Rab7 being involved in both the endocytosis and transport of sphingolipids [39]. The importance of Rab proteins in brain development and function is even more emphasized by the recent discoveries of the involvement of Rab3-associated proteins in human disease [4, 5]. Rab3A is the most abundantly expressed protein in the brain, where it is present in all synapses and involved in calcium-dependent neurotransmitter release [13]. The activity of Rab3 proteins is tightly regulated by Rab3GAP, which specifically converts active Rab3-GTP to the inactive-GDP form. Germline-inactivating mutations in the catalytic subunit of Rab3GAP (Rab3GAP1) cause Warburg Micro syndrome, an autosomal recessive disorder characterized by ocular defects (microphthalmos, microcornea, congenital cataracts and optic atrophy), neurodevelopmental defects (microcephaly, cortical gyral abnormalities, such as pachygyria and polymicrogyria, hypoplasia of the corpus callosum, severe mental retardation, and spastic cerebral palsy) and hypothalamic hypogenitalism [4]. Mutations in the non-catalytic subunit of Rab3GAP (Rab3GAP2) have been described in patients with Martsolf syndrome, an autosomal recessive disorder with a similar but milder phenotype than the Warburg Micro syndrome, including congenital cataracts, hypogonadism and mild mental retardation [5]. The precise mechanisms whereby the Rab3GAP1 and Rab3GAP2 mutations cause the human phenotype is still unclear. Rab proteins in endocrinology Abnormal trafficking of the insulin-sensitive glucose transporter GLUT4 has been described in patients with diabetes type 2. Glucose transporter 4 accumulates in the dense membrane compartments, suggesting that defects in membrane trafficking may be involved in insulin resistance [12]. Rab4 has been found to be implicated in GLUT4 biogenesis, sorting and exocytosolic movement but more studies are needed to define the exact role of this and other Rabs and Rab-associated proteins in GLUT4 trafficking [16]. Rab4 mutations in humans or mice do not yet exist. Rab27a is necessary for insulin secretion [2]. Diabetes mellitus is characterized by both low insulin and high glucagon levels. These conditions have been shown recently to be due to the expression of an inducible cAMP early repressor (ICER) [1], resulting in a repression of the transcription of the insulin gene, and associated with a diminished level of granuphilin, a Rab27a-dependent protein. However, patients with GS type II-homozygous deficiency in Rab27a do not have diabetes mellitus. The control of thyroid hormone production depends on endocytic catalysts and the tandem regulators Rab5a and Rab7, which have been found in excess in thyroid adenomas [8]. Rab proteins in nephrology Nephrogenic diabetes insipidus (NDI) is characterized by the inability to concentrate the urine in response to the antidiuretic hormone vasopressin (AVP). Under normal conditions, binding of AVP to the vasopressin receptor (V2R) leads to the insertion of aquaporin 2 (AQP2) water channels in the apical membrane of the renal collecting duct [28] (Fig. 2). The defect can be inherited and is either X-linked due to mutations in the genes encoding the V2R or autosomal recessive or dominant in mutations in the AQP2 gene. The trafficking of AQP2 in the cell and its internalization are cAMP and especially Rab11-dependent [40]. However, Rab11 mutations in humans or mice are not yet described. Fig. 2Regulation of aquaporin-2 (AQP2)-mediated water transport by vasopressin (AVP). Shown is a nephron with a magnified principal cell. In this cell, the vasopressin V2 receptor (V2R), stimulatory GTP binding protein (Gs), adenylate cyclase (AC), ATP, cAMP, and phosphorylated proteins (O-P) are indicated. This figure has been used with permission of the American Journal of Physiology and Renal Physiology Conclusion The Rab GTPases are a large family of proteins with a variety of regulatory functions in membrane trafficking. The central role of these proteins has become clear during the past decade, as part of the progress that has been made in understanding in detail the mechanistic principles of transport vesicle formation, movement, and fusion. Sequencing of the human genome has enabled researchers to realize the diversity of the Rab gene family, although the functions of the majority of the gene products are still unknown. The availability of complete genomic sequences as well as the important advances in molecular and cell biological methods that have already been made hold the promise of significant progress being booked in our understanding of Rab function in the near future. The identification of further genes involved not only in monogenic but also in common multifactorial human vesicle-trafficking disorders will result in a better understanding of this complex transport pathway but, more importantly, will also lead to opportunities to develop novel treatments.

Mol_Genet_Genomics-2-2-1705487

The fission yeast Rpb4 subunit of RNA polymerase II plays a specialized role in cell separation

RNA polymerase II is a complex of 12 subunits, Rpb1 to Rpb12, whose specific roles are only partly understood. Rpb4 is essential in mammals and fission yeast, but not in budding yeast. To learn more about the roles of Rpb4, we expressed the rpb4 gene under the control of regulatable promoters of different strength in fission yeast. We demonstrate that below a critical level of transcription, Rpb4 affects cellular growth proportional to its expression levels: cells expressing lower levels of rpb4 grew slower compared to cells expressing higher levels. Lowered rpb4 expression did not affect cell survival under several stress conditions, but it caused specific defects in cell separation similar to sep mutants. Microarray analysis revealed that lowered rpb4 expression causes a global reduction in gene expression, but the transcript levels of a distinct subset of genes were particularly responsive to changes in rpb4 expression. These genes show some overlap with those regulated by the Sep1-Ace2 transcriptional cascade required for cell separation. Most notably, the gene expression signature of cells with lowered rpb4 expression was highly similar to those of mcs6, pmh1, sep10 and sep15 mutants. Mcs6 and Pmh1 encode orthologs of metazoan TFIIH-associated cyclin-dependent kinase (CDK)-activating kinase (Cdk7-cyclin H-Mat1), while Sep10 and Sep15 encode mediator components. Our results suggest that Rpb4, along with some other general transcription factors, plays a specialized role in a transcriptional pathway that controls the cell cycle-regulated transcription of a specific subset of genes involved in cell division. Introduction RNA polymerase II (pol II) transcribes the coding genes, and its activity is the main target for transcriptional control in all eukaryotes. The pol II enzyme consists of 12 subunits, which are highly conserved during evolution. The specific functions of the smaller subunits such as Rpb4 are relatively poorly understood, and they can show differences between organisms. For example, Rpb4 is essential in mammals and fission yeast (Schizosaccharomyces pombe), but not in budding yeast (Saccharomyces cerevisiae). Here we use fission yeast as a model to learn more about particular roles of Rpb4 in genome-wide transcription. Much of our current understanding of the mechanism of transcription and the transcription machinery comes from studies using budding yeast. The distantly related fission yeast provides a valuable complementary model system to dissect the functions of different components of the transcriptional apparatus, because its transcription mechanism is more similar to that of mammals in some respects. For example, the initiation of transcription in both mammalian cells and S. pombe occur ∼30 bp downstream of the TATA box, whereas in S. cerevisiae this distance can vary between 40 and 120 bp (Li et al. 1994). Moreover, transcriptional initiation from mammalian promoters introduced into S. pombe occurs at the same site as in mammalian cells (Toyama and Okayama 1990). As in other eukaryotes, S. pombe pol II is a multi-subunit enzyme containing the 12 subunits Rpb1 to Rpb12. The four core subunits, Rpb1, Rpb2, Rpb3 and Rpb11, are homologous to the β′, β, α′ and α subunits, respectively, of the eubacterial RNA polymerase. Five subunits, Rpb5, 6, 8, 10 and 12, are shared by the three eukaryotic RNA polymerases I, II and III, whereas Rpb4, Rpb7 and Rpb9 are subunits specific to pol II (Mitsuzawa and Ishihama 2004). The functions of most of these smaller subunits are not well characterized. Rpb4 and Rpb7 form a conserved heterodimer complex in archaebacteria, yeast, plants and humans (Choder 2004). Crystallographic studies of pol II and biochemical data provide evidence for a role of the Rpb4/7 heterodimer in transcription initiation and RNA binding in S. cerevisiae (Armache et al. 2005; Bushnell and Kornberg 2003; Edwards et al. 1991; Orlicky et al. 2001). S. pombe Rpb7 interacts with an RNA-binding protein, implying a role in the coupling of RNA processing to transcription, and it associates with glyceraldehyde-3-phosphate dehydrogenase and actin (Mitsuzawa et al. 2003, 2005). The in vivo functions of Rpb4 have been characterized in S. cerevisiae, and several lines of evidence suggest that it plays an important role in the response and survival to stress (reviewed by Choder 2004). S. cerevisiae Rpb4 is also important for activated transcription from a subset of promoters and in carbon and energy metabolism at moderate temperatures (Pillai et al. 2001, 2003). Recent data show that it is required for the decay of a class of mRNAs whose products are involved in protein synthesis, and it interacts with subunits of the mRNA decay complex Pat1/Lsm1-7 that enhances decapping (Lotan et al. 2005). We are only beginning to understand the role of the Rpb4 subunit in S. pombe: it binds to the TFIIF-interacting carboxyl-terminal domain (CTD) phosphatase Fcp1, and it has been proposed to play a role in the assembly of the Fcp1–pol II complex, thereby promoting CTD phosphorylation for the reutilization of pol II in a new cycle of transcription (Kimura et al. 2002). S. pombe Rpb4 exhibits features that resemble its budding yeast counterpart, while others place it closer to its orthologs in multicellular eukaryotes. It is essential for viability, whereas in S. cerevisiae it is not required for cell growth under optimal conditions (Sakurai et al. 1999; Choder 2004). S. pombe Rpb4 contains 135 amino acids, placing it closer in size to its counterparts in humans (142 amino acids) and plants (138 amino acids) than to S. cerevisiae (221 amino acids). Stoichiometric amounts of Rpb4 associate with pol II in S. pombe and higher eukaryotes, while the fraction of pol II molecules containing Rpb4 in optimally growing S. cerevisiae cells is only about 20%. Finally, Rpb4 is more tightly associated with S. pombe pol II than with S. cerevisiae pol II (Sakurai et al. 1999). Here, we provide evidence that Rpb4 influences the growth and gene expression of S. pombe cells in a dose-dependent fashion under optimal conditions. We show that lower levels of rpb4 expression do not affect the survival of cells under several stress conditions. Our results imply a particularly important role for Rpb4 in the expression and regulation of a subset of genes involved in cell separation. The defects in cell growth and separation as well as the gene expression signatures associated with low levels of rpb4 expression are similar to the phenotypes and expression signatures of a range of transcriptional regulatory mutants with defects in cell separation. Materials and methods Strains and molecular genetic methods The S. pombe strains used in this study are listed in Table 1. JB22 was the parental haploid strain used to generate the strains expressing rpb4 under the P3nmt1, P41nmt1 and P81nmt1 promoters (Basi et al. 1993). The strains were constructed by PCR using the pFA6a-kanMX6-P3/P41/P81nmt1 constructs as described (Bähler et al. 1998), and the kanamycin-resistant transformants were selected on KsNoT media (Linder et al. 2002). Recombinant DNA methods were performed as described (Sambrook et al. 1989) and standard methodology and media for the manipulation of S. pombe were used (Moreno et al. 1991). Table 1Strains used in this studyStrainRelevant genotypeSourceJB22972 h−Leupold (1970)JB394 kanMX6-p3nmt1-rpb4 h−This studyJB395KanMX6-p41nmt1-rpb4 h−This studyJB396kanMX6-p81nmt1-rpb4 h−This study Growth experiments Pre-cultures for the growth curves were made by inoculating the respective strains in Edinburgh Minimal Medium (EMM) without thiamine or with 15 μM thiamine and growing the cultures for ∼17 h at 32°C with shaking. Subsequently, the pre-culture was diluted into 50 ml of fresh EMM medium at an OD600 = 0.05–0.1 and allowed to grow at 32°C with shaking. Samples were removed at various timepoints thereafter for measuring ODs at 600 nm, microscopic examination, and microarray hybridization experiments. The low OD measurements in cells expressing rpb4 under the P41nmt1 and P81nmt1 promoters is not caused by the phenotype, because mutants with similar phenotypes such as sep1 show strong increases in OD during growth (unpublished data). Quantitative RT-PCR experiments To measure rpb4 expression, the strains expressing rpb4 under the three promoters were grown for 24 h in EMM medium at 32°C (promoter on) or cultivated for a further 21 h in the presence of 15 μM thiamine (promoter off). RNA was isolated by phenol extraction, purified using the RNeasy kit (Qiagen), and treated with RNase-free DNase (Roche). The reverse transcription reactions were performed using Superscript III (Invitrogen). Real-time PCR was performed using the Brilliant® SYBR® Green QPCR Core Reagent Kit (Stratagene) on an Mx3000P QPCR system (Stratagene) with primers specific for rpb4 or fba1 (control) (sequences available on request). Arbitrary expression units were calculated using a standard curve made from serial dilutions of fission yeast genomic DNA. Stress response assays Serial dilutions (1/10) of rpb4 strains generated in our study were spotted on EMM plates with or without 15 μM thiamine. The plates were supplemented with 0.5 mM hydrogen peroxide or different concentrations of sorbitol (1, 2, 3 and 4 M) to test the response to oxidative and osmotic stress, respectively. These plates were incubated at 32°C for 2 days and photographed. The effect of temperature stress was tested by incubating the strains at 36°C for 2 days. Microscopy Unfixed cells were observed at 2-h intervals for 25 h by light microscopy using a Zeiss Axioscope fluorescence microscope set up for differential interference contrast (DIC) with a 40× objective and Axiovision digital imaging system. To stain the nuclei, cells were spread onto microscopic glass slides. Subsequently, the cells were fixed by heating at 70°C for 1 min and DAPI (4′,6′ diamidino-2-phenylindole) was added at a final concentration of 1 μg/ml. To view the division septa, calcofluor was added to cultures at a final concentration of 5 mg/ml and the cultures were incubated in the dark at room temperature for 5 min. The cells were washed twice with phosphate buffer saline, and 5 μl of cell suspension was spotted onto microscopic glass slides. DAPI- and calcofluor-treated cells were visualized under a Zeiss Axioscope microscope as above. Microarray experiments To compare the gene expression profiles as a function of rpb4 expression, cells were grown in the presence of thiamine to OD600 of 0.2, and 50 ml of culture was centrifuged. The remaining culture was washed four to five times with EMM and subsequently grown in EMM medium without thiamine for 17 h and centrifuged. To identify transcripts affected by lowered rpb4 expression, the desired strains were grown in the presence of thiamine for 21 and 23 h and harvested. A wild-type parental strain grown in the presence of thiamine served as a reference. Although thiamine leads only to minimal expression changes (Jenkins et al. 2005), this experimental design corrects for effects caused by the presence of thiamine. Pellets were frozen in liquid nitrogen and kept at −70°C until further processing. RNA was isolated by phenol extraction and purified by the RNeasy kit (Qiagen). cDNA probes were prepared with Superscript (Invitrogen) and labeled with Cy3 or Cy5. Detailed methods were as described (Lyne et al. 2003; http://www.sanger.ac.uk/PostGenomics/S_pombe). Microarrays were scanned with a Genepix 4000B scanner and analyzed with Genepix software (Axon Instruments). Data filtering, normalization and quality control were performed as described (Lyne et al. 2003). The data are based on two to three independent biological repeats with dye swaps. To quantify global effects on gene expression, we repeated microarray analysis spiked with external control RNAs. The expression data were normalized using Bacillus subtilis spikes added in equal quantities to the RNA samples before labeling (Lee et al. 2005). Each microarray contains 720 control elements for the spikes, which are spread across the complete grid of the array in a spatially even manner. Clustering and visualization were done with GeneSpring. Gene annotations were taken from the S. pombe GeneDB database (www.genedb.org/genedb/pombe/index.jsp). Overlaps between different gene lists were calculated using the hypergeometric distribution. All normalized data sets are available from our website: http://www.sanger.ac.uk/PostGenomics/S_pombe Results Dosage-dependent effect of Rpb4 on cell growth Rpb4 is an essential subunit of S. pombe pol II. To elucidate its in vivo function(s), we constructed haploid strains expressing rpb4 from its normal chromosomal location using the thiamine-regulated nmt1 promoter. Three different versions of this promoter were used: the wild-type P3nmt1 promoter has the highest level of promoter activity and maintains substantial basal levels of expression even under repressed conditions; the two derivatives, P41nmt1 (medium strength) and P81nmt1 (weakest strength) have reduced levels of activity due to mutations in the TATA box (Basi et al. 1993). We performed quantitative PCR of cells expressing rpb4 under the three promoters, both in the absence (nmt1 promoter ‘on’) or presence (nmt1 promoter ‘off’) of thiamine to verify changing levels of rpb4 mRNA (Fig. 1). Expression levels of rpb4 showed a range of more than 1,000-fold in the different conditions, and the relative expression levels were as expected for the three promoters of different strength. The rpb4 levels were especially low when expressed from the P41nmt1 and P81nmt1 promoters in the presence of thiamine. The fba1 gene is highly expressed and was used as a control gene. Although it shows much less variation in expression levels than rpb4, there are some changes in accordance with rpb4 levels (Fig. 1). This probably reflects effects on global transcription as a function of rpb4 expression levels (see below). Fig. 1Expression levels of rpb4 mRNA when transcribed from different promoter constructs. Cells expressing rpb4 from the P3nmt1, P41nmt1 or P81nmt1 promoter were grown for 24 h in EMM medium at 32°C (promoter on; gray bars) or cultivated for a further 21 h in the presence of thiamine (promoter off; white bars). rpb4 (left) and fba1 (right; control) mRNA levels from the same cells were measured by quantitative RT-PCR. The PCR was performed on reverse transcription reactions, which were carried out in the presence (RT+) or absence (RT−) of reverse transcriptase To analyze effects of rpb4 expression levels, the three strains were grown at 32°C either in the absence or presence of thiamine, and cell growth was monitored. In the absence of thiamine, all strains grew similar to each other and to the wild-type strain (Fig. 2a). This shows that rpb4 expressed from the weak P81nmt1 promoter under induced conditions was sufficient to sustain normal cell growth. When these strains were grown in the presence of thiamine, however, there was a direct correlation between residual promoter activity and growth rate: cells expressing rpb4 from the P3nmt1 promoter showed the same growth rate as wild-type cells, while cells expressing rpb4 from the P41nmt1 promoter and, even more so, from the P81nmt1 promoter showed reduced growth rates (Fig. 2b). Thus, in the presence of thiamine, the lowered Rpb4 levels become limiting for cell growth, and residual growth directly reflects basal expression levels from the different promoter constructs. Fig. 2Effect of Rpb4 on cell growth. Wild-type cells (filled circle) and cells expressing rpb4 from either the P3nmt1 (filled square), P41nmt1 (filled triangle) or P81nmt1 (crosses) promoter were grown in EMM medium at 32°C. OD600 was measured at the indicated times. a In the absence of thiamine, all strains grew at similar rates. b In the presence of thiamine, the growth rates were dependent on the residual levels of rpb4 being expressed from the different nmt1 promoters Effect of Rpb4 on cell growth under stress conditions Rpb4 is required for stress response in S. cerevisiae (Choder 2004). We therefore, determined if Rpb4 played a similar role under stress conditions in S. pombe. Cells expressing varying levels of rpb4 were subjected to temperature stress (36°C), osmotic stress (1 M sorbitol) or oxidative stress (0.5 mM hydrogen peroxide). We found that neither in the presence nor absence of thiamine, the cells exhibited any increased stress sensitivity compared to control cells when grown under the different stress conditions (data not shown). Similar results were obtained when cells were grown in up to 2 M sorbitol. No significant growth of either the wild-type or Rpb4 expressing cells was observed in the presence or absence of thiamine at 3 and 4 M concentrations of sorbitol. These observations indicate that low levels of Rpb4 do not affect the survival of cells under any of the tested stress conditions. Repression of rpb4 transcription causes defects in cell separation We next examined whether different levels of rpb4 expression would influence specific cellular processes. Cells expressing rpb4 from any of the three nmt1 promoters in the absence of thiamine did not lead to any obvious phenotypes even after 25 h of growth in medium without thiamine (Fig. 3a). Fig. 3Rpb4 affects cell separation and morphology. Cells expressing rpb4 from either the P3nmt1 or P81nmt1 promoter were analyzed by microscopy. a Cells grown in EMM in the absence of thiamine (promoter ‘on’) at 32°C appear like normal, wild-type cells. b When grown in the presence of thiamine (promoter ‘off’), cells expressing rpb4 from the P3nmt1 promoter appear normal, while those expressing rpb4 under the P81nmt1 promoter display defects in cell morphology and septation. Multiple phenotypes including elongated cells with single or multiple septa and growth by branching are evident. c Cells expressing rpb4 under the P81nmt1 promoter were analyzed by fluorescence microscopy in the presence of thiamine. DAPI staining revealed single nuclei in each cell compartment, while calcofluor staining showed septa (some of them aberrant) separating the cell compartments. d Control cells and mutants showing similar phenotypes as cells expressing rpb4 under the P81nmt1 promoter cells. From left to right: wild-type cells, sep1 deletion mutants, ace2 deletion mutants and sep1 ace2 double deletion mutants In thiamine-containing medium, however, cells expressing rpb4 from the weak P81nmt1 promoter became more elongated than the wild-type cells after 19 h of growth (Fig. 3b). Moreover, after continued incubation in the same medium for another 2 h (21 h with thiamine), cells also showed defects in cell separation, which became more evident after an additional 4 h of incubation (25 h with thiamine) (Fig. 3b). Most cells were elongated with either a single or multiple septa, and some cells also displayed bent morphologies and/or growth by branching. Cells expressing rpb4 from the medium strength P41nmt1 promoter displayed no obvious cell separation defects until 23 h of growth in the presence of thiamine. Similar defects as described above, however, became apparent after 23 h of growth in thiamine (data not shown). In contrast, cells expressing rpb4 from the strong P3nmt1 promoter did not exhibit any aberrations in morphology or division after growth in thiamine-containing medium for 25 h (Fig. 3b) and continued to look like wild-type cells even at 40 h of growth in thiamine-containing medium (data not shown). Thus, the basal level of expression from the strongest promoter is sufficient to prevent the cell separation defects seen with the weaker promoters. We investigated these phenotypes in more detail for cells expressing rpb4 under the P81nmt1 promoter by staining the division septa and cell nuclei (Fig. 3c). Several nuclei were present separated from each other by either a single septum or multiple, aberrant septa. Some cells also contained large deposits of septal material at random positions (Fig. 3c). We conclude that lowered rpb4 expression levels lead to filamentous growth with multiple cells remaining attached to each other. Similar phenotypes to those of cells expressing rpb4 under the P81nmt1 promoter have observed in several different mutants (see Discussion); for example, deletions in two transcription factor genes, sep1 and ace2, also lead to defects in cell separation (Fig. 3d; Bähler 2005). These findings raise the possibility that Rpb4 plays a specific role in cell separation after cytokinesis. Low levels of rpb4 transcripts most strongly affect expression of genes involved in septation We used S. pombe whole-genome microarrays to investigate the effects on gene expression caused by low levels of rpb4. Wild-type cells and cells expressing rpb4 under the different regulatable promoters were grown in the presence and absence of thiamine followed by RNA extraction. Cells expressing rpb4 under the P3nmt1 promoter did not show any significant differences in gene expression profiles when compared with wild-type cells, either in the presence or absence of thiamine (data not shown), consistent with our earlier observations that these cells show normal growth and phenotypes. If rpb4 was expressed under the weakest promoter in the presence of thiamine, however, gene expression changed substantially. As down-regulation of rpb4 is expected to have a global effect on transcription, we used external spikes for normalization (Materials and methods). This revealed that mean transcript levels were 26–36% lower compared to wild-type cells grown under the same conditions (Fig. 4a). The expression levels of rpb4 were two- to threefold lower under these conditions compared to the levels under its own promoter. Thus, lowered rpb4 expression leads to reduced overall transcript levels. Fig. 4Effects on global and specific gene expression as a function of rpb4 levels. a Microarray experiments of cells expressing rpb4 under the weakest promoter. The histograms show the distribution of gene expression ratios of P81nmt1-rpb4 cells grown in the presence of thiamine for 21 and 23 h relative to wild-type cells grown in the presence of thiamine. Genes in both histograms are colored by their relative expression levels at 21 h as indicated at upper right. External spikes have been used for normalization, revealing a global down-regulation of transcript levels by 26% (21 h) to 36% (23 h) on average when rbp4 expression is lowered. The averages of two independent repeats for each timepoint are shown. b Overlap between the genes downregulated when rpb4 is expressed under the weakest promoter in the presence of thiamine and genes downregulated in both sep10 and sep15 mutants (Lee et al. 2005). The number in brackets represents the overlap expected by chance, given the sizes of the gene sets considered and the total number of 5,180 genes. The overlap is highly significant as calculated using the hypergeometric distribution (P ∼ 1 × 10−30). c Hierarchical cluster analysis of 249 genes that are either 1.5-fold upregulated or threefold downregulated when rbp4 is expressed under the weakest promoter in presence of thiamine. Horizontal rows represent the profiles of hierarchically clustered genes. Columns represent experiments of different mutant strains and different time-points for the same strain. The transcript levels for each strain (indicated at bottom) relative to levels in wild-type cells are color-coded as indicated at right, and missing data are shown in gray. The microarray data from mcs6, pmh1, sep10 and sep15 mutants are from Lee et al. (2005), and the data from sep1 and ace2 mutants are from Rustici et al. (2004) On top of this genome-wide down-regulation of transcription, 96 transcripts were at least 1.5-fold upregulated in three of the four samples tested, while 150 genes were at least threefold downregulated in three of the four samples tested upon lowering rpb4 expression (Supplementary Tables S1 and S2). These genes were categorized according to their functions (Supplementary Table S3). The majority of the genes whose expression levels were affected by rpb4, encoded proteins involved in metabolism and transport. Many of the upregulated transcripts in cells with lowered rpb4 expression levels overlapped with genes upregulated during different environmental stresses (Chen et al. 2003) and nitrogen starvation (Mata et al. 2002) (P ∼ 1 × 10−13 to 1 × 10−36). This probably reflects an indirect effect caused by the defects in these cells. A substantial number of the downregulated genes in the category of cell cycle and cell division played specific roles in cytokinesis and cell separation. For example, the two genes eng1 and agn1 encode β-glucanase and α-glucanase, respectively; Eng1 degrades the primary division septum between the new ends of daughter cells and Agn1 hydrolyzes the old cell wall surrounding the septum leading to full separation of daughter cells (Martín-Cuadrado et al. 2003; Dekker et al. 2004). The genes adg1 and adg3 have also been shown to be involved in cell separation (Alonso-Nunez et al. 2005), and rng3 encodes a protein involved in formation of contractile ring during cytokinesis (Wong et al. 2000). Both the lists of genes being upregulated and downregulated after rpb4 down-regulation showed highly significant overlaps with lists of genes upregulated or downregulated, respectively, in sep10, sep15, mcs6 and pmh1 mutants (P ∼ 1 × 10−16 to 1 × 10−47) (Lee et al. 2005), whose genes encode different proteins involved in transcriptional regulation (see Discussion). Among the downregulated genes, the most pronounced overlap was observed with sep10 and sep15 mutants (Fig. 4b). The downregulated genes also showed some overlap with genes downregulated in sep1 and ace2 mutants (P ∼ 1 × 10−14) (Rustici et al. 2004), which encode cell-cycle transcription factors. Figure 4c shows a cluster analysis of the genes most strongly regulated by rpb4 repression and their expression levels in the various transcriptional mutant backgrounds. The sep10, sep15, mcs6 and pmh1 mutants show highly similar expression signatures to cells with compromised rpb4 expression, while the overlap with sep1 and ace2 mutants is less evident. Intriguingly, if rpb4 was overexpressed using the medium strength promoter, the expression levels of these genes were reversed, with downregulated genes becoming upregulated and upregulated genes becoming downregulated (Fig. 4c). Thus, transcript levels of this gene set are particularly responsive to the transcript levels of rpb4. Discussion We have analyzed functions of the Rpb4 subunit of pol II in S. pombe. Despite the overall conservation in structure and function of pol II across species, the Rpb4 subunit presents a special case. S. pombe Rpb4 is similar to its orthologs in higher eukaryotes with respect to its structure and essential nature (Sakurai et al. 1999), but in contrast to its human ortholog, it can form stable hybrid dimers with S. cerevisiae Rpb7 (Sakurai et al. 1999) and can rescue the temperature-sensitive phenotype associated with the lack of S. cerevisiae Rpb4 (Shpakovski et al. 2000). Thus, S. pombe Rpb4 may have intermediate characteristics useful to bridge the knowledge from S. cerevisiae and multicellular eukaryotes. We used regulatable gene expression constructs to study the in vivo roles of Rpb4. Under optimal conditions, rpb4 influences the growth of cells in a dose-dependent manner when expression drops below a certain level using the regulatable promoters (Figs. 1, 2). In S. cerevisiae, deletion of the RPB4 gene leads to slow cell growth under optimal conditions (Woychik and Young 1989; Choder 2004), while in S. pombe rpb4 is essential (Sakurai et al. 1999). We found that low levels of rpb4 expressed from the weakest promoter were sufficient for the cells to survive a range of different stress conditions including temperature stress. Accordingly, core stress genes were actually upregulated in rpb4 mutants, and S. pombe Rpb4 is therefore unlikely to play any specialized role during environmental stress. S. cerevisiae Rpb4 is dispensable for oxidative and osmotic stress (Maillet et al. 1999), but unlike in S. pombe, it is required for survival of temperature stress (Woychik and Young 1989). Fission yeast cells grown under low expression of rpb4 were elongated and showed defects in cell separation as indicated by the accumulation of division septa, some of them highly aberrant (Fig. 3). Strains with cell separation defects similar to the ones described here include mutants in components of the exocyst complex (sec6, sec8, sec10 and exo70), an anillin homolog (mid2), septins (spn3, spn4), an endo-α-1,3-glucanase (agn1), an endo-β-1,3-glucanase (eng1), calcineurin (ppb1), a MAP kinase and phosphatase (pmk1, pmp1), two transcription factors (sep1, ace2), and subunits of the mediator complex (sep10, sep11 and sep15) (Sipiczki et al.1993; Yoshida et al. 1994; Longtine et al. 1996; Toda et al.1996; Sugiura et al. 1998; Grallert et al. 1999; Wang et al. 2002; Martín-Cuadrado et al. 2003; Spåhr et al. 2003; Tasto et al. 2003; Dekker et al. 2004). In S. cerevisiae, haploid rpb4 deletion mutants exhibit an axial budding pattern with normal yeast cell morphology, but diploid rpb4 mutants are more elongated than wild-type cells during nitrogen starvation and grow in a unipolar pattern, resembling pseudohyphae formation (Pillai et al.2003). Lowered expression of rpb4 causes relatively modest reductions in global transcription, which probably leads to the reduced growth rates under these conditions (Fig. 4a). In addition, low rpb4 transcription results in substantially lowered expression of a distinct subset of genes. Among the strongly affected genes were eng1, agn1, adg1 and adg3 with known functions in cell separation, which are targets of a transcriptional cascade regulated by the Sep1 and Ace2 transcription factors (Rustici et al. 2004; Alonso-Nunez et al. 2005). Accordingly, the genes affected by lowered rpb4 expression significantly overlapped with genes affected in sep1 and ace2 mutants. Much stronger similarities, however, were evident with the gene expression signatures of several strains mutated in the sep10, sep15, mcs6 and pmh1 genes, which encode various components of the transcriptional machinery. Sep10 and Sep15 are essential components of the mediator complex of pol II, while Mcs6 and Pmh1 are components of a complex homologous to metazoan Cdk7-cyclinH-Mat1, a cyclin-dependent kinase-activating kinase (CAK) and a pol II CTD kinase associated with transcription factor IIH (Spåhr et al. 2001, 2003; Lee et al. 2005). The overlap in expression signatures between these four mutants and cells with lowered rpb4 expression was evident for both upregulated and downregulated genes. These differentially regulated genes showed reverse regulation (upregulated genes became downregulated and vice versa) when rpb4 was overexpressed from the medium strength promoter (Fig. 4c). Thus, the expression of this gene set is particularly responsive to rpb4 expression levels and also depends on a range of other factors involved in general transcription. This could either reflect a specific transcriptional pathway involving Rpb4 and the other factors (either directly or indirectly via the control of a regulator), or the affected genes could be generally sensitive to transcriptional interference. In any case, these data show that the transcriptional pathway required for cell separation in fission yeast (Bähler 2005) is strongly affected when interfering with rpb4 expression levels. We speculate that the Rpb4/Rpb7 complex is involved in linking the transcriptional output to the metabolic status of the cell. The rbp4 and rpb7 transcripts are down-regulated when cells enter stationary phase (unpublished observations), Rpb7 interacts with the glycolytic enzyme glyceraldehyde-3-phosphate dehydrogenase (Mitsuzawa et al. 2005), and global transcription is sensitive to rpb4 expression levels. It is possible that the transcriptional pathway required for cell separation is particularly sensitive for compromised transcription to adjust the growth mode to availability of nutrients: in abundant nutrients, global transcription is efficient and growth is best as single cells, while in limiting nutrients, global transcription and cell separation are compromised, and cells grow as multicellular pseudohyphae, which may allow a more efficient grazing for new nutrients as growth is directed. In the first description of S. pombe, multi-cellular pseudohyphae similar to those seen in rpb4 mutants have been observed (Lindner 1893); the natural capacity to switch to multi-cellular growth might have been lost in the yeast strains cultured in the laboratory. Taken together our data suggest that S. pombe Rpb4, besides its general role in transcription, performs a more specialized function in regulating a specific gene expression program required for cell separation at the end of the cell cycle. It is possible that this role is carried out by interacting with other general factors of the transcriptional machinery such as the mediator and CTD kinase as suggested by similarities in phenotypes and expression signatures in our study. Accordingly, a comparison of the X-ray crystallographic structure of the complete pol II, including the Rpb4/7 heterodimer, with an electron microscopic structure of the pol II-mediator complex, has indicated that Rpb4/7 may interact with the mediator (Bushnell and Kornberg 2003). Alternatively, the subset of genes affected by changes in rpb4 expression (and by manipulating other general transcription factors) could reflect genes that are especially sensitive to transcriptional stress. This interpretation does not imply any specific gene expression program, although it could be used by the cell to adjust transcription to available nutrients. Given the conservation of Rpb4, it will be interesting to see whether the human ortholog plays a similar specialized role to transcribe particular subsets of genes. Electronic supplementary material Supplementary material

Health_Care_Anal-3-1-2045690

Information Rx: Prescribing Good Consumerism and Responsible Citizenship

Recent medical informatics and sociological literature has painted the image of a new type of patient—one that is reflexive and informed, with highly specified information needs and perceptions, as well as highly developed skills and tactics for acquiring information. Patients have been re-named “reflexive consumers.” At the same time, literature about the questionable reliability of web-based information has suggested the need to create both user tools that have pre-selected information and special guidelines for individuals to use to check the individual characteristics of the information they encounter. In this article, we examine suggestions that individuals must be assisted in developing skills for “reflexive consumerism” and what these particular skills should be. Using two types of data (discursive data from websites and promotional items, and supplementary data from interviews and ethnographic observations carried out with those working to sustain these initiatives), we examine how users are directly addressed and discussed. We argue that these initiatives prescribe skills and practices that extend beyond finding and assessing information on the internet and demonstrate that they include ideals of consumerism and citizenship. Introduction In discussions about the increasing use of internet technologies in health care a new kind of patient has slowly been constructed—a patient that is reflexive and informed, with highly specified information needs and perceptions, as well as highly developed skills and tactics for acquiring information. This specific depiction of the patient transforms the patient into an active participant in his or her care, “empowered” through training, skill development and the use of technologies such as the world wide web, all of which can be provided or enabled by government organizations or other political actors. Through increased access to medical information coming from outside of the health care arena the boundary between lay and expert is breaking down, enabling patients to become experts, be more assertive in managing their own care and change the nature of their existing relationships with health care professionals [6, 27, 28]. As Henwood et al. [31] point out, however, just because the potential for empowerment exists, it does not necessarily occur. Patients conform only with difficulty to the images associated with the reflexive consumer, and the concept of the “informed patient” is empirically difficult to sustain [41]. Becoming informed requires skills related both to information and to the various media that can be used to access that information. Patients reflect low-level skills with respect to searching for information, rely on intermediaries, and experience concrete barriers during the medical encounter, all of which act as actual constraints on the emergence of informed patients [31, 32]. In literature about the questionable reliability of web-based information, the suggestion that lay information seekers are constrained in conforming to this image is too mild. This body of literature asserts the stronger position that they are actually prevented from being empowered because of the large amount of information and the questionable quality of much of that information. Central in this discussion is the concern that even if patients manage to find web-based medical information, they lack the skills necessary to evaluate the quality of that information, and therefore are especially vulnerable to harmful information in the form of errors, fraud and “quackery” [16, 17, 34, 35, 50, 51]. This concern is answered by the argument that a pressing need exists for educating patients in how to judge a website’s reliability [10, 18, 19, 23, 33, 48, 50, 54]. Taken together, these discourses construct patients-as-web-users that are simultaneously skilled and capable, but also incompetent and unskilled. The informed patient must be educated in how to search for and assess information in order to conform to the definition of a reflexive, empowered consumer. Patients are not empowered merely through access to the internet or web-based information [1, 49], but, this literature suggests, must be constructed in the process of being led to pre-selected information by health professionals and information specialists. In suggesting the need to create user tools that have pre-selected information, as well as guidelines for patients to use to check the individual characteristics (author, sources, date, underlying financial sponsorship, etc.) of the information they encounter, authors emphasize the boundary between lay and expert assessments of information, privileging the latter over the former. Existing literature prescribes that patients utilize government-provided medical/health portals, click on hyperlinked icons (seals of approval or “trust marks”) provided by non-profit or non-governmental organizations, follow checklists created by professionals or health educators, and/or download special toolbars, all of which will assist them in finding and evaluating information on the web. Because web pages provide powerful examples of pre-defining (or pre-confining) how information technologies should work, how they should be perceived and how different actors should utilize them, it is important to study the discourses that they carry [43]. However, little attention has been given to the underlying prescriptions about skills and use, as well as additional political agendas and messages about individual behavior, that individual web-based reliability initiatives, such as portals, seals and special toolbars, convey. In this article, we look not at how patients assess information (or what skills they do/not reflect), but rather, at what health educators, medical professionals and review organizations suggest the skills of a reflexive consumer should be.1 We view the various user tools, such as guidelines, checklists portals, and clickable seals, as artifacts with politics [56]—particularly, the politics of building consumer-citizens and shaping their skills, perceptions and behaviors. We begin by looking at more general literature about the construction of users and designers, followed by a review of literature that transforms potentially informed patients into reflexive consumers. Using two types of data: “front page” data (information, images, quotes gleaned directly from websites and their accompanying promotional items) and supplementary “behind the scenes” data from interviews and observations carried out with those working to sustain given initiatives, we examine how users are directly addressed and discussed. How are users enrolled by sites and promotional items? What ideals are represented in these user tools and in what kind of practices are they embedded? We argue that, although these examples claim to target “all citizens,” the prescriptions for action that these user tools carry suggest that the user envisioned by the developers of these tools is not everybody, but rather a specific, ideal type of user: the good consumer/ responsible citizen.2 In order to develop skills for finding and assessing information on the internet, patients must engage in certain practices—practices that also enable developing the skills necessary to share the responsibility for information reliability and to be an empowered health consumer and responsible citizen. The Importance of the State in Configuring Users of Technologies Studies of science, technology and society (STS) have a long tradition of reflecting on how users are or are not included in the design and implementation of different technologies.3 Woolgar [58] argues that designer preconceptions about use can shape what counts as legitimate behavior and that users are “configured,” i.e.: their identity and skills are defined and constraints upon their (possible, future) actions are set, before the technologies ever reach the anticipated user group. Because of this configuring, technologies can be read as carrying specific scripts for use [5]. Rose and Blume [52], however, have criticized much of the subsequent work on configuring users because it attributes configuration largely to market-driven design and pays too little attention to the possibility of states as providers or enablers of technology use, and therefore neglects to consider the potential role states have in user configuration. According to Rose and Blume, focusing on technologies that are developed or facilitated by the state and its institutions may highlight significant tensions between individuals as users and the state of which they are members and citizens. Policy documents can develop meanings for information technologies by providing, for example, the language of discourse about those technologies [38]. In their behavior and discourse, collective providers, much like market-like providers, presume that individuals will be active consumers of technologies, meaning that the state may enact policies that, while not overtly or explicitly configuring the user, do create or maintain an environment that helps to shape how users are configured. In using certain technologies (or by using them in the specified ways), individuals fit with their configurations and follow the technologies’ scripts and actualize their potential as “good” citizens [52]. Singleton [53] makes a similar point with her suggestion that the New Public Health in the UK seeks to train citizens in more ways than one. She asserts that training programs for healthy citizens are not just about physical health but also about practical skills and about distinguishing between active and inactive citizens. Klecun-Dabrowska and Cornford [38] have also looked at the recent discourses on health in the UK, specifically focusing on the role of telemedicine within wider national and international debates on health. They, too, saw that documents convey a vision of a refocused health care system to serve a population that has enough information resources to enable individuals to manage their own care and participating more actively in the health care process. Information, they argue, is transformed into something that is easily captured, understood and transformed to achieve wider goals—goals related to social responsibility and community values. At the same time, states want to forward their political agendas and invent new modes of indirect steering for empowering their members such that they freely, willingly and self-reflexively choose the paths toward the desired outcomes of the state. The possibility of the state to directly influence its citizens is made relative, as the state competes with other types of influences. This has led to the development of new types of steering, where governments seek to rule the social indirectly through designing, facilitating and moderating processes of self- and co-governance [8]. More communicative and cooperative modes of re-centering allow for bottom up articulation, without relinquishing everything to citizen preferences. Again, these programs are not just about the empowerment of individuals or about the distribution of expertise, but are also about distinguishing between active and inactive citizens. Although states demonstrate a readiness to engage in active teamwork and argue the need for collaborative solutions, Bang argues that they nonetheless reveal how key relationships are re-imagined and redrawn. Two things are worthy of note: first, the terms “re-imagining” and “redrawing” suggest that something new can emerge, which is a distinct possibility. However, with respect to the internet, at least, studies have shown that the boundaries that are constructed around information, for example, tend to reinforce existing social ideas [49] and geographic borders [26]. The lines that are redrawn serve only to reinforce what is already there. Second, the exclusive focus on states tends to neglect the role of other political actors. Especially in activities regarding reliable medical information, states are not alone, but are joined by non-governmental organizations (NGOs) and inter-governmental organizations (IGOs), as well as non-profit organizations voicing their own political advocacy agendas. We extend the analysis in this paper to look at both programs from federal agencies and initiatives created by other types of political institutions. These programs and initiatives provide tools for accessing and assessing information on the web, in some cases enabling information provision. (See Textbox 1 for an overview of the different initiatives that are used as examples).4 For example, the US Department of Health and Human Services provides a portal with contact information for organizations (http://www.healthfinder.gov) and the US National Library of Medicine (NLM) provides a portal with health content from the National Institutes of Health (http://www.medlineplus.gov). Outside of the US, the Health on the Net Foundation (HON) in Geneva seeks to raise awareness about ethical issues related to providing medical information, while the World Health Organization (WHO) proposes the creation of a “.health” domain afforded only to sites meeting specific criteria.5, 6Textbox 1Overview of home pages for cited examplesHealthfinder is the official Federal Gateway to health information, provided by the US Department of Health and Human Services http://www.healthfinder.govMedlineplus is a portal provided by the US National Library of Medicine http://www.medlineplus.govQuackwatch seeks to combat health fraud and is run by Dr. Stephen Barrett http://www.quackwatch.orgHealth on the Net is a Geneva-based non-governmental organization that provides diverse user tools. The focus of this study is the 8 principle HON Code of Conduct and accompanying hyperlinked seal http://www.hon.chMedCERTAIN was funded by the EU from 2000-2002under the “Action Plan on promoting safer use of the Internet by combating illegal and harmful content on global networks”. The collaborative MedCIRCLE is a follow-up that provides users with a downloadable toolbar http://www.medcertain.org; http://www.medcircle.orgTNO’s QMIC is a three-tiered trust mark for medical websites in the Netherlands. The third tier signifies content review of information http://www.qmic.nl/qmic/home.doThe World Health Organization recently released a list of approved sites for vaccine safety http://www.who.int/immunization_safety/safety_quality/approved_vaccine_safety_websites/en/ These self-ascribed roles lead to numerous questions, such as why UN NGO status (in the case of HON) or a “neutral position” (claimed by the WHO) is crucial to building claims about the reliability of medical information. Why is a ‘.gov’ domain “more reliable” than a ‘.com’ and where does the ‘.org’ domain rank in relation to other existing and proposed domains? What potential tensions arise between states and citizens (or even NGOs and citizens) in these settings? Although these questions would be difficult to answer, they do give us cause to think about the political messages that are sent to lay information seekers when they are instructed on how to search for, evaluate and use web-based medical information. Informed Patients...Reflexive Consumers...Ideal Citizens? This entire line of discourse about empowering patients and the role of web-based information has led to the semantic challenge of properly naming those non-medically trained individuals searching for information online. For example, the term patient does not encapsulate those persons who search for information regarding the health situation of a family member or friend. Miller and Reents’ [44] alternative, “information retrievers”, makes the user too passive, while the suggested informatics alternative “medical end-users,” [22] makes the route to the information too technology specific, without reflecting the social aspects of information use and broader information “landscapes” [13, 29, 32] that persons can access. Furthermore, the term medical does not reflect broader issues related to health and fails to allow for use of information by those that Kivits [36] calls “healthy internet users,” those not necessarily afflicted by illness, but nonetheless interested in health information. The concepts “citizen” and “consumer” are terms for users that are used most often by those creating different web-based reliability initiatives, working under the suggestion that these terms are neutral and avoid the semantic problems mentioned above [see, for example, 7, 9, 45]. However, these words also carry connotations; Anderson et al. even go so far as to state, “When speaking about consumers, from this model it should become clear that we mainly refer to educated patients with chronic diseases from developed countries” (2003: 72). Using these two terms more generally implies certain rights, such as Gustafson and Wyatt’s [25] assertion that consumers deserve both high quality content and a certain degree of confidence in the information they use, but such use also connotes certain responsibilities. Each of the initiatives to which the usage of these terms is attached defines particular types of participation that extend beyond gathering information to include activities in the health care process and social community at large. The consequence, of course, is that teaching patients how to search for and assess information on the web becomes more than just a process of constructing empowered patients—it extends further to inscribe notions of good consumerism and responsible citizenship. When we couple the discussion about empowerment on the practical skills that users are expected to develop, we see that it takes place within an overt and dominant biomedical discourse [24], which is framed not only by medical professionals, but also by political actors. For example, in relation to assessing web-based health information, Edgar et al. [15] describe three essential skills: the ability to conduct a search and find the “right” sites; the ability to judge the quality of information found on a given site and the ability to synthesize that information into a useful context for personal/individual health. Eng and Gustafson [16] argue, however, that the skill is more than just assessing and using information, but is actually about deciding which of the existing tools works best for the individual. Only in finding the right tool can s/he implement a personal evaluative framework and learn how to be an educated consumer. How the Initiatives Enroll Citizens and Consumers Each of the initiatives discussed in this paper attempts to enroll internet users in specific activities of finding and assessing health information. The most prominent strategy of enrollment that we see is the suggestion of abounding risks to personal health and the simple ways to combat them. Involvement—learning how to check information actively and always—becomes a matter of personal responsibility and an identifying component of reflexive information consumption. Failure to learn about user tools and use them as prescribed is to be inactive, and therefore, deviant or lacking [52, 53]. Highlighting Risks and Providing Simple Instructions for Combating Them One evident strategy for interesting patients in the need to check the reliability of the information is the construction of web-based information as information that puts individuals at risk. Because of the nature of the web, information can come from anyone, anywhere—it originates outside of a given country, in another medical tradition or health care context and as such it is foreign and worthy of suspicion. If information isn’t pre-approved, individuals could be making their decisions based on wrong information, fraud, or quackery. An individual’s health is potentially endangered because this information is always potentially just one mouse-click away and is difficult to distinguish from “truly reliable” information. In the language of this discourse, boundaries are constructed around information—these boundaries distinguish, for example, between geographic locations, but also reinforce more traditional ideas by distinguishing between lay persons and experts. The proposed need to implement user education programs for checking the reliability of the information they encounter suggests that anyone online should be aware of this potential for harm and, therefore, actively involved in policing that information [21, 35]. HON, for example, currently includes information on its website about being a “vigilant user”. Instructions about clicking and verifying proper use of its HONcode icon is prefaced by the following:Unfortunately, we cannot banish incompetence or fraud from the medical Internet. If you come across a healthcare Web site that you believe is either possibly or blatantly fraudulent and does NOT display the HONcode, please alert Quackwatch. Of course, if such a site DOES display the HONcode, alert us immediately. HON cannot prevent dishonest operators from simply cutting and pasting the HONcode seal onto their Web sites in a bid to enhance their credibility. We do conduct our own random checks on subscribers to ensure they remain compliant with the HONcode. But we also rely heavily on vigilant Web surfers to alert us to dubious sites—and they do. ...There are three quick ways users can check whether a chosen site featuring our seal is a bona fide HONcode subscriber.7 HON points out to users that it needs its users to help police information on the internet. It invites them to join in a partnership with two types of authorities, the first being itself, an overarching political organization and the second being participants from the community of medical professionals. Active policing on the part of the user is made extremely simple and practically effortless—checking information is also as easy as a single mouse click (merely clicking on a small icon) or just running through a short checklist.8 Taking a few seconds can prevent all users from encountering bad information, whereas not checking could subject the user to fraudulent or otherwise harmful information. With its clickable trust mark9 HON provides users with an easy tool to double check the background of the web information provider. This type of tool makes the action of checking information is made simple and non-time consuming—as easy as a mouse click. The MedCERTAIN project, which proposed a mechanism similar to HON’s clickable seal, except that it proposed to include content review in addition to review of ethical principles, produced a prototype with the instruction, ‘Remember to verify by simply clicking on it.’10 Similarly, the Dutch QMIC trust mark states, ‘In the blink of an eye, you can see that information is reliable and correct.’11 The user that all of these organizations presuppose is someone who, like the project reviewers, is (or should be) familiar with the ethical issues underlying the provision of any, but especially medical, information on the web. We begin to see how political actors invent new modes of indirect steering for empowering users such that they choose specific paths. They attribute to these users skills and practices more in keeping with professional levels of information assessment and action. But they also make these practices “user friendly” by emphasizing the ease with which information can be checked, with little additional thought or demands on time, and by encouraging participation in partnerships with them. Directing Paths Toward Information The various initiatives also emphasize their links to the medical community and partnerships with one another. This is evident in at least three forms: logos from other departments and organizations placed on websites, joint projects with physician’s organizations and overt statements about the best course of action that are located on websites. At the time of writing, HON’s website includes the logos of the University Hospital, the EU, the city of Geneva and Sun Microsystems. The tag lines for each individual logo identify how these different groups provide social and technical support for HON and its web activities. These logos show that HON is part of a larger national and international community that transcends medical practice, politics and businesses—and suggest that evaluating medical information on the web can only be done in this context. Some studies of patient searching behaviors have indicated that patients refer to their physician for information about where to search for information on the web. [See, for example, Cotton and Gupta 12] Recognizing the potential of taking advantages of this preference for physicians as trusted sources of website recommendations, but also recognizing that physicians probably do not have time to review scales of sites, some initiatives have developed (or furthered existing) relationships with professionals. The US National Library of Medicine (NLM) has worked with professional communities to develop prescription pads for web-based information (see Fig. 1). Fig. 1Example of one version of the prescription pad that physicians can give to patients during the consultation to refer them to web-based information about a specific health topic This project enabled NLM to inform physicians about the topics available on medlineplus. Additionally, it supports the physician and patient in post-consultation searches for health information. With this information prescription, the physician gives the patient a place to start and the term to use when searching on the web for information specific to his or her health situation—and that starting place is a government portal:In this trial, the doctors each had their own pre-printed prescription pad. The prescription pad that we used in the first part of the pilot had room for 12 URLs. There were about 40 to choose from and they chose the ones that they saw most often in their practices. What we found is that the doctors don’t want to bother with all that. They want to have their own prescription pad, they really like that. But they just want to be able to write something on it and then send people to Medlineplus.12 Finally, websites send patients back and forth between the various initiatives and tools, implicitly and explicitly endorsing one another. HON, for example, recommends that English-speaking site visitors follow a specific and narrow path to information:To find good (English-language) healthcare information, you can bypass the all-purpose commercial search engines and go straight to healthcare portals like Health on the Net (http://www.hon.ch) or Healthfinder (http://www.healthfinder.org/). These portals have already eliminated the irrelevant for you. A useful rule of thumb is first check out the governmental, not-for-profit and hospital Web sites, or those carrying an immediately recognizable and trusted name.13 It is interesting in this case that HON excludes more commercial search engines as a valid route to information. The message is that web tools that are general in scope, rather than specifically directed toward health information, or that are funded by commercial interests (or both) are incorrect choices because they contain too much extra “stuff”14 and will not enable the user to find what s/he is looking for. Reliability is redefined as an issue of what is “relevant” and proper behavior involves not wasting time sifting through information and other materials that are, according to HON de facto irrelevant. Furthermore, in emphasizing that HON and the US gateway are the best types of sites, HON privileges state (implicitly working with medical) actors over other types of actors providing medical information on the web. The suggestion, then, is that engaging in good searching practices and finding/using the “right” information means prioritizing government and medical channels. Through these statements, the organizations affirm each other, arguing not only the importance of a joint approach to reliability, but also the importance of collaboration between different countries and regions. The risks associated with misinformation are countered with initiatives that seek to rein in information, centralizing it at certain web addresses and judging it according to politically and professionally defined criteria. One example would be the criterion for levels of readability [16] applied to sites such as Medlineplus. Another would be the use of selection criteria for information that derive from similar roots [9] or the transfer of existing publishing standards to evaluate how information is produced [1]. Website providers enable access to information that is consolidated and repackaged in a uniform format for readability and then placed on the respective site under a uniform design and structure for ease of navigation. Once again finding and using good information is made “easy” in the design of technology-specific tools. What is interesting are the personalized messages and rhetoric of choice in which these standardized formats are embedded. Healthfinder is “your guide to reliable information,” while Medlineplus provides “Trusted health information for you.” Access to information gives users new choices, but these initiatives convey that responsible consumerism and reflexive use of information mean choosing narrow routes to prepackaged information. The internet offers numerous avenues to information, but “good” participation means utilizing choice to choose the routes provided by one’s own government and combining these with overarching non-governmental organizations or other non-profit organizations. Individualizing Choice Along the Way Both HON and the MedCIRCLE initiative have taken this one step further—emphasizing the need to combine top-down control of information with bottom-up user choices. They have both developed special toolbars that can be downloaded and coupled on the user’s internet browser (see Fig. 2 for the MedCIRCLE prototype). Such a toolbar allows the user to set his/her preferences for which components of information are most important and then gives a confidence rating for how well an individual website answers to these preferences. The toolbar gives a message that user preferences about the reliability of information are important and that users should actively set their own criteria for assessment. However, this is acceptable only once users are already searching within the confined space of pre-approved sites. Fig. 2Patients can download this toolbar from http://www.medcircle.org and set their own preferences for characteristics that they find important in the provision of health information The confidence rating and other information provided by the tool are only available for sites already in the HON or MedCIRCLE databases. While these types of tools suggest the importance of individual choice, they nonetheless hold individuals responsible for following the ‘right’ paths to information. In this sense, the discussion is not merely an issue of creating reflexive consumers, but also of judging non-reflexivity as irresponsible and deviant. Following the Path Back to the State The responsibility for finding good information and avoiding the risks associated with bad information on the internet becomes collaborative and distributed. Within this context, individuals are held responsible for the choices they make, how they engage with available information and how they engage with other actors in the collaborative process. This process involves new actors, such as HON and MedCERTAIN/MedCIRCLE, but also longer standing actors, such as the WHO, state governments and medical associations. Reliability is created through distributing skills, information, and practices, and is intertwined with the creation of new networks that bind together the technical and social. We see this in especially in uses of internet-based technologies to further public health education. Healthfinder, for example, sends monthly newsletters to inform its public about changes and updates to the site and about different services it provides. Each newsletter also contains a list of “health observances” for that month, each of which includes hyperlinks to special sites with more information. These observances are, where possible, coupled on the (national) holidays being celebrated in that month. For example, February is ‘the month for all kinds of hearts’ and the newsletter’s healthy observances are all related to cardiac issues, while July’s newsletter focuses on food safety tips for warm summer days and firework safety tips specifically related to the July 4th holiday. Although other months are more general, all months prescribe topics of suggested interest for the general population:All of us, in all stages of life, can find a health observance of interest this month. April’s observances range from National Donate Life Month, a request for us to consider giving another person life, to WalkAmerica, a campaign to promote the benefits of carrying babies to term, sponsored by March of Dimes, to many observances in between.15 The observances further point out special days of awareness and political activism, and also broadly publicize public health education opportunities:The National Headache Foundation is sponsoring National Headache Awareness Week June 5–11 to recognize headache pain as a real and legitimate condition and encourage those who suffer with headaches to see a physician for proper diagnosis and treatment. This year’s campaign, “Stop Migraines Before They Stop You,” will feature public education activities nationwide.”16 The links that are selected for inclusion in the newsletters link largely to other parts of the Department of Health and Human Services and to other health-related government institutions (.gov domain endings), such as the National Institutes of Health (of which NLM is also a part) or the Centers for Disease Control and Prevention. As these examples show, different actors have developed practices that place various elements in new and unique relationships with one another. Issues related to health become intertwined with, for example, federal holidays. The health practices of the individual are now inextricably linked with diverse forms of community participation. In this sense, health education and enrollment is no longer only or primarily about verifying online health information, but rather about diverse activities that extend beyond the online realm. Directly Linking Health Practices and Citizenship These initiatives are tools to help people find and/or check information and about helping lay persons become empowered, not merely through distribution of information, but also through coaching or education in how to participate more actively in the search for and use of information. Users are invited to join specific, defined communities that will, in one way or the other, protect them from misinformation and inform them about better ways to act. HON even emphasizes the strength and necessity of its partnership with lay users in able do what it sets out to do—HON cannot help lay users if the lay users do not help HON. Additionally, there is a normative message about responsible citizenship. HON, for example, responds to individuals who follow their instructions and report (suspected) misuse of the HONcode icon by thanking them for their “sense of civic duty” and healthfinder’s newsletters make the initiative much more than just a gateway, or portal, to links of organizations with information. They also serve a promotional function to keep visitors returning to the healthfinder website and getting them to link through to other government sites. Further still, they show that being an active consumer and responsible citizen involves more than merely checking the information one encounters on sites or reporting misinformation and/or misuse of seals through certain channels. The responsibility extends into all areas of daily life (from good nutrition to awareness about possible diseases) and extends beyond individuals to include not only family or friends, but also the community-based or social responsibility that accompanies participation in educative programs, activism, or fund-raising events. What begins as a prescription for how to search for and assess information on the web actually extends to the entire lifestyle in which these searching activities are contextualized (As Dessauer [14] notes, much of what we think of with cyberspace loses meaning and referents if it excludes the external context). Interestingly, despite the varied forms of the messages emanating from the different initiatives, each of the prescribed behaviors, such as entering the web at a location of geographical significance (within government borders), participating in a chain that actively polices adherence to an ethical code and reports misuse, or even downloading and using a browser with a special toolbar, emerges as simultaneously optional and obligatory [53]. Herein lie strains of the tensions that are suggested by both Rose and Blume and Singleton: initiatives are created and promoted under the guise of increasing choice and democratization, but the actual prescriptions that they carry entail more powerful suggestions about which choices (both online and off) are the “right” or “better” choices—with alternatives to those choices being (explicitly or implicitly) discouraged. Discussion In this paper, we discuss how political actors have enabled access to web-based information largely in terms of specific user tools; however, through these examples, we see that, as Markham [42, 43] argues, access to the internet is access not just to a tool, but to a place (emergent in for example, how the interface is designed or the level of engagement in a given activity) and a way of being (dependent on the degree to which the individual integrates the technology into his or her understanding of social construction). As providers or enablers of technology use, political actors at various levels (primarily states, but also NGOs and IGOs that provide different types of sites and tools) provide points of entry and create avenues not just to online information, but to online and offline social communities with specific rules for behaviour. This has implications for how technology is configured within the relationship between individuals and the state and other political actors. Especially important is that it highlights how the emphasis on claims of “neutrality” within the reliability debate is misplaced. Organizations such as HON or WHO, who claim their neutrality based on their international scope and/or their position outside of national governments, nonetheless carry specific normative ideas of acceptable politically or socially related user behavior. Likewise, individual state governments that claim to represent the needs of their own “general public” impose values on the information that they recommend to that public – for example, in privileging a .gov above all else, followed by a “.org” or “.edu” and reducing as much as possible endorsement of a “.com.” Existing literature arguing the need for collective solutions to reliability problems, together with the abundant presence of initiatives offering different options for interactions between lay end users and those providing assessment tools, indicate a readiness among political actors and some health professionals to engage in active teamwork with the public. There is even the possibility that this language of collaborative efforts suggests that the public is/can be seen as just another interest group balanced with physicians [30]. However, in the examples provided above, we see that partnership does not necessarily imply equal footing for each of the three types of actors. Organizations use the catch-word reliability as a rhetorical device to capture attention and enroll users—redirecting their information-seeking behaviors, but also emphasizing deeper-rooted values that situate individual health-related practices within a greater context of idealized citizenship. As such, in addition to prescriptions for use of specific tools, we find underlying prescriptions for practices that reflect good citizenship. What, then, comprises the construction of a ‘good’ citizen? First and foremost a good citizen is one who is interested in a healthy lifestyle and reflects this through choosing to participate in activities such as searching for information about his/her health and the health of loved ones. Secondly, in searching for this information, the citizen reflects the possession of basic practical skills by choosing to align his/her practices with political actors, thereby choosing for sites that have been created or in some way reviewed by these actors. Thirdly, the citizen actively participates in the online community in which these sites are embedded. Active participation includes helping to “police” the information online by checking links and actively reporting misuse. Additionally, active participation includes extending knowledge accrued online to participate in offline community activities (activism, education, and fund-raising). This participation also forms a reciprocal relationship between the participant and the given organization—in return for making oneself available to the community, the burden to search for, find, and evaluate information is made easier. Searching can be done using tools within which information has been pre-selected and/or information can even be sent directly to the individual’s personal e-mail. Both of these technical options for acquiring information also carry an implicit social promise of being taken care of—protection from misinformation or mis-action due to lack of information. The readiness to participate in this partnership—the commitment of the organization to helping protect its user community is made explicit. Through the creation of initiatives that respond to debates about the reality of information online, organizations prescribe both online and offline behaviors, such that these are inextricably intertwined. This leads to the emergence of new collectives, counteracting the idea that internet enables people to separate activities with ease, compartmentalizing on and offline activities and leading to division.17 These collectives further counteract notions of individualization and separation from the community, as well as arguments that certain actors can be rendered irrelevant. Existing relationships become re-conceived as new alliances between medical and political actors, where internet technologies can be used to strengthen existing public education activities. Additionally, new (types of) communities emerge: families and friends or geographically concentrated communities are supplemented by communities that share aspects of visiting the same web sites and participating in the same self-policing chains—online and offline, individuals participate in communities by watching out for one another and sharing in the same activities. Although these initiatives claim to target “all users, everywhere,” they actually idealize use such that discussions about increased choice (and how this leads to patient empowerment) also carry the normative implications that “true” empowerment is only possible when the “right” choices are made. Within these discussions, individuals have the responsibility to develop skills that make them informed consumers, and the process of acquiring these skills also enables them to contribute their part to the greater civic communities in which they engage. The additional implications connoted by the terms “citizen” and “consumer” raise the question of whether these really are more suitable as replacements for the word “patient” than other suggested alternatives. Conclusion Whereas discussions of technologies and their users have largely neglected to consider the role of states (and other political actors), discussions of the relations of states and individuals have not always considered the important reciprocal effects of and on changing technologies. Discussions of skill manage to link individuals to individual technologies, but they have not successfully coupled this back on the role of states. This paper has used the discussion about the reliability of web-based information to attempt to bridge these gaps and demonstrate the integral relationship between technologies and users, states and individuals, and individuals and the development of skills. It shows that political actors, including both governmental and non-governmental organizations, make presumptions about citizen willingness to use certain technologies and, as such, utilize soft power to encourage these citizens-as-users-of-technologies to forward certain agendas. Under the guise of democratizing ideals and suggestions of empowerment, they use policy, programs and web-based tools to encourage citizen participation. Underlying these different modes of participation, these actors are able to re-create boundaries both online and off, not only by demarcating geographical boundaries on the web, but also by supporting conventional understandings of social norms, such as the role of the medical expert as a primary leader. In so doing, they give patients the “choices” about how to behave, with strong normative undertones about rights and responsibilities associated with being informed and reflexive consumers. As others have pointed out, where medical or health-related information is discussed, an issue of power is usually at stake [37]. The power in question is generally interpreted as that of medical professionals, but this is not necessarily the only interpretation, as political actors clearly use this for their own means and ends, as well.

J_Headache_Pain-4-1-2386847

Involvement of calcitonin gene-related peptide in migraine: regional cerebral blood flow and blood flow velocity in migraine patients

Calcitonin gene-related peptide (CGRP)-containing nerves are closely associated with cranial blood vessels. CGRP is the most potent vasodilator known in isolated cerebral blood vessels. CGRP can induce migraine attacks, and two selective CGRP receptor antagonists are effective in the treatment of migraine attacks. It is therefore important to investigate its mechanism of action in patients with migraine. We here investigate the effects of intravenous human alpha-CGRP (hαCGRP) on intracranial hemodynamics. In a double-blind, cross-over study, the effect of intravenous infusion of hαCGRP (2 μg/min) or placebo for 20 min was studied in 12 patients with migraine without aura outside attacks. Xenon-133 inhalation SPECT-determined regional cerebral blood flow (rCBF) and transcranial Doppler (TCD)-determined blood velocity (Vmean) in the middle cerebral artery (MCA), as well as the heart rate and blood pressure, were the outcome parameters. No change of rCBF was observed at the end of infusion [1.2% ± 1.7 with hαCGRP, vs. −1.6% ± 3.1 with placebo (mean ± SD)] (P = 0.43). Vmean in MCA decreased to 13.5% ± 3.6 with hαCGRP versus 0.6% ± 1.8 with placebo (P < 0.005). Since rCBF was unchanged, this indicates a dilation of the MCA. hαCGRP induced a decrease in MAP (12%) (P < 0.005) and an increase in heart rate (58%) (P < 0.0001). CGRP dilates cerebral arteries, but the effect is so small that it is unlikely to be the only mechanism of CGRP-induced migraine. Introduction Based on animal research showing calcitonin gene-related peptide (CGRP) in perivascular nerves [1] and a strong vasodilator effect of CGRP on cerebral blood vessels [2], a role for CGRP in the pathogenesis of migraine pain was first suggested by reports of increased CGRP in external jugular venous blood during migraine attacks [3–5]. In one recent study, there was, however, no increase of CGRP [6]. More solid evidence was presented in a study which demonstrated that, CGRP infused intravenously in patients with migraine was able to induce a vascular type headache in great majority of patients and in some patients this headache fulfilled all diagnostic criteria for migraine without aura [7]. The final proof of the involvement of CGRP in migraine mechanisms was provided by two phase II clinical trials [8, 9], demonstrating significant efficacy of the specific CGRP antagonists, BIBN4096BS [10] and MK0974 [9]. Thus, CGRP is not only able to induce attacks, but it seems to be continuously important throughout the entire migraine attack. CGRP is one of the most potent dilators of isolated cerebral arteries known today [2, 11], but possible species differences make it difficult to predict the effect of CGRP in the human cerebral circulation. Although the effect of CGRP on the cerebral circulation has been studied in normal subjects [12], it is important to study it also in patients with migraine in whom migraine-like headache and migraine attacks are induced by CGRP [7]. The aim of the present study was therefore to investigate the cerebral hemodynamic effects of CGRP in patients with migraine outside of attack. Patients and methods Patients Twelve patients with migraine (11 females, 1 male; mean age 39.5 years, range 31–47 years; mean weight 69.7 kg, range 51–89 kg) were included. All suffered from migraine without aura according to criteria of the International Headache Society [13]. The subjects were not allowed to take medication, coffee, tea, alcohol or tobacco for 12 h before the study, and they were not allowed to take a triptan 24 h or ergotamine 48 h before the study. Exclusion criteria were as follows: use of any kind of daily medication including prophylactic headache therapy but excluding oral contraceptives; pregnancy or breastfeeding; excessive use of analgesics or alcohol; serious somatic or psychiatric disorders; ischaemic heart disease; a supine systemic blood pressure more than 160/90 or less than 110/75 mmHg at entry of study. Patients were informed that they were free to withdraw at any time and all gave written informed consent. The study was approved by the local ethics committees of Copenhagen and Copenhagen County (KA 96054) and complied with the Declaration of Helsinki. Design and procedure The patients with migraine were studied outside of attacks. The study used a double-blind, placebo-controlled crossover design. Subjects were randomized to receive 2 μg/min human αCGRP (hαCGRP; Clinalfa, Switzerland) or placebo (0.9% NaCl) infused intravenously for 20 min on 2 days separated by at least 1 week. The dose was chosen as the highest tolerated dose (because of blood pressure reduction) based on reports in the literature [14, 15]. Patients were randomly assigned by computer (Med. Stat®, version 2.12). Randomization and preparation of study drugs was done by medical staff, who were not involved in the study. Five patients started with placebo and seven with hαCGRP. This quota was chosen by the computer and not known before the study. The effectiveness of blinding was not estimated. When subjects arrived at the laboratory, a cannula (Viggo Venflon®, 1.4 mm) was placed in the right cubital vein for hαCGRP/placebo infusion. Baseline values of mean maximal blood velocity (Vmean) in the middle cerebral artery (MCA), regional cerebral blood flow (rCBF), blood pressure, heart rate and pCO2 were recorded after 30 min of rest in the supine position in a quiet room. Then, hαCGRP or placebo was infused intravenously for 20 min by a volumetric pump (Braun Perfuser). Vmean, blood pressure, heart rate and end tidal pCO2 were recorded every 10 min during and after infusion until 80 min after start of the hαCGRP infusion. rCBF was measured again during the last 5 min of infusion and at 75–80 min after start of hαCGRP/placebo infusion. At every recording, it was noticed if volunteers were flushing and information about headache presence, intensity (measured on a 0–10 scale) and characteristics were obtained before, during and after the infusions. The results concerning headache response have been published elsewhere [7]. Methods rCBF was measured with a highly sensitive, brain-dedicated, fast-rotating, single photon emission computerized tomograph (Tomomatic 232). Each study lasted 4.5 min. A mixture of atmospheric air and 133Xenon was rebreathed during the first 1.5 min through a closed system from a 4-L reservoir (740 Mbq/L). During the last 3 min, the 133Xenon mixture was expired against atmospheric air. rCBF was recorded simultaneously in two slices positioned 50 and 90 mm above and parallel to the orbito-meatal plane (OM). Each slice was 16 mm thick and the distance between the centers of slices was 40 mm. The full width half maximum resolution of the instrument is about 16 mm in the horizontal plane. rCBF was calculated according to Celsis et al. [16]. A fixed matrix of regions of interest was superimposed on the rCBF picture. The shape and size were fitted to the outlines on the brain excluding extracranial flow. Regional mean values were calculated within the predefined regions of interest. The matrix was divided into regions of interest representing the hemispheric rCBF regions and the vascular territories of supply by the anterior-, middle- and posterior cerebral arteries. The maximum whole body radiation was approximately 0.6 mSV per rCBF measurement [17]. Time averaged mean of the maximal blood velocity (Vmean) in MCA at the usual headache side was measured with transcranial Doppler (TCD) ultrasonography (2 MHz, Multidop X Doppler: DWL, Sipplingen, Germany). A mean of four consecutive values of Vmean was taken (each representing the mean value of typically four to five heart beats automatically calculated by the computer). Positions of the measurements were reproduced from day to day by recording the position in relation to the angle and distance relative to the orbito-meatal line. Blood pressure and heart rate were measured with an automatic inflatable arm cuff (Omega 1400, Invivo Research Laboratory, New York, USA). Simultaneously, with each rCBF and TCD measurement, the end expiratory pCO2 was recorded by means of a capnograph (Datex OY, CD 101). Statistics Hemodynamic responses are given as mean and standard deviation (±SD). Differences in blood velocity, rCBF, blood pressure, heart rate and pCO2 over time within the group were analyzed with analysis of variance (MANOVA, Statgraphics® 7.0), and changes were then located with a multiple range test (Confidence intervals, Statgraphics® 7.0). Differences in the delta values (baseline—the last measurement during infusion and baseline—the last measurement during the study) in Vmean, rCBF, blood pressure, heart rate and pCO2 between the effect of hαCGRP and the effect of placebo were tested with a paired t test (Statgraphics® 7.0). In all tests, P < 0.05 was considered statistically significant. Results During CGRP infusion, two volunteers experienced a substantial decrease in blood pressure (70/50 and 65/25, respectively) causing the infusion to be terminated ahead of time. Signs and symptoms were pallor, cold sweat, stomach ache, nausea and palpitation. The situation was restored after placing the volunteers in head-down tilt for 7 and 20 min, respectively. No rCBF and TCD measurements were taken during this time period, and therefore, results from these two volunteers were excluded from the calculations below. Changes in end tidal pCO2 End-tidal pCO2 decreased significantly over time when volunteers were treated with hαCGRP (P < 0.0001) but not when treated with placebo (P < 0.26, MANOVA). The pCO2 decrease at the last measurement during infusion was −4.0% ± 0.8% when treated with hαGCRP versus 0.7% ± 1.0% when treated with placebo (P < 0.05, paired t test). The peak decrease in end-tidal pCO2 induced by hαCGRP (−6.8% ± 2.6%) occurred at 25 min, 5 min after end of infusion. Transcranial Doppler measurements The TCD examination was performed on the side of usual migraine. The following data for Vmean in MCA are corrected for changes in pCO2 according to Markwalder et al. [18]: Vmean changed significantly over time when patients were treated with hαCGRP (P < 0.0001, MANOVA), but not when treated with placebo (P = 0.26, MANOVA). At the last measurement during infusion, 15 min after its beginning, Vmean compared to baseline was −13.5% ± 11.4 when treated with hCGRP, versus 0.6% ± 5.6 when treated with placebo (P < 0.005, paired t test). At the end of the in-hospital period (75 min after start of the infusion), there was no difference between the two groups (P = 0.97, paired t test) (Fig. 1, Table 1). There was a positive correlation between ΔVmean and ΔMABP (r = 0.703, P < 0.05). Fig. 1The effect of hCGRP (2 μg/min) for 20 min and the effect of placebo on the mean velocity (Vmean) in middle cerebral artery (squares) and the regional cerebral blood flow (rCBF) (triangles) in the territory of the middle cerebral artery (MCA) used for transcranial Doppler examination. X-axis: time from start of infusion (min), y-axis: changes in percent of baseline. Vmean (squares) decreased −13.5% ± 3.6 compared to baseline with hCGRP treatment versus 0.6% ± 1.8 with placebo treatment (P < 0.0001). After 75 min, there was no difference between the two groups (P = 0.97). No change of rCBF (triangles) was observed at the end of infusion (P = 0.43) or at the end of study period (P = 0.12)Table 1The maximum mean blood velocity (Vmean) in the middle cerebral artery (MCA)Time (min)Vmean in MCA (cm/s)No pCO2 correctionCorrected for pCO2 changesCGRPPlaceboCGRPPlaceboCGRP%Placebo%081.6 (10.7)79.9 (12.3)81.6 (10.7)79.9 (12.3)0015–2066.9 (13.5)*79.5 (12.4)70.7 (14.2)*80.3 (12.7)−13.5 (11.4)*0.6 (5.6)75–8076.6 (9.9)77.4 (12.7)80.6 (11.3)79.2 (13.6)−0.7 (11.6)−0.9 (6.4)Values within parentheses represent the standard deviation (SD)Data is shown with and without pCO2 correction according to Markwalder et al. [18]The pCO2-corrected data is shown as absolute data and in percent change from baseline (P < 0.001)* Significant changes compared with baseline Regional cerebral blood flow All data have been corrected for changes in pCO2: rCBF(korr) = rCBF(meas) × exp(0.04(pC02(Basal) − pCO2(n)). For comparison of Vmean and rCBF, we calculated rCBF in the territory of MCA, the artery used for TCD examination. Area-weighted data from the OM + 50 mm and OM + 90 mm were used for the results given below. There was no change over time neither when patients were treated with hαCGRP (P = 0.26) nor when they were treated with placebo (P = 0.88, MANOVA). During infusion, rCBF compared to baseline was +1.2% ± 5.5 when treated with hαCGRP versus −1.6% ± 9.8 when treated with placebo. The difference between CGRP and placebo was not significant (P = 0.43, paired t test). Furthermore, no significant difference between CGRP and placebo was found at the end of the study period (P = 0.12, paired t test) (Fig. 1, Table 2). Table 2Regional cerebral blood flow rCBF in the MCA territory recorded 50 and 90 mm above the orbitomeatal plane (OM)Cerebral blood flow (ml blood/100 g/min) Time (min)MCA (OM 50 + OM 90) in the territory used for TCD examinationNo pCO2 correctionCorrected for pCO2 changesAbsolute dataAbsolute dataData in % of baselineCGRPPlaceboCGRPPlaceboCGRPPlacebo070.8 (13.7)67.7 (9.3)70.8 (13.7)67.7 (9.3)0015–2066.8 (9.4)65.3 (11.6)71.1 (10.5)66.7 (11.4)1.2 (1.7)−1.6 (3.1) 75–8069.3 (7.9)65.0 (9.8)72.8 (8.7)66.7 (10.4)4.2 (3.0)−1.6 (2.8)Values within parentheses represent the standard deviation (SD)Data is shown with and without pCO2 correction according to Markwalder et al. [18]The pCO2-corrected data is shown as absolute data and in percent change from baseline (P < 0.001) The absolute rCBF data and the rCBF data with pCO2 correction concerning the MCA area relevant for TCD examination are given in Table 2. There was also no significant difference in pCO2-uncorrected rCBF between CGRP and placebo during (P = 0.78, paired t test) or after the infusion (P = 0.357, paired t test). There was a negative correlation between ΔrCBF(MCA) and ΔMABP (r = −0.654, P < 0.05). rCBF during hαCGRP infusion was not significantly different in any territory or at any time point. Blood pressure The mean arterial blood pressure (MABP) decreased significantly over time when patients were treated with hαCGRP (P < 0.0001), but not when they were treated with placebo (P = 0.48, MANOVA). The maximal decrease occurred at the last measurement during infusion (20 min after start of infusion) and amounted to −12.3% ± 8.1% with hαCGRP treatment versus +2.4% ± 5.9% with placebo treatment (P < 0.005, paired t test). MABP had normalized at the end of the study period (80 min from start of infusion) (P = 0.49) (Table 3). Both systolic and diastolic blood pressures decreased during infusion of CGRP compared to that during infusion of placebo (P < 0.05). Table 3The mean arterial blood pressure (MABP), systolic and diastolic blood pressure (mmHg) and heart rate (bpm) after CGRP and placebo treatmentBlood pressure (mmHg)Heart rate (bpm)Time (min)MABPSystolic BPDiastolic BPCGRPPlaceboCGRPPlaceboCGRPPlaceboCGRPPlacebo087.8 (9.6)91.6 (9.5)117.0 (11.1)119.0 (12.6)73.2 (9.5)77.9 (9.4)63.0 (6.0)64.4 (7.2)1578.0 (7.5)*90.0 (9.2)110.7 (9.6)117.2 (12.5)61.7 (7.8)76.4 (8.7)92.9 (10.2)*63.8 (6.8)2076.9 (10.1)*93.5 (8.3)112.2 (13.3)123.9 (11.6)59.3 (9.1)78.3 (7.6)98.9 (11.2)*65.5 (7.1)7586.5 (9.2)91.4 (8.7)117.6 (12.0)121.9 (13.4)71.0 (8.9)76.2 (7.2)76.1 (7.0)*62.9 (6.0)8085.3 (9.2)91.3 (9.9)116.4 (13.2)123.3 (11.3)69.7 (8.4)75.3 (9.7)75.3 (7.7)*64.4 (5.3)Values within parentheses represent ±standard deviation (±SD)* P < 0.05 Heart rate When patients were treated with hαCGRP, heart rate increased significantly over time (P < 0.0001, MANOVA), but not when they were treated with placebo (P = 0.44, MANOVA). The peak increase in heart rate after hαCGRP treatment occurred 20 min after start of infusion reaching 58.1% ± 22.7% versus 1.9% ± 5.58% with placebo treatment (P < 0.0001, paired t test). Heart rate after hαCGRP treatment was still different from that after placebo treatment at the end of the study period (P < 0.005, paired t test) (Table 3). Other signs Flushing after hαCGRP treatment was pronounced and appeared only in the face, neck and upper chest. It appeared from 10 min (time of first observation) after start of the infusion of hαCGRP in all patients until median 70 min after start of infusion (range 20–80 min). There was no flushing when the patients were treated with placebo. The median headache during infusion of hαCGRP was 1 versus 0 in the placebo-treated group (P < 0.01). During the following 11 h, all patients experienced headache after hαCGRP treatment versus the one after placebo treatment (P < 0.001). The median headache score was 4 after hαCGRP treatment and 0 after placebo treatment. For details, see [7]. Discussion The main finding of the present study was that CGRP-infusion dilated the MCA in patients with migraine, while cerebral blood flow remained unchanged. The dose of 2 μg/min of CGRP was chosen from the literature in which the same dose of 545 pmol/min was given in healthy volunteers without any reported adverse events [14]. In another study, up to 25 μg CGRP was given as an intravenous bolus injection and all subjects had facial flushing [15]. The dose of 2 μg/min for 20 min is probably the maximally tolerated dose of CGRP in patients with migraine, since it resulted in a substantial decrease of MABP and a marked increase in heart rate. Two subjects were near fainting and had to be withdrawn. In healthy volunteers, a dose of 1.5 μg/min was used without adverse events apart from flushing and without any effect on mean blood pressure [12]. It remains uncertain whether the stronger circulatory response in the present study was exclusively due to the higher dose or whether patients with migraine are more sensitive to CGRP. After 20 min infusion at 1.5 μg/min, the mean plasma level was 340 pmol/L in healthy volunteers [12]. It can be calculated from these results that the plasma level would be 450 pmol/L in our patients with migraine. The EC50 for CGRP for the dilatory effect on human pial arteries is 500 pmol/L [19]. The dose of CGRP used in the present study thus resulted in a plasma level very near to the EC50 for cerebral arteries in vitro. The blinding of the present study can be criticized. Signs (increased heart rate) and symptoms (facial flushing and feeling of warmth) during CGRP infusion made it difficult to keep the study completely blinded. However, the present design is the best one available, because no technique is available to disguise such symptoms. An interesting finding of the present study is the significant reduction of blood velocity (13.5% decrease) in the MCA, but rCBF remained unchanged during intravenous hαCGRP-infusion. A reduced velocity in an artery, with unchanged regional blood flow in its territory of supply, reflects dilation [20]. The relative diameter change can be roughly estimated from the relation: flow = mean velocity multiplied by the cross-sectional area of the artery (2πr2). When the flow is constant, velocity is a function of the reciprocal value of r2 and Va/Vb = rb2/ra2 [20]. We can thus estimate from the change in mean velocity that the hαCGRP infusion caused a 7.5% increase in MCA diameter corresponding to 17% increase of its cross sectional area. αCGRP infused in a considerably lower dose (0.6 μg/min) did not change MCA velocity or rCBF in healthy volunteers [21], while the dose of 1.5 μg/min in volunteers dilated the MCA (9% increase in diameter) to the same extent as in the present study [12]. The dilatation of MCA in the present study could, theoretically, be due to cerebrovascular autoregulation secondary to decreased blood pressure. However, conduction arteries like MCA do normally only autoregulate due to blood pressure changes to a much lesser extent than arterioles (for review, see [22]). A positive correlation (r = 0.7, P < 0.05) between changes in MCA and changes in MABP seems to support this possibility. However, this correlation could also reflect that MCA and systemic circulation vary in parallel. In healthy volunteers, using a slightly lower CGRP dose of 1.5 μg/min, blood pressure was unchanged but MCA dilated to the same extent as in the present study [12]. This strongly indicates that MCA dilation is a direct effect of CGRP. In rat MCA, using the in vitro-pressurized arteriographic model, luminal CGRP was without any dilating effect, whereas abluminal CGRP dilated the artery [23]. This indicated that CGRP was unable to cross the blood–brain barrier in rat MCA. In contrast, both in patients with migraine (Fig. 1) and in healthy volunteers [12], a dilation of MCA was observed. Theoretically, CGRP could act on an endothelial receptor, but this is unlikely because the CGRP antagonist BIBN4096BS could not block the effect on MCA in man [12]. Furthermore, in man, the calcitonin receptor-like receptor (CLR) and receptor activity-modifying protein (RAMP) are located mainly in the muscular layer of the human MCA [24]. In the endothelium, there was only a minor amount of CLR and largely absent RAMP1 [24]. There is thus no CGRP receptor (CLR plus RAMP1) on the endothelium. The present results with a dilation of MCA could indicate that the blood–brain barrier to CGRP in the MCA is less tight in man than in rats, possibly related to the large difference in size. In healthy volunteers, dilation of MCA was accompanied by a modest (14%) increase in rCBF [12], whereas in the present study, in patients with migraine, a 5% increase in rCBF was not statistically significant (Fig. 1). The discrepancy may be due to random variation. Altered cerebrovascular reactivity in patients with migraine compared to controls is another possibility. The decrease of mean MABP from 88 to 77 mmHg observed in the present study could also be a factor if autoregulation was disturbed by CGRP. In patients with subarachnoid hemorrhage, a CGRP-infusion concomitant with intravenous fluid to correct any drop in blood pressure resulted in a dilation of the MCA on the vasospasm side [25]. Blood pressure was unchanged while heart rate and cardiac output were increased and total peripheral resistance was decreased [25]. In two studies in man [14, 15], an increase in noradrenaline in plasma was observed after CGRP. We found in the present study a decrease in mean blood pressure and an increase in heart rate as would be expected after administration of a potent vasodilator such as CGRP. In conclusion, the potent endogenous migraine-inducing molecule CGRP resulted in a dilation of MCA and unchanged rCBF. CGRP can thus most likely cross the blood–brain barrier to some extent in the large human cerebral arteries. The vasodilator effect on these arteries is in our opinion, however, so small that it is unlikely to be the only mechanism of CGRP-induced migraine observed in our patients with migraine [7].

BMC_Med_Educ-4-_-328087

Learning from the problems of problem-based learning

Background The last decade has witnessed a rapid expansion of biomedical knowledge. Despite this, fashions in medical education over the same period have shifted away from factual (didactic) teaching and towards contextual, or problem-based, learning (PBL). This paradigm shift has been justified by studies showing that PBL improves reasoning and communication while being associated with few if any detectable knowledge deficits. Background Many doctors have commented that their medical education began in earnest on the first day that they entered the hospital wards as a hands-on practitioner. Claims of this kind support the view that the apprenticeship model of professional learning – which has been the backbone of training in the healing arts for thousands of years [1] – remains as central to medical career development today as ever [2]. A perennial complaint of the medical apprentice-in-training is that there are too few structured teaching activities within the busy world of postgraduate work [3], a concern which many institutions have addressed by developing formalised continuing education initiatives reminiscent of medical school courses [4-6]. Predictably, different complaints prevail at the pre-licensure phase of the training spectrum, where students often feel more motivated to acquire the (implicit) competence of the practising doctor [7,8] than to absorb large volumes of (explicit) scientific and/or humanistic theory [9,10]. Such feelings underlie an unresolved debate over the optimal balance between factual ('teaching', or content-based) and practical ('training', or performance-based) components of professional development [11] and, as such, could explain a recent drift away from the didactic emphasis of older biomedical educative approaches [12] and towards quasi-experiential, or problem-based, learning [13,14] (PBL; Figure 1A). This hypothesis cannot fully account for the PBL-led transformation of medical teaching in the 1990s, however, overlooking as it does a powerful contrary trend: the explosive proliferation of biomedical knowledge [15,16] as epitomized by the completion of the Human Genome Project [17]. Although at first sight contradictory, this reciprocal relationship between knowledge growth and didactic teaching invites a unifying explanation: namely, that the switch of educational philosophy to non-didactic methods represents a strategy for teachers and students to cope with the expansile information environment [18,19]. Figure 1 Comparison of growth rates of PBL and biomedical knowledge. (A), Relative growth of interest in PBL versus lecture-based teaching based on PubMed keyword frequencies between 1975 and 2000. The Title fields of the journal database were searched each year for the strings "problem-based learning" (open squares) or "lecture" (open diamonds); the retrieved items were then scrutinized to determine those dealing with the subject of teaching style (e.g., eponymous "lectures" were excluded). (B), Growth rates of journal articles dealing with subject matter relating to science, medicine and education. The strings "gene" (solid diamonds), "clinical" (open squares) and "medical education" (solid triangles) are shown here, illustrative of the frequencies of many other keywords searched. There are many things which a fresh medical student, unburdened by factual knowledge, can begin to learn: basic surgical methods, resuscitation interventions, generic reasoning skills, and counselling techniques, to name a few. The pivotal question, then, is not whether such context-dependent (but sequence-independent) learning will prove effective [20]; rather, it is why this reorientation of teaching philosophy has occurred at all, and at this time. Or to put the issue another way: what is the hard evidence indicating that the original educational system was broken and that the new system is likely to fix it? A secondary issue, which has been a prime concern of PBL critics [21], is whether there may prove to be long-term hidden costs payable for the clear short-term benefits afforded by the PBL teaching philosophy. Since there are major differences in the way that PBL is implemented between schools, evidence to confirm or refute such hypotheses may be impossible to assemble. By the same token, it is an oversimplification to view all PBL as having low fact-based content, just as it is to equate all older teaching methods with rote learning. Nonetheless, since PBL veers more to the active/contextual, and didactic teaching to the passive/factual, it is plausible that one bias occurs at the expense of the other. Discussion What is knowledge – anyone know? The traditional educational sequence involves theory preceding application, an accelerated model of which has long been satirized in clinical circles as "watch one, do one, teach one". As noted above, however, some applications may be learned in the absence of theoretical knowledge, just as some subsets of theoretical knowledge may be unassociated with any obvious application. Is it possible, then, to define a minimum essential "core knowledge" spectrum for the student of biomedicine? If so, should such knowledge expand in parallel with other biomedical information, or should any such expansion be restricted by its relevance to changes in clinical performance (the 'barefoot doctor' model)? In the latter case, how long can healthcare competence and credibility be maintained in the face of rising constraints on scientific core knowledge [22]? If core knowledge is indeed expanding at a rate similar to that of non-core knowledge, then the strategy of solving the broad problem of knowledge expansion by defining a narrower core can only be a temporizing measure. On the other hand, if the quantum of core knowledge is deemed non-expansile – arbitrarily defined, for example, to represent the amount of knowledge capable of being instilled in an average student by x teaching hours per week spread over y years – then any expansion of non-core knowledge will cause the core to shrink as a proportion of total knowledge. In 1984, for example, a list of two hundred drugs was hailed as a solution to information overload in the field of pharmacology [23]; but by 2000 the overload problem in this discipline was perceived to have deteriorated despite both the embracement of PBL and relentless efforts to re-define a core curriculum [24]. A key difficulty in addressing this problem is that expansion of biomedical information is asymmetric – different areas of knowledge grow at different rates which in turn vary (and are ascribed differing priorities) during different periods (Fig. 1B). In practice, most curricula cope with differential knowledge growth by adding new core modules to cover areas of rapid growth [25]; the problem with this approach is that the notion of "core" becomes fluid, invalidating the concept. Moreover, it is difficult to discard ageing core knowledge at the same rate as adding new information, since the credibility of newer information tends by its nature to be weaker than that of older content. Rigid conservation of the core leaves trainees selectively deficient in new knowledge areas, on the other hand, making them less competitive in the marketplace. Discrepancies emerging between planned (taught) and actual (learned) medical curricula [26] further weaken the practicability of paradigms based on core knowledge. The concept of core knowledge as a stand-alone solution to the problem of information inflation thus appears flawed [27]. Although at any one time certain knowledge subsets may be deemed dispensable for learning purposes, a continuous expansion of knowledge must imply a comparable expansion of knowledge essential for maintenance of professional competence [28]. Moreover, practicality should not be the sole criterion by which core knowledge is judged; a medical training system cannot succeed by simply cloning service-based doctors, but must also produce academics, researchers, visionaries and leaders able to develop the service infrastructure [29]. It is against the background of these diverse challenges that the recent growth of PBL should be appraised. The lure of the non-expert A traditional authoritative doctor who dares challenge the information explosion faces the same risk as the immovable object confronting the irresistible force. The outcome has been to shatter the image of the omniscient doctor, as well as to dent the plausibility of experts in all fields. This slide from grace of the specialist – formerly a dominant figure in the medical educational pantheon, and a revered colleague in the battle against information overload [30] – has paralleled the decline of the lecture as a teaching medium. These linked trends suggest that the internet-empowered medical customer of the 21st century (whether patient or student) now questions information promulgated by mere individuals, thus threatening clinical and teaching paradigms formerly assumed unassailable. This problem has created a niche for PBL, rejecting as it has the old curriculum's reliance upon experts and specialists [31]. The reported disadvantages of non-expert biomedical teaching with respect to knowledge transmission [32] and disease understanding [33] have been parried by numerous studies showing no detectable information deficits in PBL-trained (compared to lecture-taught) students [34-38]. Such negative data may be of limited reassurance, however, given the insensitivity of the endpoints used to measure what is in practice a rather limited curricular divergence [39]. It is likewise arguable that endpoints such as knowledge acquisition and clinical skills are surrogates, and that the most critical deliverable of medical training – namely, the quality of patient outcomes – has not been measured in any controlled trial of PBL [40]. These points lend credence to criticisms that the present-day popularity of PBL has so far been driven more by individual enthusiasm and conjecture than by objective scientific evidence [41]. Who, then, stands to gain from PBL? Medical teachers are perhaps the most immediate beneficiaries [42]. Reduced self-perceptions of fallibility may be one attraction for teachers, as new PBL supervisors find that their educational contributions are no longer falsifiable by their pupils. In addition, responsibility for providing a sufficient knowledge base can be passed from teacher to student under the PBL auspices of 'self-directed learning' [43]. Yet another benefit relates to the lack of formal preparation required to initiate a PBL session [44] – an advantage which suggests a gain in efficiency. But does this bear scrutiny [45]? Things have changed Efficiency can be calculated by dividing (productivity) by (time and effort). What do we mean in this context by productivity? A half-century ago, the only responsibility of a medical school was to produce clinicians to serve the local community; today, however, teaching activities incorporate postgraduate specialist education, continuing medical education, professional and career development, public and patient awareness programmes, education-related research, conference and workshop organization, national and international collaborative initiatives, professional accreditation and audit activities, development of electronic teaching resources, and so on. Hence, a modern faculty's teaching productivity is not able to be gauged exclusively (or even predominantly) by the number and quality of its outgoing medical graduates, but rather must be judged by the sum total of its useful educational output. This raises short- and long-term issues as to the most efficient ways to utilize faculty time and effort: traditional teaching service activities (e.g., tutorials, mentoring) must compete with more ambitious developmental activities (e.g., production of journal articles, books, software or web resources). Small-group tutorials are a time-honoured teaching modality, but the opportunity cost is high; while there must surely remain a place for personalized teaching, it seems doubtful whether the modern academic system can tolerate the luxury of an accelerating trend in this direction [46]. In contrast, the traditional apprenticeship training approach seems cost-effective, relying as it does upon the learner assisting a professional in the execution of his/her paid duties. In this context it is worth noting that the development of PBL – growth of which during the 1990s coincided with similar trends favouring noncognitive-based medical school admissions [47-49] and humanities-rich preclinical experience [50-52] – was spawned a quarter of a century ago in a regional medical school in Canada [53]. One need scarcely point out that the 1975 academic environment responsible for this educational breakthrough bears little resemblance to the market-driven imperatives that preoccupy most medical faculty members today, both in Canada [54-56] and elsewhere [57-62]. A changing environment not only justifies, but mandates, adaptation; if the 1990s trends do indeed represent a retreat from an information-dominated world, then the substitution of a PBL-dominated philosophy could be fraught with significant longterm perils. From words to actions Solutions lie in compromise. Such change is painful because it involves the abandonment of ideals formerly attainable; the vision of a one-size-fits-all medical school becomes no longer practical, and ever more difficult decisions will be needed as to what style(s) of graduate is most urgent for a faculty to produce. This process of curricular differentiation has started, but the pace is set to quicken as medical markets emerge and diverge, and as competition for faculty survival sharpens. To what extent, though, should these divisive educational decisions be made by markets, faculties, students, patients or governments? Contrary to popular thought, there will remain a strong need – and possibly an enlarging one – for a subset of highly-trained medical graduates from a knowledge-intensive learning environment who are capable of assimilating the complexities of science, informatics, humanities and logistics that comprise modern medicine. Since the proportion of individuals and faculties suitable for this leadership mission looks set to decline, however, a larger number will need to accept the equally daunting compromise of skills prioritization. Teachers cannot teach without students, but students can learn without teachers. This belated insight has transformed the role of teachers into that of learning facilitators, akin to a culture of "thinking apprenticeship". Paradoxically, in an age when even complex skills such as landing aircraft are learned using robotic simulators, the trend in medical education has switched back to labor-intensive small-group teaching under the guise of PBL. This at first seems all the more curious given the unprecedented availability of alternative technologies for teaching clinical reasoning, the increasing importance of an adequate knowledge base in an ever more sophisticated professional environment, the growing pressures on faculties to use limited fiscal resources in the most cost-effective manner, and the novel opportunities for commercializing educational activities and products via the development of software and web-based resources. The rise in PBL popularity over the 1990s thus suggests a retreat from the fallout of the biomedical information explosion. Although this response seems rational enough as a short-term adaptive measure, it should not be regarded as a solution to the problem of knowledge expansion. Just as PBL was originally pioneered as a reaction against complacency in traditional pedagogy, so must today's medical schools reject expediency and confront the unresolved information-management challenges of 21st-century medical education. The formulation of more efficient techniques for imparting factual knowledge, a greater emphasis on directing limited resources to the production of reusable teaching tools [63], and a willingness to experiment with differentiated medical curricula that prioritise graduate skill subsets, can all play a role in driving educational reform as a positive and ongoing adaptive process. Summary The knowledge explosion of the last two decades has been accompanied by a decreasing reliance on didactic teaching. This educational paradigm shift has been led by widespread embracement of PBL, the original rationale of which was to improve students' ability to reason and communicate. In recent years, however, PBL has grown more rapidly in apparent response to information overload in medical school curricula, and may thus be viewed as a symptom of the problem of biomedical knowledge expansion. The challenge of defining the right balance between what is taught, what is learned, and what remains unlearned will not disappear. Although few knowledge deficits have been detected in today's PBL-educated students, a decreasing concern with the adequacy of the professional knowledge base could yet erode the future credibility of the medical profession. By continuing to rely on popular PBL escape clauses such as 'self-directed learning' and 'information management', today's medical educators risk losing sight of this longterm threat. The era of active learning began thousands of years ago with the first apprentice. We now live in a new era with new challenges, one of which is exponential information expansion. PBL provides one way for faculty and students to cope with this challenge, but sidesteps deeper issues relating to the widening core of essential professional knowledge. Innovative curricular experiments using educational strategies complementary to PBL would therefore appear timely. Abbreviations PBL, problem-based learning Competing interests None declared. Pre-publication history The pre-publication history for this paper can be accessed here:

Qual_Life_Res-3-1-2039864

The effect of walking and vitamin B supplementation on quality of life in community-dwelling adults with mild cognitive impairment: a randomized, controlled trial

Objectives To examine the effect of walking and vitamin B supplementation on quality-of-life (QoL) in community-dwelling adults with mild cognitive impairment. Introduction Especially in older people, both mental and physical function decrease due to multiple age related changes, which in turn may affect quality of life (QoL). The most obvious decrease in mental function is cognitive decline, which is a common aspect of aging. However, in some cases decline is more serious than expected for a certain age. This is specified as Mild Cognitive Impairment (MCI). MCI is considered to be a potential transitional stage between normal cognitive function and Alzheimer’s disease, characterized by (1) subjective memory complaint (2) objective memory impairment (3) normal mental status (4) intact activities of daily living (ADL) (5) absence of dementia [1]. Independent of the latter four criteria, subjective memory complaints are related to lower QoL [2]. Moreover, MCI is associated with poor physical health and high risk of ADL dependence [3, 4]. Since both cognitive and physical decline belong to the most important determinants of QoL in community dwelling elderly subjects [5], subjects with MCI are likely to be susceptible to a decrease in QoL. The number of adults with MCI is increasing considerably due to the aging population. For multiple reasons, it is important to prevent a decrease in QoL. Apart from the personal benefits, a high rated QoL also reduces medical consumption and helps to maintain independency as long as possible [6]. This in turn may relieve significant others, caregivers and medical society in general. For this reason, attention should be paid to possible interventions contributing towards a higher level of overall QoL and it’s mental and physical components. In this respect, physical exercise and vitamin supplementation are interesting interventions worth investigating. Regular participation in moderate intensity aerobic training is reported to be beneficial in improving QoL and wellbeing, which is an important aspect of QoL [7, 8]. Since walking is the most prevalent physical activity among older adults [9], improving QoL by increasing the time spent on moderate intensity walking seems promising. Indeed, a community based walking program significantly improved both the physical and mental components of health-related QoL in older adults (n = 582) [10]. Inconclusive evidence has been reported on the influence of vitamin B supplementation on QoL. Different aspects of QoL were not responsive to short term supplementation (range 4–12 weeks) with different doses and combinations of B vitamins in men and women [11–13]. Not much is known about QoL in community dwelling elderly with MCI. Moreover, no trials on the effect of exercise and vitamin B supplementation on QoL have been carried out yet in adults with MCI. The FACT-study (Folate physical Activity Cognition Trial) was developed to examine the effect of these interventions on cognition [14]. Aspects of QoL were measured as a secondary outcome. In the present paper, the effectiveness of 1 year moderate intensity walking (two sessions of 60 min per week) and daily vitamin supplementation (5 mg folate, 50 mg vitamin B6 and 0.4 mg B12) on both overall QoL and it’s health-related components is examined in community dwelling older adults with MCI. We hypothesize that 1 year moderate intensity walking benefits QoL. Concerning the effect of vitamin supplementation, this paper should be considered as explorative. Methods Study design The study was designed as a randomized, placebo controlled intervention trial, based on a two-by-two factorial design. The study-protocol has been described in detail elsewhere [14] and was approved by the VU University Medical Center medical ethics committee. Written informed consent was obtained from all participants. Participants In a medium-sized Dutch town community dwelling subjects aged 70–80 years with MCI were identified using a population based two-step-screening [15]. The operational criteria for MCI according to the criteria of Petersen et al. [1] and additional inclusion criteria for the RCT are described in Table 1. Subjects were not paid to participate in the study. Table 1Inclusion and exclusion criteria for participation in the trial Operationalization of Petersen criteria for MCI (1–5) and additional inclusion criteria for the RCT (6–12)1. Memory complaints (answer yes to question ‘do you have memory complaints’, or at least twice sometimes at cognition scale of Strawbridge [16]2. Objective memory impairment; 10 WLT delayed recall ≤5 + percentage savings ≤ 100 [17]3. Normal general cognitive functioning; TICS ≥ 19 + MMSE ≥ 24 [18, 19]4. Intact daily functioning: no report of disability in activities of daily living on GARS-scale, except on the item ‘taking care of feet and toe nails’ [20]5. Absence of dementia; TICS ≥ 19 + MMSE ≥ 246. Being able to perform moderate intensity physical activity, without making use of walking devices, e.g., a rollator or a walking frame 7. Not using vitamin supplements/vitamin injections/drinks with dose of vitamin B6, B11 or B12 comparable to vitamin supplement given in intervention8. Not suffering from epilepsy, multiple sclerosis, Parkinson’s disease, kidney disorder requiring haemodialysis, psychiatric impairment9. Not suffering from depression as measured by the GDS (cut off ≤5) [21]10. Not using medication for rheumatoid arthritis or psoriasis interfering with vitamin supplement11. No alcohol abuse (men < 21 consumptions a week, women < 15 consumptions a week)12. Not currently living in a nursing home or on a waiting list for a nursing homeMCI = Mild Cognitive Impairment, RCT = Randomized Controlled Trial, 10 WLT = 10 Word Learning Test, TICS = Telephone Interview for Cognitive Status, MMSE = Mini Mental State Examination, GARS = Groningen Activity Restriction Scale, GDS = Geriatric Depression Scale Randomization After the baseline interview, subjects were randomly assigned to the interventions using the statistical computer program SPSS. Intervention groups were: (1) walking program or placebo activity program; and (2) vitamin B supplementation or placebo supplementation. Randomization was stratified for physical activity level at baseline in minutes per day as measured by the LASA physical activity questionnaire [22]. For the flow of participants see Fig. 1. Fig. 1Flow chart. TI = Telephone Interview, WP = Walking Program, PAP = Placebo Activity Program, FA/B12/B6 = Folic Acid, Vitamin B12, Vitamin B6 supplementation, SO = significant other, T6 = follow-up after 6 months, T12= follow-up after 12 months, *reason for exclusion: only baseline data available Exercise intervention Subjects assigned to the walking program (WP) participated twice a week 60 min in group-based moderate intensity walking during 1 year. Each session consisted of a warming up, moderate intensity walking exercises and a cooling down. The WP was based on ‘Sportive Walking,’ an existing aerobic walking program [23] aimed at improving cardiovascular endurance. Therefore, duration and intensity of the walking exercises increased gradually during the program. Sessions took place outdoor in municipal parks. Subjects not assigned to the WP participated in a placebo activity program (PAP) with the same frequency, session duration and program duration. However, the PAP consisted of low intensity exercise, such as light range of motion movements and stretching. Sessions were divided into five themes: relaxation, activities of daily living, balance, flexibility, posture and a combination of all. For each theme three sessions were developed and the entire series of 18 sessions was repeated during the intervention period. The PAP was carried out in community centers. Both programs were supervised by qualified and trained instructors. Attendance to both programs was assessed by the percentage of attended sessions. Vitamin supplementation (FA/B12/B6) Subjects in the vitamin supplementation group took one pill containing 5 mg vitamin B11 (Folic Acid), 0.4 mg vitamin B12 (Cyanocobalamin) and 50 mg vitamin B6 (Pyridoxine-hydrochloride) daily during 1 year. This vitamin supplement is available on prescription in The Netherlands. Subjects randomized to the control group took an identically looking placebo pill. The pills were packed in blister packs for 1 week, which were labeled for each day of the week. Compliance with the vitamin supplementation was verified by pill counts in returned blister packs during the intervention. Outcome measures Baseline data on sociodemographic and background variables were collected using a postal screening questionnaire. The measurement of other baseline variables, as reported in Table 2, has been described elsewhere [14]. In the present manuscript a distinction was made between ‘overall quality of life,’ referring to a subjects overall enjoyment of life and ‘health-related quality of life,’ referring to health-related factors affecting quality of life. The term QoL was used as an umbrella term for both overall and health-related QoL. The population-specific Dementia Quality of Life questionnaire (D-QoL) [24] was used to assess overall QoL and the generic Short Form 12 (SF12) [25] to assess health-related QoL. The D-QoL is a 29 item measure especially developed for elderly with cognitive decline and dementia. The participant is asked about how much they enjoyed activities that were reported to be important for elderly such as ‘watching animals.’ Moreover, the frequency of certain positive and negative feelings such as ‘lovable’ or ‘worried’ were asked for. Finally, they were asked to rate their overall quality of life. The participant was instructed to choose the best fitting answer from five item response scales. The answers were divided into five domains of QoL measuring sense of aesthetics, feelings of belonging, negative affect, positive affect/humor and self esteem. A mean score ranging from one to five was calculated for these subscales and for the total D-QoL. A higher score indicated better quality of life. Median internal consistency reliability of the D-QoL was 0.80 and median test–retest reliability was 0.72 in a sample of 95 older adults with different stages of cognitive decline [24]. The SF12 consists of 12 items measuring eight concepts of both mental and physical health, i.e., physical functioning, role-physical, bodily pain, general health, energy, social functioning, role emotional and mental health. These concepts are subdivided into two summary scores using a norm-based criterion: i.e., mental and physical component summary scales (SF12-MCS and SF12-PCS). The mean score is 50 with a standard deviation of 10. For example, a score of 60 corresponds to a-QoL rating of one standard deviation above the average ratings in the general population. Test retest reliability for the SF-12 MCS was 0.76 in a sample of 187 adults in the United Kingdom and 0.77 in a sample of 232 adults in the United States. Reliability coefficients of the SF-12 PCS in these populations were 0.86 and 0.89, respectively [25]. In the present study, measurement took place during a personal interview at baseline and after 6 and 12 months. Both the D-QoL and the SF-12 were administered by a trained interviewer who was unaware of the participants’ group allocation. Table 2Baseline characteristics of participants (n = 152)Exercise interventionVitamin interventionWP (n = 77)PAP (n = 75)FA/B12/B6 (n = 78)Placebo-pill (n = 74)Age (Mean (SD))75 (2.9)75 (2.8)75 (2.8)75 (2.9)Gender (% male)48*645655MMSE (Median (10th–90th ‰))29 (26–30)29 (27–30)29 (25–30)29 (27–30)Education (% low/middle/high) 61/22/1752/29/1957/26/1755/26/19Marital status (% living together)75686973Physical activitya (min/day) (Median (10th–90th ‰))44 (10–155)39 (11–120)45 (13–155)38 (9–111)Vitamin status (% deficient FA/B12/B6)b46/8/048/8/049/9/045/7/0Homocysteine statusc (% hyperhomocysteinemia)27232723Blood pressure (% hypertension)d27*142516BMI (kg/m2) (Median (10th–90th p‰))26.7 (23.1–31.5)26.6 (23.5–32.7)26.5 (23.3–32.8)26.7 (23.5–31.2)Smoking (% smokers)13151711Number of self-reported diseases (% 0,1,2)e52/42/669/27/466/28/655/41/4WP = Walking Program, PAP = Placebo Activity Program, FA/B12/B6 = Folic Acid, Vitamin B12, Vitamin B6 supplementation, MMSE = Mini Mental State Examination, Education: low = no education, primary education, lower vocational training; intermediate = intermediate level secondary education, intermediate vocational training; high = higher level secondary education, higher vocational training, university training. BMI = Body Mass Indexa ≥3.0 metabolic equivalents b Cut off points: FA red blood cell <337 nmol/l or FA plasma < 6,3 nmol/l; B12 ≤ 150 pmol/l; B6 < 20 nmol/lc Homocysteine > 14 mmol/ld Hypertension = diastole ≥ 90 and systole ≥160e Cardiovascular disease, chronic obstructive pulmonary disease, diabetes, epilepsy, multiple sclerosis, Parkinson’s disease, psychiatric disease, renal failure requiring dialysis and/or rheumatoid arthritis * Significantly different from PAP (P < 0.05) Statistical analyses Differences between groups on baseline characteristics were tested using independent t-tests (normally distributed variables), Mann–Whitney U tests (not normally distributed variables) and χ2 (categorical variables). Within group differences were tested using dependent t-tests. Subsequently, data were analyzed according to a modified intention-to-treat principle, based on data from all randomized participants who provided data at baseline and at least one follow-up measurement. To evaluate the effects of the walking program and the vitamin supplementation on QoL, longitudinal regression analysis was used. The two follow-up measurements were defined as dependent variable and multi level analysis with two levels was used, (1) time of follow-up measurement (values corresponding with performance after six and 12 months intervention); (2) individual. According to the study protocol [14], the effect of both interventions was examined independently from each other. Data were analyzed using a crude and an adjusted model. Independent variables were exercise intervention and vitamin intervention. By analyzing both interventions in the same model, results were adjusted for the possible influence of the other intervention. Moreover, all analyses were adjusted for baseline performance on the outcome measure by adding this as a covariate. In the adjusted model, education, baseline activity level, baseline vitamin status, attendance to the exercise program and compliance with the supplementation were added as covariates. Interaction between gender and the WP or FA/B6/B12-supplementation was checked in the adjusted model. In the case of significant interaction, results were reported for men and women separately. In the case of no interaction, gender was added to the adjusted model as an additional covariate. Also, in the ‘adjusted model’ an interaction effect of the exercise program with attendance to the exercise program was checked. Finally, data were analyzed according to the per protocol principle, including all participants who attended at least 75% of the sessions. This cut-off point is in concordance with previous exercise intervention studies in older adults [26, 27]. Data were analyzed using SPSS for Windows (release 12.0.1). A significance level of 5% was used for between group comparisons and of 10% for interaction terms. For all analyses, regression coefficients and 95% confidence intervals for the adjusted models were reported, with the regression coefficients directly indicating the difference in QoL ratings between the WP and the PAP or the FA/B12/B6-supplementation versus placebo supplementation. In the case of significant interaction, regression coefficients and the 95% confidence intervals of the interaction terms were reported. Results Patient characteristics Hundred-seventy-nine participants were randomized to the interventions. Twenty-seven of them were excluded from the analyses, because they only provided baseline data. These subjects were more often married (71% vs. 52%, P = 0.05) and less often current smokers (0% vs. 14%, P = 0.04) than the remaining 152 participants who provided QoL data at baseline and at at least one follow-up measurement. The latter 152 participants were included in the analyses (see Fig. 1). Their mean age (SD) was 75 (2.9) years. Fifty-six percent was male. Additional baseline variables are described per factor in Table 2. Compared to the PAP, the WP included fewer men (48% in WP vs. 64% in PAP) and more subjects with hypertension (27% in WP vs. 14% in PAP). Ratings for both overall and health-related QoL at baseline and after 6 and 12 months intervention are presented in Table 3. No baseline differences were observed on these measures, except for a higher rating of D-QoL self-esteem in subjects in the FA/B12/B6-group compared to subjects in the placebo-supplementation group. Table 3Means (standard deviations) of QoL ratings at baseline and after 6 and 12 months in older adults with MCIaWP PAPFA/B12/B6 Placebo T0 (n = 77)T6 (n = 77)T12 (n = 71)T0 (n = 75)T6 (n = 75)T12 (n = 67)T0 (n = 78)T6 (n = 78)T12 (n = 71)T0 (n = 74)T6 (n = 74)T12 (n = 67)D-QoL sumscore3.5 (0.26)3.5 (0.29)3.5 (0.27)3.5 (0.32)3.5 (0.34)3.5 (0.34)3.5 (0.32)3.5 (0.32)3.5 (0.33)3.4 (0.24)3.5 (0.31)3.5 (0.27)D-QoL aesthetics3.5 (0.63)3.5 (0.64)3.6 (0.60)3.5 (0.70)3.5 (0.71)3.5 (0.65)3.5 (0.64)3.5 (0.68)3.6 (0.61)3.4 (0.68)3.5 (0.67)3.6 (0.64)D-QoL belonging3.7 (0.50)3.7 (0.49)3.7 (0.44)3.8 (0.45)3.7 (0.47)3.7 (0.46)3.8 (0.50)3.6 (0.50)3.6 (0.48)3.7 (0.44)3.8 (0.45)3.8 (0.40)D-QoL negative affect2.7 (0.45)2.7 (0.46)2.8 (0.50)2.7 (0.55)2.8 (0.54)2.8 (0.52)2.7 (0.54)2.8 (0.47)2.8 (0.53)2.7 (0.47)2.7 (0.53)2.8 (0.49)D-QoL positive affect3.8 (0.39)3.7 (0.46)3.8 (0.40)3.8 (0.40) 3.7 (0.44)3.8 (0.43)3.8 (0.41)3.7 (0.47)3.8 (0.44)3.8 (0.39)3.8 (0.43)3.8 (0.39)D-QoL self esteem3.6 (0.45)3.8 (0.41)3.8 (0.40)3.7 (0.48)3.7 (0.49)3.8 (0.48)3.8 (0.48)*3.8 (0.48)3.9 (0.48)3.6 (0.43)3.7 (0.43)3.7 (0.38)SF12-MCS54.6 (6.85)55.6 (6.40)55.3 (4.39)54.7 (8.07)55.0 (7.34)55.3 (6.24)55.5 (7.49)55.9 (6.91)55.8 (4.90)53.8 (7.36)54.6 (6.86)54.8 (5.76)SF12-PCS48.2 (7.15)48.1 (7.57)50.5 (6.13) 48.7 (7.86)48.8 (8.47)49.8 (7.04)47.9 (8.20)47.4 (8.79)49.8 (6.68)49.1 (6.67)49.6 (7.00)50.6 (6.49)MCI = Mild Cognitive Impairment, WP = Walking Program, PAP = Placebo Activity Program, FA/B12/B6= Folic Acid, Vitamin B12, Vitamin B6 supplementation, D-QoL = Dementia Quality of Life, SF12-MCS = Short Form 12 Mental Component Summary, SF12-PCS = Short Form 12 Physical Component Summarya Higher rating indicates better QoL* P < 0.05 (difference between FA/B12/B6 and placebo) Attendance to the WP and the PAP Overall median attendance to the exercise programs (10th−90th percentile) was 63 (0–89) percent and did not differ between the WP and the PAP. Especially in the first weeks, a considerable number of subjects discontinued participation, mostly because they did not want to participate in the exercise programs after all. Most frequent reasons for discontinuation of the program after the first weeks were health-related problems. No adverse events of the WP or PAP itself were reported. Adherent subjects attending at least 75% of the sessions (n = 51) were more often living together (82% vs. 65%, P = 0.03) and less physically active than non-adherers (n = 101), (median [10th–90th percentile] was 36 [13–82] vs. 44 [10–169] min/day, P = 0.02). At baseline, adherers also had lower ratings of D-QoL-belonging (3.6 [0.41] vs. 3.8 [0.49], P = 0.02) and higher SF12-MCS values (56.5 [5.6] vs. 53.7 [8.1], P = 0.02). Other baseline and QoL characteristics did not differ significantly. Compliance with the (FA/B12/B6)supplementation Four participants did not return the blister packs. On the basis of pill counts in returned blister packs, median compliance (10th–90th percentile) with the FA/B12/B6-supplementation was 100 (97–100) percent and compliance with placebo-supplementation was 100 (35–100) percent. Even though median compliance in both groups was 100%, compliance in the placebo-group was significantly lower (P < 0.05). Eight subjects, one in the FA/B12/B6-group and seven in the placebo-group, did not take (vitamin)supplementation. Seven of them decided immediately after randomization not to participate in the interventions. The other wanted to participate in the exercise intervention only. Two participants discontinued taking vitamin pills during the trial after reporting sleep problems and increased forgetfulness; one participant discontinued taking the placebo pills after reporting not feeling well. Modified intention to treat analyses Results of the walking program and FA/B6/B12 supplementation are presented in Table 4. With respect to overall QoL, no positive significant main effect of the WP or FA/B6/B12 supplementation was found. A significantly detrimental effect of FA/B6/B12 supplementation was observed on D-QoL-belonging, (beta (95%CI) = −0.18 (−0.29; −0.07), P < 0.01). A positive interaction between the WP and attendance to the WP was observed on D-QoL-belonging and D-QoL-positive affect. With each percent increase in attendance, D-QoL-belonging increased with 0.003 points (P = 0.04) and D-QoL-positive affect with 0.002 points (P = 0.06) in the WP compared to the PAP. With respect to health-related QoL, an interaction between the WP and gender was observed on the SF12-MCS (P = 0.06) and therefore analysis for the SF12-MCS was stratified for gender. No main effects of the WP or FA/B12/B6-pills were observed. However, in men in the WP, SF12-MCS increased with 0.03 points with each percent increase in attendance (P = 0.08). Table 4Results of longitudinal multi level analyses on the effect of the WP and FA/B6/B12 supplementation on change in QoL (adjusted model) WP versus PAP Beta (95%CI)P-valueFA/B12/B6 versus placebo Beta (95%CI)P-valueD-QoL sumscore0.04 (−0.03;0.10)0.25−0.06 (−0.12;0.004) 0.07D-QoL aesthetics0.06 (−0.07;0.20)0.37−0.07 (−0.20;0.07)0.33D-QoL belonging0.00 (−0.11;0.11)0.96−0.18 (−0.29; −0.07)0.00D-QoL negative affect−0.02 (−0.12;0.08)0.650.04 (−0.05;0.14)0.37D-QoL positive affect0.04 (−0.04;0.13)0.34−0.04 (−0.12;0.04)0.33D-QoL self esteem0.08 (−0.02;0.18)0.110.00 (−0.10;0.11)0.94SF12-PCS0.66 (−1.23;2.54)0.49−0.73 (−2.65;1.19)0.45SF12-MCS*Men−0.82 (−2.24;0.60)0.250.25 (−1.31;1.81)0.76Women1.66 (−1.50;4.81)0.301.32 (−1.93;4.56)0.42WP = Walking Program, PAP = Placebo Activity Program, FA/B12/B6= Folic Acid, Vitamin B12, Vitamin B6 supplementation, D-QoL = Dementia Quality of Life, SF12-MCS = Short Form 12 Mental Component Summary, SF12-PCS = Short Form 12 Physical Component Summary* Interaction WP and gender Per protocol analyses Subgroup analyses were performed in subjects attending 75% or more of the WP or PAP sessions (n = 51, 33 men and 18 women). No between group differences were observed for FA/B12/B6-pills versus placebo-pills. A significant positive effect of the WP compared to the PAP was observed on D-QoL-positive affect, beta (95%CI) = 0.23 (0.06; 0.39), P < 0.01 and a borderline significant positive effect on D-QoL-self esteem, beta (95%CI) = 0.17 (0.001; 0.34), P = 0.05. Discussion No positive main effect of walking or daily FA/B6/B12 supplementation was observed on QoL in community-dwelling adults with MCI. However, ratings of overall QoL (i.e., feelings of belonging, positive affect) and the mental component of health-related QoL improved slightly with increasing attendance to the walking program. In a subgroup that attended at least 75% of the sessions, a beneficial effect of the walking program was observed on positive affect and self esteem. To our knowledge, this is the first intervention study on QoL in community-dwelling adults with MCI. While memory complaints are reported to be negatively associated with QoL in healthy older adults with subjective memory complaints [2], QoL ratings in our study population were already quite high at baseline. Baseline ratings on the DQOL sumscore and subscales fell ample above the midpoint of the scale, except for negative affect. Baseline scores on the SF-12MCS fell around a half standard deviation above the average in the general population and SF-12PCS fell about the average ratings. QoL-ratings have been reported to decrease as the severity of cognitive decline increases [28]. The possibility exists that MCI as operationalized in the present study may not have been serious enough to negatively influence overall and health-related QoL. In spite of the high baseline values, the QoL scales still allowed for further improvements, i.e., there was no ceiling effect. However, it has been discussed before that QoL may represent a stable concept, which is difficult to change or that existing measures may not be responsive to subtle changes [29]. The relationship between physical activity and QoL has been studied extensively. However, it is difficult to draw a clear conclusion, since various definitions and operationalizations of QoL circulate. Moreover, comparisons between studies are being complicated by the wide variety of study populations and features of exercise intentions such as intensity, exercise mode, frequency and session and total duration [30]. However, Rejeski et al. [31] concluded in a review including 28 studies, of which 11 RCT’s, that physical activity positively influenced aspects of health-related QoL. In a recent meta-analysis of Netz et al. [7] including 36 studies, a small positive effect of exercise was observed on wellbeing in healthy older adults. In that meta-analysis four components of wellbeing were considered, including aspects that were also measured in the FACT-study, such as positive and negative affect, perception of physical fitness and physical symptoms. In the present study no main effects of the WP were observed in the modified intention to treat analyses. First, a possible explanation for the lack of effect may be that only participants with good QoL were able to attend enough sessions. In contrast to an earlier study, no baseline differences in number of chronic diseases, physical health-related QoL and endurance were observed between adherers (attending ≥75% of the sessions) and non-adherers (attending <75% of the sessions) [32]. However, adherers rated their mental health-related QoL at baseline significantly better than non-adherers. The difference was three points, which approximately equaled a difference of 5%. The possibility exists that subjects with lower mental health-related QoL were inclined to attend less sessions. Nevertheless, it is not likely that this biased our results, because non-adherers and drop-outs from the exercise programs were included in the modified intention to treat analyses. In future studies in subjects with cognitive decline, session attendance may be improved by informing subjects extensively about the study aims and the consequences of participation. Moreover, if possible with respect to logistic and financial issues, we advice to schedule time and staff for the close personal follow-up of temporary drop-outs. Second, it has been reported that the association between physical activity and QoL is lower among older adults who function at or above the norm [31]. By applying inclusion criteria for the present trial (e.g., community dwelling, no ADL disabilities, being able to perform moderate intensity physical activity), we presumably selected physically healthy and active subjects. This is supported by the high baseline activity levels. Two-thirds of the participants reported to be physically active at moderate intensity for 30 min or more per day. Subjects meeting this guideline are reported to have better health-related QoL than physically inactive adults [8]. Additionally, Netz et al. found that larger effects of exercise on wellbeing were observed in sedentary adults [7]. However, in the present study, no interaction between the walking program and baseline physical activity level was observed (results not presented), indicating that inactive participants did not benefit more from the WP than active participants. Therefore, it is not likely that baseline physical activity level was a main cause of the lack of main effects. Finally, inconclusive evidence is available about the intensity and exercise mode of physical activity required to benefit QoL. Netz et al. [7] concluded that aerobic training of moderate intensity was most beneficial for wellbeing. In a cross-sectional study, it was also observed that moderate intensity physical activity was positively related to health-related QoL [8]. In contrast, in a review by Spirduso and Cronin [33] no evidence of a relationship between exercise intensity and the rate of improvement in QoL was found. If the former would be true, the possibility exists that the contrast between both programs in the present study would not have been large enough to induce differences in QoL. If the latter would be true, participants would have benefited from participation in both exercise programs regardless of intensity. Both programs may either have added to better self-efficacy, or may prevented a decline in self-efficacy. The walking program by training cardiovascular endurance; the placebo activity program by training e.g., balance and ADL. Self-efficacy refers to somebody’s belief that one has the capabilities to successfully manage situational demands and is mentioned to be a mediating mechanism for the effect of physical activity on QoL [7, 30, 34, 35]. Thus, the presence of the low intensity placebo activity program in our study may have contributed towards the lack of between group differences. Nevertheless, several outcomes improved with increasing attendance to the walking program. In the per protocol analyses a beneficial effect was observed on positive affect. Self esteem also tended to improve. However, observed differences were small and approximated 5% differences from baseline QoL ratings. As a rule of thumb, a minimal change of 5% has been mentioned to signify clinical relevance. To obtain a change of 5% by increasing attendance, the required increase in attendance would be 62% for D-QoL-belonging and 94% for D-QoL-positive affect and the SF12-MCS. Therefore, it can be questioned whether the observed effects are clinically relevant. No effect of the FA/B12/B6 supplementation was observed except for a negative effect on feelings of belonging. However, no theoretical rationale exists for this effect. Our findings are in line with previous RCT’s on the effect of vitamin B supplementation on aspects of QoL. Deijen et al. [11] observed no effect of supplementation with 20 mg vitamin B6 for 3 months on mood in healthy men (n = 76). Also no effect of supplementation with 750 μg folate, 15 μg vitamin B12 or 75 mg vitamin B6 daily for 35 days was observed on mood in women aged 65 or over (n = 75) [11]. Finally, no effect on health-related QoL was observed of a weekly injection with 1 mg vitamin B12 for 4 weeks in adults with vitamin B12 deficiency (n = 140) [13]. These findings may find its origin in the used operationalizations and measures of QoL that include very few items that directly relate to nutrition. Amarantos et al. [36] underline the need to develop QoL measures including items that relate nutrition to QoL. To conclude, the walking program and vitamin B supplementation were not effective in improving QoL in community-dwelling older adults with MCI within 1 year. However, increasing attendance to moderate intensity physical activity may benefit certain aspects of QoL.

Neurosci_Lett-2-1-2330064

Contribution of P2Y1 receptors to ADP signalling in mouse spinal cord cultures

Mixed neuronal and glial cell spinal cord cultures from neonates express ADP sensitive P2Y1,12&13 receptors. ADP (10 μM) evoked increases in intracellular calcium that were essentially abolished by the P2Y1 receptor antagonist MRS2179 (10 μM), responses were also absent in preparations from P2Y1 receptor deficient mice however UTP (100 μM) evoked calcium rises were unaffected. ADP also evoked a robust increase in extracellular signal-regulated protein kinase (ERK) phosphorylation that was of similar magnitude in the cultures from wild type and P2Y1 receptor deficient mice. These results suggest that ADP acts through P2Y1 receptors to mediate an increase in intracellular calcium but not to stimulate ERK phosphorylation in the spinal cord. Nucleotides released from neurons, glia and damaged cells play a signalling role in the nervous system. P2Y receptors are a family of G-protein coupled nucleotide receptors, and several of the eight subtypes (P2Y1,2,4,6,11,12,13,14) are expressed in the nervous system [11]. P2Y1 receptors have been shown to be involved in calcium signalling in a range of neurons and glial cells (e.g. [3,4,9,12]). In addition signalling through recombinant P2Y1 receptors stimulates mitogen-activated (MAP) kinases [10] and in native cells P2Y1 receptors have been implicated to play a role in extracellular signal related kinases (ERK) activation and stretch induced injury in astrocytes [8]. In this study the aim was to determine whether ADP evoked calcium and MAP kinase responses in cultured spinal cord neurons and glial cells and use knockout mice to determine the contribution of P2Y1 receptors to ADP evoked signalling. Wild type or P2Y1 receptor deficient C57 neonatal mice [7] (3–7 days) of either sex were decapitated and the spinal column was removed and washed twice in a calcium/magnesium-free Hanks balanced salt solution (HBSS) containing HEPES (0.3 M), sodium pyruvate and penicillin/streptomycin (50 μg/ml each). The spinal cord was then cut into small pieces and placed in ice cold HBSS. The tissue pieces were then placed in a HBSS-trypsin solution (2.5%) and incubated at 37 °C for 30 min. The tissue was then transferred to a HBSS-trypsin inhibitor solution (1 mg/ml) for 5 min before being replaced with 1 ml Neurobasal media containing glutamine and B27 supplement. Mechanical trituration was used to dissociate the spinal cord cells. The resulting cell suspension was seeded onto poly-d-lysine-coated 16 mm coverslips and maintained at 37 °C in a humidified incubator with a 95% O2/5% CO2 atmosphere. After 2 h an additional 2 ml of Neurobasal media was added and cells left to grow for 2–5 days. Spinal cord cultures were ester loaded with a calcium sensitive dye, Fluo-3 acetoxymethyl ester (Fluo-3-AM, final concentration 1 μM) for 30 min at 37 °C in a humidified incubator. The coverslips were then washed in extracellular solution (in mM; 150 NaCl, 10 HEPES, 10 Glucose, 2.5 KCl, 2.5 CaCl2, 1 MgCl2, pH 7.3) to remove any extracellular dye and maintained at room temperature. Coverslips were mounted in a perfusion chamber on the stage of a Fluoview FV300 confocal microscope and were continuously perfused with extracellular solution. The argon laser excitation of the dye was at 488 nm and the emissions were captured at wavelengths greater than 510 nm by Olympus Fluoview v4.2 software at a frequency of 0.5 Hz for a continuous time period of 280 s. Agonist application was via a U-tube [2]. ADP stocks were treated with hexokinase to remove any ATP contamination. Agonist applications were at intervals of ∼10 min. Antagonists were perfused for 8 min before test solution application so as to allow complete equilibration. Cells displaying ‘puff’ artefacts to just extracellular solution were discounted from analysis [1]. Baseline measurements of fluorescence were calculated as a mean of the 10 s of recording prior to agonist application. Peak changes in fluorescence for neuronal and glial cells were expressed as a ratio of the baseline (increases in fluorescence are expressed as a self-ratio, baseline value is 1). Spinal cord cells were cultured for 2–5 days as described previously and agonists were applied to individual wells and incubated for 10 min. The cells were subsequently lysed in a solution (in mM; 10 β-glycerophosphate, 1 EDTA, 1 EGTA, 50 Tris–HCl, 1 benzamidine, 1 sodium orthovanadate, 0.2 PMSF, 50 sodium fluoride, 1 mg/ml pepstatin A, 1 mg/ml leupeptin, 0.1% β-mercaptoethanol, 1% Triton X-100). The samples were then mixed with sodium dodecyl sulphate sample buffer (containining β-mercaptoethanol) at a ratio of 1:1 and denatured for 5 min at 95 °C prior to loading on a 10% sodium dodecyl sulphate polyacrylamide gel. The protein was then transferred onto a nitrocellulose membrane and subsequently blocked with blocking medium (5% non-fat milk powder and Tris-Tween 20, buffered salts-TTBS) overnight at 4 °C. Nitrocellulose membranes were then incubated in 5% blocking medium containing either anti-phospho-ERK antibody (Cell Signaling Technology, MA, USA) at a dilution of 1:1000 or anti-non-phosphorylated ERK antibody (Cell Signaling Technology, MA, USA) at a dilution 1:1000. Immunoreactive bands were visualised by incubation of nitrocellulose membranes with anti rabbit peroxidase conjugate at 1:1500 for 2 h at room temperature and then detected with chemiluminiscence (ECL Plus, Amersham, UK) and recorded onto hyperfilm (Amersham, UK). Western blots were analysed using Image J, NIH, USA—http://rsb.info.nih.gov. Phospho ERK intensity was corrected for loading using non-phosphorylated ERK blotting and expressed as fold over basal. Total RNA was isolated from either intact spinal cord or spinal cord using the RNeasy Mini Kit (QIAGEN, Valencia, CA). The RNA was treated with DNase I (Sigma) to remove any DNA contamination. Half of the purified RNA was reverse-transcribed to cDNA using SuperScript II-reverse transcriptase (Invitrogen, Carlsbad, CA), and the remainder of the RNA was used as a negative control to ensure that there was no genomic contamination. The primers have been described previously [1]. PCR reactions were carried out with 2 μl of cDNA and a final concentration of 500 nM of each primer. Amplification took place in a Techne Genius PCR machine (Techne, Cambridge, UK) using the following protocol: denaturation at 94 °C for 5 min; repeated cycles of denaturation at 94 °C, followed by annealing at 57 °C and extension at 72 °C, each for 30 s; final extension at 72 °C for 10 min. Thirty-five cycles were used. The amplification products were examined by electrophoresis on a 1% agarose gel and visualized using ethidium bromide under UV light. Statistical tests were preformed using an unpaired t-test with a p-value of <0.05 considered as significant. Data are expressed as mean ± standard error of the mean. Spinal cord cells cultured from neonates for 3–7 days contained both neuronal and glial cells that could be distinguished by their morphology and staining with glial fibrillary acidic protein; approx 43% of cells were glia and 57% neuronal (n = 576 cells from 4 coverslips). We have previously used calcium imaging to look at P2Y receptor signalling in mouse superior cervical ganglia [1] and similar protocols were used in this study. ADP (10 μM) evoked robust increases in intracellular calcium for both neurons (64% cells responding with peak self-ratio amplitude of 2.20 ± 0.14, n = 120 cells from 12 coverslips) and glial cells (57% of cells responded with a peak self-ratio amplitude of 2.17 ± 0.2, n = 120 cells from 12 coverslips). Responses were not sustained and generally decayed during the time-course of application (self-ratio at end of 30 s application 1.34 ± 0.08 and 1.27 ± 0.04 for neurons and glia, respectively, n = 20 for each). The selective P2Y1 receptor antagonist MRS2179 reduced ADP evoked calcium rises in a concentration-dependent manner and these effects were reversed on antagonist washout (Fig. 1). ADP evoked calcium rises were essentially abolished in spinal cord cultures from P2Y1 receptor deficient mice with small residual responses in 11% of neurons (self-ratio 1.14 ± 0.05, n = 100 cells from 10 coverslips) and 17% of glial cells (self-ratio 1.21 ± 0.05, n = 115 cells from 11 coverslips)(Fig. 2). Responses to the P2Y2/4 receptor agonist UTP (100 μM) in cultures from P2Y1 receptor deficient mice were the same as for WT for neurons (45 and 48% responding with amplitudes of 1.96 ± 0.09 and 1.79 ± 0.08, n = 60, 114 from 6 and 12 coverslips, respectively for WT and P2Y1 receptor deficient mice) and glial cells (58 and 62% responding with amplitudes of 1.88 ± 0.13 and 1.67 ± 0.04, n = 60, 119 from 6 to 12 coverslips, respectively for WT and P2Y1 receptor deficient mice) indicating that Gαq coupled receptor calcium signalling was not affected in the P2Y1 receptor deficient mouse (Fig. 2). Taken together these results define a role of the P2Y1 receptor in calcium signalling in spinal cord neurons and glial cells. These studies support the role of P2Y1 receptors in calcium signalling in the nervous system [3,9,12] and corroborate previous studies using sympathetic ganglion cultures from the P2Y1 receptor deficient mouse [1]. P2Y1 receptor stimulation has also been reported to mediate ERK phosphorylation in recombinant [10] and native systems [8] and this may contribute to changes in the spinal cord following injury. In spinal cord cultures ADP evoked a time-dependent phosphorylation of p44/42 ERK that was detected at 5 min, peaked at 10 min, and declined back to basal levels by 30 min (Fig. 3A). When just media was added to the cultures there was no change in ERK phosphorylation (Fig. 3B) indicating that ERK stimulation does not result from a mechanical artefact. Adenosine (10 μM) also had no effect demonstrating that adenosine receptors are not involved. We therefore compared ADP evoked ERK phosphorylation in spinal cord cultures from WT and P2Y1 receptor deficient mice. Surprisingly ADP (10 μM) evoked similar levels of ERK phosphorylation (∼3-fold increase over basal) in WT and P2Y1 receptor deficient cultures (Fig. 3) demonstrating that a receptor other than the P2Y1 receptor mediates this ADP response. Of the currently eight identified P2Y receptor genes ADP is an agonist at Gαq coupled P2Y1, and Gαi coupled P2Y12 and P2Y13 receptors [11]. ADP is also an agonist at the human P2Y11 receptor; however a corresponding mouse homologue of the P2Y11 receptor has not been detected suggesting that mice do not express P2Y11-like receptors [11]. Transcripts for P2Y1,12&13receptors were detected in 3–7 days old spinal cord cultures (data not shown). Amplification of RNA for ADP sensitive P2Y12&13 receptors raised the possibility that these may contribute to the ERK phosphorylation. AR-C66931MX [6] is an antagonist of P2Y12 and P2Y13 receptors with a reported pA2 of 8.7 at P2Y12 receptors and 0.01 μM reduced ADP responses by ∼80% at recombinant P2Y13 receptors (see review [11]). AR-C66931MX (10 μM a supra-maximal concentration at P2Y12 receptors) had no effect on ADP evoked ERK phosphorylation indicating that P2Y12 receptors did not mediate the ERK phosphorylation. AR-C66931MX may have complex actions at P2Y13 receptors, possibly dependent on cell type, and has been described as an activator of P2Y13 receptor dependent high density lipoprotein endocytosis in hepatocytes [5] questioning the use of this compound as a selective antagonist at the P2Y13 receptor. Thus it remains to be determined whether the P2Y13 receptor, or a novel ADP sensitive receptor, mediates the ADP evoked ERK phosphorylation in the spinal cord.

Ann_Biomed_Eng-2-2-1705493

Dependence of Intramyocardial Pressure and Coronary Flow on Ventricular Loading and Contractility: A Model Study

The phasic coronary arterial inflow during the normal cardiac cycle has been explained with simple (waterfall, intramyocardial pump) models, emphasizing the role of ventricular pressure. To explain changes in isovolumic and low afterload beats, these models were extended with the effect of three-dimensional wall stress, nonlinear characteristics of the coronary bed, and extravascular fluid exchange. With the associated increase in the number of model parameters, a detailed parameter sensitivity analysis has become difficult. Therefore we investigated the primary relations between ventricular pressure and volume, wall stress, intramyocardial pressure and coronary blood flow, with a mathematical model with a limited number of parameters. The model replicates several experimental observations: the phasic character of coronary inflow is virtually independent of maximum ventricular pressure, the amplitude of the coronary flow signal varies about proportionally with cardiac contractility, and intramyocardial pressure in the ventricular wall may exceed ventricular pressure. A parameter sensitivity analysis shows that the normalized amplitude of coronary inflow is mainly determined by contractility, reflected in ventricular pressure and, at low ventricular volumes, radial wall stress. Normalized flow amplitude is less sensitive to myocardial coronary compliance and resistance, and to the relation between active fiber stress, time, and sarcomere shortening velocity. Introduction Coronary arterial inflow varies in time during the cardiac cycle. Systolic inflow is smaller than diastolic inflow, demonstrating that the pulsatility of coronary flow is not caused by variation of the arterial venous pressure difference. Instead, the phasic pattern of coronary inflow has been attributed to changes in cross-sectional area of the myocardial vessels. For example, a transient decrease in cross-sectional area affects coronary flow in two ways. First, coronary inflow decreases and outflow increases because blood is squeezed out of the coronary bed. Second, both coronary inflow and outflow decrease because of the increase of coronary resistance. Vessel cross-sectional area depends on the local coronary pressure, the vessel wall mechanical properties and the embedment of the vessel in the myocardial tissue. The relation between vessel cross-sectional area and transmural pressure is nonlinear,11 and changes due to autoregulation. The embedment in the myocardial tissue can be represented by myocardial wall stress, which consists of two contributions: the stress in the collagen matrix, through which the vessel is tied to the surrounding tissue, and the intramyocardial pressure in the interstitial fluid. In the end, the pattern of coronary flow is mainly determined by contraction of the myofibers in the cardiac wall, since myofiber contraction determines the level of both myocardial wall stress and coronary pressure. Many experiments have been performed to elucidate the interplay between the factors governing coronary flow. It has been observed that the pulsatile component of the coronary inflow signal (1) is about proportionally related to left ventricle (LV) contractility, flow amplitude being about zero when contractility is about zero,16 (2) is virtually independent of systolic LV pressure in isovolumic beats, executed at various LV volumes, for pressures up to about 13 kPa,15,17 and (3) is about the same in isobaric beats at low LV pressure as in isovolumic beats at LV pressures up to about 13 kPa.15 It was observed also that minimum systolic flow (4) is virtually independent of LV pressure for pressures below 13 kPa, but (5) decreases with increasing LV pressure for pressures above about 13 kPa.20 In addition, it has been found that (6) systolic intramyocardial pressure may exceed left ventricular pressure in subendocardial layers in low afterload beats,18 and (7) that an increase of coronary perfusion pressure leads to an increase of intramyocardial pressure.18 Mathematical models have been proposed as a tool to interpret the experimental data. In early models, the interaction between the coronary vessel and the myocardium was modeled through the intramyocardial pressure only. In the waterfall model9 and the intramyocardial pump model,1,2,6,22 this pressure was assumed to be determined completely by left ventricular pressure, with intramyocardial pressure decreasing linearly from left ventricular pressure at the endocardial surface to zero at the epicardial surface. These models can explain the pulsatility of coronary inflow under normal physiological conditions, and the experimental observations (1) and (5), listed above. In later models, the interaction between vessel and tissue was still modeled through intramyocardial pressure only, but intramyocardial pressure was assumed to depend both on left ventricular pressure and on transverse tissue stress, i.e. stress perpendicular to the muscle fiber direction. In a finite element model of the beating heart, incorporating transverse tissue stress, observation (6) was replicated.12 Transverse stress is also included in models by Beyar et al., together with nonlinear characteristics of the properties of the coronary bed, and exchange of fluid between the coronary vessels and the myocardial interstitium.5,29,30 In these models experimental observations (1) through (6) are replicated. Vis et al.26,27 extended the model for interaction between the vessel and the cardiac wall, by assuming that the coronary vessel is subject to an extravascular pressure that depends both on intramyocardial pressure and local tissue stress. So the effective compliance of the coronary vessel depends both on the compliance of the vessel wall and the stiffness of the myocardium, thus implementing an idea introduced before as the time-varying elastance concept.15,17 In a finite element model of one vessel in the myocardial wall Vis et al. investigated the influence of contractility, pressure, and circumferential wall stretch on vessel area, for static diastole and systole. With the model, observations (6) and (7) were reproduced. With the increasing complexity of the models, more experimental observations have been replicated. Simultaneously, the number of model parameters has increased, making it difficult to perform a detailed parameter sensitivity analysis and identify the critical model parameters. Therefore, our aim was to study the primary relations between left ventricular pressure and volume, wall stress in fiber and transverse direction, intramyocardial pressure and the coronary blood flow, with a mathematical model with a limited number of parameters, and to assess the sensitivity of the model results to the model parameter settings. Material and methods The complete model consists of four parts, describing ventricular wall mechanics, myocardial constitutive properties, intramyocardial pressure, and the coronary and systemic circulation.Figure 1.Passive material behavior. Top: passive stress as a function of stretch ratio λ for uniaxial stretch along fiber direction (σm,f) and radial direction (σm,r). Bottom: passive left ventricular pressure–volume relation according to model (solid line) and experimental data;19 experimental data given for average minimum and maximum volume (dashed line), and minimum and maximum volume ± 1 SD (dash-dot lines). Ventricular Wall Mechanics The model of ventricular wall mechanics describes how left ventricular pressure and volume are related to local tissue properties, i.e. fiber stress and strain, and radial wall stress and strain. The model is based on a previously published model,3 which is extended to describe the influence of radial wall stress. While the original model is derived for arbitrary ventricular geometries with rotational symmetry, here we will derive the equations for the special case of a thick walled sphere. We consider the sphere to consist of a set of nested thin spherical shells. In each shell, stresses satisfy the condition of force equilibrium in radial direction: where σr, σc, and σl denote the radial, circumferential and longitudinal component of the tissue stress tensor, respectively, and r indicates the radial position in the wall. We neglected shear stress components in view of the low shear stiffness of the tissue. The myocardial tissue was assumed to be incompressible, and to consist of myofibers, embedded in a collagen matrix. The Cauchy stress tensor in tissue is written as: where pim denotes the intramyocardial pressure, , the stress in the collagen matrix, and σa the stress generated in the myofibers along the myofiber direction We assume that the fibers are located in spherical shells at an angle α with the circumferential direction, and adopt the assumption by Arts,3 that the stress in the collagen matrix in the plane of a shell is completely determined by the passive stress σm,f in the fibers. Then it holds: where σm,r represents the radial wall stress, generated in the collagen matrix, and the total fiber stress σf is introduced as the sum of passive and active fiber stress. Substitution in (1) yields: This equation describes how, in each shell, fiber stress σf and radial wall stress in the collagen matrix σm,r contribute to the variation of total radial wall stress σr in radial direction. Together, all shells increase radial wall stress from zero stress at the outer surface (r = ro) to minus left ventricular pressure at the inner surface (r = ri): The relation between wall stress and left ventricular pressure is found by integrating Eq. (4) from the endocardial to the epicardial surface. To evaluate the integral, we first adopt the assumption by Arts3 that σf is constant across the wall. Secondly, we assume that a representative position can be found, such that: Evaluation of the integral and substitution of the boundary conditions (5) yields: where ri and ro are rewritten in terms of the cavity volume Vlv and wall volume Vw, and is introduced as a short notation of Apart from the term this relation is identical to that derived in Arts et al.3 To complete the model relating wall mechanics to cavity mechanics, Eq. (7) is complemented by a relation between ventricular volume and tissue strain. We choose the passive ventricle at zero transmural pressure as a reference state. In this state, we assume a sarcomere length ls0 and a cavity volume Vlv0. It has been shown,3 that the fiber stretch ratio λf can be approximated by: This ratio corresponds to the circumferential stretch ratio at the outer surface of a shell that contains the left ventricular cavity volume and one third of the wall volume. At the same location, the radial stretch ratio λr equals: where we assumed incompressibility of the myocardial tissue. Equations (7)–(9) describe how global ventricular properties Vlv and plv are related to local tissue properties λf, λr, σf and Myocardial Constitutive Properties The model of ventricular mechanics is completed with constitutive laws for fiber stress and radial stress. The active fiber stress σa was modeled to depend on contractility c, sarcomere length ls, time elapsed since activation ta and sarcomere shortening velocity vs as: with contractility c (0 ≤ c ≤ 1), scaling constant σar, and: Here, ls,a0 denotes the sarcomere length below which active stress becomes zero and ls,ar represents the sarcomere length to which the reference stress σar is referred to. Times ta and tmax denote the time elapsed since activation, and the duration of the twitch, respectively. Velocity v0 represents the unloaded sarcomere shortening velocity, while cv governs the shape of the stress–velocity relation. Parameter values in the active stress law (Table 1) are chosen in agreement with experimental data.8,13Table 1.Reference settings for parameters in the model of the left ventricle.ParameterValueUnitParameterValueUnitVw20010−6 m3σar55103 PaVlv,06010−6 m3c1–ls,01.910−6 mls,a01.510−6 mσf00.9103 Pals,ar2.010−6 mσc00.2103 Patmax40010−3 scf12–v01010−6 m s−1cr9–cv0– The passive stress along the fiber direction is modeled as: where it was assumed that no stress can be transmitted in compression. Passive transverse stress is modeled similarly. Since the radial direction at ventricular level coincides with the transverse direction at the tissue level, we write the transverse stress model in terms of radial stress and stretch ratio: Settings of parameter values σf0, cf, σr0 and cr are based on experimental data on the pressure volume relation of the passive left ventricle.19 In the latter study, the pressure volume relation was described by a logarithmic relation, involving chamber stiffness at positive and negative pressures, and maximum and minimum attainable volume. The bottom panel of Fig. 1 shows the experimental data, and our model prediction for parameter settings in Table 1. Uniaxial stress–strain responses are shown in the top panel of Fig. 1.Figure 2.Schematic representation of the left ventricle (LV) in the systemic circulation and coronary circulation; Rart, Rper and Rven represent the systemic arterial, peripheral and venous resistance, respectively; MV and AV represent the mitral and aortic valve, respectively; Cart and Cven the systemic arterial and venous compliance; Lven and Lart the inertia of the blood in the venous and arterial system; Rart,c, Rmyo,1, Rmyo,2 and Rven,c represent the coronary arterial, the two intramyocardial and venous resistances; Cart,c, Cmyo,c and Cven,c the coronary arterial, intramyocardial and venous compliance; part,c, pmyo,c and pven,c the pressure in the coronary arteries, myocardium and veins, respectively; pao and pla the aortic and left atrial pressure, respectively; qao, qm represent the AV flow and MV flow; qart,c and qven,c the coronary arterial inflow and venous outflow; represents the intramyocardial pressure at the representative radial position When simulating isovolumic and isobaric contractions, the pressure drop across the coronary circulation is switched from pao− pla to a constant perfusion pressure pper.Figure 3.Hemodynamics in the beating heart. Left: simulation of the physiological state with aortic pressure driving the coronary circulation. Right: simulation of isolated heart experiment with a constant perfusion pressure and maximum vasodilation. From top to bottom: ventricular pressure plv, intramyocardial pressure (bold) and radial wall stress aortic valve (AV) flow qao (bold) and mitral valve (MV) flow qm; left ventricular cavity volume Vlv; coronary perfusion pressure pper, myocardial coronary pressure pmyo,c (bold) and left atrial pressure pla; coronary inflow (bold) and outflow per 100 g wall volume; volume of myocardial coronary bed Vmyo,c.Figure 4.Hemodynamics at maximal vasodilatation and constant perfusion pressure, expressed in time courses of ventricular pressure plv, ventricular volume Vlv, radial wall stress intramyocardial pressure and coronary arterial inflow Left: isovolumic beats for ventricular volumes ranging from 120 to 20 ml. Middle: isovolumic beats at 60 ml for contractility parameter c ranging from 1.0 to 0.1. Left: isobaric beats for systolic pressures ranging from 16 to 2 kPa. Intramyocardial Pressure To derive the expression for the intramyocardial pressure, again we consider the ventricular wall to be composed of a number of nested shells. The pressure in between two shells is assumed to be a fraction β of the left ventricular pressure. In analogy to Eq. (5), this pressure is in equilibrium with radial tissue stress σr: In the model, we assume a linear variation of β with the transmural position in the wall, with β = 1 at the endocardial surface, and β = 0 at the epicardial surface. Again stress at the shell containing the LV cavity volume and one third of LV wall volume is considered representative. For this location at a radius in the wall we find: where we introduced the notation and the radial positions ri, ro and are defined as: Equation (17) describes intramyocardial pressure as a function of LV pressure and volume: the volume dependency enters the equation through the radial stress σm,r [Eq. (15)], which depends the radial stretch ratio λr [Eq. (9)], which in turn depends on LV volume [Eq. (8)]. Systemic and Coronary Circulation The model of left ventricular wall mechanics is incorporated in lumped parameter models for the coronary and systemic circulation (Fig. 2). The aortic (AV) and mitral valve (MV) are modeled as an ideal diode. Vessels are modeled with constant resistances R, inertances L and capacitances C. The pressure drop Δp across each of these components is given by with V the volume in the capacitance and q the flow through a resistance or an inertance. The pressure–volume relation of the capacitance represents a linearization around the physiologic working point, V0 representing the volume at zero pressure. Values of parameters in the circulation model were based on literature (Table 2).Table 2.Reference settings for parameters in the circulation model; coronary resistance values in parentheses represent maximum vasodilation.Systemic circulationCoronary circulationParameterValueUnitParameterValueUnitRart5106 Pa s m−3Rart,c700 (200)106 Pa s m−3Rper120106 Pa s m−3Rmyo,1900 (100)106 Pa s m−3Rven5106 Pa s m−3Rmyo,2900 (100)106 Pa s m−3Cart2010−9 m3 Pa−1Rven,c200106 Pa s m−3Cven80010−9 m3 Pa−1Cart,c0.0310−9 m3 Pa−1Vart,050010−6 m3Cmyo,c1.410−9 m3 Pa−1Vven,0300010−6 m3Cven,c0.710−9 m3 Pa−1Lart60103 Pa s m−3Vart,c0610−6 m3Lven60103 Pa s m−3Vmyo,c0710−6 m3Vblood500010−6 m3Vven,c01010−6 m3 The connection between the model of LV mechanics and the coronary circulation model is made through the intramyocardial pressure, that acts on the myocardial capacitance Cmyo,c. The values of the coronary capacitances were based on measurements by Spaan et al.:23 0.0022, 0.091 and 0.045 ml mm Hg−1 100 g−1 LV in large coronary arteries, myocardial coronary bed, and coronary veins, respectively. Zero pressure volumes were chosen such that under normal physiological conditions, time-averaged coronary volume was about 15 ml 100 g−1 of LV tissue, distributed over arterial, myocardial and venous vessels in a ratio of 1:2:2.23 In the coronary circulation, resistance values during normoxia and hyperemia were derived from Chilian et al.7 In that study, total coronary resistance under normal and vasodilated conditions was measured to be 66 and 14 mm Hg min g ml−1, respectively. Distribution of resistance over the arterial, myocardial and venous compartment was measured to be 25, 68 and 7% under normal conditions, and 42, 27 and 31% under maximal vasodilation. Systemic parameters are chosen to yield representative function curves for a 70 kg adult, at a heart rate of 75 bpm. LV wall volume was set to 200 ml, and cavity volume at zero pressure was set to 30% of this volume. The arterial load was modeled by a three-element windkessel model, consisting of a characteristic aortic impedance Rart, an arterial compliance Cart, and a peripheral resistance Rper. The peripheral resistance was chosen to yield realistic time-averaged values of aortic pressure and aortic flow. Next, arterial capacitance was chosen to yield realistic values of minimum and maximum aortic pressure. Total blood volume was set to 5000 ml. The blood volume at which mean systemic pressure is zero was assumed to be equal 70% of total blood volume, about 85% of which is contained in the venous system. Venous capacitance was chosen such that the additional 30% of blood volume leads a mean systemic pressure of about 2 kPa. Simulations Performed With the model, a number of simulations were performed. First we considered the normal physiological situation, with normal vessel tone and coronary flow driven by the difference between aortic and left atrial pressure. Next we investigated the changes induced in the isolated heart setup used by Krams et al.,16,17 by the combination of maximum vasodilation, constant perfusion pressure and zero coronary outflow pressure. Then, under these conditions we simulated isovolumic and isobaric beats, for various settings of LV volume, pressure and contractility. Following Krams et al.16,17 and Pagliaro et al.,20 results were analyzed in terms of minimal systolic coronary flow and the normalized coronary flow amplitude (NFA): Finally, we investigated sensitivity of NFA to changes in myocardial radial stiffness, active material properties, coronary myocardial resistance, and coronary capacitance.Figure 5.Analysis of coronary inflow data from simulations in Fig. 4, expressed in normalized arterial coronary flow amplitude (NFA), defined in Eq. (22), and minimal coronary arterial inflow Left: NFA (top) and (bottom) in isovolumic beats as a function of maximum left ventricular pressure for contractility parameter c varying from 0.1 to 1.0; in top panel experimental data15 are given by the dashed line (reference), dash-dotted line (high contractility) and dotted line (low contractility). Middle: NFA and in isovolumic beats as a function of contractility parameter c for LV volumes of 40, 60, and 100 ml. Right: NFA and in isobaric beats as a function of maximum ventricular pressure for contractility parameter c ranging from 0.6 to 1.0. Results The Normal Beating Heart To illustrate the behavior of the model, first the normal physiological situation was simulated, with normal vessel tone and coronary flow driven by the difference between aortic and left atrial pressure. Ventricular and intramyocardial pressure rise to 16.1 and 9.1 kPa, respectively (Fig. 3, left panel). Intramyocardial pressure is almost completely determined by ventricular pressure, since radial wall stress is virtually zero. Maximum aortic and mitral flow are 646 and 307 ml/s, respectively. Stroke volume is 75.6 ml at an ejection fraction of 64%. Perfusion pressure equals aortic pressure and varies between 11.4 and 16.3 kPa. Myocardial coronary pressure varies between 3.0 and 12.3 kPa. Left atrial pressure remains about constant at 1.45 kPa. Mean coronary arterial inflow and venous outflow are 2.2 ml s−1 100 g−1 LV wall volume. Coronary arterial inflow displays a two peaked pattern, the peaks occurring in early filling and in ejection. Diastolic inflow is larger than systolic inflow. Coronary arterial outflow shows only one peak per cycle, occurring during ejection. Myocardial coronary volume varies between 10.7 and 12.5 ml. Constant Perfusion Pressure and Maximal Vasodilation This simulation represents the experimental situation in the isolated heart with a constant perfusion pressure of 10 kPa and maximal vasodilation. Myocardial coronary pressure varies between 1.5 and 10.0 kPa (Fig. 3, right panel). Arterial inflow varies between −0.1 and 14.2 ml s−1 100 g−1 LV wall volume, with a mean of 8.3 ml s−1 100 g−1 LV wall volume. In the arterial inflow curve, the positive flow peak in early systole has disappeared. Myocardial coronary volume varies between 5.8 and 12.4 ml. The Isovolumic Beating Heart The isovolumic experiments by Krams et al.15–17 were simulated at a constant perfusion pressure of 10 kPa and maximal vasodilation. Maximum LV pressure decreases with decreasing LV volume (Fig. 4, left panel). As LV volume decreases below 60 ml, the volume at which pressure in the passive LV is zero, diastolic LV pressure becomes negative. At these volumes, radial wall stress becomes positive, while being constant during a beat. With decreasing volume, maximum intramyocardial pressure and coronary arterial inflow decrease, whereas minimum coronary arterial inflow increases. The normalized coronary flow amplitude (NFA) [Eq. (22)] decreases with decreasing LV pressure (Fig. 5, top left panel), but pulsatility persists at zero ventricular pressure. NFA decreases with decreasing contractility as well. Minimum coronary arterial inflow increases with decreasing LV pressure (Fig. 5, bottom left panel). At low pressures, the slope of the pressure–flow relation decreases. For a constant LV volume of 60 ml, developed LV pressure decreases with decreasing contractility (Fig. 4, middle panel). At this volume of 60 ml, radial wall stress is zero. Maximum intramyocardial pressure decreases with maximum LV pressure. Maximum coronary arterial inflow decreases slightly, whereas minimum coronary arterial inflow increases strongly. With decreasing contractility, normalized coronary flow amplitude (NFA) decreases about linearly, while minimal arterial inflow increases linearly (Fig. 5, middle panel). For other volumes, a similar behavior is found. The Isobaric Beating Heart Simulations of hemodynamics in the isobaric beating heart were performed at a filling pressure of 1 kPa, a constant perfusion pressure of 10 kPa, maximal vasodilation and contractility parameter c = 1. With decreasing LV pressure during ejection, minimal LV volume decreases and radial wall stress increases (Fig. 4, right panel). Reduction of maximum LV pressure from 16 to 12 kPa is accompanied by a reduction of maximum intramyocardial pressure to 7.6 kPa. Maximum arterial inflow decreases, whereas minimum arterial inflow increases. Further reduction of LV pressure below 12 kPa hardly affects maximum intramyocardial pressure, due to the increasing contribution of radial wall stress. Maximum and minimum arterial inflow remain about constant. These changes are reflected in the right panel of Fig. 5. With decreasing left ventricular pressure, initially NFA decreases and minimum arterial inflow increases. Below a threshold pressure of about 10 kPa, these quantities become about constant. A similar behavior is found for lower contractilities. Sensitivity Analysis Results of the sensitivity analysis are shown in Fig. 6. Sensitivity to settings of parameters in the model for radial wall stress was investigated, since in our model radial wall stress is an important determinant of intramyocardial pressure. If the radial stress parameter cr0 [Eq. (15)] is decreased or increased by a factor 2, the pulsatile character of the coronary inflow at lower pressures persists, both for isovolumic and isobaric beats (Fig. 6, top panel). In isovolumic beats, maximum LV pressure increases with decreasing cr0. At the lowest LV volume simulated, 15 ml, maximum left ventricular pressure does not drop below 14 kPa, when cr0 is set to zero. In isobaric beats, at high pressures NFA is not affected by changes in cr0. Below an ejection pressure of 10 kPa, NFA increases with increasing radial stiffness. When cr0 is set to zero NFA decreases proportionally with left ventricular pressure, until pulsatility is lost at zero left ventricular pressure. Sensitivity to settings of parameters in active stress model was investigated, since myofiber contraction is the origin of both left ventricular pressure and radial wall stress [Eq. (7)], and thus of intramyocardial pressure and of coronary flow impediment. In both isovolumic and isobaric beats, NFA increases slightly if the twitch duration tmax [Eq. (12)] is reduced from 400 to 300 ms (Fig. 6, 2nd panel from top). Changing the linear relation between fiber stress and sarcomere shortening velocity into a hyperbolic one (by setting cv = 1 in Eq. (13) does not affect NFA in isovolumic beats, where shortening velocity is zero. NFA in isobaric beats is hardly affected. Increasing the sarcomere length ls,a0, below which active stress is zero [Eq. (11)], from 1.5 to 1.6 μm yields a reduction in NFA in isovolumic and isobaric beats with a ventricular pressure below about 10 kPa (Fig. 6).Figure 6.Sensitivity of relation between maximum left ventricular pressure and normalized flow amplitude (NFA) to settings of model parameters, for isovolumic beats (left) and isobaric beats (right). Results for reference parameter settings are indicated with ‘ref’. Top: radial stiffness parameter cr0 set to twice (‘high’) or half (‘low’) the reference value, or set to zero (‘zero’). Next: variation of active stress model: twitch duration reduced from 400 to 300 ms (‘tim’), stress–velocity relation changed from linear into hyperbolic (‘vel’), and stress-free sarcomere length increased from 1.5 to 1.6 μm (‘len’). Next: all coronary resistances Rart,c, Rmyo,1, Rmyo,2 and Rven,c set to twice (‘high’) or half (‘low’) their reference values. Bottom: coronary myocardial capacitance Cmyo,c set to twice (‘high’) or half (‘low’) the reference value. Sensitivity to coronary resistance was studied to mimic change in resistance due to variation in perfusion pressure. A two-fold simultaneous increase or decrease of all resistances in the coronary circulation model (Rart,c, Rmyo,1, Rmyo,2 and Rven,c) does affect NFA more in isovolumic beats than in isobaric beats (Fig. 6, 3rd panel from top). Still, changes in NFA are limited, and pulsatility at low pressures is maintained. Finally, sensitivity to myocardial compliance Cmyo,c was studied, since pulsatility of coronary inflow is closely related to changes in coronary volume. When myocardial compliance Cmyo,c is set to twice or half its reference value, NFA-values change quantitatively, but the phasic nature of coronary flow at low left ventricular pressure remains, both in isovolumic and isobaric beats (Fig. 6, bottom panel). Again, the effect is larger in isovolumic beats than in isobaric beats. Discussion Model Setup The aim of this study was to design a model with a limited number of parameters for investigation of the primary relations between left ventricular pressure and volume, wall stress in fiber and transverse direction, intramyocardial pressure and the coronary blood flow. Central to the model are Eq. (7), which shows how fiber stress, that ultimately drives the cardiac cycle, is converted into both LV pressure and radial wall stress, and Eq. (17), that describes how LV pressure and radial wall stress contribute to intramyocardial pressure. Finally, it is the variation of intramyocardial pressure that causes a variation in arterial coronary inflow through a change in coronary volume, represented by intramyocardial compliance. Since the use of a simple model is not without danger, we will address the impact of the main model simplifications. In the model of ventricular wall mechanics spatial homogeneity of fiber stress and strain was assumed. The plausibility of this assumption has been illustrated in finite element models of ventricular mechanics.21,24 The assumption that in plane stress is dominated by fiber stress [Eq. (3)] is realistic during systole, when myofibers are active and the main flow impediment occurs. Obviously, radial wall stress is spatially inhomogeneous, since the ventricle is thick walled. Yet we introduced a representative stress computed at a representative radial position enclosing one third of the wall volume, in evaluating the integral of σm,r over the wall thickness [Eq. (6)]. In principle, since the transmural variation of radial strain is known once LV cavity volume is known, using the constitutive Eq. (15), the integral in Eq. (6) can be determined exactly. Due to the strong increase of radial wall strain from the outside to the inside surface and the nonlinearity of the constitutive law, its value would be dominated by the inner layers. However, in those layers the assumption of a solid wall with a smooth endocardial surface is in error: the endocardial surface shows many invaginations, relieving any radial stress that would build up in the subendocardial tissue. The choice to use representative positions, strains and stresses is compatible with the aim of the study, to investigate determinants of the coronary flow signal. Obviously the model cannot describe spatial differences in coronary perfusion,4 which are clinically relevant in relation to vulnerability to ischemia.10 In view of the use of a representative intramyocardial pressure, and the use of average values for the coronary compliances and resistances, the flow in the model should be regarded as a mean flow over the myocardial wall. In the constitutive model for the cardiac tissue, the description of the active stress component is chosen as simple as possible, while maintaining the basic dependence of active stress on time, sarcomere length, and sarcomere shortening velocity. Within the limitations of the model, parameter settings are chosen to mimic experimental data.8,13 The model could be extended to incorporate the sigmoidal relation between stress and sarcomere length, or the increase of twitch duration with increasing sarcomere length. However, we prefer the current version with a limited number of parameters, in view of the sensitivity analysis. The model for passive tissue stress is simplified, since we do not model a complete three-dimensional state of stress. Yet, the behavior shown in Fig. 1 agrees well with experimental data.28 We approximate extravascular pressure by intramyocardial pressure, thus neglecting the contribution of local tissue stress.26 In fact this assumption is made in many other models, including the waterfall model,9 the intramyocardial pump model,1,2,6,22 and the models by Huyghe et al.12 and Beyar et al.5,22,30 In addition, in our model intramyocardial pressure is determined completely by the model of LV wall mechanics. Thus, we can not replicate the observation that intramyocardial pressure depends on perfusion pressure.18 We feel this simplification is allowed in our simulations of isovolumic and isobaric beats, in which we assumed a constant perfusion pressure. In the coronary circulation model resistances are constant. This is an approximation to the real situation, in which vessel diameter and therefore vessel resistance depend on vessel transmural pressure. The nature of the approximation is two-fold. First, we neglect the chronic change in resistance due to a chronic change in perfusion pressure. This is not critical for the value of NFA, as is apparent from our sensitivity analysis where we applied a two-fold change of coronary resistance (Fig. 6). Second, we neglect cyclic changes in resistance during the cardiac cycle, which seems incompatible with the ability in the model to temporarily store blood in the coronary capacitance. Yet, the phasic nature of coronary flow is reproduced in the model. We explain this by noting that a change in vessel diameter during the cardiac cycle is associated with change in coronary volume, that affects coronary inflow irrespective of the location in the coronary tree where the volume change takes place. However, the associated change in resistance is important in the smallest vessels only, due to the nonlinear relation between vessel diameter and resistance. Obviously, the effect of changes in coronary resistance during the cycle may show up in aspects of the coronary flow signal, that are not captured by the normalized flow amplitude NFA. The simplifications in the model may be seen as limitations, if one has the goal to explain all experimental observations available. In this study, we consider it a strength of the model, in view of our aim to investigate primary interactions and dependence on parameter settings. Results In contrast to the situation in all other organs, coronary inflow of the heart muscle occurs mainly in diastole and is significantly impeded in systole. Experimentally, the amplitude of the coronary inflow signal was found to be only weakly coupled to systolic LV pressure in isovolumic beats at various LV volumes.15 This behavior is illustrated by the three fits to experimental data sets, shown in top left panel of Fig. 5. As in the experiment, in our model pulsatility of coronary inflow is maintained at low left ventricular pressure, although dependence of NFA on pressure is slightly higher than in the experiment. The mechanism by which pulsatility of coronary inflow is maintained in isovolumic beats is explained as follows (Fig. 4, left panel). Total LV pressure is the sum of diastolic pressure in the passive ventricle, and the extra pressure, generated by muscle contraction. Diastolic pressure decreases with decreasing volume and becomes negative as volume decreases below the resting volume of 60 ml. We note that in the experiments by Krams et al. negative diastolic pressure was induced as well, by applying suction to the balloon, inserted in the LV cavity.16 Since LV volume is constant during a beat, radial wall stress and its contribution to LV pressure are constant as well. The extra pressure, related to muscle contraction, reduces with decreasing volume, but at the lowest volume of 20 ml it still is about 15 kPa, and maximum total LV pressure is still about 11 kPa. The fact that the variation in LV pressure, which determines pulsatility of coronary flow, decreases more slowly with decreasing volume than maximum LV pressure, is reflected in the value of NFA shown in Fig. 5. The relation between maximum LV pressure and minimum coronary inflow, shown in the bottom left panel of Fig. 5, qualitatively corresponds with experimental data by Kouwenhoven et al.14 and Pagliaro et al.20 In the latter study, minimum coronary inflow was found to decrease with increasing maximum LV pressure for pressures above 13 kPa, but was virtually independent of systolic LV pressure below 13 kPa. It was suggested that at low pressures ‘the shielding effect of the contracting ventricle prevents ventricular pressure from being transmitted in the myocardial wall’.20 In our model, this shielding effect is identified as radial wall stress. Experimentally, NFA in isobaric beats at low LV pressure was found to be about equal to NFA in isovolumic beats.15 This finding is replicated in our model (Fig. 5, right panel). In contrast to the isovolumic case, the relation between maximum pressure and NFA is non monotonous, with distinct behavior at high and low pressures. At high pressures, LV volume remains mostly above the passive resting volume of 60 ml, and intramyocardial pressure is dominated by LV pressure. At low pressures, the LV ejects to far below the passive resting volume of 60 ml, and the decreasing contribution of LV pressure to intramyocardial pressure is compensated by the increasing contribution of radial wall stress. It is to be noted that in the isobaric beats both contributions to intramyocardial pressure, radial wall stress and left ventricular pressure, vary in time, while in the isovolumic beats radial wall stress is constant and cyclic flow impediment is due to the varying contribution of left ventricular pressure only (Fig. 4). This explains the two peaks in the intramyocardial pressure-signal in Fig. 4: the first one is related to the rise in left ventricular pressure, while the second one arises from the rise in radial wall stress, related to the decrease in left ventricular volume. Another experimental observation is that NFA is about proportionally related to LV contractility, expressed by maximum elastance, NFA being about zero when contractility was about zero.17 This observation is replicated by our model, in which NFA changes about proportionally with the contractility parameter c, irrespective of the volume setting (Fig. 5). For low contractilities, minimum coronary inflow approaches the theoretical value of 8.3 ml s−1 100 g−1 of tissue, obtained from the coronary perfusion pressure of 10 kPa, the total coronary resistance of 6 × 108 Pa s m−3, and a wall volume of 200 g. Finally, it has been observed that intramyocardial pressure may exceed left ventricular pressure in low afterload beats.18 This is also the case in our model (Fig. 4, right panel), and explained again from the contribution of radial wall stress at the low ventricular volumes associated with the low LV ejection pressures. Sensitivity Analysis Sensitivity of NFA to the radial stiffness parameter cr0 in isovolumic beats is low (Fig. 6, top left). This can be best understood from Eq. (17), which shows that only the constant level of radial wall stress is affected by changing cr0, while the variation of left ventricular pressure, and thus of intramyocardial pressure and arterial inflow, is unaffected. The increase in left ventricular pressure with decreasing stiffness is explained from Eq. (7), which shows that fiber stress is converted into both LV pressure and radial wall stress. With decreasing radial stiffness, an increasing part of fiber stress is converted into LV pressure. In the extreme case of zero radial stiffness, minimum left ventricular pressure is about 14 kPa, which is higher than the pressure measured in the experiment (Fig. 6).15 Variation of NFA with radial stiffness is more prominent in isobaric beats at low LV pressure (Fig. 6, top right). An increase of radial stiffness causes a more rapid increase of radial wall stress with decreasing volume below the equilibrium volume. Thus the pressure at which the contribution of radial wall stress to intramyocardial pressure takes over from the contribution of ventricular pressure is increased (17), and pulsatility of coronary inflow is maintained. At zero radial stiffness, intramyocardial pressure is determined fully by left ventricular pressure, explaining loss of pulsatility of coronary flow with decreasing ejection pressure, in disagreement with experimental findings.15 Sensitivity of NFA to settings of the active stress model (Fig. 6) is low, as far as the velocity and time dependence is concerned. NFA was more affected by increasing the stress-free sarcomere length from 1.5 to 1.6 μm. In fact this variation has the same effect as decreasing contractility at low sarcomere lengths. Thus, in this case the decrease of NFA is explained from the decrease in variation of left ventricular pressure, which is reflected in a decreased variation of intramyocardial pressure. The increase of NFA with increasing myocardial compliance (Fig. 6), both in isovolumic and isobaric beats, is explained from the fact that, at the same variation of intramyocardial pressure, the volume change of the myocardial bed increases. Hence the pulsatility of arterial inflow is increased as well. A decrease of myocardial compliance has the opposite effect. The first order effect of a variation of coronary resistance is a change in both the average and pulsatile component of coronary flow, as was illustrated already in Fig. 3. In this approximation, the relative changes in average and pulsatile flow component are equal, causing NFA to remain unaffected. However, since coronary compliance is kept constant, the relative contribution of flow related to volume change of the coronary bed becomes less when decreasing coronary resistance, causing the decrease of NFA, observed in Fig. 6. Increasing coronary resistance has the opposite effect. Relation with Other Models In comparison to the waterfall model9 and the intramyocardial pump model,1,2,6,22 our model is more advanced in its description of intramyocardial pressure, adding the effect of radial wall stress. Radial wall stress becomes important only below the zero-pressure volume of the passive ventricle. Thus, the early models still are applicable in normal physiological conditions, in which ventricular volume is higher than this zero-pressure volume during the major part of the cardiac cycle. In comparison to the models by Beyar et al.5,29,30 our model is simpler, emphasizing primary interactions, at the cost of information on transmural variation of coronary flow and long-term exchange of fluid between the coronary vessels and the interstitium. However, the reduced complexity yields increased insight into primary determinants of coronary flow impediment. Main limitation of our model with respect to the model presented by Vis et al.,26,27 lies in neglecting the direct contribution of tissue stress to the extravascular pressure and effective compliance of the coronary vessel. The approach of Vis et al. requires a detailed analysis of the equilibrium between the pressure in a cavity in the LV wall, and the stress in the surrounding tissue. However, such an analysis would compromise the simplicity, that we aim for in this study. Conclusion In conclusion, a mathematical model of the interaction between coronary flow and cardiac mechanics is presented, with a limited number of model parameters. The model replicates the experimental observations, that the phasic character of coronary inflow is virtually independent of maximum left ventricular pressure, that the amplitude of the coronary flow signal depends linearly on cardiac contractility, and that intramyocardial pressure in the left ventricular wall may exceed left ventricular pressure. The normalized amplitude of coronary inflow is mainly determined by contractility, reflected in dependence of active fiber stress on sarcomere length, and maintained at low ventricular volumes by radial wall stress. The sensitivity of the NFA to myocardial coronary compliance and resistance, and to the relation between active fiber stress, time, and sarcomere shortening velocity is low.

Ann_Biomed_Eng-2-2-1705509

Remote Non-invasive Stereoscopic Imaging of Blood Vessels: First In-vivo Results of a New Multispectral Contrast Enhancement Technology

We describe a contactless optical technique selectively enhancing superficial blood vessels below variously pigmented intact human skin by combining images in different spectral bands. Two CMOS-cameras, with apochromatic lenses and dual-band LED-arrays, simultaneously streamed Left (L) and Right (R) image data to a dual-processor PC. Both cameras captured color images within the visible range (VIS, 400–780 nm) and grey-scale images within the near infrared range (NIR, 910–920 nm) by sequentially switching between LED-array emission bands. Image-size-settings of 1280 × 1024 for VIS & 640 × 512 for NIR produced 12 cycles/s (1 cycle = 1 VIS L&R-pair + 1 NIR L&R-pair). Decreasing image-size-settings (640 × 512 for VIS and 320 × 256 for NIR) increased camera-speed to 25 cycles/s. Contrasts from below the tissue surface were algorithmically distinguished from surface shadows, reflections, etc. Thus blood vessels were selectively enhanced and back-projected into the stereoscopic VIS-color-image using either a 3D-display or conventional shutter glasses. Introduction Visually guided procedures provide instant feedback and meantime afford insights that would otherwise be difficult or even impossible to obtain.26 This is a primary reason for mankind’s continuous desire to extend vision beyond the boundaries of the human eye, which has resulted in many successful diagnostic imaging modalities like X-ray imaging, endoscopy, thermography, ultrasound scans and MRI that all found their way into the clinic. While investigating the principal feasability of a camera for imaging blood oxygenation levels, we noticed that our multispectral images also contained information about subcutaneous vasculature, with improved contrast in the near infrared.24 Near infrared imaging of superficial blood vessels in itself is not new; soon after infrared photography2,9,25 early electronic cameras were used for experiments.13,15 Since then, numerous near infrared imaging methods have been developed using transillumination mode and/or reflection mode.5,6,10,14, 22 Our multispectral camera, however, provides the opportunity to obtain normal color images within the visible range (VIS) which are pixel-to-pixel matched with images obtained in the invisible near infrared range (NIR). By combining the image information content of both spectral bands with a processing algorithm, we developed a new technique that allows selective enhancement of superficial blood vessels with selectable pigment suppression within a normal color image.23 The underlying principle does not necessarily require stereoscopic image acquisition to derive increased vessel contrasts from the tissue. When, however, developing a new medical imaging technique, deriving image information from patient tissue is not the only issue. Especially for visually guided procedures it is also crucial to put effort in an ergonomic human interface and avoid conflicts between the proprioceptional and visual perceptions of the user. The added value of stereoscopic image recordings already was recognized and successfully used to document medical cases more than a century ago.4 We likewise reasoned that offering enhanced blood vessel contrast at the cost of depth perception would restrict the usefulness of our technique and therefore decided to realize a stereoscopic version of our vessel contrast enhancement device. Visually guided procedures typically require a combination of well trained eyes and specific fine-motoric skills, since the coordination between eyes and hands is task dependent.16 For eye-hand coordination, brain processes for perception and action interact so closely that they cannot be separated and the influence of visual illusions (like human depth perception) to motoric tasks becomes stronger when input takes place via lower levels in the brain.3 The match between a human interface device and the visual and motoric brain processes strongly defines whether a technology can be applied intuitively or not. A new technology can be classified as intuitive if it speeds up the learning process for novices in a certain skill, without impairing the performance of persons already skilled in the existing art.20 Thus a useful increase in vascular contrast should neither imply a task-impairing decrease in depth perception (like in monoscopic techniques) nor distort spatial clues for human vision (like shadows) nor introduce false depth information (like projection parallax). As a first reconnaissance of practical feasibility, we applied our new technique to several visual clinical procedures for which we expected that improved visualization of blood vessels would be of interest, being:blood withdrawal;vein inspection in dark skin;detection of veins through iodide;inspection of varicose veins and nevi pigmentosum. Methods Instrumental Setup The instrumental hardware setup is schematically drawn in Fig. 1. Two synchronized identical single-chip CMOS-cameras (Vector Technologies Belgium, custom built) were equipped with apochromatic lenses and two identical custom-built dual-band LED-arrays (O2-View, the Netherlands). These LED-arrays were current controlled and each had two individually programmable channels for the emission of in total ±1.2 Cd visible white light with adjustable color temperature (consisting of 40 broadband white LEDs with a yellow accent and 20 broadband white LEDs with a blue accent) as well as one programmable channel for the emission of near infrared radiation (20 LEDs, 920 nm, max. 32 mW per LED). The light sources were constructed so that the geometrical beam profiles of VIS and NIR matched very closely (and thus also any resulting shadows and/or reflections). Left (L) and Right (R) image data was simultaneously acquired and streamed to a dual processor PC equipped with a stereoscopic monitor (Sharp LL-151–3D). It was also possible to connect a conventional cathode ray tube (CRT) monitor equipped with shutter glasses, for comparison of stereoscopic representation. Figure 1.Experimental setup. Two CMOS-cameras, with apochromatic lenses and dual-band LED-arrays, simultaneously stream Left (L) and Right (R) image data to a dual processor PC. Both cameras captured color images within the visible range (VIS, 400–780 nm) and grey-scale images within the near infrared range (NIR, 910–920 nm) by sequentially switching between LED-array emission bands. A schematical representation of image aquisition and data structure is drawn in Fig. 2. The CMOS-camera detector chips were equipped with a Bayer filter mosaic to obtain an RGB color image within the VIS. The camera pixels, however, were also sensitive to near infrared radiation because all three filter channels of the Bayer mosaic (red, green and blue) were designed as highly transparant within the NIR. Thus four NIR-pixels were acquired for each VIS RGB color pixel group. By sequentially switching between emission bands of the LED-arrays, images within the Visual range (VIS, 400–780 nm) and Near Infrared range (NIR, 910–920 nm) were acquired in an alternating fashion. VIS and NIR image-size-settings could be varied independantly. For 8-bit encoding depth, at image-size-settings of 1280 × 1024 for VIS and 640 × 512 for NIR, 12 cycles/s were obtained (1 cycle = 1 VIS L&R-pair + 1 NIR L&R-pair). At the cost of decreasing image-size-settings downto 640 × 512 for VIS and 320 × 256 for NIR, camera speed could be increased up to 25 cycles/s. Encoding depths of 10-bit and 12-bit were also available, but only used for stills (due to the lower obtainable framerate). Figure 2.Schematic diagram of image aquisition. The sequentially acquired alternating VIS and NIR raw image frames form 3D-matrices for the Left and Right channel. The NIR image size is smaller than the VIS image size to increase framerate while maintaining an overview of the imaged area. Enlarged details illustrate the NIR transparency of the Bayer pattern RGB-filters which are applied to obtain a VIS color image. The time domain axis is expressed in image cycles. Along this image cycle axis, the control signals for NIR and VIS LEDs (synchronized with respectively NIR and VIS camera exposures) are visualized. Data Aquisition Multispectral stereoscopic movies were recorded in several typical clinical settings for which the technique was considered as possibly useful. LED currents and diaphragm settings were chosen so that for each movie saturated pixels were avoided. A preview mode allowed aiming, adjustment of converging angle α and focusing of the cameras. After software triggering the stereo-camera streamed a sequence of 8-bit digitally encoded image cycles to PC-memory (using auto-incremental numbering). All images were automatically saved on a fast SATA harddisk-array. Camera and light source settings were automatically stored in a text file and located in the same directory. All patients and volunteers gave their informed consent for filming as well as for publishing the resulting image material. No diagnosis or therapy was based upon any of our results. Data Processing General Aspects Processing was performed with custom developed software (programming language C++) using the stored text file with camera and light source settings as additional input values. Figure 3 schematically represents this process. Figure 3.Schematic diagram of image processing. The images captured within the visible range (VIS) and the images captured within the near infrared range (NIR) are combined, which reveals blood vessel patterns below the skin. The left and middle column focus on edge-enhancement and suppression of superficial artifacts, the right column serves to fill-in the blood vessel lumen. For raw VIS & NIR images as well as processed results see figures 6, 7, 8 and 9. The processing method allowed discrimination between image information obtained from the tissue surface versus image information obtained from within the tissue. In order to achieve this, a distinction was made between shadows, reflections and absorption contrasts for both VIS and NIR. Suppressing Shadows on the Surface Since NIR and VIS beams were matched closely, shadows produced by irregular shapes at the tissue surface (e.g. skin structure, skin folds, nevi, hair, etc.) or by objects between the light sources and the tissue (fingers, needles, surgical tools, etc.) also matched well in both wavelength ranges. Our algorithm excluded such matching shadows from enhancement and left all useful aspects of shadows unaffected (e.g. depth clues). Useful Effects of Shadows and Lighting Geometry Due to the fact that the VIS and IR shadows matched very closely, the algorithm was able to selectively enhance contrast from below the surface while leaving shadows on the surface unaffected (thus not distorting these important depth clues). Due to the shallow angle of the lightbeams, the skin texture was pronounced. Due to the lighting from two sides, objects that were brought towards the tissue surface (e.g. needles, scalpels, probes, etc.) when lighted from both sides, could produce two separate (not too heavy) shadows. These shadows met and typically formed a “V” pattern when an object touched the surface in the middle of the field of view, thus providing extra information for depth perception. Enhancing Absorption Contrast of Blood Vessel Walls Edge detection by a Prewitt image filter was performed on each VIS and NIR image. Pixel positions containing edge information above a certain adjustable threshold in both spectral regions were classified as surface artefacts and excluded from enhancement. The boundary regions of absorption contrasts caused by structures below the tissue surface, produced edges that were mainly present in the NIR image. Enhancement was selectively performed only for pixel positions where the NIR image contained more edge information than the corresponding VIS image. These “valid” pixels identified the vessel boundaries. The positions of these valid pixels were stored in a 1st NIR mask. Discarding Reflections on the Surface Shiny areas that produced reflections and/or saturated pixels also matched in both wavelength ranges. This characteristic allowed to calculate a 2nd NIR mask in which superficial reflections were also excluded from enhancement. Neighboring pixels of identified saturated pixels were excluded from enhancement. The radius of this exclusion region was programmable, but due to the favorable anti-blooming behavior of the CMOS-camera chips, suppressing direct neighbor pixels appeared sufficient. Enhancing Absorption Contrast of Blood Vessel Lumen The “content” of blood vessels was separately enhanced by raising pixel values of the normalized NIR image to the power of N (with N user adjustable between 0.5 and 2.5) while discriminating NIR pixels below a freely adjustable noise threshold and excluding information from identified shadows. Multiplication with the 2nd NIR mask then produced a final enhancement mask for subsequential backprojection into the VIS image by pixel-to-pixel multiplication. Suppressing Contrasts Originating from Melanin Pigment Superficial contrasts within the VIS, originating from melanin pigment concentrations, could either be filtered out or left unchanged. Figure 4 illustrates the normalized distribution of the intensities for the aquired separate spectral bands (R, G, B and NIR) in relation to the intensity (IVIS) of the composed visible RGB-image. Intensities were calculated using the Intel ippiRGBtoGray function.11 Four clouds of data points can be discerned, being IR/IVIS (red channel), IG/IVIS (green channel), IB/IVIS (blue channel) and INIR/IVIS (NIR channel). By calculating the ratio of (IR/IVIS )/(INIR/IVIS ) and comparing the result for each pixel with an adjustable threshold, it can be decided whether or not to apply backprojection to a pixel. This concerns the pixels located within the overlapping region of IR/IVIS and INIR/IVIS within Fig. 4. The resulting difference is clear when comparing between Fig. 9c and d with regard to the visualization of nevi pigmentosum and hair. Figure 4.Relative intensity distribution of the different spectral bands. Dimensionless normalized distribution of the intensities for the 4 aquired red (IR), Green (IG), Blue (IB) and near infrared (INIR) individual spectral bands expressed in ratio to the normalized intensity (IVIS) of the composed RGB-image. The four data clouds (R, G and B labeled by their natural colors and NIR labeled as pink) show a generally marked separation, but especially for a number of pixels within the red and infrared the data clouds partly overlap. This indicates regions where the VIS contrast potentially is superior to the NIR contrast. By comparing an adjustable threshold with the calculated ratio of (IR/IVIS)/(INIR/IVIS) it can be decided which spectral band provides superior contrast for the pertaining pixel and thus whether or not it is used for enhanced backprojection. Data Presentation The images could either be displayed on an auto-stereoscopic liquid crystal display (LCD) monitor or on a conventional CRT monitor equipped with shutter glasses. The auto-stereoscopic LCD-monitor (Sharp LL-151–3D) was equipped with software controllable switching between stereoscopic (or 3D) and normal (monoscopic or 2D) mode. Monitor resolution was 1280 × 768 in monoscopic mode (XGA). In stereoscopic mode the available pixels were split-up in two separate images (L&R) by activation of a vertical LCD parallax barrier. The principle behind this technology is illustrated in Fig. 5. Figure 5.Principle of applied autostereoscopic LCD-monitor (reprinted with permission from Sharp). In 2D mode, only one camera-channel is displayed (either from the L or R camera) and the parallax barrier is not actuated. Both eyes of an observer therefore receive the same image and a conventional flat image with full resolution is seen. In 3D mode, both camera channels are displayed (L&R) and the parallax barrier is actuated. The left eye and right eye of an observer now receive different images, and a stereoscopic (in-depth image) with halve resolution is seen. The CRT-monitor (iiyama vision master 21) was used at a resolution setting of 1280 × 1024 in combination with wireless shutter glasses (e-Dimensional) and thus provided stereoscopic information without sacrificing resolution. The enhancement algorithm occurs on a pixel-to-pixel basis and does not affect resolution. Our device offered several imaging modes for data presentation:Monocopic raw VIS preview (normal full color vision)Monoscopic raw NIR preview (greyscale)Off-line stereoscopic looped VIS view with and without enhanced blood vessel back-projection (with freely adjustable enhancement settings).Off-line stereoscopic looped raw NIR view or enhanced NIR view (with freely adjustable enhancement settings)Stereoscopic stills in all modesMonoscopic stills in all modes (freely switchable between Left and Right) All modes offered the possibility for pause and scrolling forward or reverse frame-by-frame. By means of virtual slider controls and virtual pushbuttons the user interface allowed freely adjustable settings for shadow suppression, pigment suppression, noise threshold and vessel lumen fill-in contrast. Results Results for Blood Withdrawal During routine blood withdrawal the inserted needle tip remained only slightly visible on the VIS image due to skin surface deformation (see Fig. 6a). On the raw NIR image, however, the inserted needle remained visible within the tissue for a few mm with some metallic reflection. Subcutaneous bleeding during needle removal could be detected in the raw NIR image while the needle tip was still in the tissue (see Fig. 6b). This was clearly highlighted by backprojection in the VIS image (see Fig. 6c). The processing settings used to obtain this backprojection mainly laid the accent upon enhancing the absorption contrast of blood vessel lumen, whereas edge enhancement was set to minimum. Reflections on the thumb nail did not lead to image distortion and the blood volume in the nail bed showed more contrast. Figure 6.Routine blood withdrawal. Image pairs showing unprocessed images for VIS (a) and NIR (b) as well as the result after application of the new image processing method (c). Note the forked shadow (which is not effected by the enhancement algorithm), the clearly visualized subcutaneous bleeding and the improved visibility of the needle tip. Results for Dark Skin Hardly any vascular contrast is present within the VIS image (see Fig. 7a). The NIR image, however, is not affected by skin pigmentation and reveals a subcutaneous vascular pattern (see Fig. 7b). The combined information results in an enhanced image (see Fig. 7c). Figure 7.Influence of skin pigmentation. A dark skin color (a) has no significance for the applied NIR wavelength of 920 nm. Blood vessels provide good contrasts (b) and the resulting enhanced image (c) offers an improved visualization of the vasculature. Results for Vein Detection Through Iodide Iodizing skin portions before surgery is a common clinical procedure which darkens the skin and lowers the visibility of blood vessels. Our method offers imaging right through iodide. To demonstrate this, we filmed a glass Petri dish, placed horizontally upon a volunteers arm while filling it up to a 3 mm thick layer of iodide solution. From the VIS image no vessels could be detected through the iodide solution (see Fig. 8b) and hardly any through the empty Petri dish (see Fig. 8a). The raw NIR image, however, clearly showed superficial blood vessels even through the Petri dish (see Fig. 8c). Figure 8.Vein detection through iodide. Within the visible range, the superficial vasculature is only vaguely discernable (a). After filling the Petri-dish with a 3 mm thick layer of iodide solution, this fully blocks out the tissue view within the visual range (b), whereas a clear view of the vasculature remains possible at the applied NIR wavelength of 920 nm (fig 8c). Results for Varicose Vein and Nevi Pigmentosum Inspection The device was used to film varicosis patients during dermatological outpatient clinic. The visualization of varicose veins could be drastically improved. Even in cases where the blood vessels were covered by subcutaneous fat, they could be detected quite well (see Fig. 9). By either de-activating or activating the first NIR mask in the processing algorithm (see Fig. 3) it was also possible to choose whether pigment contrasts in the visual range were left intact or filtered out. The difference is clear when comparing between Fig. 9c and d with regard to the visualization of nevi pigmentosum and hair. By clicking a software screen button, this suppression of pigment contrast could be freely switched on and off during viewing. Figure 9.The VIS image does not contain much information about the underlying vascular pattern (a). The NIR image, however, clearly shows what’s hiding beneath the surface (b). Note that, when building the enhanced image, the nevus which is present in the VIS image can freely either be suppressed as a surface contrast (c) or be kept visible (d). Discussion From the figures presented in this article, it is clear that all experiments resulted in improved visualization of superficial blood vessels. Based upon the absorption coefficients of Hb versus HbO2, arteries provide superior contrasts at 920 nm. The presented images, however, show enhanced venous contrasts, because the larger arteries are buried deeper underneath the skin. Stereoscopic movies provide a more lifely impression of the underlying, but presentation of stereoscopic movies is not possible in a printed journal. We therefore offer the possibility to download viewer software and movies via the internet http://www.erasmusmc.nl/ThoraxcenterBME/html/research/additional/bloodvesselcamera.htm For both the autostereoscopic LCD and the CRT display plus shutter glasses, we found that the perception of depth was less for stills than for movies. This qualitative observation matches research findings on stereoscopic display techniques in X-ray technology8 and is also consistent with the differences between static and dynamic stereoacuity described by Mathias and Rudolf Sachsenweger.17 Stereoscopic movies also showed an increase in apparent image sharpness compared to stereoscopic stills, which may be explained by Shipley’s description of stereoscopic contour integration over blur.18 Compared to the autostereoscopic LCD-display, the increase in perceived sharpness was much stronger when using a CRT with shutter glasses. For monoscopic stills, however, the LCD-display (switched to monoscopic mode) was superior to the CRT. These observations are in agreement with the technology overview by Szold19 and can be explained by the sacrifice of resolution when using the LCD display in autostereoscopic mode, which price has not to be paid when using the CRT with shutter glasses. The beneficial effect of separately adjustable enhancement paths for blood vessel contours and blood vessel lumen, which we experienced during the development of our processing algorithm, is consistent with the findings of Yin et al. that surface-features and edge processes make different contributions in determining an object’s unity and shape.27 The perceived usefulness of stereoscopic information, however, also supports the concept formulated by Tse that ‘‘mergeable’’ volumes, rather than relatable contours, are the critical elements in completion.21 With our technique we have aimed to avoid parallax errors and loss of vessel contrast by shadows which are inherent to other blood vessel contrast enhancement techniques projecting the vascular pattern via a projector onto the skin.28 We exclude parallax errors because the user truly looks beneath the tissue surface trough two camera’s. Absence of vessel contrast in shadows is avoided by lighting from two sides and by discrimination of superficial artifacts versus contrasts originating from below the surface. This ability to selectively enhance contrasts from beneath the surface, while preserving the natural depth clues of shadows on the surface, supports both static and dynamic transmittance anchoring of the visual system which is crucial for depth perception.1 Preservation of natural shadows is furthermore important for the correct interpretation of depth clues by occlusion (e.g. from a hand or an instrument positioned between observer and tissue) thus precluding errors in depth perception.12 Since only carefully balanced white light is projected on the skin, there is no impairment of color perception. It is up to the user to freely switch between normal full color vision, color vision plus superimposed blood vessel backprojection and stereoscopic near infrared greyscale vision (with or without enhancement features). The embodiment described in this article still has some drawbacks:Due to the fact that the pertaining configuration aquires VIS and NIR images sequentially, motion artifacts can lead to a backprojection shift.The deeper the vessels lie under the surface, the less we can visualize them. In the pertaining configuration, the depth range is limited to about 1 mm. Based upon the absorption coefficients of Hb and HbO2, arteries provide superior contrasts at 920 nm. The presented images, however, show enhanced venous contrasts, because the larger arteries are buried deeper underneath the skin. The above mentioned drawbacks require further development. Fortunately, many technological improvents still can be added. Allthough the device worked quite well under normal ambient lighting conditions (100–200 Lux), the use of a switchable filter might further improve spectral separation. Furthermore, known techniques like application of crossed polarizers, in combination with more powerful light sources, can offer a considerably larger penetration depth.7 Conclusions Compared to inspection with the naked eye under normal lighting conditions, the tested stereoscopic blood vessel contrast enhancer offered improved visualization in all investigated settings, providing the best stereoscopic image quality when using the CRT monitor with shutter glasses and the best monoscopic image quality when using the LCD set to monoscopic mode. Our technique supports perception of depth, 3-dimensional motion and discrimination between tissue surface and underlying structures. It also has potential as an educational tool by offering the possibility to look and record trough the eyes of an experienced specialist. Further improvements on penetration depth, frame-rate and focus depth-of-field form targets for momentary ongoing further research.

Exp_Brain_Res-3-1-1914290

Regularity of center-of-pressure trajectories depends on the amount of attention invested in postural control

The influence of attention on the dynamical structure of postural sway was examined in 30 healthy young adults by manipulating the focus of attention. In line with the proposed direct relation between the amount of attention invested in postural control and regularity of center-of-pressure (COP) time series, we hypothesized that: (1) increasing cognitive involvement in postural control (i.e., creating an internal focus by increasing task difficulty through visual deprivation) increases COP regularity, and (2) withdrawing attention from postural control (i.e., creating an external focus by performing a cognitive dual task) decreases COP regularity. We quantified COP dynamics in terms of sample entropy (regularity), standard deviation (variability), sway-path length of the normalized posturogram (curviness), largest Lyapunov exponent (local stability), correlation dimension (dimensionality) and scaling exponent (scaling behavior). Consistent with hypothesis 1, standing with eyes closed significantly increased COP regularity. Furthermore, variability increased and local stability decreased, implying ineffective postural control. Conversely, and in line with hypothesis 2, performing a cognitive dual task while standing with eyes closed led to greater irregularity and smaller variability, suggesting an increase in the “efficiency, or “automaticity” of postural control”. In conclusion, these findings not only indicate that regularity of COP trajectories is positively related to the amount of attention invested in postural control, but also substantiate that in certain situations an increased internal focus may in fact be detrimental to postural control. Introduction By now, it is well established that maintaining and controlling an upright posture requires a certain amount of attention (for a review see Woollacott and Shumway-Cook 2002). The relation between attentional resources and the processing of information from somatosensory, visual, and vestibular systems is readily apparent in cases of reduced or conflicting sensory information (e.g., Redfern et al. 2001, 2004; Shumway-Cook and Woollacott 2000; Teasdale and Simoneau 2001). On the one hand, the degree of attention, or cognitive involvement, required for controlling posture increases with task difficulty. This has been (indirectly) illustrated by, for example, Lajoie et al. (1993) and Vuillerme and Nougier (2004), who both found that the reaction time of a verbal response to an auditory stimulus increased with the difficulty of the postural task. On the other hand, the amount of attention required to perform a secondary suprapostural task is known to influence posture (e.g., Balasubramaniam et al. 2000; Huxhold et al. 2006; Pellecchia 2003; Riley et al. 2003). The dependency of posture on attention is even more prominent in the presence of pathology or aging, when both peripheral and central changes occur that decrease the (physical) capability needed to maintain balance during standing or walking (e.g., Brown et al. 1999; Lindenberger et al. 2000; Marchese et al. 2003; Rankin et al. 2000; Redfern et al. 2004). In balance-impaired elderly individuals, for example, the performance of a secondary cognitive task (i.e., dual-task paradigm) may promote postural instability and even falls (Barra et al. 2006; Brauer et al. 2001). Hence, the amount of attention required for maintaining and controlling upright posture is indicative of the degree of “automaticity” of postural control and, for that reason, has been advocated and used as an important tool in clinically oriented studies (cf., Geurts et al. 1991; Melzer et al. 2001). Center-of-pressure (COP) fluctuations measured while standing on a force platform provide a complex output signal of the postural control system in which various pertinent cognitive, perceptual, and motor processes are reflected. Recently, COP measures pertaining to the dynamical structure of COP fluctuations have helped to understand the inherent complexity of the postural control system and its constituent processes (e.g., Baratto et al. 2002; Collins and De Luca 1993; Newell et al. 1993; Pascolo et al. 2005; Peterka 2000; Yamada 1995a; cf., Riley and Turvey 2002 for a review). Inspired by this development in modern posturography, Roerdink et al. (2006)—in a study on the functional recovery of posture in stroke patients—proposed a direct relation between the amount of attention invested in postural control and the regularity of COP fluctuations. COP trajectories were more regular (as indexed by reduced sample entropy) in stroke patients than in healthy elderly and became less regular when performing a secondary cognitive task while standing. These results were interpreted to imply that postural sway regularity is positively correlated with the degree of cognitive involvement in postural control. Interestingly, the regularity of the COP fluctuations decreased with rehabilitation, whereas postural stability (as indexed by the largest Lyapunov exponent) increased, suggesting that the required degree of cognitive involvement in postural control decreased during the course of rehabilitation. Hence, these progressively more irregular COP fluctuations (as indexed by an increase in sample entropy) may be interpreted as an increase in the efficiency or “automaticity” of postural control. This interpretation is in line with physiological studies showing that a decrease in “complexity” or “irregularity” of a physiological time-series is indicative of a decrease in healthiness or effectiveness of the physiological control system (cf., Goldberger et al. 2002), a phenomenon known as “dynamical diseases” (cf. Belair et al. 1995). In other words, increased COP regularity may be explained as an indication of an increasingly ineffective postural control strategy. The aim of the present study was to examine the role of attention in the regulation of posture. To this end, the amount of attention invested in postural control was manipulated experimentally in a large group of young healthy adults. In particular, we increased the difficulty of the postural task through visual deprivation under the assumption that an increase in task difficulty is associated with an increase in cognitive involvement in postural control, creating a so-called internal attentional focus (cf., Andersson et al. 1998; Teasdale et al. 1993; Teasdale and Simoneau 2001). In contrast, we used a concurrent attention demanding cognitive task in order to withdraw attention from postural control, creating an external focus of attention (cf., Huxhold et al. 2006). These manipulations of attention allowed us to examine the proposed direct relation between COP regularity and the amount of attention invested in postural control (Roerdink et al. 2006). In line with this relation, we hypothesized that: (1) increasing postural task difficulty (i.e., by standing with eyes closed, creating an internal focus) increases the cognitive involvement in postural control and hence the regularity of COP fluctuations, and (2) reduced attention to postural control (i.e., by performing a cognitive dual task, creating an external focus) decreases the regularity of COP fluctuations. In addition, based on the suggestion that posture is mainly controlled in the direction of the largest postural sway (Roerdink et al. 2006) and the common finding that in healthy young adults postural sway is largest in the sagittal plane relative to the frontal plane (e.g., Gatev et al. 1999; Winter et al. 1998), we expected COP regularity to be largest in the sagittal plane. Regularity of COP trajectories was quantified by the sample entropy (Richman and Moorman 2000; Roerdink et al. 2006). In order to examine the structure of COP fluctuations in more detail, we further used a combination of more traditional (i.e., based on summary statistics) and dynamical measures that are all defined operationally in terms of readily interpretable features of motor control (see also Table 1): standard deviation of COP time-series (indexing variability or the amount of postural sway), sway-path length of the normalized posturogram (indexing the amount of twisting and turning of the COP trajectory), largest Lyapunov exponent (Rosenstein et al. 1993; indexing local stability), correlation dimension (Grassberger and Procaccia 1983; indexing the number of active, dynamical degrees of freedom involved in postural control and hence its dimensionality, e.g., Kay 1988) and the scaling exponent (e.g., Peng et al. 1995; indexing long-range correlations in COP time-series). Based on previous research, we expected visual deprivation to decrease local stability and to increase variability of COP time-series (cf., Roerdink et al. 2006; Woollacott and Shumway-Cook 2002; Yardley et al. 1999a). Moreover, we anticipated the attention manipulations to induce adjustments in the dynamical structure of postural control leading to changes in scaling exponent, sway-path length and dimensionality of COP fluctuations. Table 1Abbreviations and meaning of interest of the calculated COP measuresVariableMeaning of interestSample entropy, SEnNegatively related with the regularity of COP trajectoryStandard deviation, σ (mm) Positively related with the variability of COP trajectorySway-path length, SPn (s−1)Positively related with the curviness of COP trajectoryLargest Lyapunov exponent, λmaxNegatively related with the local stability of COP trajectoryDimensionality, D2Positively related with the number of active control variablesScaling exponent, αLong-range correlations: α = 0.5 for uncorrelated data (i.e., white noise),α = 1.5 for “Brown noise”, the integration of white noise Methods Participants and procedures A total of 30 healthy young adults (10 males, 20 females; mean age = 24 years, range = 19–30 years), without known motor impairments or movement-related disorders, volunteered to participate in the experiment. Participants stood barefoot on a 1 × 1 m custom-made strain gauge force plate1 with their arms hanging relaxed alongside their body. The medial sides of the heels were separated by about 8 cm and each foot was placed with the toes outward at a 10° angle from the sagittal midline (i.e., standard Romberg position). In order to examine the role of attention in the regulation of posture we carried out two manipulations. On the one hand, we increased the postural task difficulty by inviting participants to stand with eyes closed, while on the other hand, we withdrew attention from postural control by inviting them to perform a cognitive dual task. In particular, the participants were invited to stand upright with (1) eyes open (EO-ST), (2) eyes closed (EC-ST), (3) eyes open while performing a dual task (EO-DT) and (4) eyes closed while performing a dual task (EC-DT). The dual task consisted of uttering backwards names read out aloud by the investigator (e.g., “Simon” had to be repeated as “nomis”). The sole aim of this cognitive dual task was to withdraw attention from the postural task. Therefore, the participants were instructed to perform the task to the best of their ability. No feedback on the accuracy with which they performed this task was provided. Each participant performed the four task conditions in random order and once in reverse order, resulting in a total of eight recordings. COP trajectories were collected for 35 s at a sampling rate of 100 Hz. After the local ethics committee had approved the study, all participants gave their informed consent prior to their participation. Data analysis In order to examine whether posture is actively controlled in the direction of largest postural sway (Roerdink et al. 2006), we analyzed both the registered x (mediolateral ML) and y (anterioposterior AP) COP time-series. After omitting the first 5 s of each recording, leaving 30 s of data for further analyses (i.e., 3,000 samples), the time-series were bi-directionally filtered (second-order low-pass Butterworth filter, cut-off frequency of 12.5 Hz) to eliminate low amplitude measurement noise.2 First, we calculated the conventional standard deviation σ of x and y COP trajectories to quantify the variability, or amount, of postural sway. Subsequently, we normalized the x and y trajectories to unit variance (i.e., by dividing the time-series in question by their respective standard deviation) and calculated the sway path defined as the length of the COP trajectory traveled per second. The applied normalization procedure enabled us to determine the sway path in the normalized posturogram, abbreviated as SPn, providing a scale-independent measure of the amount of twisting and turning of the COP trajectory. Apart from these summary statistics of postural sway (i.e., SPn and σ), which by definition ignore the temporal structure of the COP time-series, we assessed COP dynamics by means of sample entropy, largest Lyapunov exponent, correlation dimension and scaling exponent, which will be briefly explained in the following (for a more extensive description we refer to Roerdink et al. (2006) and references therein). Table 1 presents the six different measures and their interpretation. Sample entropy, SEn Healthy physiological systems are often characterized by an irregular and complex type of variability, whereas disease or aging is often associated with greater regularity and less complexity (cf., Goldberger 1996; Goldberger et al. 2002; Pincus et al. 1991). A method to quantify the regularity of time-series is the sample entropy analysis (Lake et al. 2002; Richman and Moorman 2000). Sample entropy indexes the regularity of a given time-series, and is used to analyze complex stochastic systems that (by definition) include both deterministic and random processes (Pincus 1991). Specifically, sample entropy calculates the probability that a sequence of data points, having repeated itself within a tolerance r for a window length M, will also repeat itself for M + 1 points, without allowing self-matches (see Lake et al. 2002; Richman and Moorman 2000).3 Smaller sample entropy values are associated with greater regularity. In the present study, a decrease in sample entropy (i.e., more regular sway fluctuations) was interpreted as a decrease in the effectiveness of postural control. Largest Lyapunov exponent, λmax The largest Lyapunov exponent4 provides a measure of the local stability of a dynamical system (e.g., Abarbanel et al. 1996), i.e., the system’s sensitivity to initial conditions or its resistance to small internal perturbations, such as the natural fluctuations that occur while maintaining an upright stance. It quantifies the exponential divergence or convergence of initially nearby trajectories in state space as time progresses (e.g., Rosenstein et al. 1993). If nearby points diverge, they produce instability. The exponent λmax indexes this instability: positive values of λmax indicate either the presence of deterministic chaos (i.e., a form of variability that is brought about by an underlying lawful nonlinear dynamical structure) or complete randomness (i.e., noise), implying that nearby points diverge rapidly, reflecting local instability and lack of predictability. In order to distinguish a deterministic component from genuine randomness, it is necessary to validate results against surrogate data (cf., Theiler et al. 1992). Dimensionality, D2 The dimensionality of all COP time-series was calculated using the correlation dimension algorithm of Grassberger and Procaccia (1983).5 The correlation dimension provides an index of the number of independent degrees of freedom (equations of motion) that are required to reproduce the time evolutionary properties of the COP time-series. Note that this analysis of the dynamical degrees of freedom is different from the analysis of the (mechanical) degrees of freedom of the joints as commonly applied in the study of motor control, and that no straightforward or uniform relation exists between the number of component degrees of freedom in motion and the dimension of the organizational dynamic in controlling those components (Newell and Vaillancourt 2001). Scaling exponent, α In order to determine whether the measured COP time-series were characterized by the presence of long-range correlations, we applied a fractal analysis method for biological time-series called detrended fluctuation (DFA) analysis (Peng et al. 1995).6 The scaling exponent α as determined by this method indicates the presence or absence of (long-range) correlations in the COP trajectories, as defined by Peng et al. (1995). For uncorrelated data, (e.g. white noise) α = 0.5. An α greater than 0.5 and less than or equal to 1.0 indicates persistent long-range power-law correlations. In contrast, 0 < α < 0.5 indicates a different type of power-law correlation such that large and small values of the time series are likely to alternate. For α > 1 correlations exist but cease to be of a power-law form; α = 1.5 indicates Brown noise, i.e., integrated white noise. The scaling exponent α can also be viewed as an indicator that describes the roughness of the time series: the larger the value of α, the “smoother” the time series (Peng et al. 1995). Surrogate analysis To test for spurious effects and to distinguish between deterministic features and randomness, the scaling exponents, dimension estimates, Lyapunov exponents and entropy values were also computed for surrogate data (Theiler et al. 1992). In particular, we generated both time- and phase-randomized surrogate data of the filtered COP time-series (Fig. 1). In time-randomized surrogate data, the distribution of the original data is being preserved (i.e., mean, variance, etc. are unaltered), whereas the temporal correlations in the COP time-series are destroyed. The absence of temporal correlations will result in a scaling exponent α close to 0.5 and very large values for dimension and sample entropy. Phase-randomized surrogate data are obtained by randomizing the data’s Fourier phases. In contrast to time-randomization, this procedure does not alter the spectral power distribution and preserves the data’s auto-correlation function. Consequently, scaling exponents of phase-randomized and original data should match, whereas estimates of correlation dimension and sample entropy should be largely increased in the surrogate data.Fig. 1An example of the surrogate analysis, as applied to all time-series. Surrogate data of a measured COP trajectory (upper panel) were constructed by randomizing the Fourier phase (middle panel) and the temporal order (lower panel) Statistical analysis For all dependent variables, the first and second trials were averaged. To test for differences between AP and ML COP measures and to examine the effect of the different experimental tasks, we used a repeated measures analysis of variance (ANOVA) design with within-subject factors vision (2 levels: standing with eyes open EO, and standing with eyes closed EC), dual task (2 levels: standing without cognitive dual task performance, i.e., single task, ST, and standing with cognitive dual task performance, DT), and plane (2 levels: sagittal and frontal plane).7 Subsequently, we tested for differences between surrogate and original data using a design involving a within-subject factor surrogate (3 levels: original data and time- and phase-randomized surrogate data). To assess the strength of the (main and interaction) effects, we determined the eta squared (η2), a commonly used measure of effect size in AVOVAs, reflecting the proportion of variance in the dependent variable that is attributable to each effect. Subsequently, η2 was converted into Cohen’s f according to: An effect size (f) of > 0.4 was considered to reflect a strong effect (Cohen 1988). The analyses were performed using SPSS (SPSS, Inc., Chicago, IL, USA). Results The result section is organized as follows. We first report possible differences in the dependent variables between the eyes open and eyes closed conditions (i.e., main effect of vision, hypothesis 1). Second, we describe the effect of experimentally withdrawing attention from postural control by comparing single task and dual task conditions (i.e., main effect of dual task, hypothesis 2). Third, we report whether significant vision × dual task interaction effects were present. Finally, we present the effects of plane, which may reveal possible directional differences in control. In this context, we also report the effects of randomization on the dependent variables to ensure that the observed changes in the dynamics of COP trajectories were genuine effects. Table 2 presents the results of the vision × dual task ANOVA for the six dependent posturographic measures (i.e., interindividual means corresponding to the main effects of vision and dual task, collapsed over x and y time-series, as well as F, P and f values for main and interaction effects). Significant vision × dual task interaction effects are presented in Fig. 2. Table 2Main and interaction effects of vision and dual task (i.e., collapsed over x and y time-series) of sample entropy (SEn), standard deviation (σ), sway-path length of the normalized (by the standard deviation) posturogram (SPn), largest Lyapunov exponent (λmax) and scaling exponent (α) of COP time-series for 30 healthy individualsConditionMeanVision (EO vs. EC)ConditionMeanDual task (ST vs. DT)Vision × dual taskaF(1, 29)PƒF(1, 29)PƒF(1, 29)PƒSEnEO0.723.83=0.060*0.36ST0.701.45ns0.256.72<0.050.48EC0.70DT0.72σEO3.5211.82<0.0050.64ST3.892.45ns0.293.18=0.0850.33EC4.01DT3.64SPnEO4.275.28<0.050.43ST4.1313.57<0.0050.686.98<0.050.49EC4.52DT4.66λmaxEO1.5636.23<0.0011.12ST1.710.10ns0.064.26<0.050.38EC1.88DT1.73D2EO2.2323.58<0.0010.90ST2.2045.70<0.0011.266.15<0.050.46EC2.48DT2.51αEO1.3913.70<0.0010.69ST1.3924.57<0.0010.921.80ns0.25EC1.34DT1.35* Significant vision × plane interaction (F(1, 29) = 5.48, P < 0.05, ƒ = 0.44), which was caused by the fact that, in contrast to that in the frontal plane, the effect of vision was significant in the sagittal plane (F(1, 29) = 6.47, P < 0.05, ƒ = 0.47)a  See Fig. 2 for mean values of the conditions EO-ST, EC-ST, EO-DT and EC-DTFig. 2Interindividual averages, collapsed over x and y time-series, of sample entropy (SEn), standard deviation (σ), sway-path length of the normalized (by the standard deviation) posturogram (SPn), local stability (λmax) and dimensionality (D2) for the four experimental conditions: standing with eyes open (EO-ST), eyes closed (EC-ST), eyes open while performing a cognitive dual task (EO-DT) and eyes closed while performing a cognitive dual task (EC-DT). The asterisks indicate significant (P < 0.05) differences between conditions Increased postural task difficulty (EO vs. EC) As is apparent from the significant main effects of vision in Table 2, standing with eyes closed resulted in more regular sway fluctuations, as indexed by a decrease in SEn. In addition, sway variability (σ), dimensionality (D2) and sway-path length (SPn) were increased, whereas α and local stability (as indexed by an increase in λmax) decreased. Decreased attention to posture (ST vs. DT) Collapsed over x and y time-series and visual conditions, no significant main effects involving dual task were found for σ, SEn or local stability. In contrast, a main effect of dual task was found for both SPn and D2 in that performing a cognitive dual task brought about an increase in both variables as compared to the single task condition (Table 2). Moreover, as shown in Table 2, performing a dual task resulted in a significant decrease of α indicating that dual task performance brought about changes in the time-varying structure of sway fluctuations. Vision × dual task interaction effects Significant dual task × vision interaction effects (Table 2) revealed that the effects of visual deprivation on SEn, σ and SPn were different for single and dual task conditions (see Fig. 2). In particular, the observed effect of visual deprivation on σ and SEn was significant only for the single task condition, whereas the observed increase in SPn only existed for dual task performance (see Fig. 2). On the other hand, as can be appreciated from both Table 2 and Fig. 2, the effects of introducing a cognitive dual task on SEn, σ and SPn depended on eye closure. Although no significant main effect of dual task was found for SEn, there was a significant dual task × vision interaction. This effect occurred because SEn increased significantly from 0.68 to 0.72 through the introduction of a dual task when standing with eyes closed, whereas such an increase was absent when standing with eyes open (0.72 for both single and dual task conditions). Similarly, when standing with eyes open, dual task performance had no significant effect on σ, whereas dual task performance resulted in a decrease in σ when standing with eyes closed (see Fig. 2). The observed significant dual task × vision interaction effects for λmax were brought about by the fact that the increase (i.e., decrease in local stability) when standing with eyes closed was stronger for the single task condition than for the dual task condition. A similar asymmetric effect was found for D2, in that the increase when standing with eyes closed was stronger for the dual task condition than for the single task condition. Moreover, as can be observed in Fig. 2, the effect of performing a dual task (i.e., increase in D2) was larger when standing with eyes closed than with eyes open. Effects of plane and randomization Collapsed over conditions, significant differences between sagittal and frontal plane were found for all variables, except α. In particular, sway variability, σ, and local stability, λmax, were significantly larger in the sagittal than in the frontal plane (F(1, 29) = 20.94, P < 0.001, f = 0.82 and F(1, 29) = 15.19, P < 0.005, f = 0.72, respectively). In contrast, sample entropy, SEn, and dimensionality, D2, were significantly lower in the sagittal than in the frontal plane (F(1, 29) = 145.38, P < 0.001, f = 2.31 and F(1, 29) = 16.01, P < 0.001, f = 0.74, respectively). A significant vision × plane interaction was found for SEn (F(1, 29) = 5.48, P < 0.05, f = 0.44), which was caused by the fact that SEn was smaller when standing with eyes closed (EC-ST, see Fig. 2) for COP fluctuations in the sagittal plane, whereas this was not the case for COP fluctuations in the frontal plane. Figure 3 shows the results of the interindividual means of the surrogate analyses. Sample entropy measures of both phase- and time-randomized surrogate data were significantly higher compared to the original COP time-series (F(2, 58) = 10985.1, P < 0.001, f = 22.34). Whereas scaling exponents of phase-randomized surrogate data and original COP time-series did not differ, correlations were completely absent when the data were time-randomized as evidenced by α values around 0.5 (F(2, 58) = 5436.7, P < 0.001, f = 14.12). As a result of very high-dimensional noise in the time-randomized data, no embedding dimension could be estimated and hence no dimensionality estimates and Lyapunov exponents could be determined for the time-randomized surrogate data.8 Randomizing the phases of original COP data significantly increased the dimensionality (F(1, 29) = 50.07, P < 0.001, f = 1.31). In addition, the Lyapunov exponent of the phase-randomized surrogate data was significantly higher than that of the original time-series (F(1, 29) = 53.44, P < 0.001, f = 1.36). Fig. 3Grand means, collapsed over all conditions, planes and participants, of sample entropy, scaling exponent, dimensionality and local stability for the original (OR) COP time-series and their phase-randomized (PHASE) and time-randomized (TIME) surrogate counterparts. The error bars represent the interindividual standard deviations. The asterisks represent significant (P < 0.05) differences between the surrogate data and the original time-series Discussion The present experiment was conducted to investigate the role of attention in the regulation of posture. Specifically, we examined whether an increase in postural sway regularity (i.e., as indexed by a decrease in SEn) is representative of an increase in cognitive investment in postural control. We hypothesized that COP trajectories become more regular (i.e., SEn decreases) when task difficulty is increased (EC vs. EO) and, conversely, become less regular (i.e., SEn increases) when an attention-demanding cognitive dual task is introduced (DT vs. ST). We further expected that these changes in regularity of COP fluctuations would be accompanied by changes in variability, local stability, sway-path length, dimensionality and scaling exponent reflecting functional modifications of postural control. For the proper interpretation of the present findings, however, it was necessary to ascertain that the observed structure (and changes herein) of the COP fluctuations did not result from noise, but was indeed brought about by deterministic processes. Therefore, we will first discuss the results of the surrogate analyses before discussing the respective effects of vision, dual task and plane on the dynamical structure of postural sway. Surrogate analyses Although nonlinear estimates of dynamical structure are not readily interpretable in an absolute sense, they can be meaningfully interpreted by comparing them across conditions, (see e.g., Newell et al. 1993), as well as with surrogate data. For example, entirely random data are characterized by large (theoretically infinite) dimensionality and large λmax values, whereas chaotic/deterministic data have smaller dimensionality and smaller λmax values. In the present analysis, the surrogate data had greater dimensionality and larger λmax values than the original data, implying that the latter had considerable deterministic structure (cf., Theiler et al. 1992). Moreover, both phase- and time-randomized surrogate data showed increased sample entropy values. Hence, the original COP fluctuations clearly had a deterministic component, which was evidenced further by the fact that the scaling exponents became 0.5 after time-randomization (resulting from a loss of temporal correlations in the shuffled time-series), but remained unaffected by phase-randomization (i.e., preserving temporal correlations). These findings are consistent with those of previous studies suggesting that COP fluctuations are (largely) of deterministic origin (e.g., Doyle et al. 2004; Riley et al. 1999; Yamada 1995b), and testify to the relevance and need of including dynamical measures in posturography. Increasing postural task difficulty (EO vs. EC) In the present study, standing with eyes closed brought about an increase of sway variability, which was accompanied by an increase in dimensionality and λmax, implying that local stability decreased. Possibly, the observed increase in dimensionality may serve as a mechanism to enrich information so as to facilitate the control of standing and to cope with the reduced (local) stability (cf., Riley and Clark 2003; van Emmerik and van Wegen 2002). These observations are in line with the common notion that visual deprivation increases the task difficulty of postural control, and, consequently, requires cognitive monitoring of postural control. We expected that this increase in cognitive investment would be accompanied by a decrease in SEn (i.e., an increase in regularity), which was indeed the case (see Fig. 2, compare EO-ST with EC-ST), thus confirming hypothesis 1. These findings are consistent with the results of other studies showing that COP variability tends to increase as experimental task conditions become increasingly difficult, whereas the temporal structure of postural sway tends to become increasingly regular (Riley and Clark 2003; Roerdink et al. 2006; Thurner et al. 2002). In the present study, visual deprivation resulted in qualitatively similar changes in the structure of COP trajectories as reported in the study of Roerdink et al. (2006) for stroke patients (i.e., increased σ, λmax, D2, and decreased SEn and α with respect to healthy elderly adults). In healthy young adults, the qualitatively similar changes in COP dynamics with visual deprivation as compared to standing with eyes open may, likewise, be interpreted to indicate that postural control is performed less automatically and effectively. Whereas in stroke patients such modifications in postural control may be due to a defect or slowing down of the central processing of sensory information (cf., Teasdale et al. 1991; Woollacott et al. 1986), in young healthy adults a more regular sway, resulting from standing with eyes closed, implies increased “active” monitoring of postural control with increasing task difficulty (cf., Andersson et al. 1998; Nashner and McCollum 1985; Redfern et al. 2001; Teasdale et al. 1993; Teasdale and Simoneau 2001). If the proposed relation between cognitive investment in postural control and postural sway regularity does indeed exist, then the performance of a cognitive dual task should result in less regular COP trajectories. That this was indeed the case will be discussed in the next two subsections. Withdrawing attention from posture The performance of a concurrent dual task led to changes in scaling exponents and an increase in dimensionality, reflecting cognition-invoked adjustments of postural control. These changes under dual task performance may have served to enrich the information captured in sway fluctuations without increasing the amount of sway (i.e., variability remained unaltered). This interpretation is amplified by the observation that the sway-path length of the normalized posturogram increased, indicating more twisting and turning in the COP trajectories. Interestingly, despite the fact that attention was withdrawn experimentally from postural control, local stability remained unaltered. In contrast to what we expected, no main effect of dual task was found for SEn. However, this finding does not necessarily militate against the proposed relation between the regularity of COP fluctuations and the amount of attention directed to postural control, as will be argued in the following subsection. Vision × dual task interaction Especially noteworthy in this context and in view of our expectations is that while standing with eyes closed postural sway regularity decreased (i.e., sample entropy increased) when performing a cognitive dual task (compare EC-DT and EC-ST in Table 2 and Fig. 2). This finding is consistent with hypothesis 2 and supports the proposed positive correlation between COP regularity and the degree of attention involved in postural control. Apparently, the fact that regularity remained unaltered (i.e., 0.72) when performing a dual task while standing with eyes open (EO-DT) implies that, for young healthy adults, standing with eyes open is not very attention demanding. Conversely, during the more challenging task of standing with eyes closed (EC-ST) COP fluctuations became more regular (i.e., 0.68; Fig. 2). However, the finding that sample entropy again increased to its “normal” (EO-ST) level when withdrawing the focus of attention from the postural task (EC-DT) indicates that the increased cognitive monitoring of posture during EC-ST had a detrimental effect. Visual deprivation increased the awareness of the postural task (i.e., creating an internal focus), resulting in efforts to actively (consciously) control posture and, as such, preventing the postural control system to work in a relatively automatic and efficient manner (viz., Andersson et al. 2002; Hunter and Hoffman 2001; McNevin and Wulf 2002; Milton et al. 2004). Similarly, when standing with eyes open the attention-demanding dual task had no significant effect on σ, whereas when standing with eyes closed the dual task did result in a decrease in σ (see Fig. 2), corroborating the findings of e.g., Andersson et al. (2002), McNevin and Wulf (2002) and Morioka et al. (2005). In contrast, many authors have found an increase in postural sway variability when performing a cognitive dual task (see Shumway-Cook and Woollacott 2000). It has been suggested that this effect of dual task on the amount of sway may, in part, result from articulation (Dault et al. 2003; Yardley et al. 1999b). In this context, it is important to note that the present finding that variability actually decreased with the introduction of a cognitive dual task indicates that articulation played no significant role in the present study. Sagittal versus frontal plane The increased variability of the COP trajectories in the sagittal plane (as compared to the frontal plane) was accompanied by reduced local stability and greater regularity (i.e., a decrease in sample entropy). These observations are consistent with the findings of Roerdink et al. (2006), which showed that in healthy elderly adults local stability was reduced in the sagittal plane, whereas regularity and variability were elevated in this plane. Based on these results, they suggested that posture is mainly controlled in the direction of largest postural sway (i.e., sagittal plane), which required a certain amount of attention as reflected by the regularity findings. The present observation that during standing with eyes closed (EC-ST) regularity increased significantly in the sagittal plane, whereas no effect was found for the frontal plane, is in line with this suggestion: young healthy adults mainly control posture in the sagittal plane, which becomes particularly attention demanding when task difficulty is increased (see also Pellecchia 2003; Riley et al. 2003). Conclusion The present study showed that the amount of attention invested in postural control is positively correlated with sway regularity. Specifically, the present study showed that increasing postural task difficulty by means of visual deprivation (EC-ST) not only resulted in an increase of COP variability and a decrease in local stability, but also in more regular COP trajectories. These findings could be taken to imply that the participants actively monitored their posture in order to cope with the increased postural task difficulty. However, when the amount of attention invested in postural control was experimentally reduced by introducing a cognitive dual task (EC-DT), both regularity and variability of sway fluctuations returned to values observed when standing with eyes open (EO-ST). This finding suggests that during standing with eyes closed (EC-ST) the increase in monitoring posture was due to an increase in awareness of the postural task (i.e., internal attentional focus), preventing postural control from working in an automatic and efficient manner (e.g., Hunter and Hoffman 2001). All in all, it is fair to conclude that the methodological and analytical approach adopted in the present study allows for disentangling whether or not there is an increase in cognitive involvement and to what extent an increase in cognitive involvement has a detrimental or beneficial effect.

Plant_Mol_Biol-3-1-1805039

Global genome expression analysis of rice in response to drought and high-salinity stresses in shoot, flag leaf, and panicle

To elucidate genome-level responses to drought and high-salinity stress in rice, a 70mer oligomer microarray covering 36,926 unique genes or gene models was used to profile genome expression changes in rice shoot, flag leaf and panicle under drought or high-salinity conditions. While patterns of gene expression in response to drought or high-salinity stress within a particular organ type showed significant overlap, comparison of expression profiles among different organs showed largely organ-specific patterns of regulation. Moreover, both stresses appear to alter the expression patterns of a significant number of genes involved in transcription and cell signaling in a largely organ-specific manner. The promoter regions of genes induced by both stresses or induced by one stress in more than one organ types possess relative enrichment of two cis-elements (ABRE core and DRE core) known to be associated with water stress. An initial computational analysis indicated that novel promoter motifs are present in the promoters of genes involved in rehydration after drought. This analysis suggested that rice might possess a mechanism that actively detects rehydration and facilitates rapid recovery. Overall, our data supports a notion that organ-specific gene regulation in response to the two abiotic stresses may primarily be mediated by organ-specific transcription responses. Introduction Drought and highly saline soils are among the most serious challenges to crop production in the world today. This is particularly the case in developing countries, where these abiotic stresses severely limit crop growth and productivity. Both traditional breeding and genetic engineering of crop plants have been utilized to improve drought and high-salinity tolerance or resistance with the goal of increasing agricultural productivity in affected regions. Understanding plant responses to abiotic stresses at the genomic level provides an essential foundation for future breeding and genetic engineering efforts. Recent research on drought and high-salinity responses in Arabidopsis implied that a large proportion of the genome is involved in drought (Shinozaki et al., 2000, 2003) or high-salinity stress responses (Xiong et al., 2002; Zhu, 2001, 2002). In several cases, it has been shown that alteration of individual gene expression level can significantly impact responses to drought (Garg et al., 2002; Haake et al., 2002) or high-salinity stresses in plants (Kasuga et al., 1999; Shi et al., 2003; Xu et al., 1996; Zhang et al., 2004). Genome-wide identification of genes regulated by drought or high-salinity conditions has manifold significance. First, it provides a more comprehensive understanding of the transcriptional responses to those stresses. Second, it provides a starting point for further elucidating the role of individual genes in stress responses, which will be of great value in crop engineering. Third, it aids in the identification of stress responsive promoters and responsible cis-elements within them that are important both for basic study and crop engineering applications. DNA microarrays provide a high throughput means of analyzing genome expression, which has been used to study patterns of gene expression in response to drought or high-salinity stresses in several plant species (Seki et al., 2003, 2004). Initially, a microarray containing ∼1,300 full-length cDNA clones from Arabidopsis was used to study gene expression under drought and cold stresses. This study resulted in the identification of 44 and 19 cDNA clones as drought and cold-inducible genes, respectively (Seki et al., 2001). Other studies employed an improved microarray containing around 7,000 Arabidopsis full-length cDNA clones to profile gene expression in response to abscisic acid (ABA) treatment (Seki et al., 2002a) as well as cold, drought, and high-salinity stresses (Seki et al., 2002b). Another study employed an Affymetrix GeneChip covering approximately 8,100 genes from Arabidopsis to monitor changes in gene expression under salt, osmotic, and cold stresses. This study revealed that resulting expression changes varied significantly between root and leaf, with only minor overlap (Kreps et al., 2002). Similar studies have also been performed in barley to assess the drought and high-salinity gene expression responses using a microarray containing 1,463 DNA elements (Ozturk et al., 2002). Rice (Oryza sativa) is a model plant for cereal crops and has perhaps the richest set of resources available for plant genomic studies (Feng et al., 2002; Goff et al., 2002; RCSC, 2003; Sasaki et al., 2002; Yu et al., 2002). In rice, the high-salinity stress response has been analyzed with a microarray containing 1,728 cDNA clones from a root cDNA library of salt-tolerant rice (var Pokkali) (Kawasaki et al., 2001). Another study with a microarray containing 8,987 DNA elements detected 509 possible abscisic acid (ABA) or gibberellin-responsive genes (Yazaki et al., 2003). Later, the same group used a 22,000 rice cDNA based oligonucleotide array to identify ABA and GA responsive genes, made comparison of ABA-responsive genes and their putative responsible promoter elements between rice and Arabidopsis (Yazaki et al., 2004). This analysis gave an initial global view of the ABA- and GA-responsive genes in rice. A total of 73 genes induced by cold, drought, high-salinity and ABA treatment were further identified using a microarray containing 1,700 full-length cDNA clones (Rabbani et al., 2003). Recently, an Affymetrix rice genome array containing 55,515 probe sets was used to profile the transcriptomes of rice strains with salt-tolerant and salt-sensitive genotypes, revealing genome-wide differential transcription under salinity treatments during vegetative growth (Walia et al., 2005). Taken together, previous studies in rice have identified various genes regulated by different stress conditions. However, a systematic comparison of whole-genome expression responses to drought and high-salinity stresses in various organs has not yet been performed. The availability of the complete genome sequences of two rice sub-species (Yu et al., 2005) makes the construction of a whole-genome microarray possible. A whole-genome 70mer oligomer microarray for rice was developed and successfully employed to obtain genome expression profiles (Ma et al., 2005b; Jiao et al., 2005). Here, we use this whole-genome microarray to monitor expression changes for a total of 36,926 genes in response to drought and high-salinity stresses in shoots at the four-tiller stage and in flag leaves and panicles at 1 week prior to heading. This analysis revealed the extent of reprograming and alteration of cellular pathways in response to drought and high salinity stresses. The possible underlying mechanism for the observed reprograming of the genome expression was examined. Materials and methods Plant material and stress treatments Plant material used in this study was O. Sativa L. ssp. indica (cv. Minghui 63). Shoot samples were selected at the four-tiller stage (vegetative stage) and flag leaf and panicle samples were selected at one-week-before-heading (reproductive stage). The plants used for shoot samples were grown in hydropolic half-strength Hoagland solution up to the four-tiller stage, and then plants allowed to reach the reproductive stages were grown in soil. For drought treatment samples, plants were taken out from the Hoagland solution directly and their roots were put on filter paper for drought treatment. Under this condition, the phenotype of shoot was observed. Materials at the following stages were collected: Stage 1 (D1), when leaves were slightly rolled and leaf relative water content (RWC) was around 90–95%; Stage 2 (D2), when leaves were half-rolled and leaf RWC was around 80–85%; Stage 3 (D3), when leaves were completely rolled and leaf RWC was around 70–75%. For most of the plants, usually it took half to 1 h to reach stage 1, and then after another 2 h or 4 h, they could reach stage 2 or stage 3 respectively. Rehydration samples were prepared by first subjecting the rice plants to drought treatment until leaves were completely rolled (D3), then supplying enough water and samples were collected after 48 h. For each treatment stage, three independent replicates were collected. For sample collection under high-salinity treatment (200 mM NaCl), rice plants were collected when plants reached stages 1, 2, or 3 as outlined above for drought treatment. Rice plants were taken out from Hoagland solution and were directly put into Hoagland solution with 200 mM NaCl. For most of plants, usually it took 5 h to reach stage 1, and then after another 12 and 24 h, they could reach stage 2 and stage 3 respectively. For each stage, three independent replicates were collected. Samples from normal grown plants corresponding to each drought and high-salinity stages and rehydration stage were also collected at the same time as respective controls. Upon collection, samples were frozen immediately in liquid nitrogen and stored in the –80°C freezer. The estimation of RWC in rice plants was prior reported method (Bare and Weatherley, 1962). RNA isolation, probe labeling, and hybridization RNA preparation, fluorescent labeling of probes, slide hybridization, washing and scanning were performed as described previously (Ma et al., 2005b). Total RNA was prepared from frozen samples using the RNAwiz reagent (Ambion). Each RNA preparation was used to generate a labeled cDNA probe for hybridization. There were at least three high-quality replicate data sets for each experiment, with each data set obtained from an independent biological sample. The cDNA synthesis, probe labeling, and microarray hybridization, microarray slide washing, and array scanning were performed following previously described protocols (Jiao et al., 2005). Hybridized microarray slides were scanned with an Axon GenePix 4000B scanner, and independent TIFF images for both Cy3 and Cy5 channels were used for subsequent analysis. Microarray and initial data analysis The microarray construction was described previously (Ma et al., 2005b, Jiao et al., 2005). Microarray data were preprocessed with Limma R Package (Smyth, 2005). We first performed Loess normalization to balance the dye-bias, and to remove the print-tip effect within each array. Quantile normalization was performed to remove the experimental variances across replicate arrays. Based on the distribution of negative control oligos within each slide, we determined an experimental threshold for minimum non-specific hybridization intensity, and the gene oligos above 90-percentile of negative oligos were set as to be detected in each experiment. To identify the genes that respond to drought and salinity stress, we used the Limma R implementation to perform hypothesis tests by fitting a linear model to the expression data. During pair-wise comparison between each treated sample versus control, a moderated t-statistics were computed, by using an empirical Bayes method to shrink the gene-wise sample variances towards a common value (Smyth, 2004), and the differential level was represented by a log (2) of intensity ratio. The P value adjustment used in false discovery rate control for multiple testing is Benjamini and Hochberg method (Benjamini and Hochberg, 2000; Reiner et al., 2003). The genes with significantly differential expression were selected by the adjusted P values less than 0.05. RT-PCR analysis of genes The expression profiles were further quantified by RT-PCR and compared to results obtained by chip hybridization. The first strand of cDNA was generated from 1 μg of total RNA isolated independently from each sample in a 100 μl volume and 1 μl was used as template in each PCR reaction (25 cycles of 1 min at 94°C, 1 min at 58°C, 1 min at 72°C). A total of eight drought-induced genes were selected for RT-PCR analysis (the primers of these genes are listed in Table S18). The Actin1 gene of rice was used as a control for RT-PCR experiments (forward primer, 5′-cgcagtccaagaggggtatc-3′; reverse primer, 5′-tcctggtcatagtccagggc-3′). Functional classification GO terms used in rice gene functional annotations were downloaded from the BGI-RIS database at http://rise.genomics.org.cn (Zhao et al., 2004). For biochemical pathway analysis, we classified genes by associated biochemical pathway(s) using the AraCyc database (http://www.arabidopsis.org/tools/aracyc) for Arabidopsis, which is based on MetaCyc pathway collections (Mueller et al., 2003). A rice gene was considered to be associated with a biochemical pathway if it had an Arabidopsis homolog in that pathway. To compare the expression profiles of conserved genes between rice and Arabidopsis, we selected reciprocal best-matched gene pairs between rice and Arabidopsis, which are based on a TBLASTN search of both rice and Arabidopsis gene sequences. This method used to identify highly homologous genes between rice and Arabidopsis has been described in previous reports (Jiao et al., 2005; Ma et al., 2005b). For the diagrams in Fig. S3, the expression ratios of the selected rice genes at D3 and S3 stage only were used. Analysis of cis-acting regulatory elements Differentially expressed genes supported by available full-length cDNA sequences were used to further elucidate regulatory cis-elements in the putative promoter regions. After mapping the FL-cDNAs on indica rice 12 pseudo-molecules, 2 kb DNA sequences upstream from the 5' end of the available cDNA sequences were extracted and searched against a plant cis-regulatory database (Higo et al., 1999) (http://www.dna.affrc.go.jp/htdocs/PLACE/signalscan.html). The abundance of known ABRE core and DRE core elements in each query set were counted. Identification of novel cis-regulatory elements was performed using the Improbizer tool (cis-Site Seeker) by Jim Kent (http://www.cse.ucsc.edu/∼kent/improbizer/), which is based on ab-initio prediction of consensus binding sites among each co-regulated gene set. The upstream 2 kb regions of co-regulated FL-cDNA genes were used for common motif searching. To evaluate the significance for each putative novel motif, we generated an equal number of 2 kb contrast sequences in each query set by randomly arranging the four nucleotides. We performed the standard Student’s t-test on motif score tables, comparing query set versus contrast set. The known ABRE motif predicted by Improbizer has a pretty high confidence level (above 0.999), and the other two novel motifs were calculated to above a 0.95 confidence level. Motif consensus sequences were displayed by WEBLOGO program at http://weblogo.berkeley.edu/ by Crooks et al. (2004). Gel shift assays Nuclear protein extracts and gel mobility shifts were performed as described previously (Gupta et al., 1998). Oligomers were synthesized using the following sequences: OsJRFA070715: GCAGATACTGCAGCCAACCTCTCT, OsJRFA070715M: GCAGATACTGAAGCCAACCTCTCT. The complement sequence for each oligo was also synthesized. After denaturing and annealing, double-stranded probes were used for labeling. Clustering and chromosomal location analysis of genes involved in stress responses Unsupervised classification of stress-response genes was based on log-transformed ratios of all up- or down-regulated genes detected in at least one stage of drought treatment. We performed hierarchical clustering analysis to display the expression pattern and tree diagram at different stress stages by employing Cluster and TreeView (Eisen et al., 1998). Classification of rehydration-induced genes was based on the log-transformed ratios of 48-h rehydration and three stages of drought treatment in the three organs, and 807 genes in flag leaf, 281 genes in shoot, and 224 genes in panicle were grouped with the above method. To analyze the genome-wide distribution of rehydration-induced genes in the three organs and drought and high-salinity induced genes in shoot, as well as identify potential clusters of co-regulated genes, we mapped these genes to the indica 12 chromosomes using the BLAT program. The expression profiles in Fig. 9 were drawn using a customized Perl script plus a scalable vector graphics (SVG) bundle. Results Identification of drought and high-salinity stress responsive genes using a rice whole genome oligomer microarray A previously reported rice genome microarray containing 70mer oligomer probes for 36,926 unique genes or gene models (Ma et al., 2005b; Jiao et al., 2005) was used for genome-wide profiling of gene expression responses to drought and high-salinity stresses. Total RNAs were isolated from four-tiller stage shoots and from flag leaves and panicles (as described in Experimental procedures) at three different time points (or sampling stages) after initiation of drought or high-salinity stress treatments (Fig. S1). The sampling stages for both drought and high-salinity treatments were based on morphological phenotypes (leaf rolling state) and physiological states (relative water content, RWC, see Bare and Weatherley, 1962). Stage one is defined as exhibiting slightly rolled leaves with 90–95% RWC. Stage two is defined as exhibiting halfway rolled leaves with 80–85% RWC. Stage three is defined as exhibiting completely rolled leaves and with 70–75% RWC. For drought treatment, the three stages were designated as D1, D2, and D3, while the three stages for the high-salinity treatment were designated as S1, S2, and S3 (see Fig. S1). The cDNA probes produced from the stress-treated and corresponding untreated control sample pairs were hybridized to microarray slides. For each sample point, three replicates were performed with dye-swap to correct for uneven dye effects. The microarray data processing includes LOESS normalization within slide and quantile normalization between slides to remove the systematic errors (see Experimental procedures for details). To identify the differentially expressed genes, Limma R implementation was used to perform pair-wise comparison, and a moderated t-statistics was computed for each gene of treated sample versus control sample. Genes with FDR-adjusted P value less than 0.05 were considered to indicate significant differences (see Experimental procedures for details). Numbers of genes up- or down-regulated by drought or by high-salinity stress at each stage in each organ were shown in Fig. 1. In total, 582, 1,257 and 614 drought up-regulated genes and 795, 646 and 1,305 drought down-regulated genes were identified in flag leaf, shoot and panicle, respectively; and 1,676, 817 and 1,310 high-salinity up-regulated genes and 1,270, 1,323 and 2,284 high-salinity down-regulated genes were identified in flag leaf, shoot and panicle, respectively (Fig. 2A, B). The complete lists of these genes are available in the supplemental data (Tables S2–S13). Fig. 1The number of differentially expressed genes at each stage of drought and high-salinity treatments in three organs. Differentially expressed genes (both induced or repressed) are defined as those with P < 0.05. D1, D2 and D3: three stages of drought stress, S1, S2, and S3: three stages of high-salinity stress. y-axis shows the gene number. (A) Number of genes induced or repressed in flag leaf. (B) Number of genes induced or repressed in panicle. (C) Number of genes induced or repressed in shoot.Fig. 2Comparison of drought and high-salinity genome expression responses in the three rice organs. (A) Total number of drought and high-salinity stress inducible genes and the number of genes induced in response to both stresses in each of the three rice organs. Genes induced at least at one stage under drought or high-salinity treatments are included in the analysis. y-axis shows the gene number. (B) The total number of drought and high-salinity stress repressed genes and the number of genes repressed in response to both stresses in each of the three rice organs. Genes repressed at least at one stage under drought or high-salinity treatments are included in the analysis. y-axis shows the gene number. (C) Clustering analysis of all drought and high salinity responsive genes in each of the three rice organs. The differentially expressed genes (both induced or repressed) with P < 0.05 at least at one stage of drought treatment are included. The median ratio (treated/untreated sample) is log2 transformed and subject to complete linkage hierarchical clustering. D1, D2, and D3: three stages of drought treatment; D3R: 48-h water recovery after drought; S1, S2, and S3: three stages of high-salinity treatment. In total, 1,377, 1,903, and 1,919 genes from flag leaf, shoot, and panicle, respectively, were included in this analysis. (D) Relatedness of the genome expression patterns across selected stress-treated rice organs. A complete-linkage hierarchical clustering analysis of overall relatedness for expression ratios from selected organs at the stage 3 of both abiotic stress treatments. The abbreviations for different sample types are the same as in panel C Most of the previously known genes responsive to both drought and high-salinity stress have been recovered from our microarray analysis. Those include the LEA protein (OsJRFA063984), aquaporin (OsIRUA001311), OsNAC1 (OsJRFA108080), dehydrin rab 16b (OsIFCC035025) and DREB1 (OsJRFA067313). Many other responsive genes have been identified for the first time in this study (Table S1C). Among those newly revealed genes, some are expected to be involved in general cell function or known to be involved in other stress responses, while others have putative or unknown function. The annotation for some of those genes suggested that they include transcription factors from multiple families, heat shock proteins, various stress (drought, high-salinity, disease, cold, and ABA) responsive genes, protein kinases, transporters, photosynthesis enzymes, and other metabolic pathways (Table S1C). The diversity of affected processes suggests a high level of complexity in regulation. There is a significant overlap between genome expression profiles in response to drought and high-salinity stress To compare the gene expression profiles of the drought and high-salinity stress responses, genes commonly regulated by both stresses at each organ type were examined. A total of 322, 415, and 174 genes were up-regulated and 215, 173, and 372 genes were down-regulated by both stresses in flag leaves, shoots and panicles, respectively (Fig. 2A, B). The common genes represent 55, 33 and 28% (induced) and 27, 27 and 29% (repressed) of all drought-responsive genes in flag leaf, shoot and panicle, respectively. These results indicate that about one-third, even half in flag leaf, of the drought responsive genes in each organ are also regulated by high salinity stress. To further compare gene expression under the two abiotic stresses, all drought regulated genes were selected for cluster analysis (see Experimental procedures). As shown in Fig. 2C, up to half of the genes up- or down-regulated by drought stress also exhibit a similar expression pattern under high-salinity stress. The remainder of the drought responsive genes exhibits minimal expression changes or distinct expression patterns in response to high-salinity stress. Among genes specifically induced by drought stress, some have known functions while others were previously predicted to have functions related to drought stress or the ABA response. For example, our experiments showed that genes with putative functions including NAM, HLH, G-box binding, Zn-finger, AP2 transcription factors, and protein kinases (including MAPK family genes) were all affected by drought stress. Quite a few known drought-responsive genes such as DREB1A, LEA protein, WSI76 protein, MAP65 (microtubule associated protein), and ubiquitin were also included. These genes were specifically induced by drought and may function exclusively in the response to drought stress in rice. To gain an overall view of the effects of drought and high-salinity stress on various functional gene groups, gene ontology (GO) categories for drought and high-salinity stress induced genes in three organs were examined (Fig. S2). For the majority of these functional categories, the relative numbers of genes responsive to the two stresses were similar. We also examined the effects of drought and high-salinity on rice genes coding enzymes in known metabolic pathways using shoot as a model (Fig. S3). Some pathways are similarly regulated by the two stresses, including the Calvin cycle, TCA cycle variation I, brassinosteroid biosynthesis II, gibberellin biosynthesis, and IAA biosynthesis I, and the de novo biosynthesis of pyrimidine ribnucleotides/pyrimidine deoxyribonucleotides/purine nucleotides (Fig. S3 and data not shown). While other pathways respond differentially to drought or high-salinity stresses, the representative pathways include sterol biosynthesis, sugars and polysaccharides, ABA biosynthesis, lignin biosynthesis, octane oxidation, cutin biosynthesis, and starch degradation (Fig. S3). For example, some steps in these pathways were repressed or invariable under high-salinity stress, but were induced by drought stress. It is noticeable that one step in the ABA-biosynthesis pathway (corresponding to Arabidopsis NCED3) showed slightly inhibited in gene expression under high-salinity stress (only at stage 3), but was induced under drought treatment (Fig. S3c). Our informatic analysis suggested that OsJRFA107649 is the closest Arabidopsis NCED3 homolog, with 66% identity (E = 1.8E–195). From our analysis, OsJRFA107649 expression ratios (log2 transformed) were 0.278 (D1), 3.461 (D2) and 2.53(D3) under drought treatment and 3.626 (S1), 2.283 (S2) and –1.958 (S3) under high salinity treatment. The expression ratios of rice gene at D3 and S3 stage were used for the diagram in Fig. S3. It is interesting to note that under high salinity this gene was initially induced and then inhibited at stage 3 (S3 ratio: –1.958), while under drought treatment this gene was induced starting at stage 2 and maintained after stage 3 (D3 ratio: 2.53). This example illustrated a distinct response kinetics for different responsive genes in response to drought and high-salinity stresses. By comparing transcriptomes under drought and high-salinity stresses in three distinct organs, we found that drought responsive genome expression in flag leaf is closer to that for high salinity stress than to the drought expression response observed in any other organ (Fig. 2C, D). In general, there seems to be a closer relationship between the transcriptome-level responses to the two abiotic stresses in the same organ than between or among transcriptomes in distinct organs in response to the same stress. However, the extent of overlap in the responses to drought or high-salinity stresses varies for the different rice organs. For example, it is more divergent in the responses to the two abiotic stresses in panicle, while flag leaves show the greatest overlap. Distinct groups of genes are induced during rehydration after drought stress In its natural environment, a plant’s ability to respond to rehydration after drought stress is important for its survival, although little is known about the gene expression changes that occur during rehydration. Our experiments showed that after a period of drought (third stage), all the rice leaves rolled and turned yellow, and the RWC dropped to 70–75%. When we applied water to those severely stressed plants, rice leaves started to unroll after 5 h, and exhibited normal flatness within 24-48 hours of rehydration. We collected tissue samples 48 h after rehydration and extracted total RNA for gene expression analysis (Fig. 2C). Comparing to samples from normal grown rice plants, in flag leaf, shoot and panicle, 807, 281 and 224 genes were up-regulated after 48 h of rehydration following drought stress, respectively (Fig. 3, Table S14–S16). We further compared gene expression profiles during drought treatment with those after rehydration (Fig. 3A–C). The expression patterns of those genes seem to belong to three groups, with flag leaf exhibiting the greatest number of gene expression changes following rehydration. Group I, which included 71, 98 and 68 genes from flag leaf, shoot and panicle, respectively, was induced during both drought treatment and rehydration. Group II, which included 71, 8 and 21 genes in flag leaf, shoot and panicle, respectively, was repressed during drought stress but induced by rehydration (for shoot and panicle, see Table 1). Group III, which included all remaining 665 genes in flag leaf, 175 genes in shoot, and 135 genes in panicle, showed no significant expression changes under drought stress but were up-regulated during rehydration. We suspected that group I genes must be induced by drought stress, but have yet to return to normal levels after 48-h rehydration. Those genes in groups II and III were specifically induced by rehydration. Fig. 3Cluster analysis of genes exhibiting elevated expression after 48-h rehydration following drought. (A) (B) and (C) Cluster analysis of the genes exhibiting significant elevation of expression (with P < 0.05 as threshold) after 48 h of rehydration treatment for flag leaf, shoot, and panicle. The median ratio is log2 transformed and subject to complete linkage hierarchical clustering. A total of 807 genes in flag leaf (A), 281 genes in shoot (B), and 224 genes in panicle (C) are included in this analysis. D1, D2, and D3 represent three stages of drought treatment, and D3R represents 48-h rehydration of the D3 stage. (D) The histogram shows the number of genes in each of the three rice organs induced under drought treatment at least at one drought stage, after 48-h rehydration following the D3 stage, and the number of genes with overlapping expression in various pairs of organs. y-axis shows the gene numberTable 1Genes induced by 48-h rehydration and inhibited by drought stress in shoot and panicleGene namePutative functionsD1D2D3D3RShootOsIFCC006273Unknown0.596−1.611−3.4971.920OsIFCC021002Glycosyl hydrolases family 170.440−1.105−1.7752.470OsJRFA107373Unknown−1.521−0.487−4.3482.470OsJRFA101949AP2 domain0.216−0.166−2.4562.026OsJRFA106016Kelch motif−0.315−0.582−2.6041.865OsJRFA070715Peroxidase−0.189−0.499−2.4562.215OsJRFA065471Folate/biopterin transporter0.331−0.533−2.5953.095OsIFCC041459Chitinase−0.460−0.136−2.1251.972PanicleOsIFCC023423Photosystem I reaction center subunit IV−0.067−0.293−2.6102.206OsIFCC032344Photosystem I reaction center subunit n−0.141−0.626−1.7002.390OsIFCC001915Photosystem II polypeptide−0.119−0.934−1.7741.787OsJRFA061606Pathogenesis–related protein PR-10a−1.811−0.007−1.8414.117OsJRFA107373Unknown−3.6930.395−1.9785.605OsJRFA109398Chlorophyll a/b binding protein−1.867−1.348−1.8991.905OsIFCC025509Photosystem II 10 kDa polypeptide PsbR−1.077−0.966−1.7351.831OsIFCC038501Chloroplast precursor−1.087−0.966−1.8051.837OsIFCC022062Photosystem I reaction center subunit VI−0.540−0.206−1.9961.811OsIFCC024081Chlorophyll a/b-binding protein−0.595−0.685−1.8142.298OsIFCC014461Unknown−0.794−0.660−1.7221.790OsIFCC017469Unknown−0.879−1.619−1.7642.417OsIFCC012276Catalase−1.567−0.594−1.8932.592OsJRFA060135Unknown−0.922−0.491−2.5401.793OsIFCC033400Pathogenesis-related protein PR-10a−0.115−0.032−2.5502.008OsJRFA059435Magnesium-protoporphyrin IX monomethyl esteraerobic oxidative cyclase−0.344−0.348−1.8372.851OsIFCC022709Protease inhibitor/seed storage/LTP family−0.091−0.300−1.6841.842OsJRFA061968Geranylreductase−0.334−0.244−1.8042.001OsJRFA071762Aluminium-induced protein0.217−0.594−1.8491.752OsIFCC029079Blight-associated protein p12 precursor−1.1100.379−1.8482.980OsJRFA106991Unknown−0.899−0.142−1.8741.715Log2 transformed ratios of all genes at three stages of drought stress and rehydration stage were listed The down-regulation of group II genes in response to drought led us to speculate that they may play an important role in conferring drought stress tolerance, whereas the up-regulation of group II and group III genes during rehydration could be important for recovery. In both group II and III genes, we found that two classes of genes were over-represented among the rehydration-inducible genes in shoot and panicle: transporter genes and photosynthesis-related genes. These transporter genes included: OsJRFA065471 (folate/biopterin transporter), OsJRFA066919 (putative potassium transporter), OsJRFA067899 and OsIFCC019970 (ABC transporter, putative), OsJRFA068003 and OsJRFA106202 (Transmembrane amino acid transporter protein), OsJRFA068765 (H-ATPase), OsIFCC040482 (phosphate:H+ symporter), OsJRFA072183 (Sodium:sulfate symporter transmembrane region), OsJRFA102086 (putative lipid transfer protein) and OsJRFA103807 (aquaporin). The photosynthesis related genes cover most of the gene components of the two photosystems, such as genes for putative chlorophyll a/b-binding protein, Photosystem I or II reaction center subunits, and plastocyanin. These photosystem genes and transporter genes represent a large portion of all genes induced by rehydration, and their physiological roles in plants fit well with a potential contribution toward plant recovery from drought stress. Limited overlap of stress responsive genes among rice organs We also examined the degree of overlap in expression of stress responsive genes in two or more rice organs under drought or high-salinity stress. Venn diagrams revealed that only a small portion of genes was shared between each pair of organs (Fig. 4). The greatest overlap occurred between shoot and flag leaf, which shared more inducible genes than shoot and panicle or flag leaf and panicle under both drought and high-salinity conditions. Under drought stress, one-third of the genes up-regulated in flag leaf (197/582) were also induced in shoot (Fig. 4A), while under high salinity stress, over half of the induced genes (494/817) in shoot (mostly young leaves) were also up-regulated in flag leaf (Fig. 4B). Only 21 and 23% of genes up-regulated in panicle were also induced in flag leaf under drought and high salinity stresses respectively. The percentage of genes shared between shoot and panicle is similar to that between flag leaf and panicle (Fig. 4A, B). Fig. 4Comparison of gene expression patterns among the three rice organs in response to drought and high-salinity stresses. (A) and (B): Venn diagram of drought (A) and high-salinity (B) induced genes among the three rice organs. (C) and (D): Venn diagram of drought (C) and high-salinity (D) repressed genes among the three rice organs Interestingly, only a small fraction of stress responsive genes were expressed in all three organs examined. For example, 42 and 151 genes were induced in all three organs under drought and high-salinity stress respectively. In general, most of those genes exhibited high levels of expression as well as strong inducibility. Among them, 27 genes were induced by both drought and high-salinity stress in all three organs. This group of 27 genes includes a protein kinase (OsJRFA058518), chlorophyll a/b binding protein (OsJRFA062972), ABA-responsive protein (OsJRFA063578, OsIFCC018156), LEA protein (OsJRFA063984), dehydrin (OsIFCC035028), CHY zinc finger protein (OsIFCC003263), and homeobox protein (OsIFCC018343) (Table 2).Table 2Genes up- or down-regulated by both drought and high-salinity stresses in all three organsGene namePutative functionsFlag leafShootPanicleD(h)S(h)D(h)S(h)D(h)S(h)Up-regulatedOsJRFA058518Protein kinase domain2.5911.9393.9393.4433.1343.553OsJRFA058851Unknown5.4303.9005.8944.6002.0423.878OsJRFA062356Unknown4.9592.4774.6994.6603.3145.727OsJRFA062972Chlorophyll a/b binding protein4.7582.5421.8952.9312.2912.204OsJRFA063156Unknown1.8371.7434.4053.9212.6062.866OsJRFA063334Unknown3.2082.1404.5213.1881.8092.167OsJRFA063578ABA-responsive protein2.7433.3524.1693.7331.8592.748OsJRFA063889Unknown3.3403.2632.1863.3442.0613.712OsJRFA063984LEA protein5.5134.7303.7634.6293.9716.773OsJRFA068381Unknown2.6481.8303.6302.8331.8201.786OsJRFA070577Unknown3.0362.8773.0652.8991.8832.372OsJRFA070872Unknown8.4857.7333.4466.3463.9634.107OsJRFA071812Unknown2.7482.2683.0152.2441.7672.559OsJRFA072568Unknown3.0013.1212.4113.3261.7142.005OsJRFA106307Unknown4.2843.4934.3992.5112.3162.960OsJRFA106562Unknown3.7974.0092.5443.4672.0103.978OsJRFA107065Unknown2.7701.7514.8593.6482.9842.292OsJRFA108083Unknown2.2742.5974.8534.2413.0774.631OsIFCC031279Alpha-galactosidase3.4073.0873.8193.5382.3892.206OsIFCC036408Unknown3.1373.9873.0383.1114.7424.192OsIFCC035028Dehydrin7.9747.6895.1486.7678.0456.382OsIFCC018156Abscisic acid-induced protein5.1677.2243.2434.9112.3292.327OsIFCC003263CHY zinc finger1.7092.1471.9942.2823.7911.956OsIFCC018343Homeobox domain4.2903.9553.4323.9592.6833.017Down-regulatedOsIFCC033098Similar to Arabidopsis F16L1.3 protein−2.676−4.719−2.542−2.346−1.701−2.519OsIFCC015113Phosphoribulokinase /Uridine kinase family−2.456−2.689−1.863−1.717−1.853−1.899The highest ratio of each gene at three stages under drought (D(h)) or high-salinity (S(h)) were log2 transformed and listed Analysis of stress down-regulated genes (Fig. 4C, D) revealed a similar small overlap among the three organs. Only 68 and 129 out of 795 total genes down-regulated in flag leaf under drought stress were also inhibited in shoot and panicle, respectively. For high-salinity stress, 135 and 214 out of 1,270 down-regulated genes in flag leaf were also repressed in shoot and panicle respectively. Only 16 and 38 genes were down-regulated in all three organs under drought and high salinity stress respectively, while only two genes were inhibited in all three organs under both drought and high salinity stress (Table 2). The shoot samples we used consisted largely of young leaves and were thus physiologically closer to flag leaf than panicle. Indeed, shoot and flag leaf generally exhibited similar patterns of gene up-regulation in response to drought and high-salinity stress. However, this does not appear to be the case for repressed genes, where overlap between shoot and flag leaf gene expression is minimal. To validate the microarray data, RT-PCR analysis was used to verify the transcription response of representative genes from microarray results. For this purpose, we picked a group of eight representative genes and designed primers for RT-PCR analysis (Table S18). The cDNA templates were synthesized from total RNA samples prepared from shoot, flag leaf and panicle under drought and unstressed controls. The semi-quantitative RT-PCR results for the eight representative genes were shown in Fig. 5. The expression patterns of all eight genes reflected changes observed by microarray analysis fairly and accurately for all three organs examined, except in three cases where RT-PCR failed to confirm the microarray results (Fig. 5), indicating that our microarray data is reliable. Fig. 5RT-PCR analysis of the representative drought-induced genes among the three rice organs. Total RNA samples were prepared from the shoot (S1, S2, and S3), flag leaf (F1, F2, and F3), and panicle (P1, P2, and P3) and taken from plants at three stages of drought treatment and the untreated control plants (S0, F0, and P0). The corresponding log2 transformed median ratio of microarray data is shown at the bottom of each mRNA blot line. N/A: no expression. The * corresponds to cases where the microarray data was not confirmed by the mRNA blot results Enrichment of ABRE and DRE cis-regulatory elements in abiotic stress-induced gene promoters Previous reports have suggested that several well-characterized drought-, high-salinity and cold inducible gene promoters contain two common cis-regulatory elements, the ABA-responsive element (ABRE) and the dehydration-responsive element (DRE) (Seki et al., 2001, 2002a, b; Yazaki et al., 2003; Rabbani et al., 2003; Dubouzet et al., 2003; Nakabayashi et al., 2005; Maruyama et al., 2004; Yamaguchi-Shinozaki and Shinozaki, 2005). ABRE and DRE confer ABA-dependent and ABA-independent gene expression in response to water stress. It has been reported that a single copy of ABRE in the promoter region only induces relatively minor elevation of expression level, while multiple copies of ABRE strongly activate expression (Shen and Ho, 1995; Nakabayashi et al., 2005). To elucidate the relationship between copy numbers of DRE and ABRE cis-regulatory elements in gene promoter regions and abiotic stress-induced gene transcription, we divided the genes into groups based on their expression patterns. Genes induced specifically by drought or high-salinity stress in each of the three organs defined six groups, while genes induced by both drought and high-salinity stress in each organ defined another three groups. Common inducible genes shared by at least two organs under drought or high-salinity stress defined another nine groups. We only selected those genes among all those groups with full-length cDNA sequences available for further promoter analysis. In brief, the 2 kb upstream regions (–1 to –2,000 bp) preceding the ATG start codon of genes were selected for promoter motif analysis (see Experimental procedures). A similar number of rice genes with full length cDNA available that lack stress responsive expression were used as control. Copy number of ABRE and DRE core motifs on each promoter region were counted. The percentages of genes in each group with 1 to 6 copies of ABRE core or DRE core elements were then calculated. Compared to control genes, copy numbers of the ABRE and DRE core motifs in the promoter regions of genes with organ-specific expression in response to drought or high-salinity stress were not markedly different (Fig. 6A, D). The promoter regions of genes responsive to both drought and high-salinity stress in each organ were enriched for ABRE and DRE core motifs compared to the promoters of control genes (Fig. 6B, E). For example, only 25% of control genes have over four copies of ABRE core motif in their promoter regions, but 48, 44 and 48% of genes induced by both drought and high-salinity stresses in flag leaf, panicle and shoot respectively have over four copies of the ABRE core motif in their promoter regions (Fig. 6B). Fig. 6Distribution of the ABRE and DRE core motif sequences among different groups of drought and high-salinity stress responsive genes. In all panels, the x-axis shows the copy number of ABRE or DRE core motifs and the y-axis shows the relative number of genes containing different copy numbers of ABRE or DRE core motifs.(A) ABRE core motif distribution among promoters of the genes induced only by drought or only by high-salinity stress in each single rice organ. (B) ABRE core motif distribution among promoters of genes induced by both drought and high-salinity stress in each rice organ. (C) ABRE core motif distribution among promoters of genes induced in two or three organs. (D) DRE core motif distribution among promoters of genes induced only by drought or only by high-salinity stress in each rice organ. (E) DRE core motif distribution among promoters of genes induced by both drought and high-salinity stress in each rice organ. (F) DRE core motif distribution among promoters of genes induced in two or three organs or by both stresses in more than one organ. The following 18 gene groups (with number of genes with full-length cDNA gene number) were analyzed. (1) Genes induced by drought stress only in flag leaf (F-D123, 82). (2) Genes induced by high-salinity stress only in flag leaf (F-S123, 433). (3) Genes induced by drought stress only in shoot (S-D123, 272). (4) Genes induced by high-salinity stress only in shoot (S-S123, 77). (5) Genes induced by drought stress only in panicle (P-D123, 139). (6) Genes induced by high-salinity stress only in panicle (P-S123, 420). (7) Genes induced by both stresses in flag leaf (F-D123S123, 190). (8) Genes induced by both stresses in shoot (S-D123S123, 252). (9) Genes induced by both stresses in panicle (P-D123S123, 77). (10) Genes induced by drought stress in both flag leaf and shoot (F-S-D123, 123). (11) Genes induced by drought stress in all three organs (F-S-P-D123, 59). (12) Genes induced by high-salinity stress in all three organs (F-S-P-S123, 97). (13) Genes induced by high-salinity stress in both flag leaf and shoot (F-S-S123, 309). (14) Genes induced by drought stress in both flag leaf and panicle (F-P-D123, 59). (15) Genes induced by high-salinity stress in both flag leaf and panicle (F-P-S123, 184). (16) Genes induced by drought stress in both shoot and panicle (S-P-D123, 51). (17) Genes induced by high-salinity stress in both shoot and panicle (S-P-S123, 124). (18) Genes induced by both stresses in all three organs (F-S-P-D123S123, 20) When genes expressed in at least two organs under drought or high-salinity stress were compared to control genes, the ABRE and DRE core motif copy numbers also showed significant differences (Fig. 6C, F). For example, 50% of these genes contain over four copies of the ABRE core motif in their promoter regions, compared with 25% for the control genes (Fig. 6C). A similar pattern was found for the DRE core element (Fig. 6F). Genes induced in an organ-specific manner by drought or high-salinity stresses did not contain a significantly different number of copies of ABRE or DRE core motifs compared to control genes (Fig. 6A, D). Thus it is plausible that the stress-induced expression in this fraction of genes may rely less on transcriptional activation mediated by ABRE and DRE motifs. Unidentified organ-specific cis-regulatory elements may exist in the promoter regions of these genes and play a more important role in transcription activation under drought or high-salinity stress. Identification of two novel cis-regulatory elements that respond to rehydration after drought stress As shown in Table 1, there are 8 and 21 genes with full-length cDNA sequences repressed by drought stress but induced by rehydration in shoot and panicle, respectively. An analysis of the upstream region of these eight shoot genes identified a novel motif (motif-SP, GGCAGCCG) located near to the translation start codon (–73 to –250) in five of them (Fig. 7A). To investigate the potential functional role of this novel motif, we performed a gel shift mobility assay which involved incubating a 24 bp probe containing motif-SP from one of those five genes (OsJRFA070715) and a control containing a single base mutation at an invariable position with nuclear extracts (see Experimental procedures). Only nuclear extracts from drought-treated shoot (at the third stage), but not from unstressed shoot or shoot under rehydration, showed a sequence specific binding activity with the probe (Fig. 7C). A single base mutation at an invariable position in the motif-SP core of the probe completely abolished protein binding activity, suggesting high specificity of interaction with the core motif included in this 24 bp sequence. This specific binding activity was only observed in drought-treated plants, suggesting that it is involved in transcriptional repression under drought conditions. Fig. 7A novel promoter motif associated with genes repressed by drought but induced after rehydration in rice shoot. (A) The core sequence of the motif and the position of the motif in each promoter. (B) The expression pattern of a representative gene containing motif-SP. The histogram shows the log2 transformed ratio at three drought stress stages (D1, D2, D3) and 48-hour rehydration (D3R). (C) Gel shift assay of motif-SP. Nuclear proteins were extracted from shoots of untreated plants (C), plants at stage 3 of drought treatment (D3), and plants after 48-h rehydration following stage 3 (D3R). Core sequences of the probe OsJRFA070715: TGCAGCCA, and core sequence of the probe with a single base point mutation OsJRFA070715M: TGAAGCCA We also searched promoter regions of 21 genes inhibited by drought stress but induced by rehydration in panicle. Three conserved motifs were found through this blind search in 11 of the 21 genes: ABRE motif, the above-mentioned motif-SP, and another novel element, motif-P (Fig. 8). The presence of the ABRE motif suggests that a gene may respond to drought stress by employing an ABA-dependent pathway, while motif-SP may play an important role in drought-stress repression and subsequent activation during rehydration. We failed to detect specific binding activity using similar nuclear extracts (data not shown) and thus the role of motif-P remains to be defined.Fig. 8Three motifs associated with genes repressed during drought and induced by rehydration in panicle. The core sequence logos and the position of the motifs in each of the 11 promoters are listed Transcription factor genes under drought and high salinity stress are expressed in a largely organ specific manner We further examined transcription factor genes whose expression are regulated by drought and high-salinity stresses. Among all genes induced by drought or high-salinity treatment in the three organs, a total of 186 genes (Table S17) were predicted to be transcription factors with a DNA-binding domain. These transcription factors belong to various families, including AP2, bHLH, MYB, HB, NAC, zinc finger, MADS, bZIP, WRKY and HSF families. These transcription factors could also be divided into several groups depending on their expression patterns. Among the 186 transcription factor genes, 12 genes were induced in all three organs in at least one stage following either drought or high-salinity stress, whereas the remaining 174 genes were mainly up regulated in one or two organs. Transcription factors induced only in one individual organ were listed in Table 3. It is interesting to note that over half of the transcription factors were expressed in at least two organs, while other transcription factors were activated in an organ-specific manner. This suggests that the expression of different transcription factors may play a key role in common or organ-specific gene expression in response to drought or high-salinity conditions. Table 3Transcription factor genes induced in each organ by drought or high-salinity stressGene name Putative functionPanicle: induced by high-salinity onlyS(h)D(h)OsIFCC018668bHLH transcription factor1.673−0.850OsIFCC029156Helix-loop-helix DNA-binding domain1.720−0.599OsJRFA110611No apical meristem (NAM) protein2.234−1.500OsJRFA105079CCAAT-box binding factor HAP5 homolog3.521−0.850OsJRFA070817RING zinc finger protein1.708−0.685OsIFCC039583Zinc finger, C3HC4 type (RING finger)2.2780.530OsJRFA108605Helix-loop-helix DNA-binding domain4.7130.240OsJRFA108208AP2 domain2.429N/AOsJRFA101136C3HC4-type zinc finger5.559−0.753OsIFCC008718No apical meristem (NAM) protein2.542N/AOsJRFA066984Dof domain, zinc finger1.8460.168OsJRFA110661Zinc finger C-x8-C-x5-C-x3-H type1.732N/AOsJRFA106969Myb-like DNA-binding domain2.010N/AOsIFCC016263Zinc finger, C2H2 type2.636N/APanicle: induced by drought onlyS(h)D(h)OsIFCC042866AP2 domainN/A3.050Panicle: induced by both drought and high-salinityS(h)D(h)OsJRFA107283NAM-like protein7.7732.080Shoot: induced by drought onlyS(h)D(h)OsJRFA105599DRE-binding protein 1AN/A1.784OsIFCC031932WRKY DNA -binding domain−2.7462.186OsIFCC031182Myb factorN/A2.565OsIFCC042758Helix-loop-helix DNA-binding domain−0.0242.367OsJRFA100208Helix-loop-helix DNA-binding domain−0.8363.132OsJRFA107524Dof domain, zinc fingerN/A2.294OsJRFA106333Helix-loop-helix DNA-binding domain0.4301.762OsJRFA106282WRKY DNA -binding domainN/A2.026OsIFCC043271Helix-loop-helix DNA-binding domain0.2452.259OsIFCC000715Myb-like DNA-binding domain0.9492.617OsJRFA110587Similar to DNA-binding protein WRKY30.9054.075OsJRFA107146AP2 domain0.6252.746OsIFCC038336Zinc finger transcription factor ZF10.5292.407Shoot: induced by both drought and high-salinityS(h)D(h)OsJRFA072192Zinc-finger protein2.0742.059OsIFCC000984WRKY DNA -binding domain1.7442.153Shoot: induced by high-salinity onlyS(h)D(h)OsIFCC017057B3 DNA binding domain2.026N/AOsJRFA067496TRAF-type zinc finger2.1990.801Flag leaf: induced by drought onlyS(h)D(h)OsIFCC029554Zinc finger, C2H2 type−0.5532.032Flag leaf: induced by high-salinity onlyS(h)D(h)OsIFCC001054Zinc finger C-x8-C-x5-C-x3-H type1.7050.582The highest ratio of each gene among three stages of drought (D(h))or high-salinity (S(h)) were log2 transformed and listed Minimal co-regulation of neighboring genes in drought or high-salinity stress responses Co-regulation of adjacent genes has been observed in several organisms (Hurst et al., 2004), including human (Caron et al., 2001; Lercher et al., 2002), Drosophila (Spellman and Rubin, 2002), Arabidopsis (Birnbaum et al., 2003, Ma et al., 2005a), and yeast (Cohen et al., 2000). All these studies identified co-regulation of neighboring gene clusters on chromosomes (chromatin domains). In rice, analysis of organ-specific gene expression revealed that about 10% of the genome belongs to chromatin domains that exhibit this co-expression pattern (Ma et al., 2005b). However, similar analysis of light-regulated genes failed to reveal significant co-regulation (Jiao et al., 2005). To investigate possible co-regulation of neighboring genes in the drought and high-salinity responses, we mapped rehydration induced genes in three organs and drought and high-salinity induced genes in shoot to all 12 indica chromosomes (Figs. 9, S4). This analysis indicated that significant gene clustering among rehydration, drought, and high-salinity induced genes did not occur at the chromosomal level, similar to the pattern characteristic of light-regulated genes (Jiao et al., 2005). Among the 12 rice chromosomes, only three regions showed significant enrichment of 10 or more co-regulated genes within any given 2 Mb domain. For example, the 33–35 Mb region on chromosome 2, the 11–13 Mb region in chromosome 3, and the 31–33 Mb region in chromosome 6 (Fig. 9B) contain significant enrichment of rehydration regulated genes in all three organs, although even in those cases genes are not directly adjacent. Fig. 9Distribution of drought, high salinity and rehydration regulated genes in a representative rice chromosome. (A) Map of chromosome 3 genes differentially expressed at the first stage of drought, first stage of high-salinity stress, and 48 h after rehydration in shoot. (B) Map of chromosome 3 genes differentially expressed at 48 h after rehydration in the three rice organs Discussion This study provides new insight into the rice response to drought and high salinity stresses at the whole genome level. Using a whole genome microarray, we monitored the expression of 36,926 unique or known rice genes or gene models in three different organs under drought and high salinity stress. Our work thus offers the first comprehensive picture of genome expression modulation in response to drought and high salinity stress in three distinct rice organs. Genome expression reprograming showed significant overlap between drought and high salinity responsive genes Using the criteria outlined above, a total of 2,957 and 2,090 rice genes showed significant up- or down-regulation in response to high salinity stress and drought stress in at least one of the three organs. Our analysis suggested that 927 out of 2,090 (44%) genes induced by drought were also up regulated by high-salinity stress in the same organs. This number is consistent with previous studies reported for Arabidopsis (Seki et al., 2002a, b; Shinozaki et al., 2003). In rice, an even higher percentage of drought induced genes were also up-regulated under high-salinity stress, as reported previously on a smaller scale (Rabbani et al., 2003), which covered only cDNAs from plants under drought, cold and high-salinity treatment. Thus our analysis should be more representative of the overall rice genome response to drought and high salinity stresses without such bias. However, drought and salt stress conditions in the field may be more subtle and less regular as in our experimental conditions, thus our results are only meant to illustrate some particular situations rice genome expression responses to stresses. It is interesting to note an overlap of up to about half of the genes induced or inhibited by drought and high-salinity stress in rice. This observation is consistent with current understanding that these two stresses affect plants in overlapping but not identical ways. At a physiological level, both stresses cause water depletion in the above-ground portion of the plants and induce similar morphological responses (Fig. S1). At the molecular level, half of the transcriptional factor genes identified in our results were shared by both stresses in each organ, which is consistent with the observation that about half of the genes that respond to the two stresses were shared at the whole genome level. Reprograming of genome expression in response to drought and high-salinity stresses is largely organ specific In this study, we examined whole genome expression profiles under drought and high-salinity conditions in three organs: four-tiller stage shoot, filling stage flag leaf and panicle. Rice plants at these two growth stages, particularly the late one, are sensitive to drought and high-salinity stress. It is well known that drought or high-salinity stress at the heading and early panicle stages can severely compromise rice growth and development and reduce crop yield even with late rehydration. It is evident that the rice genome is subject to significant reprograming with regard to which portion of genome is expressed under drought or high salinity stress. Our results showed that only a limited number of drought and high salinity responsive genes were shared between any two organs. We found that only 13.5 and 20.5% of drought-induced genes in panicle were shared with those induced in shoot and flag leaf respectively, and 33.8% of drought-induced genes in shoot were also activated in flag leaf (Fig. 4). The percentages of genes shared between any two organs under high-salinity conditions were similar to those under drought stress, suggesting that responses to drought or high-salinity stress in different organs were independently regulated. This observation is similar to the observed small overlap of cold responsive gene expression between root and leaf in rice (Kreps et al., 2002). Genome expression reprograming under either drought or high salinity stress entails a large number of genes involved in many aspects of cellular function. A GO analysis of stress responsive genes (Fig. S2) indicates that, in most categories, three organs activated similar scale (number) of genes in response to drought or high salinity stress, while only small percentage of responsive genes were overlapped between or among organs (Fig. 4). This suggests that responsive genes under the same category are largely distinct individuals under distinct organ types. It is thus possible that homologous or functionally similar gene family members are responsive to the same stresses in each organ. This is consistent with our observation of organ-specific transcription factor gene expression in response to both drought and high salinity stresses (Table 3). The fact that genes specifically induced in each organ do not exhibit enrichment of DRE and ABRE motifs (Fig. 6) suggests that organ-specific transcriptional factor gene expression may be responsible for activating organ-specific downstream genes in a secondary transcriptional response to stress. For these genes, it is likely that certain organ-specific promoter elements mediate this response pathway. According to the GO analysis, under high salinity stress, about 10% storage proteins were up-regulated in panicle, while in flag leaf and shoot only less than 1% storage protein genes were induced (Fig. S2). This kind of difference among organs may play a role in organ specific response to two stresses; further function analysis of these differentially regulated genes among organs are needed. Rice may possess specific mechanisms to facilitate plant recovery during rehydration after drought According to our microarray analysis, rice genes induced by 48-h rehydration were divided into three groups according to their expression patterns (Fig. 3), somewhat similar to a previous report for Arabidopsis (Oono et al., 2003). This observation suggested that monocotyledonous and dicotyledonous plants share similar gene expression responses to rehydration after drought. Analysis of the genome-wide distribution of rehydration-regulated genes in the three organs showed no significant co-regulation of neighboring genes at the chromosomal level. This observation is similar to the case of light-regulated genes (Jiao et al., 2005) but different from general gene transcription from organ samples (Ma et al., 2005b). This distinction in the modulation of genome expression may reflect different mechanisms employed in response to different developmental or environmental signals. Up to now only one possible cis-element has been reported to be involved in the rehydration process after dehydration in Arabidopsis (Satoh et al., 2002; Oono et al., 2003). In rice, at the whole genome level we identified 807, 281 and 224 genes induced by rehydration in flag leaf, shoot and panicle separately (Fig. 3). These genes provided us an important starting point to study rehydration mechanisms and to search for novel cis-regulatory promoter elements associated with dehydration or rehydration. Two novel cis-regulatory elements (motif-SP and motif-P) were identified (Fig. 8). Motif-SP was also found in shoot-specific genes with similar expression patterns in response to drought and rehydration. Gel shift assays provided evidence that motif-SP may function as a cis-element to mediate the drought-induced repression and late de-repression (activation) during rehydration in rice. Further functional analysis of the promoter elements may substantiate the role of those novel promoter motifs. Previous studies reported that drought stress suppressed plant photosynthesis systems and significantly modulated the activity of some membrane transporters (Rizhsky et al., 2002; Ramachandra et al., 2004; Johansson et al., 2004; Becker et al., 2003). Here microarray analysis revealed that genes involved in photosynthesis and genes encoding transporters were repressed or maintained at low levels of expression under drought but were then strongly activated after rehydration. The repression of metabolic genes during drought stress allows the plant to conserve energy and subsist on less water, conferring better drought tolerance. When supplied plenty of water upon rehydration, activation of these genes could aid in the recovery of full photosynthesis activity and transmembrane solute/water exchange, thus helping plant resume its normal growth and development quickly. Electronic supplementary material Below are the electronic supplementary materials. Fig. S1 (JPG 3,830 kb) Fig. S2 (JPG 4,490 kb) Fig. S3 (JPG 2,773 kb) Fig. S4 (JPG 15,611 kb) ESM 5 (XLS 7,081 kb) ESM 6 (XLS 37 kb) ESM 7 (XLS 48 kb) ESM 8 (XLS 71 kb) ESM 9 (XLS 40 kb) ESM 10 (XLS 39 kb) ESM 11 (XLS 73 kb) ESM 12 (XLS 92 kb) ESM 13 (XLS 72 kb) ESM 14 (XLS 49 kb) ESM 15 (XLS 74 kb) ESM 16 (XLS 74 kb) ESM 17 (XLS 122 kb) ESM 18 (XLS 48 kb) ESM 19 (XLS 22 kb) ESM 20 (XLS 19 kb) ESM 21 (XLS 23 kb) ESM 22 (XLS 10 kb)

J_Chem_Ecol-4-1-2373416

The Chemistry of the Postpharyngeal Gland of Female European Beewolves

Females of the European beewolf, Philanthus triangulum, possess a large glove-shaped gland in the head, the postpharyngeal gland (PPG). They apply the content of the PPG to their prey, paralyzed honeybees, where it delays fungal infestation. Here, we describe the chemical composition of the gland by using combined GC-MS, GC-FTIR, and derivatization. The PPG of beewolves contains mainly long-chain unsaturated hydrocarbons (C23–C33), lower amounts of saturated hydrocarbons (C14–C33), and minor amounts of methyl-branched hydrocarbons (C17–C31). Additionally, the hexane-soluble gland content is comprised of small amounts of an unsaturated C25 alcohol, an unknown sesquiterpene, an octadecenylmethylester, and several long-chain saturated (C25, C27) and unsaturated (C23–C27) ketones, some of which have not yet been reported as natural products. Surprisingly, we found a dimorphism with regard to the major component of the PPG with some females having (Z)-9-pentacosene, whereas others have (Z)-9-heptacosene as their predominant component. The biological relevance of the compounds for the prevention of fungal growth on the prey and the significance of the chemical dimorphism are discussed. Introduction Hymenoptera possess a huge variety of exocrine glands (e.g., Hölldobler and Wilson 1990). The chemistry and function of different types of these have been studied for a number of social species, whereas comparatively little is known from solitary bees and wasps. Recently, a postpharyngeal gland (PPG) has been described from a species of digger wasp (Strohm et al. 2007), the European beewolf, Philanthus triangulum Fabricius 1775 (Hymenoptera: Crabronidae, formerly Sphecidae, Melo 1999). The occurrence of this gland is surprising since the PPG was assumed to be restricted to ants (Hölldobler and Wilson 1990; Schoeters and Billen 1997; Lenoir et al. 1999) where it functions in generating the colony odor (e.g., Hefetz et al. 1992, 1996; Soroker et al. 1994, 1995, 1998; Vienne et al. 1995; Dahbi et al. 1998; Lenoir et al. 1999, 2001; Oldham et al. 1999; Soroker and Hefetz 2000; for a review of other proposed functions, see Eelen et al. 2006). In beewolves, the PPG has a unique function in protecting the larval provisions from microbial attack (Strohm and Linsenmair 2001; Herzner and Strohm 2007; Herzner et al. 2007a). Female European beewolves hunt and paralyze honeybees, bring them to their nest burrow, and provision one to six bees in a brood cell as larval food for one progeny. Due to the humid and warm conditions in the brood cell, the highly nutritive provisions are prone to detrimental microbial attack (Strohm and Linsenmair 2001). Early fungus infestation inevitably destroys the food resources, and larvae are killed by fungal toxins or starve to death. Observations in special cages (Strohm and Linsenmair 1994–1995) revealed that beewolf females intensively lick the bodies of the paralyzed bees and apply the secretion from the PPG to the bees’ surface (Strohm and Linsenmair 2001; Herzner et al. 2007a). This treatment has the effect of delaying fungus growth for 2 to 3 days, which is a highly relevant effect given the short larval feeding period of only 8 to 11 days. The primary mechanism of this delay is not a direct chemical effect of the secretion on fungi, but the prevention of water condensation on the bees that in turn impairs the germination and growth of fungal spores (Herzner and Strohm 2007). Male European beewolves also have a PPG that is even larger than in females (Herzner et al. 2007b). However, it serves as a reservoir for the scent marking pheromone that males apply to their territories to attract females (Kroiss et al. 2006). Despite the advanced understanding of the function of the secretion of the PPG of female European beewolves, there has been no detailed investigation of its chemistry. Therefore, we analyzed the chemical composition by using combined gas chromatography-mass spectrometry (GC-MS), gas chromatography-Fourier transform infrared spectroscopy (GC-FTIR), fractionation, and derivatizations. Materials and Methods Sampling Beewolf females were taken either from a field population close to the Biocenter of the University of Würzburg, Germany, or from a laboratory population kept at the University of Würzburg (daughters of field caught females). They were all mated with actively provisioned nests with honeybees and were between 1 and 4 weeks old. To identify the chemicals of the PPG, females that were freshly killed by CO2 were decapitated, and their PPGs were removed from the heads by grasping the hypopharynx with tweezers and gently pulling the attached gland out through the mouth (Strohm et al. 2007). The glands were immersed immediately in 0.25 ml n-hexane (Fluka Chemie GmbH, Buchs, Switzerland) that had been distilled prior to use. For the identification of the components in the PPG, the glands of four females were pooled. This enabled us to identify minor components that were not reliably detectable in the extracts of individual PPGs. In order to have an easier method to obtain the content of the PPG, we investigated whether extracts of whole heads differed in composition from dissected glands. There was a large amount of hydrocarbons in the PPG, and extracts of entire heads were identical to extracts of the dissected glands. There were no additional peaks detectable and the proportions were the same (see Herzner et al. 2007b for an analogous procedure for male PPGs). Thus, as an easier alternative to the dissection of the PPG, entire heads were extracted. To obtain data on PPG content variability, we analyzed individual extracts made from the heads of 37 females. Heads of freshly killed females were cut off and extracted in distilled hexane for 4 h. An internal standard (octadecane, Sigma, St Louis, MO, USA) was added to assess the absolute amount of compounds in the PPG (although octadecane could be detected in pooled samples, it was not detectable in individual samples and was therefore employed as an internal standard). Identification Identification of the chemicals was accomplished by GC-MS. Head extracts were fractionated by solid phase extraction (SPE, Chromabond, unmodified silica, 3 ml, 500 mg, Macherey-Nagel, Düren, Germany) with hexane as the first and dichloromethane as the second eluent to separate nonpolar and polar fractions. Alkanes were characterized by comparison of mass spectra and retention indices with those of purchased standard alkanes (Aldrich, Deisenhofen, Germany). Corresponding alkenes were tentatively identified by their typical mass spectra, their retention indices, and (depending on availability) with commercially available (Aldrich) and synthesized standards. Dimethyl disulfide (DMDS) derivatization was carried out to determine the position of double bonds according to the method of Dunkelblum et al. (1985). The configurations of the double bonds were determined by using GC-FTIR (Attygalle 1994). Methyl-branched alkanes were identified by using MS databases and diagnostic ions, and by determining retention indices (Carlson et al. 1998). Details on the identification of polar compounds are given in the results section.For the identification of the hydrocarbons, we used a Hewlett Packard HP 6890 Series GC System coupled to a Hewlett Packard HP 5973 Mass Selective Detector (Agilent Technologies, Böblingen, Germany). The GC was equipped with a DB-1 fused silica capillary column (30 m × 0.25 mm i.d., 0.25 μm film thickness; J & W, Folsom, CA, USA). Temperature was programmed from 100°C to 300°C with a 6°C/min heating rate, held for 20 min at 300°C. Helium was used as carrier gas with a constant flow of 1 ml/min. Injection was carried out at 300°C in the splitless mode for 2 min. The electron impact mass spectra (EI-MS) were recorded with an ionization voltage of 70 eV and a source temperature of 230°C. The software ChemStation (Agilent Technologies) for windows was used for data acquisition.GC-FTIR spectra were obtained by using an HP 5890 GC (Agilent Technologies) coupled to an FTS 575C Tracersystem (BioRad, Hercules, CA, USA). The GC was equipped with the same column as described above. Temperature was programmed from 80 to 270°C with a 4°C/min heating rate. Helium was used as carrier gas with a constant flow of 1–2 ml/min. Injections were carried out by using a split/splitless injector at 250°C in the splitless mode for 60 s. Injection volume was 0.1 μl. IR spectra were recorded by scanning 256 times in a frequency range from 4,000 to 700 cm-1 with a resolution of 1 cm-1. Data system was a Dell Optiplex GX110-PC with BioRad WinIR Pro (Version 2.7) Tracer Software and Sadtler IRSearchMaster. Data Handling and Statistical Analysis Because the relative amounts of compounds constitute compositional data, they were transformed to logcontrasts according to Aitchison (1986): Zij = log10(Yij/g(Yj)) where Zij is the standardized peak area i for individual j, Yij is the peak area i for individual j, and g(Yj) is the geometric mean of all peaks for individual j, prior to statistical analysis. Several peaks had to be combined for the quantitative analysis of individuals because they were not always clearly separated in the chromatograms. To test for differences in the proportions of components between groups, we subjected the transformed data to exact tests for two independent groups (SPSS 13.0, SPSS Inc. 2004). Results Identification and Analysis of Hydrocarbons in the PPG The chromatograms of the crude hexane extracts of the PPG showed a total of 53 peaks that represented 62 different compounds (Fig. 1, Table 1). The mean amount of all compounds was 337 ± 292 μg (minimum, 36.7 μg; maximum, 1,410 μg, N = 37). All alkanes as well as (Z)-9-tricosene, (Z)-9-pentacosene, and (Z)-9-heptacosene were identified by comparison of mass spectra and retention indices with those of available standards. The location and configuration of double bonds of other alkenes were determined with DMDS derivatives and GC-FTIR data (band at 721 cm-1: cis configuration of RCH = CHR′; Attygalle 1994). Some components could, however, not be completely characterized (location of methyl groups or double bond, configuration of double bond) due to the small amounts in the extracts. One component could not be identified at all. There were more unsaturated (81.5%) than saturated compounds in the nonpolar fraction of the PPG extract from beewolf females. Fig. 1Total ion chromatograms of the hexane extract of the PPG from individual female European beewolves, Philanthus triangulum, with a (Z)-9-pentacosene (C25-type) and b (Z)-9-heptacosene (C27-type) as the predominant hydrocarbon peak. Numbers correspond to the numbers in the peak list (Table 1). Some compounds listed in Table 1 were present in quantities too low to be visible in these chromatogramsTable 1List of compounds in the postpharyngeal gland of females of the European beewolf, Philanthus triangulumPeak no.RICompound nameDiagnostic MS ions11400Tetradecanea19821500Pentadecanea21231600Hexadecanea22641700Heptadecanea24051710X-Methylheptadecane–61800Octadecanea25471814X-Methyloctadecane–81900Nonadecanea26891906Sesquiterpene41, 69, 81, 93, 107, 120, 133, 222102000Eicosanea282112080(Z)-9-Octadecenylmethylestera222, 264, 296122100Heneicosanea296132200Docosanea310142267Δ-X-Tricosene322152273(Z)-9-Tricosenea322162286Δ-7-Tricosene322172300Tricosanea32418233811-; 9-Methyltricosane168/169, 196/197; 140/141, 224/2251923437-Methyltricosane112/113, 252/2532023525-Methyltricosane84/85, 280/281212371Δ-X-Tetracosene336222400Tetracosanea338232436Δ-14-Tricosen-6-one99, 115, 127, 181, 265, 336242464Δ,Δ-X,Y-Pentacosadiene348252471(Z)-9-Pentacosenea350262492Δ-Pentacosene350272500Pentacosanea35228253713-; 11-; 9-Methylpentacosane196/197; 168/169, 224/225; 140/141, 252/253292538Δ-15-Tetracosen-7-one94, 113, 141, 181, 195, 265, 3503025437- Methylpentacosane112/113, 280/281312554Δ-16-Pentacosen-8-ol67, 348322572(Z)-9-Hexacosene364332600Hexacosanea366342639Δ-16-Pentacosen-8-one127, 155, 195, 209, 265, 364352668Pentacosan-8-one127, 143, 155, 267, 283, 366362672(Z)-9-Heptacosenea378372700Heptacosanea380382714NI–39273513-; 11-; 9-Methylheptacosane196/197, 224/225; 168/169, 252/253; 140/141, 280/281402740Δ-17-Heptacosen-9-one141, 169, 209, 223, 265, 378412800Octacosanea394422840Δ-18-Heptacosen-10-one155, 171, 183, 223, 237, 265, 392432868Heptacosan-10-one127, 143, 155, 171, 267, 283, 295, 394442872(Z)-9-Nonacosene406452900Nonacosanea40846293315-; 13-; 11-Methylnonacosane224/225; 196/197, 252/253; 168/169, 280/281473000Triacontanea422483076(Z)-9-Hentriacontene434493100Hentriacontanea43650313415-; 13-Methylhentriacontane224/225, 252/253; 196/197, 280/281513200Dotriacontanea450523275(Z)-9-Tritriacontene462533300Tritriacontanea464RI retention index; X, Y position of the methyl-group or double bond not known; Δ configuration of the double-bond not known; NI not identifiedaIdentification with available standards. Identification of the Constituents in the Polar Fraction The polar fraction contained a group of uncommon constituents. Hydrogenation of double bonds with H2 and palladium on carbon (Attygalle 1998) as well as DMDS adducts were used to obtain sufficient information by mass spectrometry to identify these compounds. 18-Heptacosen-10-one was identified by comparison with an already published mass spectrum (Yasui et al. 2003). The mass spectrum of this compound showed a molecular ion of m/z 392 and two diagnostic fragment ions at m/z 155 for [C9H19–CO]+ and at m/z 265 for [C17H33–CO]+ indicating the carbonyl position at C10. After hydrogenation of this unsaturated ketone, the resulting 10-heptacosanone as well as the 10-heptacosanone of the untreated extract showed identical mass spectra (m/z: 155 [C9H19-CO]+ , 267 [C17H35-CO]+, 394 M+) and identical retention indices (2,868) with that already published by Yasui et al. (2003). 14-Tricosen-6-one was characterized by its diagnostic masses (m/z: 99 [C6H11-CO]+, 265 [C17H33-CO]+, 336 M+). Also, hydrogenated 14-tricosen-6-one was tentatively identified as 6-tricosanone by comparison with a commercially available MS library (NIST 2.0, Stein et al. 2001). The double bond positions in these and additional unsaturated ketones were determined from mass spectra of DMDS adducts. Additionally, three unsaturated and one saturated ketone were tentatively identified as 15-tetracosen-7-one, 16-pentacosen-8-one, 17-hexacosen-9-one, and 8-pentacosanone. The corresponding alcohol 16-pentacosen-8-ol was tentatively identified by its diagnostic masses after hydrogenation of the polar fraction with lithium aluminum hydride (Attygalle 1998; Table 1). (Z)-9-Octadecenylmethylester was characterized by using a commercially available standard (Aldrich). Finally, a compound with a typical sesquiterpene mass spectrum was found in small amounts in the polar fraction of the extract. We could not determine the type of functional group in this sesquiterpene. Chemical Dimorphism One alkene dominated the composition of the PPG, and surprisingly, this major compound differed among individuals (Fig. 2, Table 2). Among the 37 females whose PPG content was analyzed, 29 had (Z)-9-pentacosene and eight had (Z)-9-heptacosene as their major peak (from here on referred to as the C25- and C27-type, respectively). The frequency of the two types differed significantly from equality (χ2 = 15.4, df = 1, P < 0.001). By using exact tests for independent samples to assess whether the proportions of other components were also affected by the major peak, we revealed that 14 of the 21 peaks differed significantly between the C25- and the C27-type (Table 2). Generally, the C27-type had larger proportions of components with longer chain lengths. Fig. 2Frequency distribution (population of N = 37) of individual female European beewolves, Philanthus triangulum, based on the proportion of (Z)-9-heptacosene in their PPG secretion (values transformed to logcontrasts (Aitchison 1986): Zij = log10(Yij/g(Yj)) where Zij is the standardized peak area i for individual j, Yij is the peak area i for individual j, and g(Yj) is the geometric mean of all peaks for individual j). There is a clear bimodal distribution, i.e., some individuals have a large proportion of (Z)-9-heptacosene, whereas some have a small proportion; there are no intermediate typesTable 2Mean proportions of selected hydrocarbons and ketones from the postpharyngeal glands of individual female European beewolves, P. triangulum, that had either (Z)-9-pentacosene (C25-type) or (Z)-9-heptacosene (C27-type) as the major componentaCompound nameC25-typeC27-typeDifferencesbP values(Peak no. in Table 1)MeanSDMeanSDPent-HeptHeneicosane (12)0.200.140.180.130.020.086Docosane (13)0.160.090.130.080.040.001Tricosenes (14, 15, 16)0.360.270.080.070.28<0.001Tricosane (17)14.063.7412.933.021.130.003Methyltricosanes (18, 19, 20)0.100.100.050.040.050.003Tetracosene (21)1.190.650.140.081.06<0.001Tetracosane (22)0.170.100.140.070.020.003Pentacosenes (25, 26)77.487.968.295.5069.19<0.001Pentacosane (27)2.050.922.821.20−0.780.335Methylpentacosanes (28)0.060.030.080.03−0.020.704Hexacosene (32)0.130.051.620.72−1.50<0.001Hexacosane (33)0.050.030.050.020.000.06716-Pentacosen-8-one (34)1.130.730.250.150.88<0.001Heptacosene (36)1.000.3969.808.86−68.80<0.001Heptacosane (37)0.380.220.310.120.070.032Octacosane (41)0.030.030.030.020.000.18218-Heptacosen-10-one (42)1.025.181.730.75−0.70<0.001Nonacosene (44)0.040.060.950.43−0.90<0.001Nonacosane (45)0.300.300.320.11−0.020.550Hentriacontene (48)0.010.040.040.02−0.03<0.001Hentriacontane (49)0.070.080.060.020.010.062aTable entries include mean and one standard deviation (SD) of the proportion (%) of the components for females of the C25 (N = 29) and of the C27-type (N = 8), as well as the difference (C25–C27) and the significance level for the difference according to an exact test (P, significant differences in bold).bDifferences were calculated from the original data, rounding of values produced some rounding error. Discussion The PPG secretion of beewolf females contains predominantly unbranched unsaturated long chain hydrocarbons (C23–C33, mainly either C25 or C27), smaller amounts of saturated hydrocarbons (C14–C33), and small or trace amounts of methylalkanes (C17–C31), unsaturated ketones (C23–C27), saturated ketones (C25, C27), a sesquiterpene, an unsaturated C18 methylester, and an unsaturated C25 alcohol. The secretion of the PPG of beewolf females is involved in the preservation of their honeybee prey that serves as larval food (Herzner and Strohm 2007). Females apply large amounts of the PPG secretion to the prey prior to oviposition (Strohm and Linsenmair 2001; Herzner et al. 2007a). The prevalence of unsaturated hydrocarbons in the PPG of beewolf females is probably related to the preservation function. The preservation seems to be mainly accomplished by a physical mechanism (Herzner and Strohm 2007): the secretion prevents the condensation of water on the bees and in this way renders the microclimatic conditions unsuitable for the growth of fungi. Possibly, the PPG secretion covers structures on the paralyzed bee that would otherwise function as effective nuclei for the condensation of water. Scanning electron microscopy revealed that the PPG secretion forms a contiguous layer over the whole surface of the prey (Herzner and Strohm 2007). A hydrocarbon mixture of predominately alkenes might be an ideal means to build up such layers, because at the temperatures that prevail in beewolf brood cells, the unsaturated hydrocarbons might be in a more or less liquid state and can be spread easily over the bee surface. The composition of hydrocarbons on the cuticle of paralyzed honeybees is dramatically modified due to the treatment by beewolf females (Herzner et al. 2007a, b). Untreated honeybees and most other insects bear predominantly saturated straight or branched hydrocarbons (e.g., Howard and Blomquist 1982, 2005; Schmitt et al. 2007). Less frequently, alkenes constitute large proportions of cuticular hydrocarbons (e.g., on a termite, Howard et al. 1978; on an aphid parasitoid, Liepert and Dettner 1996; on workers of European hornets, Ruther et al. 2002; on stingless bee foragers, Abdalla et al. 2003; on diapausing butterfly pupae, Kaneko and Katagiri 2004; or on nesting females of the burrowing bee, Simmons et al. 2003). The reasons for the prevalence of saturated hydrocarbons in some species and unsaturated hydrocarbons in other species are unknown. Possibly, the physicochemical properties of the surface can be adjusted to specific requirements by a particular mixture of saturated and unsaturated compounds (e.g., Gibbs and Pomonis 1995; Gibbs 1998). In addition to the ubiquitous alkanes and alkenes that we found in the PPG, we also identified long chain unsaturated ketones, some of which had not yet been described as natural products. Only (Z)-18-heptacosen-10-one and 10-heptacosanone had previously been reported as components of a contact sex pheromone from females of the white-spotted longicorn beetle, Anoplophora malasiaca (Yasui et al. 2003). The function of these compounds in the secretion of beewolf females is unclear. Besides a not-completely identified sesquiterpene, (Z)-9-octadecenylmethylester, and 16-pentacosen-8-ol, these ketones are the only identified components in the PPG of beewolf females that have a functional group. They might be likely candidates to exhibit an antifungal effect. However, in bioassays, no direct antifungal effect of the PPG content could be detected (Herzner et al. 2007a). Male European beewolves have some of the same ketones and a slightly shorter unsaturated ketone in their marking secretion (Schmitt et al. 2003; Kroiss et al. 2006). Their exact function in males is also unknown (see below). The estimated amount of hydrocarbons in the secretion of the PPG differed considerably among females. This might be due to differences in physiological status, size, and age. Interestingly, the mean (approximately 330 μg) and maximum (approximately 1,400 μg) amounts match the quantities found on honeybees that were embalmed with the PPG secretion by beewolf females. Bees are each embalmed with approximately 110 μg (Herzner et al. 2007a) and females provision on average three bees per day and a maximum of ten bees per day (Strohm and Linsenmair 1997). Thus, an average female has available the necessary amount of PPG secretion for the embalming of the average number of bees caught on 1 day. Likewise, the maximum amount found in some individuals would be sufficient to embalm the maximum number of bees that the most successful females hunt per day. Moreover, according to a three-dimensional reconstruction based on histological sections, the maximum volume of the gland was estimated to be 3–4 μl (Strohm et al. 2007). The comparatively large size of the gland is explained by the need to provide enough secretion for the treatment of several prey items per day. The supply with such large amounts of unsaturated hydrocarbons might be costly and might represent a considerable part of the cost of parental care in this species (Strohm and Linsenmair 1999, 2000; Strohm and Marliani 2002). Remarkably, beewolf females show a striking dimorphism with regard to the chemical composition of the PPG secretion. The major component is either (Z)-9-pentacosene or (Z)-9-heptacosene, and there are no intermediate individuals. Both compounds are widespread among aculeate Hymenoptera and other insects (e.g., Ruther et al. 2002; Simmons et al. 2003). Most other components of the PPG secretion also differ between the two morphs, in that C27-types tend to have larger proportions of long-chain compounds. Possibly, the whole metabolism of hydrocarbons is adjusted to longer chain lengths in the C27-types. Why females have either (Z)-9-pentacosene or (Z)-9-heptacosene as the major compound is unclear. Preliminary analyses (E. Strohm, G. Herzner, M. Kaltenpoth, unpublished data) suggest that there is no effect of age or physiological status on the expression of the major component. One proximate explanation could be that conditions during development differ between the females and cause differential gene activation [for example, the synthesis of (Z)-9-heptacosene might be induced by high temperatures because of the presumably higher melting point, e.g., Gibbs et al. 1998; Rouault et al. 2000]. However, our study specimens were bred under identical temperature conditions in the same climate chamber with the same diet (honeybees from the same population) and, nevertheless, showed this dimorphism. Furthermore, an analysis of beewolf females from different populations ranging from northern Germany to the southern valleys of the Alps not only showed both types of females, but also revealed that they occurred in similar proportions (Strohm et al. 2008). Together with the lack of intermediate individuals, this might suggest that the dimorphism has a genetic basis. Such a dimorphism would have to be balanced because otherwise one morph would disappear at least from some populations either because it has a selective disadvantage or because of genetic drift. One possible explanation for a balanced dimorphism is a spatial difference in the suitability of the two alleles. Such spatial heterogeneity might either be generated by differences in abiotic or biotic conditions. There are specialized cuckoo wasps that enter the nests and oviposit on the bees (Strohm et al. 2001). These wasps seem to employ chemical mimicry in order not to leave traitorous signs in the nest. Most notably, the chemical mimicry only refers to the C27-type of females (Strohm et al. 2008). Thus, the C27-type might be disadvantaged with regard to the rate of parasitism, but this could be balanced by other advantages. Again, this hypothesis is weakened by the fact that the proportions of both types of individuals are similar over a wide geographical range despite probable differences in the abundance of the cuckoo wasps. Compared to the content of the PPG of males of the European beewolf (Schmitt et al. 2003; Kroiss et al. 2006), females have fewer components with functional groups and lower proportions thereof, whereas the composition of the alkanes and alkenes is similar. The function of the secretion of the male PPG is also quite different. Males use this secretion to scent mark their territories and to attract receptive females (Simon-Thomas and Poorter 1972; Evans and O’Neill 1988; Strohm 1995; Strohm and Lechner 2000; Schmitt et al. 2003). The marking secretion is stored in and delivered from the PPG (Kroiss et al. 2006). Therefore, the compounds with functional groups may play a role in the attraction of females (Herzner et al. 2005; Kroiss et al. 2006) and may also convey some additional information on male quality and suitability as a mate (Herzner et al. 2006; Kaltenpoth and Strohm 2006). In summary, females of the European beewolf, P. triangulum, have large amounts of mainly unsaturated hydrocarbons in their PPG. The composition of the secretion is probably shaped by its function as an antifungal coating of the prey, paralyzed honeybees. Thus, the function of the PPG of beewolf females differs entirely from this gland in ants. However, the general chemistry is consistent with that found in ants. This supports earlier arguments based on morphology, ultrastructure, and behavioral context (Strohm et al. 2007) that the PPGs of these two taxa are homologous. Comparison of physiological aspects of the PPGs of ants and beewolves, as well as the investigation of other aculeate Hymenoptera is necessary to obtain further insights into the evolution and function of the PPG in this group of insects.

Dev_Genes_Evol-4-1-2292473

Analysis of the Tribolium homeotic complex: insights into mechanisms constraining insect Hox clusters

The remarkable conservation of Hox clusters is an accepted but little understood principle of biology. Some organizational constraints have been identified for vertebrate Hox clusters, but most of these are thought to be recent innovations that may not apply to other organisms. Ironically, many model organisms have disrupted Hox clusters and may not be well-suited for studies of structural constraints. In contrast, the red flour beetle, Tribolium castaneum, which has a long history in Hox gene research, is thought to have a more ancestral-type Hox cluster organization. Here, we demonstrate that the Tribolium homeotic complex (HOMC) is indeed intact, with the individual Hox genes in the expected colinear arrangement and transcribed from the same strand. There is no evidence that the cluster has been invaded by non-Hox protein-coding genes, although expressed sequence tag and genome tiling data suggest that noncoding transcripts are prevalent. Finally, our analysis of several mutations affecting the Tribolium HOMC suggests that intermingling of enhancer elements with neighboring transcription units may constrain the structure of at least one region of the Tribolium cluster. This work lays a foundation for future studies of the Tribolium HOMC that may provide insights into the reasons for Hox cluster conservation. Introduction Hox clusters arose near the origins of the animal kingdom (Larroux et al. 2007; Ryan et al. 2007). The last common ancestor of the protostomes and deuterostomes is thought to have had a cluster of at least seven genes characterized by a common transcriptional orientation and by colinearity in the order of the genes and their expression domains along the anterior–posterior axis (reviewed in Garcia-Fernandez 2005). In various metazoan lineages, Hox clusters have gained or lost genes by duplication and deletion but often have maintained their chromosomal order, transcriptional orientation, and both spatial and temporal colinearity of expression patterns (reviewed in Ferrier and Minguillon 2003), suggesting that Hox cluster organization has been subject to strong constraints during evolution. Classical model systems, such as Drosophila and Caenorhabditis elegans, have provided many important insights into the developmental functions of Hox genes but do not provide particularly good examples of Hox cluster conservation. The Hox cluster of Drosophila melanogaster is split into two parts (the Antennapedia (ANTC) and bithorax (BXC) complexes), shows changes in transcriptional orientation of some genes, and includes interspersed genes of independent origin as well as Hox-derived genes that have evolved novel developmental roles (reviewed in Ferrier and Minguillon 2003). These alterations suggest that the constraints keeping the Hox cluster intact may have been lost in the lineage leading to Drosophila. Additional Hox cluster rearrangements (breaks, microinversions, and gene transpositions) have been found in other Drosophila species (Negre et al. 2003; Negre and Ruiz 2007; Von Allmen et al. 1996) as well as in the silk moth Bombyx mori (Yasukochi et al. 2004). The Hox genes of C. elegans (reviewed in Aboobaker and Blaxter 2003) and the tunicate Oikopleura dioica (Seo et al. 2004) have undergone even more extreme loss and rearrangement such that none of their remaining Hox genes are clustered. In most cases, the Hox genes of these organisms still show spatial but not temporal colinearity. Rapid development seems to be the common denominator among most of these organisms, perhaps making temporal colinearity of Hox genes unnecessary, or even undesirable (Ferrier and Holland 2002; Ferrier and Minguillon 2003; Negre et al. 2005). While studies of disrupted Hox clusters have provided some insights into Hox cluster maintenance, a more complete understanding will require analysis of organisms where they are still intact. Studies of vertebrate Hox clusters have uncovered several potential mechanisms that may promote temporal colinearity and therefore constrain the organization of these clusters (reviewed in Kmita and Duboule 2003). These include progressive changes in chromatin state along the length of the cluster, varying affinity of regulatory elements to a gradient of signal, and the presence of global enhancer elements outside the cluster that regulate multiple genes within the cluster. However, it is not clear whether these mechanisms apply to other organisms. Duboule (2007) has suggested that the modern vertebrate Hox clusters are actually more organized than the ancestral cluster. Some of the mechanisms constraining the organization of vertebrate Hox clusters likely evolved concomitant with the co-option of Hox genes for functions such as limb development (Duboule 2007; Kmita and Duboule 2003) and, therefore, may not be applicable to other lineages. Based on this model, we might expect to gain a better understanding of the ancestral constraints on Hox clusters by studying a less organized but still intact cluster. Such clusters have been described in organisms as diverse as the cephalochordate amphioxus (Garcia-Fernandez and Holland 1994; Minguillon et al. 2005), sea urchins (Cameron et al. 2006), and the insects Apis (Honey Bee Genome Sequencing Consortium 2006; Dearden et al. 2006) and Anopheles (Holt et al. 2002; Negre and Ruiz 2007). Evidence also suggests that the red flour beetle, Tribolium castaneum, has an intact Hox cluster. Conventional cloning and sequencing of the portion of the cluster corresponding to the Drosophila Antennapedia complex has shown that this region of the homeotic complex (HOMC) is intact in Tribolium (Brown et al. 2002). Genetic mapping also suggests that the integrity of the Tribolium Hox cluster has been maintained (Beeman 1987). Moreover, the genetic methodologies possible with Tribolium, as well as the application of RNAi, have provided a comprehensive description of the full repertoire of Hox genes and their functions (e.g., Beeman et al. 1993; Beeman et al. 1989; Brown et al. 2000; Shippy et al. 2000; Shippy et al. 2006; Stuart et al. 1991; Stuart et al. 1993; Tomoyasu et al. 2005). The Tribolium genome has recently been sequenced, giving us the opportunity to explore the structure of its Hox cluster in detail. Here, we present an analysis of several Hox mutations along with the transcriptional profile of the cluster during embryonic development. We discuss these results with respect to potential mechanisms of Hox cluster organization and constraint. Materials and methods Sequence and transposable element analysis Sequence analysis was performed using Vector NTI Advance 10 (Invitrogen). Basic Local Alignment Search Tools (BLASTs) against Tribolium genome sequence (Tcas_2.0) were performed at http://www.hgsc.bcm.tmc.edu/blast/blast.cgi?organism= Tcastaneum or http://www.ncbi.nlm.nih.gov/genome/seq/BlastGen/BlastGen.cgi?taxid= 7070, and subsequent analysis was performed using Genboree (http://www.genboree.org/java-bin/login.jsp) or NCBI Map Viewer (http://www.ncbi.nlm.nih.gov/mapview/). The entire HOMC sequence was used as a BLASTn query against a collection of expressed sequence tags (ESTs) provided by Dr. Yoonseong Park (Department of Entomology, Kansas State University, Manhattan, KS, USA). Transposable elements were identified and classified using Censor to search the arthropod subset of Repbase (Kohany et al. 2006). Array design and probe synthesis Sequence for the Tribolium HOMC was taken from the Tcas_2.0 Baylor HSGC assembly. The tiled region consists of ∼810,000 bases from LG2 (2,290,000 to 301,000) stretching between the two non-Hox genes flanking the complex. NimbelGen designed ∼50 mer oligos covering this region at two densities: (1) one feature per 91 bp and (2) one feature per 70 bp. An additional region spanning 5 kb on either side of the putative homolog of dme-miR-iab-4 was tiled at a higher density of one feature every 10 bp. Visualization and scaling of tiling data was performed using Integrated Genome Browser (Affymetrix, http://www.affymetrix.com/support/developer/tools/download_igb.affx). Tribolium 0–72-h-old embryos, grown at 30°C in standard media, were collected by sieving, dechorionated for 2 min in 100% bleach, and homogenized in 200 μl of Trizol using a teflon pestle. Total RNA was then extracted using the standard Trizol protocol (Invitrogen). dsDNA was prepared from ∼10 μg of total RNA with random hexamers according to Kapranov et al. (2002), with the following modifications. Primers were annealed using a 20-min ramp to 15°C, and the first strand reaction was not subdivided for second strand synthesis. The resulting cDNA was used as template by NimbelGen for labeling and hybridization (Squazzo et al. 2006). Fluorescent in situ hybridization Probe labeling, embryo fixation, and RNA FISH were performed according to Kosman et al. (2004) with the following modifications. Tribolium embryos were dechorionated in 100% bleach for 2 min and agitated for 45 min to 1 h. Embryos were devitellinized by alternating 1 min vortexing with 1 min shaking for 5 min after the addition of cold methanol, followed by passage through an 18-gauge syringe three to five times. Primers used for making the TcNC-1 and iab-4 probes are as follows: TcNC5′:AGATAAGATATAATGAGGTGTAGAGTTG, TcNC3′: TGATTAACATGGACGGCTTCATTAG, iab-45′: CATCCTATGCACATGCGTTC, iab-43′: CGTTTTAATGGGTGCATCGT. Dig-labeled RNA probes were detected using sheepαDIG (Roche) primary and donkeyαsheep Alexa Fluor 555 (Molecular Probes) secondary antibodies. Genetics Beetles were cultured at 30°C on whole wheat flour supplemented with 5% brewer’s yeast as described by Beeman et al. (1989). Strains used were: Ga-1 and Ga-2 (wild type); mxpDch-3/Ey; Cx6/AEs; ptlKT76/+; Cx61/AEs; ptlD60/Ey and Dfd1/AEs. Eyeless (Ey; Beeman et al. 1996) and AbdominalExtra sclerite (AEs; Beeman et al. 1989) are dominantly marked balancer chromosomes that suppress crossing over within the HOMC. Cuticle preparations were performed as described by Shippy et al. (2000). For documentation, cuticles were placed in 9:1 lactic acid/ethanol on a depression slide and covered with a coverslip. Images were captured at several focal planes using a Nikon DXM1200F digital camera and combined into a single image using Auto-Montage software (Syncrosopy). RNAi Parental RNAi for ptl/Tc-Antp was performed by injection of dsRNA into the abdomens of female pupae. Eggs were collected from injected females at 3-day intervals, aged to hatching, and subjected to cuticle preparation. Analysis of ptlD60 breakpoints Eggs were collected after overnight incubation at 30°C and allowed to develop for 3 days. Genomic DNA was isolated from individual ptlD60 homozygous larvae as described by Gloor et al. (1993). Polymerase chain reaction (PCR) surveys of the HOMC were used to identify likely breakpoint positions, and fragments spanning these putative breakpoints were amplified using universal PCR (Beeman and Stauth 1997; Sarkar et al. 1993). PCR products were cloned and sequenced at the Kansas State University DNA Sequencing Center. The resulting sequences were compared to the Tribolium genome sequence to characterize the breakpoints. GenBank accession numbers for these sequences are as follows: ptlD60 A (EF591668) and ptlD60 B (EF591669). Analysis of mxpDch-3 breakpoints Tribolium genomic DNA was isolated from mxpDch-3/Ey, AEs/Ey, and Ga-1 pupae as described by Brown et al. (1990), with the exception that DNA was not purified on a CsCl gradient. Digested DNAs were separated on a 0.7% agarose gel by field inversion gel electrophoresis and transferred to GeneScreen nylon membrane (NEN Life Sciences). To look for restriction fragment length polymorphisms associated with mxpDch-3, the blot was probed with pBmxp2.1, a 5.2 kb HindIII fragment containing the 5′ end of the mxp/Tc-pb coding region. Inverse PCR (Ochman et al. 1988) was used to clone breakpoints associated with mxpDch-3. mxpDch-3/Ey genomic DNA was digested with a restriction enzyme (EcoRI, HindIII, and RsaI for breakpoint fragments A, B, and C, respectively). After circularization of the fragments, two rounds of PCR were performed with primers designed from known sequence. Resulting fragments were cloned using the TOPO-TA Cloning Kit (Invitrogen), sequenced, and submitted to GenBank under the following accession numbers: Dch3 A (EF591670), Dch3 B (EF591671), Dch3 C (EF591672). Sequences were compared to the Tribolium genome sequence to determine the location of breakpoints. Analysis of ptlKT76 transposon insertion site The ptlKT76 piggyBac-insertion site was amplified by vectorette PCR as described by Lorenzen et al. (2007). The resulting product was sequenced by Elim Biopharmaceuticals, Inc. (Hayward, CA, USA) and the sequence was submitted to GenBank as accession number EU056827. Results The Tribolium Hox cluster has retained an ancestral organization Several bacterial artificial chromosome (BAC) clones encompassing the ANTC-like region of the Tribolium Hox cluster were previously sequenced and annotated (Brown et al. 2002). Using the newly assembled Tribolium genome sequence, we have performed a similar analysis of the BXC-like portion of the cluster and find that this region contains the Tribolium orthologs of Ultrabithorax (Ubx; Bennett et al. 1999), abdominal-A (abd-A; Shippy et al. 1998) and Abdominal-B (Abd-B). As in Drosophila, the transcription units in this part of the complex are larger than those of the ANTC-like portion due to the presence of longer introns. As expected from previous molecular and genetic studies (Beeman 1987; Brown et al. 2002), all of the Tribolium Hox genes map to a single cluster on LG2. This cluster spans approximately 756 Kb within a single scaffold of the assembled genome sequence. A few small sequencing gaps are present in the assembly, but more than half can be filled by other available sequences (i.e., the three BAC clones previously sequenced for the ANTC-like portion of the cluster and four BACs from the BXC-like region sequenced for verification of the shotgun genome assembly; Tribolium Genome Consortium 2008). The total length of the filled gaps is approximately 2,635 bp (mismatches in the sequence flanking the gaps lead to some ambiguity), which is only slightly longer than the estimated total length of these gaps (1,938 bp). Thus, estimation of sequencing gaps in the HOMC region appears to be quite accurate. Two of these gaps are immediately adjacent to transposable element insertion sites and may result from difficulties in assembling repetitive DNA. These two sites account for about 1,300 bp of the total gaps in the HOMC. In two other cases, gaps in the genome assembly are associated with tandem duplications that are not present in the BAC assemblies: an approximately 160-bp duplication between Tc-Deformed (Tc-Dfd) and Tc-zen1 and an approximately 8.5-kb duplication between prothoraxless/Tc-Antennapedia (ptl/Tc-Antp) and Tc-fushi tarazu (Tc-ftz). We designed primers to amplify across the regions in question using Ga-2 genomic DNA (the same inbred strain that was used for the Tribolium genome sequence). In both cases, the size of the resulting fragment is consistent with that predicted from the BAC sequence (data not shown), suggesting that these gaps and duplications are artifacts of the genome assembly process. It is important to note that these artifacts affect only a small fraction of the HOMC sequence, but they underscore the increased quality of finished versus draft sequences. The single Tribolium Hox cluster contains orthologs of all eight Drosophila Hox genes, as well as orthologs of the Hox-derived genes, fushi tarazu and zerknüllt (zen; Tribolium Genome Consortium 2008 and Fig. 1). (In the case of zen, Tribolium has two paralogs apparently resulting from a recent duplication in the beetle lineage, independent of the zen duplication that occurred in the Drosophila lineage (Brown et al. 2002)). These genes are arranged in the same order on the chromosome as their counterparts in other insects. As in Apis (Dearden et al. 2006; Negre and Ruiz 2007), but in contrast to Drosophila and Anopheles (Negre and Ruiz 2007), the Hox and Hox-derived genes in the Tribolium cluster are all oriented in the same direction (Fig. 1). In addition, the two miRNA genes (miR-10 and miR-iab-4) that have been described in other insect Hox clusters are found at conserved positions in the Tribolium HOMC (Tanzer et al. 2005 and Fig. 1). Fig. 1Embryonic transcription across the complete Tribolium Hox complex. The tiling array consists of ∼50,000 50 bp probes that estimate degree of transcription. Relative intensities for each probe are represented as peaks correlated with a consensus annotation of the Tribolium Hox complex (below). Peak height, shown as Percentile Probe Intensity (PPI), corresponds to the level of transcription for a particular probe. The nucleotide position for each segment is displayed in the upper left and right corners of the panel (numbers correspond with linkage group 2, release Tcas_2.0). New ESTs (cyan) are displayed in the annotation track along with transposable elements (gray). For annotated genes and ESTs, the arrow indicates the direction of transcription. Red arrows indicate the location of two RNA-FISH probes The ANTC and BXC clusters of Drosophila melanogaster contain a number of non-Hox, protein-coding genes. In contrast, there is no evidence for non-Hox, protein-coding genes in the Tribolium HOMC (The Tribolium Genome Consortium 2008). Here, we corroborate those findings by using several methods to address whether unrelated genes might be interspersed among the Tribolium Hox genes. First, we searched the Tribolium genome for orthologs of genes that are located within the D. melanogaster clusters and determined that none of these genes are located within the Tribolium HOMC. Second, we analyzed predicted proteins within the region to determine whether any of them have recognizable orthologs in other species. Other than the Hox and Hox-derived genes, we found no evolutionarily conserved proteins among either the GLEAN predictions or the GNOMON ab initio predictions that map to the Hox cluster. Third, we searched a collection of Tribolium ESTs for expressed sequences within the HOMC. By this approach, we identified three non-Hox EST clusters that appear to represent noncoding transcripts as well as evidence for a mariner transposase gene (see below), but no other protein-coding genes were found. Finally, we analyzed the embryonically transcribed sequences identified by a tiling array to determine if any were likely to encode proteins. Again, we found no evidence of non-Hox-related protein-coding genes other than those within transposable elements. Although there are caveats to these analyses (e.g., gene prediction methods are imperfect, the tiling array represents only the embryonic transcriptome and EST coverage is incomplete), our results strongly suggest that the protein-coding genes in the Tribolium Hox complex (excluding genes within transposable elements) are all either Hox or Hox-derived genes. Comparison of transposable element density in the Hox complexes of various animals has led to the suggestion that higher abundance of transposable elements in Hox clusters is correlated with loss of structural integrity. Mammalian Hox complexes have a reduced number of transposons compared to other regions of the genome (Ferrier and Minguillon 2003). Moreover, when transposons are present, they seem to be preferentially inserted into nontranscribed regions of the clusters (Mainguy et al. 2007). In contrast, transposons occur fairly frequently in the split Drosophila clusters (Fried et al. 2004). Though the prediction of three transposable elements in the Tribolium Hox complex (Fig. 1) may be an underestimate, the same method predicts fivefold more in the Drosophila Hox clusters. Additionally, only three transposable elements (all mariners) have been found in the larger but intact Apis Hox complex (Dearden et al. 2006). These numbers are consistent with the apparent inverse correlation between transposon number and the level of Hox cluster organization. Taken together, these observations suggest that with respect to gene content, order, and orientation, the Tribolium Hox cluster closely resembles the putative ancestral Hox cluster (Garcia-Fernandez 2005). Thus, the constraints preserving the integrity of the Hox cluster may still be in force in Tribolium. To determine whether these constraints extend outside the Tribolium Hox cluster, we examined synteny beyond the cluster itself. As previously described, Tc-chaoptic partially overlaps the 3′ UTR of Tc-labial (Tc-lab) on the opposite strand (Nie et al. 2001). Working outward, the first gene on the same strand as the Hox genes is Tc_00927, a dolichyl glycosyltransferase orthologous to D. melanogaster CG4542. Beyond the Tc-Abd-B locus is a cluster of putative serine carboxypeptidase genes (Tc_00887, Tc_00664, Tc_00665, and Tc_00666) and the ortholog of D. melanogaster CG3909 (Tc_00886). We identified the orthologs of these genes in D. melanogaster, Anopheles gambiae, and Apis mellifera and determined their map positions. None of these genes map near the Hox cluster in any of the other insects. Likewise, orthologs of the genes adjacent to lab and Abd-B in D. melanogaster do not map near the Hox clusters of the other three insects. These results suggest that the constraints preserving the Hox cluster act only on the Hox genes themselves and not the surrounding region. The Tribolium Hox cluster produces numerous noncoding transcripts We developed a tiling microarray covering the Tribolium Hox complex to identify the transcription units active during a broad window of Tribolium embryonic development. Tiling array signal intensity profiles were compared with previously described Hox cDNA structures. Though the tiling density is not fine-scaled enough to effectively resolve intron–exon boundaries, there is a near-perfect correlation between tiling array-predicted transcription and the position of exons in the well-characterized Hox genes. The only caveat is that the 5′ exons of maxillopedia/Tc-proboscipedia (mxp/Tc-pb) and Tc-Abd-B exhibit weaker signals than the other exons of these genes. The most likely explanation is that the 5′ exon is present only in a small subset of the transcripts derived from the gene (i.e., a minor spliceoform). During the first 3 days of development, numerous regions of the Hox complex, including intergenic and intronic noncoding regions, are actively transcribed (Fig. 1). Interestingly, neither of the two most likely noncoding candidates, the previously described miRNAs, is robustly identified on the tiling array. Transcription at the tca-miR-10 locus is not detected, and transcription at the tca-miR-iab-4 locus is weak. It may be that tca-miR-10 is not expressed during the stages examined, whereas in situ hybridization assays show that tca-miR-iab-4 is strongly expressed during part of the developmental window examined (Fig. 2b). Fig. 2Expression pattern of two HOMC noncoding transcripts in Tribolium embryos. Probe positions are shown in Fig. 1. a Expression pattern from a 1-kb probe located ∼86 kb 5′ of the start of ptl/Tc-Antp. b The expression pattern of the Tribolium homolog of the iab-4 miRNA (tca-miR-iab-4) The most intense hybridization signals are detected in the central 250 kb of the Hox complex, encompassing the ptl/Tc-Antp and Ultrathorax/Tc-Ultrabithorax (Utx/Tc-Ubx) genes. Strikingly, transcription in this region is almost equally intense for coding and noncoding loci, and for both the introns and exons of the protein-coding genes. There are hundreds of discrete regions (500 bp or longer) where signal intensity is many times greater than for verified Hox gene exons. It is not possible to determine from the single time point we have analyzed if any of these discrete regions are part of larger transcripts. To verify that the observed signals in the tiling array represent authentic transcription, RNA fluorescent in situ hybridization (FISH) was performed with a representative 1-kb region between ptl/Tc-Antp and Utx/Tc-Ubx (TcNC-1 in Fig. 1). This region is expressed in a Hox-like pattern with distinct anterior and posterior borders in the posterior region of the elongating germ band (Fig. 2a). Interestingly, signal is detected primarily in two spots per nucleus, presumably at the sites of nascent transcription. This suggests either rapid degradation or processing of a primary transcript as would be seen for a pri-miRNA or an intron. In our search for additional genes within the Tribolium Hox cluster, we identified three ESTs that appear to represent noncoding transcripts. One seems to be a chimeric artifact, arising from the fusion of a tca-miR-10 precursor and part of a 28s rRNA gene. The second, represented by two independent cDNAs, maps between ptl/Tc-Antp and Utx/Tc-Ubx while the third is located within the first intron of Utx/Tc-Ubx. The last two are transcribed from the strand opposite the Hox genes and are correlated with regions of strong signal in the tiling array analysis. The mxpDch-3 mutation affects regulation of both mxp/Tc-pb and Cx/Tc-Scr Because complex regulatory regions may act as a constraining force keeping Hox clusters intact, we analyzed the mxpDch-3 mutation, which was shown to have unusual effects on the expression of mxp/Tc-pb (Shippy et al. 2000); mxpDch-3 homozygotes lack most, if not all, normal mxp/Tc-pb expression, but both heterozygotes and homozygotes display strong ectopic mxp/Tc-pb expression in a pattern reminiscent of Cephalothorax/Tc-Sex combs reduced (Cx/Tc-Scr) expression (albeit with an apparent posterior shift of some domains). This ectopic expression is sufficient to rescue some aspects of mxp/Tc-pb function so mxpDch-3 is not an mxp/Tc-pb null (Shippy et al. 2000). Interestingly, we find that mxpDch-3 fails to complement a null allele of Cx/Tc-Scr (and, in fact, appears to be null for Cx/Tc-Scr) but complements null alleles of Tc-Dfd and ptl/Tc-Antp (data not shown). To better understand this complex mutation, we characterized the breakpoints associated with mxpDch-3. mxpDch-3 is associated with a chromosomal rearrangement involving at least four breakpoints. Although we have not ruled out the presence of additional breakpoints, the simplest interpretation of our data is that a fragment of the HOMC (including Tc-zen1, Tc-zen2, Tc-Dfd, Cx/Tc-Scr, and Tc-ftz) has been removed from the HOMC and inserted between a fragment of LG9 and a non-HOMC fragment of LG2 (Fig. 3a). This scenario is consistent with previously reported pseudo-linkage of LG2 and LG9 associated with mxpDch-3 (Beeman et al. 1996) and provides an explanation for the mid-embryonic lethality of the mxpDch-3 homozygotes (Shippy et al. 2000). That is, non-HOMC breakpoints interrupt both the TFIIA-L ortholog, which is a component of the transcriptional machinery (Yokomori et al. 1993), and a homolog of groucho, which is an important transcriptional corepressor in Drosophila (Jimenez et al. 1997; Paroush et al. 1994). We conclude that the breakpoint between Tc-ftz and ptl/Tc-Antp is likely to account for the loss of Cx/Tc-Scr function associated with mxpDch-3, probably by separating the Cx/Tc-Scr transcription unit from some or all of its regulatory units. The rearrangement probably juxtaposes these regulatory elements with the mxp/Tc-pb transcription unit, providing a likely explanation for the Cx/Tc-Scr-like expression of mxp/Tc-pb in mxpDch-3 embryos. Fig. 3Rearrangements of the HOMC. In these schematic diagrams, the positions of cloned breakpoint fragments are underlined. Wild-type chromosomal position (not to scale) on LG2 (purple) is indicated by a gradient of color to illustrate the effects of inversions. a In the mxpDch-3 rearrangement, an approximately 150-kb fragment of the HOMC has been transposed between fragments of LG9 and LG2. b The ptlD60 mutation is a large inversion that splits the Hox cluster into two parts. Small fragments at each end of the inversion appear to have been deleted, including part of the ptl/Tc-Antp locus Cx/Tc-Scr regulatory elements map near or within ptl/Tc-Antp Beeman et al. (1993) observed that mutations in Cx/Tc-Scr and ptl/Tc-Antp often partially fail to complement one another. Because this is precisely the type of genetic interaction that would be predicted if the separation of Hox genes has deleterious effects, we decided to analyze ptlD60, an allele of ptl/Tc-Antp that shows such effects. The ptlD60 mutation, which results in transformation of the larval legs toward antennae as well as reductions of some labial and thoracic tissue, has been proposed to be a null allele of ptl/Tc-Antp (Beeman et al. 1993). However, adults transheterozygous for ptlD60/ptlD2 have a very different phenotype from that produced by larval ptl/Tc-Antp RNAi (Tomoyasu et al. 2005) and, instead, resemble adults in which both ptl/Tc-Antp and Cx/Tc-Scr have been knocked down (Tomoyasu, personal communication). This result raises the possibility that the ptlD60 mutation affects the function of both genes. To address this issue, we performed parental RNAi with ptl/Tc-Antp and found that the resulting larvae (Fig. 4b) show a phenotype almost identical to that of ptlD60 homozygotes (Fig. 4c), suggesting that the ptlD60 mutation primarily affects the function of ptl/Tc-Antp during embryonic development. However, there is a subtle difference between the ptlD60 and ptl/Tc-Antp RNAi phenotypes in the positioning of the T1 appendages, which are located near the ventral midline in the RNAi larvae but more laterally in the mutants. This difference is likely attributable to partial loss of Cx/Tc-Scr function in ptlD60 because Cx/Scr is responsible for the midline position of the labial appendages in Tribolium (Shippy et al. 2006) and other insects (Hughes and Kaufman 2000; Pattatucci et al. 1991; Rogers et al. 1997). Thus, ptlD60 appears to be not only a null allele of ptl/Tc-Antp but also a hypomorphic allele of Cx/Tc-Scr. Fig. 4Cuticle and enhancer trap phenotypes of ptl/Tc-Antp mutations. The antennae (ant) and the labial (lab) and thoracic (T1–T3) segments are denoted where relevant. a–e Cuticle preps displaying the phenotypes of wild-type (Ga-1; a), ptl/Tc-Antp RNAi (b), ptlD60/ptlD60 (c), ptlKT76/ptlKT76 (d), and ptlKT76/ptlD60 (e) first instar larvae. Enhancer trap-driven EGFP expression in a ptlKT76 embryo (f) appears in a very similar pattern to Cx/Tc-Scr expression (purple) in a wild-type embryo (g). A ptlKT76 larva (h) and pupa (i) also display EGFP enhancer trap expression in parts of the labial and first thoracic segments To understand why the ptlD60 mutation affects both ptl/Tc-Antp and Cx/Tc-Scr, we characterized its mutant lesion(s). We found that ptlD60 is associated with an inversion of about 6.6 Mb, with breakpoints in ptl/Tc-Antp and a distant region of LG2 (Fig. 3b). In addition, there are small deletions at each end of the inversion (approximately 3.1 kb of the ptl/Tc-Antp transcription unit including all of exon 2 and approximately 1.5 kb at the other end). Consistent with this conclusion, we find that ptlD60 can act as a crossover suppressor for LG2, reducing recombination between Reindeer (a mutation near one end of LG2) and the HOMC from its normal value of 35–40 cM (Beeman et al. 1996) to approximately 20 cM. These results suggest that breakpoints within the ptl/Tc-Antp gene affect the function of both ptl/Tc-Antp and Cx/Tc-Scr, probably by disrupting the function of Cx/Tc-Scr regulatory elements (see “Discussion”). Additional evidence for the presence of Cx/Tc-Scr regulatory elements in the vicinity of ptl/Tc-Antp comes from a piggyBac-insertion line recovered during an insertional mutagenesis project. The KT076 line carries a homozygous lethal insertion in the last intron of ptl (Fig. 5). Crosses between KT076 heterozygotes produce a class of embryos (putative homozygotes) with the T1 and T2 legs partially transformed toward antennae (Fig. 4d), a phenotype consistent with partial loss of Ptl/Tc-Antp function. However, individuals carrying the insertion display an embryonic enhancer trap expression pattern (Fig. 4f) very similar to the expression pattern of Cx/Tc-Scr (Curtis et al. 2001; Fig. 4g), despite the fact that the insertion site is about 87 kb upstream of Cx/Tc-Scr. KT076 larvae and pupae also show weak enhancer trap patterns consistent with predicted Cx/Tc-Scr domains (Fig. 4h–i). As expected from the phenotype of homozygotes, KT076 fails to complement ptlD60 as assayed by adult viability, and crosses of KT076 to ptlD60 heterozygotes produce embryos with a phenotype similar to, but slightly weaker than, that of ptlD60 homozygotes (Fig. 4e). In contrast, KT076 fully complements both the embryonic phenotype and the adult viability of Cx61, a null allele of Cx/Tc-Scr (Shippy et al. 2006), indicating that Cx/Tc-Scr function is not compromised by the insertion (data not shown). These results suggest that KT076 is a hypomorphic allele of ptl/Tc-Antp, and it will, hereafter, be referred to as ptlKT76. Together with the data from the mxpDch-3 and ptlD60 mutations, the Cx/Tc-Scr-like enhancer trap phenotype of ptlKT76 provides strong evidence that Cx/Tc-Scr regulatory elements are located near, and probably within, ptl/Tc-Antp. Although additional experiments will be necessary to pinpoint the location of regulatory elements and verify this conclusion, it is intriguing to think that overlap of regulatory elements of one Hox gene with the transcription unit of another Hox gene might be an important mechanism of Hox cluster constraint. Fig. 5Overlap of Cx/Tc-Scr regulatory elements with the ptl/Tc-Antp locus. The gene structure of ptl/Tc-Antp (coding sequence is shaded gray) and the positions of mutant lesions are shown in the diagram. The inferred positions of Cx/Tc-Scr regulatory elements in the ptl/Tc-Antp region are indicated below the diagram Discussion The Hox clusters of several insects have now been completely sequenced. While breaks in the cluster seem to have occurred several times in the Drosophila lineage, the clusters of Apis, Anopheles, and Tribolium are intact. This suggests that many insect clusters are still subject to constraints to maintain their organization. Of the insects with intact clusters, Tribolium is, by far, the most genetically tractable and has a strong history of Hox gene studies, thus offering the best system for understanding the constraints acting on an intact insect Hox cluster. Below, we discuss insights provided by our analysis of the Tribolium HOMC into mechanisms that might be responsible for Hox cluster integrity. Temporal colinearity Among organisms for which both Hox cluster sequence and expression data are available, the presence of an intact cluster appears to be correlated with temporal colinearity of Hox gene expression, while disrupted clusters are associated with lack of temporal colinearity (Monteiro and Ferrier 2006). This observation has pushed temporal colinearity to the forefront of discussions about Hox cluster maintenance. However, several questions remain to be answered. Does temporal colinearity really require an intact Hox cluster? Is temporal colinearity required for proper Hox gene function in organisms with intact clusters? If temporal colinearity is a key constraint on Tribolium Hox cluster integrity, we might expect rearrangements to affect the function of most or all Hox genes. However, our analysis of Tribolium Hox cluster mutations provides no evidence for such global effects. The ptlD60 inversion splits the complex into two parts but all of the Hox genes except ptl/Tc-Antp and Cx/Tc-Scr appear to function normally. In addition, the mxpDch-3 rearrangement results in the translocation of several HOMC genes (Tc-zen, Tc-Dfd, Cx/Tc-Scr, and Tc-ftz) to a new chromosomal location. At least one of these genes, Tc-Dfd, is functional because mxpDch-3 fully complements a Tc-Dfd null allele. Likewise, the genes remaining in the HOMC (with the exception of mxp/Tc-pb) apparently function normally. These limited effects of Hox cluster rearrangements are similar to what has been reported for Drosophila (e.g., Abbott and Kaufman 1986; Pultz et al 1988). Although additional experiments will be required to determine whether Tribolium Hox genes exhibit temporal colinearity, our results suggest that constraints on the Tribolium HOMC are more likely to act locally. Regulatory elements The reported breaks and transposition sites in the Hox clusters of Drosophila species are all located in intergenic regions near the 3′ end of a gene (Negre et al. 2005; Negre and Ruiz 2007) and, thus, are presumably less likely to separate a gene from its regulatory elements, which are predominantly located 5′ of each Drosophila Hox gene. Interestingly, the one exception to this rule is the region between abd-A and Abd-B, which contains regulatory regions for both genes and is not split in any of the Drosophila species examined so far (Negre and Ruiz 2007). These observations led to the conclusion that Drosophila Hox clusters have a modular organization (with each gene and its regulatory elements representing a separate module) and that Drosophila Hox genes are still partially clustered simply because the regulatory regions are so large that there are relatively few positions where breaks can occur without disturbing a module. That is, most of the remaining linkage in Drosophila Hox clusters (with the possible exception of abd-A and Abd-B) is due to “phylogenetic inertia,” and, given enough time, the clusters will completely disperse (Negre and Ruiz 2007). The question naturally arises whether phylogenetic inertia could also be the reason for the intact Hox clusters of insects like Tribolium. Lewis et al. (2003) suggested that unusual features of recombination (Ranz et al. 2001) may make drosophilids more tolerant of Hox cluster rearrangements than are most insects. If this is the case, intact clusters may just be the consequence of slower rates of chromosomal rearrangement (Negre and Ruiz 2007). Alternatively, constraints on Hox cluster maintenance may still be functional in Tribolium. Our analysis of mutant breakpoints in the ptl/Tc-Antp region indicates that, in at least one case, Tribolium Hox genes are not modular. That is, at least some of the regulatory elements controlling Cx/Tc-Scr expression are apparently located within the pt/Tc-Antp gene (Fig. 5). Most embryonic enhancers of Cx/Tc-Scr are predicted to lie between the mxpDch-3 and ptlD60 breakpoints because the mxpDch-3 allele appears to lack all Cx/Tc-Scr function, while ptlD60 has almost normal embryonic Cx/Tc-Scr function. This conclusion is further supported by the expression of mxp/Tc-pb in a Cx/Tc-Scr-like pattern in mxpDch-3 mutants, presumably due to juxtaposition of the mxp/Tc-pb transcription unit with sequence between Tc-ftz and ptl/Tc-Antp. Adult Cx/Tc-Scr regulatory elements are likely located within, or 5′ of, ptl/Tc-Antp because the ptlD60 rearrangement seems to have a stronger effect on the adult functions of Cx/Tc-Scr. The presence of Cx/Tc-Scr regulatory elements near, or within, ptl/Tc-Antp is supported by the observation that a piggyBac insertion within the ptl/Tc-Antp transcription unit (ptlK76) shows an enhancer trap phenotype that appears to be driven by Cx/Tc-Scr regulatory elements. In Drosophila, elements which drive Scr-like expression patterns have been found in the region between ftz and Antp (Gorman and Kaufman 1995). However, these elements are apparently redundant because breakpoints in this region only slightly reduce Scr expression levels within its normal domain. Moreover, a deficiency which removes most, if not all, of the Antp transcription unit has no effect on embryonic Scr expression. Thus, it is possible that the modularity of the Drosophila Hox cluster is a recent innovation resulting from changes in the position of cis-regulatory elements. This newly acquired modularity might have allowed breaks in the Hox cluster in the Drosophila lineage. In contrast, overlap of regulatory elements with neighboring genes might still act as a constraint on the integrity of the Hox cluster in Tribolium. Noncoding transcripts Although noncoding transcripts within Hox clusters have been recognized for many years (Cumberledge et al. 1990; Lipshitz et al. 1987; Sanchez-Herrero and Akam 1989), there has recently been renewed interest in these enigmatic RNAs. At least two of these ncRNAs, miR-196 and miR-iab-4, can have homeotic function and attenuate the actions of protein-coding Hox genes (Hornstein et al. 2005; Ronshaugen et al. 2005). In addition to the miRNAs, numerous long noncoding RNAs have been identified within the Hox clusters of both flies and mammals (Bae et al. 2002; Rinn et al. 2007). ncRNAs are implicated in a vast array of processes, including regulation of transcription, translation, epigenetic control of chromatin, mono-allelic expression, dosage compensation, and silencing (reviewed in Mattick and Makunin 2006). In the Drosophila BXC, these transcripts have been implicated in the control of Hox gene expression, although there is some controversy as to whether they promote (Sanchez-Elsner et al. 2006) or repress (Petruk et al. 2006) Hox gene transcription. Mainguy et al. (2007) found evidence for extensive noncoding transcription in the mammalian Hox clusters and proposed that polycistronic and antisense transcription might play a role in keeping Hox genes clustered. Our transcriptional profiling data demonstrates that Tribolium also shows considerable noncoding transcription in the Hox complex. While the tiling array has provided a revealing snapshot of transcription levels during embryonic development, much additional work will be necessary to characterize the actual transcripts. For example, the high levels of transcription in the ptl/Tc-Antp and Utx/Tc-Ubx regions could be the result of several individual transcripts or one long transcript. Interestingly, dicistronic transcripts spanning the Antp and Ubx orthologs have been reported in crustaceans (Shiga et al. 2006) and centipedes (Brena et al. 2006). If such transcripts have an important function, they might constrain linkage in at least some parts of the Hox cluster. The data available thus far indicate that noncoding transcripts are prevalent in both intact and broken Hox clusters. However, it is not clear whether these transcripts perform the same functions in all organisms. Perhaps, noncoding RNAs play a different or more critical role in organisms with intact clusters. Future studies in this area are likely to provide important insights into Hox gene function and possibly into Hox cluster conservation. Conclusions The results presented here provide a foundation for further studies of the constraints acting on Hox clusters. While it is important to keep in mind that multiple factors may have contributed to the maintenance of Hox clusters during evolution (Kmita and Duboule, 2003), the intact structure of the Tribolium Hox cluster and the suite of tools now available for this insect makes it an ideal candidate for such research.

Knee_Surg_Sports_Traumatol_Arthrosc-3-1-2082657

Description of the attachment geometry of the anteromedial and posterolateral bundles of the ACL from arthroscopic perspective for anatomical tunnel placement

The anterior cruciate ligament (ACL) consists of an anteromedial bundle (AMB) and a posterolateral bundle (PLB). A reconstruction restoring the functional two-bundled nature should be able to approximate normal ACL function better than the most commonly used single-bundle reconstructions. Accurate tunnel positioning is important, but difficult. The purpose of this study was to provide a geometric description of the centre of the attachments relative to arthroscopically visible landmarks. The AMB and PLB attachment sites in 35 dissected cadaver knees were measured with a 3D system, as were anatomical landmarks of femur and tibia. At the femur, the mean ACL centre is positioned 7.9 ± 1.4 mm (mean ± 1 SD) shallow, along the notch roof, from the most lateral over-the-top position at the posterior edge of the intercondylar notch and from that point 4.0 ± 1.3 mm from the notch roof, low on the surface of the lateral condyle wall. The mean AMB centre is at 7.2 ± 1.8 and 1.4 ± 1.7 mm, and the mean PLB centre at 8.8 ± 1.6 and 6.7 ± 2.0 mm. At the tibia, the mean ACL centre is positioned 5.1 ± 1.7 mm lateral of the medial tibial spine and from that point 9.8 ± 2.1 mm anterior. The mean AMB centre is at 3.0 ± 1.6 and 9.4 ± 2.2 mm, and the mean PLB centre at 7.2 ± 1.8 and 10.1 ± 2.1 mm. The ACL attachment geometry is well defined relative to arthroscopically visible landmarks with respect to the AMB and PLB. With simple guidelines for the surgeon, the attachments centres can be found during arthroscopic single-bundle or double-bundle reconstructions. Introduction The anterior cruciate ligament (ACL) consists of two functional bundles [2, 4, 16]. The anteromedial bundle (AMB) originates anteroproximal in the intercondylar notch, close to the over-the-top position at the posterior edge of the notch, from the deep high part of the femoral attachment area and inserts anteromedial on the anterior intercondylar area of the tibia. The posterolateral bundle (PLB) originates more posteriorly and distally in the notch, from the shallow low part of the femoral attachment area and inserts posterolateral on the anterior intercondylar area of the tibia. The ACL reconstruction aims at restoring normal knee function. Most ACL replacements are performed with the isometric single-bundle technique. Isometric positioning of a single-bundle graft results in replacement of the AMB only. Although tensioned over the complete range of motion, the fibres are mostly tight in flexion. The AMB is the major constraint for anterior tibial displacement of the flexed knee [30], but cannot restore normal knee laxity and kinematics near extension [4, 28, 41]. In an effort to improve knee mechanics, double-bundle anatomic ACL reconstructions are now developed with reconstruction of both AMB and PLB [9, 10, 13, 19, 28, 35, 40, 46, 49, 50, 52, 53]. As presented in previous studies, a reconstructed PLB is able to restore stability in knee angles where an isometrically placed graft fails [28, 41, 50]. Additional restraint against anterior displacement in 15° of flexion [48] as well as prevention of the pivot shift is demonstrated [28, 50]. Therefore, a reconstruction with two bundles should be able to approximate normal ACL function over the complete range of motion [39, 41, 51]. Tunnel positioning is an important factor for clinical success of ACL reconstructions. Incorrect tibial [23] and femoral [29, 55] tunnel placements result in abnormal knee mechanics. Anatomical placement restores normal knee function better than isometric placement [36, 55]. However, accurate tunnel placement seems difficult. Misplacement between 25 and 65% of the tibial and femoral tunnels is reported [8, 27, 47]. Double-bundle ACL reconstructions require an anatomical placement of the bone tunnels. It is difficult to identify the ACL remnants in chronic ACL-injured knees. Therefore, detailed information about the approximate native position is essential to determine proper anatomic tunnel placement for the two bundles during arthroscopy [17]. The anatomical position has been the subject of many studies [3, 5, 12, 14–16, 20, 26, 33, 34, 37, 38, 44]. Only a few recent studies have described the anatomic positions of the AM and PL bundles [11, 32, 45, 52]. Due to the two-dimensional and limited view on the arthroscopic monitor, the landmarks and descriptions used in the above-mentioned studies seem not sufficient for correct positioning of the two separate bundles in all planes during arthroscopic surgery. This study is aimed at acquiring quantitative geometric data of the ACL attachments on tibia and femur, such that these data can be used in an arthroscopically guided procedure for reconstruction of the ACL. We hypothesized that reliable guidelines to find the centres of the AMB, PLB and ACL relative to arthroscopically visible landmarks can be established. For the purpose of an anatomically accurate reconstruction of the ACL, the variations should be equal to or less than reported in other studies. As regard to the dimensions of drilled tunnel holes, normally 10 up to 12 mm, 95% (mean ± 2SD) of the attachment centres should be within this range. Therefore the a priori set assumption is that the maximally acceptable SD is 2.5 up to 3 mm. Methods Dissection Thirty-five intact human cadaveric knee joints of elderly donors preserved in formalin, without signs of gross bony deformity, previous fracture or degenerative disease and with intact knee ligaments were dissected. Because of local post-dissection handling procedures, no exact data on gender and age of the donors were available, but they were older than 60 years. The muscles and anterior capsule were removed. The ligaments were left intact in order to preserve controlled motion of the knee. The femur was fixated in a clamp, the tibia was moved, resulting in flexion and extension of the knee joint. During this repeated passive movement, the two functional components of the ACL were identified, based on a visually detectable difference in their tensioning patterns as described by Girgis et al. [16]. In 90° of flexion, an anterior load was manually applied. This caused tension in the fibres of the medial tibial attachment site, the AMB. The fibres of the PLB remained slack. This procedure enabled a separation of the two bundles at the tibial attachment from ventral. The femur was turned in the clamp to enable a posterior approach and the posterior cruciate ligament (PCL) was removed. In this position the initial division of the ACL at the tibial attachment was visible and used to extend the division towards the femoral origin. The tibia was moved towards extension. The PLB-fibres, inserting at the lateral femoral ACL attachment site, tightened, enabling to complete the separation as far as the femoral attachment. The outline contours of both AMB and PLB attachment areas were marked, with a waterproof felt pen. Anatomical position To quantify the position of the centres of the attachment sites relative to arthroscopically visible landmarks, three-dimensional (3D) measurements were made with a 3Space Fastrak electromagnetic tracking system (Polhemus Navigation Sciences, Colchester, VT, USA). The x, y and z co-ordinates of each measured point were recorded with an accuracy of 0.35 mm [42]. On the femur and tibia, the attachment sites of the AMB, PLB and the entire ACL were digitized by means of a collection of points placed at equal distances on the marked outlines. The 3D position of the centres of the ACL, AMB and PLB attachments were calculated by the geometric mean of all points on the outlines. The 3D distances between the centres of the two bundles were calculated. On both femur and tibia, an arthroscopically visible landmark was digitized that served as the origin of a local coordinate system. The absolute two-dimensional (2D) positions of the centres relative to these landmarks were calculated. The position in the third dimension was determined by the surface geometry of the femoral condyle and the intercondylar tibial area, respectively. Absolute positions can depend on different knee sizes and the dimensions of the femoral notch and the intercondylar tibial area. To correct for this, the relative centre positions were also calculated. Additional reference points near the attachment area on femur and tibia were digitized. Between these points two reference lines were defined to create a 2D coordinate system. The distances defined the dimensions of the femoral notch and intercondylar tibial area. Absolute centre positions were transformed in positions relative to the reference lines (%) within femur and tibia. To detect whether the absolute position of the attachment centres was actually influenced by knee size, statistical analyses were performed. The relation between absolute position and knee size can be demonstrated with a correlation coefficient. The Pearson’s correlations between the distances of the absolute positions and the length of the reference lines, representing knee size, were calculated. Correlations with a Pearson’s correlation coefficient r > 0.6 and a significance level P < 0.05, for a 95% alpha level were considered relevant and confirmed the relation between knee size and absolute centre position. Femur In the anatomical nomenclature the femoral ACL attachment uses anterior/posterior and proximal/distal positions, relative to the extended knee [17] (Fig. 1a). Since this study aimed at describing the positions of the AMB and PLB of the femoral attachment from an arthroscopic perspective, the arthroscopic nomenclature was used as recommended by the ‘‘ESSKA Scientific Workshops’’ in 1998 [4]. The definitions, shallow/deep and high/low, refer to the position along the wall and from the roof of the intercondylar notch in a 90° flexed knee (Fig. 1b). Fig. 1The orientation in the femoral notch, used in this study, is based on the recommendations of the ESKKA 1998 [4]. The notch depth (ND) is directed from shallow to deep in anatomic distal–proximal direction. The notch height (NH) is directed from low to high in anatomic dorsal–ventral direction. The notch wall is located at the medial side of the lateral condyle. The notch roof is the connection between the two condyles The digitized points on the femur are represented in Fig. 2. The main femoral landmark was derived from the over-the-top position, located at the posterior edge of the intercondylar notch. The most proximal high deep point (D) on the lateral condyle was found at the 10.30 o’clock position in the arc of the femoral notch, in a right knee (at 1.30 in a left knee) [17] (Fig. 2). The circumferences of the ACL attachments were digitized, as was the additional reference point, the most distal high shallow point (S). This point is positioned on the distal cartilage edge of the lateral condyle in the anterior notch outlet (Fig. 2a). The femoral coordinate system was defined with two reference lines (Fig. 3). The first-defined femoral reference line (from D to S) divided the notch roof from the notch wall and defined the length of the lateral notch depth (ND). The lowest, dorsal point (L) on the posterior joint cartilage edge was found using a line parallel to the line DS. The second reference line (from L to H) was defined from point L perpendicular to crossing point H, high in the notch at the line DS. It indicated the magnitude of the notch height (NH). After creating this coordinate system, the calculated attachment centres (C) of the ACL, AMB and PLB were projected on the line DS at point P. The absolute centre position (mm) was composed of the distance between the main femoral landmark, the high deep point D and the point P and the distance between the point P and the centre C. The relative (%) position was calculated by dividing the distance DP by DS (notch depth) and distance CP by LH (notch height). Finally the individual results were displayed in a diagram to define the distribution of the centres relative to the means and the advised tunnel position. Fig. 2a Distal view at a left femur. The high shallow and high deep points of the cartilage border were determined by placing the stylus at the point of an imaginary rectangular corner, indicating the separation between wall and roof, this corresponds with the 1.30 o’clock position in a left knee (10.30 o’clock in a right knee). b View at the medial side of a left lateral femoral condyle. The points that were digitized: AMB (white rounds) and PLB (black rounds) attachments; the cartilage border (grey asterisks), with the most distal, high shallow point (black asterisk) and most proximal, high deep point (white asterisk) indicating the separation between the notch wall on the lateral condyle and the notch roofFig. 3A schematic arthroscopic femoral view through the anteromedial portal. The femoral coordinate system is indicated. The first reference line, from the high deep point (D) to the high shallow point (S) on the cartilage edge separated the notch wall on the lateral condyle from the notch roof and defined the notch depth (ND). The second reference line, a perpendicular line, from the lowest point on the posterior cartilage edge (L) to the crossing point on the line DS (H) and defined the notch height (NH). The calculated attachment centre of the ACL (C) was projected on line DS (P). The absolute distances DP and CP were calculated, as were the distances relative to the reference lines (DP/DS and CP/LH). This was also done for the AMB and PLB, for the sake of clearness, only the centre of the entire ACL is depicted Tibia The digitized points on the tibia are represented in Fig. 4. The medial tibial spine (M) was determined as the main tibial landmark. The circumferences of the ACL attachments were digitized, as were the additional reference points. Those were, besides the lateral tibial spine (L), the most anterior points on the margin of the articular surfaces of the medial tibial condyle (MA) and lateral tibial condyle (LA). The tibial coordinate system was defined as follows (Fig. 5). The first tibial reference line (from M to L) was defined between the medial and the lateral tibial spine representing the interspinal distance (ID). A line connecting the anterior points MA and LA indicated the anterior margin of the articular surface of the medial and lateral condyles of the tibia. The second reference line (from M to Q) connected the medial spine (M) perpendicular with the anterior margin at crossing point Q. The length of the second reference line represents the length of the anterior intercondylar area (AL). After creating this coordinate system, the calculated attachment centres of the ACL, AMB and PLB were projected on the line ML at point P. The absolute centre position (mm) was composed of the distance between the major tibial landmark, the medial tibial spine M and the point P, and the distance between the point P and the centre C. The relative position (%) was calculated by dividing the distance MP by ML (interspinal distance) and distance CP by MQ (anterior length). Finally the individual results were displayed in a diagram to define the distribution of the centres relative to the means and the advised tunnel position. Fig. 4Proximal view at the articular surfaces of the tibial condyles and the anterior intercondylar area. The points that were digitized: AMB (white rounds) and PLB (black rounds) attachments; the lateral tibial spine (light grey oval) and medial tibial spine (dark grey oval) and the most anterior points on the margin of the articular surface of the medial and lateral tibial condyles (white asterisks)Fig. 5A schematic arthroscopic tibial view. The tibial coordinate system is indicated. The first reference line, from the medial spine (M) to the lateral spine (L), defined the interspinal distance (ID). The second reference line, from the medial spine (M) perpendicular to the crossing point (Q) at the anterior line between the most anterior points on the margin of the articular surface of the medial and lateral tibial condyles defined the anterior intercondylar length (AL). The calculated attachment centre (C) was projected on line ML (P). The absolute distances MP and CP were calculated, as were the distances relative to the reference lines (MP/ML and CP/MQ). This was also done for the AMB and PLB, for the sake of clearness, only the centre of the entire ACL is depicted Surface To determine the surface dimensions of the bony attachment areas, a line was fitted through the digitized points on the outlines of both the AMB and PLB attachments on femur and on the tibia. The enclosed surface of all areas was calculated (in mm2), with the aid of a Delaunay triangulation based on the Qhull algorithm as provided by Matlab® (version 7 The MathWorks, Inc, more details provided on http://www.mathworks.com). The surface of the entire ACL attachment was calculated as the sum of AMB and PLB. Also the percentage of AMB and PLB surfaces was calculated relative to the ACL attachment surface. Statistical analyses to detect differences in size of the attachment surfaces were performed. Femoral AMB and femoral PLB, tibial AMB and tibial PLB and finally femoral ACL and tibial ACL were compared. A 2-tailed Student’s t test for paired data was used. Statistical significance was defined as P < 0.05, for a 95% alpha level. Results Femur The oval-shaped attachment of the anterior cruciate ligament was situated on the medial surface of the lateral femoral condyle. It was positioned deep in the notch, covering most of the proximal half of the wall. The fibres of the deep low border attached to the edge of the joint cartilage, following the contour of this edge posteriorly on the condyle. In only 4 of the 35 specimens, the attachment site was completely limited to the medial wall of the lateral condyle and had no footprint in the notch roof. In 31 femurs, a small part of the deep high AMB attachment extended into the intercondylar notch roof. The notch depth (DS = 31.8 ± 2.6 mm) was the largest distance of the femoral dimensions. On average it was 2¼ times the length of the notch height (LH = 14.3 ± 1.5 mm). In shallow–deep direction measured along the notch, relative to the notch depth, the centres of the two bundles were more close to each other, than in high–low direction measured from the roof, relative to the notch height. The mean position of all AMB centres along the roofline of the notch was 7.2 mm shallow from the high deep corner at ¼ of the ND-line (Table 1; Fig. 6). In 9 of the 35 femurs the centre of the AMB was not situated on the condyle wall, but above the transition line on the notch roof. Therefore, the average centre was high in the notch at 1/10 of the NH-line, 1.4 mm from the roof. The centre of the PLB was positioned slightly more shallow at less than 2 mm from the AMB (Table 1; Fig. 6). However, it was situated clearly lower, about 5 mm, on the femoral condyle wall, approximately halfway the notch height. The centre position of the entire ACL was in the middle between the AMB and the PLB at ¼ of both reference lines. Approximately 96% of the mean centres were inside a 12 mm drill hole, if positioned at the mean centre of the ACL attachment. The Pearson’s correlation coefficients between the absolute positions and the reference lines were smaller than 0.3 and not significant (P > 0.05). Table 1The positions of the ligament centres in the femoral notch Absolute distances (mm) mean (SD)Relative distances (%) mean (SD)DPCPDP/DSCP/LHAMB7.2 (1.8)1.4 (1.7)23 (6)10 (12)PLB8.8 (1.6)6.7 (2.0)28 (6)47 (13)ACL7.9 (1.4)4.0 (1.3)25 (5)28 (9)DP The absolute distance from the high deep point D until point P, the projection of the mean ligament centre at the DS-line, in the notch depth direction; CP the absolute distance from the mean ligament centre until point P at the LH-line, in notch height direction; DP/DS The relative position of the centre to the notch depth; CP/LH The relative position of the centre to the notch heightFig. 6a A two-dimensional graph of the medial side of a right lateral femoral condyle with the individual centres, and the mean centres with the 95% Confidence Interval of the AMB (red squares), PLB (blue diamonds) and ACL (green dots). The position of a 12 mm drill hole at the ACL centre is also displayed. b A schematic arthroscopic view through the anteromedial portal with the mean centres and the 95% Confidence Interval areas of the AMB (red square with line), PLB (blue diamond with line) and ACL (green dot with line) The mean centres of the AMB and PLB were situated 6.2 ± 1.2 mm from each other. The AMB attachment area, 45% of the total ACL, was significantly (P = 0.005) smaller than the PLB attachment area (Table 2). Table 2The surface dimensions of the ACL and the two bundles at femur and tibiaAbsolute surface area (mm) mean (SD)Relative surface to total ACL (%) mean (SD)ACLAMBPLBAMBPLBFemur184 (52)81 (27)103 (39)45 (11)55 (11)Tibia229 (53)136 (37)93 (33)59 (9)41 (9) Tibia The tibial attachment area of the ACL was situated between the medial and lateral tibial condyle covering the medial part of the anterior intercondylar area. It was stretched out from the region between the tibial spines to anterior with various extensions, more or less shaped as a footprint. The fibres of the AMB inserted medially along the cartilage edge of the articular surface of the medial tibial condyle. The PLB covered the lateral side of the attachment area and was bounded by the attachment of the anterior horn of the lateral meniscus. The interspinal distance (ML = 12.9 ± 1.6 mm) between the medial and the lateral spine was on the average half times the length of the anterior intercondylar area (MQ = 24.8 ± 2.7 mm). The average centre of the AMB was situated closest to the medial spine, lateral at ¼ of the ID. The PLB was 4 mm more lateral, approximately halfway between the medial and lateral spine (Table 3; Fig. 7). The ACL was in between at 2/5 of the ID. In anterior direction the AMB, ACL and PLB were close to each other, just over 1/3 of the anterior length. Approximately 94% of the mean centres were inside a 10 mm drill hole, if positioned at the mean centre of the ACL attachment (Table 3; Fig. 7). The Pearson’s correlations coefficients between the absolute positions and the interspinal distance (reference line ML) were smaller than 0.3 and not significant (P > 0.05). The Pearson’s correlations coefficients between the absolute positions and the anterior length (reference line MQ) were significant (P < 0.05), however smaller than 0.6. Table 3The positions of the ligament centres at the anterior intercondylar tibial areaAbsolute distances (mm) mean (SD)Relative distances (%) mean (SD)MPCPMP/MLCP/MQAMB3.0 (1.6)9.4 (2.2)23 (12)37 (9)PLB7.2 (1.8)10.1 (2.1)55 (13)38 (9)ACL5.1 (1.7)9.8 (2.1)39 (12)38 (8)MP The absolute distance from the medial spine (M) until the point P, the projection of the mean ligament centre at the line ML in lateral direction; CP the absolute distance from the mean ligament centre until point P on the line ML, in anterior direction; MP/ML The relative position of the centre to the interspinal distance; CP/MQ The relative position of the centre to the anterior lengthFig. 7a A two-dimensional graph of a right anterior intercondylar tibial area with the individual centres, and the mean centres with the 95% Confidence Interval of the AMB (red squares), PLB (blue diamonds) and ACL (green dots). The position of a 10 mm drill hole at the ACL centre is also displayed. b A schematic arthroscopic view with the mean centres and the 95% Confidence Interval areas of the AMB (red square with line), PLB (blue diamond with line) and ACL (green dot with line) The mean centres of the AMB and PLB were situated 4.5 ± 0.1 mm from each other. The tibial AMB attachment area of 59% was significantly larger than the PLB attachment area (P < 0.001). The tibial attachment area of the ACL and of the AMB was significantly larger than on the femur (P < 0.001) (Table 2). Discussion Incorrect tunnel placement, tibial [22, 25] as well as femoral [43], is seen as one of the most important causes of clinical failure in single-bundle ACL reconstructions [1]. In double-bundle ACL reconstruction, exact anatomic tunnel placement seems to be even more essential. Two bundles must be accurately placed relative to the surrounding structures and relative to each other. Although exact tunnel positioning is important, it seems difficult, even for experienced surgeons [8, 27, 47]. A clear description of the anatomic centres with guidelines to determine the correct tunnel position during arthroscopic procedures can improve the accuracy. The femoral positions of the two distinct bundles found in this study, the AMB deep high and the PLB shallow low in the notch, are broadly in line with literature [11, 20, 32, 45, 52]. Compared to others, the landmarks that are used in the present study are more easy to locate during arthroscopy. Harner et al. [20] quantified the cross-sectional shape and area of the femoral and tibial attachments of both components in 10 knees without describing the positions relative to landmarks. Yasuda et al. [52] limit their study to the femoral attachment, using five specimens. They describe the centre of the PLB 5 to 8 mm anterior to the edge of the joint cartilage, on the vertical line through the contact point between the femoral condyle and the tibial plateau in a 90° flexed knee. Because this point depends on the position of the knee, it can be difficult to locate accurately during an arthroscopic procedure. Colombet et al. [11] examined seven specimens, collecting especially data of the attachment dimensions. The femoral results presented by the studies of Mochizuki et al. [32] and Takahashi et al. [45] are more suitable for practical use during arthroscopy; however, no exactly defined landmarks were used. The tibial results of Takahashi et al. [45] cannot easily be transferred to the arthroscopic situation. Finally, above-mentioned studies did not present the centres of the ACL attachment. Tunnel positioning in the femoral notch is often determined by the ‘clock’ method [13, 18, 35, 36, 52, 54]. However, this only determines the high–low position from the roof, in the transversal plane along the arc of the femoral notch [17, 54]. The accuracy in the sagittal plane, deep–shallow along the roof of the notch, seems to affect the functional outcome more [17, 21, 29, 55, 56]. This position is often determined with a femoral guide placed behind the posterior edge of the intercondylar notch, at the over-the-top position, not to be confused with the Resident’s ridge [24]. This seems sufficiently accurate for determining the AMB centre, when a guide with a 7 mm offset is placed in the 10.30 or 1.30 o’clock position. The 7 mm offset of the guide places the tunnel in the sagittal plane close to the AMB centre position, found in this study, i.e. 7.2 mm shallow along the roof. Although we did not translate the measured distance in a clock position, the 10.30 clock position seems close to the measured position in the transversal plane, 1.4 mm low from the roof on the condyle wall. This corresponds with the results of Mochizuki et al. [32] and Yasuda et al. [52]. However, the clock method is sensitive to subjective interpretation of positioning the face of the clock [7]. It is not sufficient to find the correct position for the PLB, especially in the sagittal plane. Therefore Colombet et al. [11] defined an extra guideline to position the PLB tunnel: 8 mm lower and ‘shallower’ relative to the AMB centre, found with the clock method. The present study is more precise: 1.3 mm more shallow along the notch and 5.3 mm lower from the roof on the condyle wall, relative to the above-described position of the AMB centre. These positions can best be approached through an anteromedial arthroscopic portal [6]. Various methods can be used to determine the correct drill hole position at the tibia. Some authors prefer placement of the tibial tunnel based on avoiding graft impingement against the roof of the femoral notch with knee extension [22, 34, 44]. Others prefer guides that use the posterior cruciate ligament (PCL) attachment as reference point [31]. Based on these methods, positioning in anteroposterior direction, the sagittal plane, is defined. However, placement of the tunnel in mediolateral direction, the transversal plane, which is also important [26], is not determined. Some studies describe the mediolateral position, relative to the width of the tibial plateau [23, 45]. However, this guideline cannot be used during an arthroscopic procedure. The results of this study can be used for arthroscopic positioning in both directions. The division of the AMB and PLB on the tibia in anteroposterior direction is similar to Harner et al. [20], resulting in a medial AMB and lateral PLB. This deviates only marginally from other definitions, where the division is in an anteromedial and a posterolateral part [3, 16, 38]. The position of the centres in anteroposterior direction correspond with the results of Takahashi et al. [45], who used similar reference points to define the anterior margin of the articular surface of the tibial condyles. The centres of the two bundles were more close to each other on the tibia, than on the femur. The size of the femoral ACL attachment 184 ± 52 mm2 was similar to the result of Odensten and Gillquist (200 mm2) [38]. Other studies found smaller attachment areas, 132 mm2 [45] and 113 mm2 [20]. This difference may be caused by a different measuring method, using the largest projection in a 2D image plane. Our study measured the actual three-dimensional attachment area. The partition of the femoral attachment area was nearly equal with 45%, 55%, respectively, for the AMB and PLB. This was comparable to Harner et al. [20] (AMB = 52%, PLB = 48%) and Takahashi et al. [45] (AMB = 50%, PLB = 50%). The average tibial attachment area found in this study was larger than the femoral attachment. This is in line with the results of Girgis et al. [16] and Fuss et al. [15]. On the other hand, Takahashi et al. [45] found that the femoral attachment was larger than the tibial attachment (132 vs 119 mm2). Harner et al. [20] and Odensten and Gillquist [38] found similar sizes for the tibial and femoral attachment areas. In our study the AMB occupied 59% of the tibial attachment, nearly similar to the 57% of Takahashi et al. [45], slightly more than the 52% Harner et al. [20] found. There was a large variation in knee sizes and dimensions of the reference lines. However, we did not find a correlation between the absolute distance of the attachment centres and the size of the reference lines, i.e. knee size. The absolute positions of the attachment centres were more or less similar for large and small knees. Therefore both absolute as well as relative data can be used as guidelines to find the anatomical positions. At the arthroscopic view, the absolute positions are more useful than the relative positions. As regards to the dimensions of drilled tunnel holes (10–12 mm), the standard deviations of the attachment centres within 3 mm are acceptable. The femoral positions had less variations than the tibial positions. The positions for the entire ACL attachment had less variations than the attachment positions for the separate bundles. The results of this study seem to produce accurate guidelines to find the anatomic tunnel position during arthroscopic reconstruction. However, some limitations must be mentioned. The subdivision of the ACL in AMB and PLB is not based on anatomically distinct fibre bundles surrounded with fibrous issue [5, 15, 33, 38]. Fibres of both bundles are twisted around each other [33]. Division of the ACL in separate bundles is not easy [3]. Therefore utmost care was taken to divide the two functional bundles according to a previously described method [11, 20]. The method, based on observed tension variation during passive flexion–extension movement was reported to be consistent [32]. This is expressed in the standard deviations (SD) of the mean attachment centres. The variation of the population is expressed by the SD of the ACL centre. The SD of the mean PLB and mean AMB centres show this variation combined with the division error. For the tibia the SDs of the PLB and AMB attachment centres are equal to that of the ACL. For the femur the SDs of the two bundle centres are slightly larger. However, these results do not point to large errors in the identification of the two bundles. The variations of attachment centre positions are actually rather small and similar to other studies [11, 45]. The age of the cadaveric knee specimens was more than 60 years. However, only knees without severe arthritic signs were evaluated. It seems likely that the data can be transferred to the, mostly younger, population receiving an ACL reconstruction [3, 11, 34, 44]. Perfusion fixation preserves the outer contours of shapes i.e. ligament bundles, so neither the identification nor the measurements of the various bundles can be assumed to have been hampered by the fixation. Conclusions This study was performed because no earlier study described the ACL, AMB and PLB attachment centres relative to arthroscopically visible landmarks. This resulted in quantitative data of the positions of the attachment centres of the ACL and its two bundles, relative to bony landmarks on femur and tibia, visible through an arthroscope. Using these results, the surgeon will be able to determine the anatomical position of the ACL and the two functional bundles during arthroscopy without additional images or fluoroscopic support. The results can be applied for anatomic single-bundle or anatomic double-bundle techniques.

Int_Urogynecol_J_Pelvic_Floor_Dysfunct-4-1-2259253

The diagnostic strength of the 24-h pad test for self-reported symptoms of urinary incontinence in pregnancy and after childbirth

The clinical impact of incontinence in pregnancy and after childbirth is growing because some studies report the efficacy of physiotherapy in pregnancy and because obstetric choices are supposed to have significant impact on post-reproductive urinary function (Goldberg et al. in Am J Obstet Gynecol 188:1447–1450, 2003). Thus, the need for objective measurement of urinary incontinence in pregnancy is growing. Data on pad testing in pregnancy are lacking. We assessed the clinical relevance of the 24-h pad test during pregnancy and after childbirth, compared with data on self-reported symptoms of urinary incontinence and visual analogue score. According to the receiver operating characteristic curve, the diagnostic value of pad testing for measuring (severity of) self-reported incontinence during pregnancy is not of clinical relevance. However, for the purposes of research, pad tests, combined with subjective/qualitative considerations, play a critical role in allowing comparisons across studies, quantifying the amount of urine loss and establishing a measure of severity. Introduction Incontinence is reported frequently in pregnancy and after childbirth [2–4]. It has been suggested that urinary incontinence in pregnancy is a predictor of the chance to develop post-partum urinary incontinence [5]. In that respect, prevention such as physiotherapy during pregnancy is advised in women with positive symptoms of urinary incontinence in pregnancy [6–9]. Based on questionnaires on symptoms of post-partum urinary incontinence, Goldberg et al. found a strong protective effect of cesarean delivery against the development of post-partum urinary incontinence and highlighted the impact of obstetric choices on post-reproductive urinary function [1]. When the possible consequences of the fact that a pregnant women reports urinary incontinence grow, such as an advise for preventive physiotherapy or an advice regarding the mode of delivery, there is a greater need for objectivity in diagnosing the problem. Pad testing yields an objective measurement of fluid loss over a certain period. In non-pregnant women, the diagnostic value of pad testing for self-reporting of symptoms of urinary incontinence has been questioned [10, 11]. These data are lacking for pregnant women. Most common used types of pad tests are the 1-h and the 24-h test [12–15]. The 24-h pad test is almost certainly more representative to the patients’ day-to-day experiences and is more likely to correlate with self-reported symptoms [16]. The 24-h pad testing has been studied and described exclusively in non-pregnant women. In this study, we focussed on the diagnostic strength of pad testing to measure (the severity of) urinary incontinence in pregnancy and after childbirth. The aims of this study were (1) to describe pad weight gain as measured by the 24-h pad test in a cohort of pregnant women and (2) to assess the clinical usefulness of the 24-h pad test in pregnancy and after childbirth in terms of the relationship between objective urine loss and the self-reported symptoms of urinary incontinence. Materials and methods One hundred and seventeen women who attended the outpatient clinics of the University Hospital Groningen and the Martini Hospital Groningen enrolled for the study, with a mean age of 30 years (range 17–41). All women were of Caucasian origin, except three who were of Mediterranean origin. All women were nulliparous and had no history of incontinence, pelvic operations or neurological disease. Written informed consent was obtained from all participating women. The study was approved by the medical ethical committees of both hospitals. For this study, the women were investigated at 28–32 and at 36–38 weeks of pregnancy and 6 weeks and 6 months post-partum. At each visit, all women completed a questionnaire and a visual analogue score (VAS) on symptoms of urinary incontinence. Complaints of more than five on the VAS scale 0–10 were defined as severe complaints. Women were asked to classify their incontinence as mainly: (1) stress urinary incontinence, involuntary leakage on effort or exertion or on sneezing or coughing, and (2) urge urinary incontinence, involuntary leakage accompanied by or immediately preceded by a strong desire to void [17]. Three pads were packed, each in a plastic bag, and weighed by the investigators before and after use. The women received a written instruction and were free to wear one, two or three pads. It was emphasised that the bag should be closed carefully every time a pad was changed to prevent evaporation. If the bags were open or less than three pads were returned, the test was excluded from evaluation. Pads were given to all women, to be worn for 24 h preceding their appointment. The outcome of the 24-h pad test was recorded as the weight gain as measured by a verified spring balance. Weighing was done by the first or second author, within 3 days after the pad test was carried out. According to the literature, pads were assigned as wet if the total weight gain per 24 h was ≥9 g [11]. Statistical analysis Pad test results have a non-parametric distribution. For continuous variables, non-parametric tests are used. Numeric data are analysed by cross tabulation, chi-squared test and risk analysis. Pearson correlation test was used to identify significant relationships between variables. Data are presented as median or numbers. Results Pregnancy, pad test At 28 weeks of pregnancy, 115 of 117 patients (98%) returned their pads according to the protocol. The median weight gain was 5 g (range 0–36). At 38 weeks, data were available from 98 women (84%). Two patients had withdrawn from the study because of inconvenience, while 17 pads were not, or were not according to the protocol returned. At this stage of pregnancy, the median weight gain was also 5 g (range 0–22). Distributions of pad test weight gain are given in Fig. 1. Fig. 1Distribution of 24-h pad test results at 28 and 38 weeks of pregnancyPad test results at 28 and 38 weeks of pregnancy were related (r = 0.452, p < 0.0001). Twenty-seven out of 115 (23%) and 17out of 98 pads (17%) were wet at 28 and 38 weeks of pregnancy, respectively, and again, results at 28 and 38 weeks were related (r = 0.317, p < 0.001). Pregnancy, questionnaire At 28 weeks of pregnancy, 35 of 117 women (30%) reported incontinence (stress, 28 of 35, and urge, 7 of 35), and at 38 weeks, 40 of 115 (35%) did (stress, 33 of 40 [82%], and urge, 7 of 40 [8%]). Reported incontinence at 28 and 38 weeks of pregnancy was related (r = 0.55, p < 0.001). Severe complaints of incontinence were reported in 17 of 117 (15%) and 22 of 115 cases (19%), which were also related (r = 0.482, p < 0.001). Pregnancy, pad test and questionnaire Women with self-reported incontinence at 28 and 38 weeks of pregnancy had a median pad weight gain of 6.0 (range 0–36) and 6.0 g (range 0–22), respectively, while women without self-reported incontinence had a pad test result of 4.0 (range 0–22) and 4.0 g (range 0–22), respectively, showing no differences between these groups of women. To evaluate the diagnostic value of the pad test for measuring self-reported incontinence, the sensitivity and specificity for several cutoff levels for the pad test was calculated, graphically known as the receiver operating characteristic (ROC) curve. The ROC curve for pad tests and self-reported incontinence at 28 weeks of pregnancy did not differ from the reference line (area 0.575 vs 0.5, p = 0.207), showing a non-diagnostic test (Fig. 2). At 38 weeks of pregnancy, the ROC curve showed a significant difference from the reference line (area 0.663 vs 0.5, p = 0.008), the optimum cutoff point at 5.5 g with a sensitivity of 0.629 and a specificity of 0.619 (Fig. 2). When stratified for stress incontinence or for severe symptoms of urinary incontinence, the ROC curve did not improve. Fig. 2ROC curve, the diagnostic value of the pad test at different cutoff levels (g/24 h) for measuring self reported incontinence at 28 (a) and 38 (b) weeks Puerperium, pad test At 6 weeks post-partum, 80 of 117 patients (68%) returned their pads according to the protocol. The median weight gain was 3 g (range 0–40). At 6 months post-partum, pad data were available from 76 patients (65%). Six patients had withdrawn from the study because of inconvenience, three patients had their delivery preterm, 32 pads were not, or were not according to the protocol returned. These women did not want to participate in the pad study anymore or did not return their pads according to the protocol. The women that withdrew from the pad test study were asked to fill up the questionnaires on symptoms of urinary incontinence to check for bias. The women did not differ for self-reported symptoms on urinary incontinence from the women that continued the study on pads. In the study group, the median weight gain was again 3 g (range 0–40). Distributions of the pad test weight gain are given in Fig. 3. Pad test results at 6 weeks and 6 months post-partum were related (r = 0.792, p < 0.0001). Seven out of 80 (9%) and 3 out of 76 pads (4%) were wet at 6 weeks and 6 months post-partum, respectively, and again, the results at 28 and 38 weeks were related (r = 0.320, p = 0.017). Fig. 3Distribution of 24-h pad test results at 6 weeks and 6 months post-partum Puerperium, questionnaire At 6 weeks post-partum, 21 of 115 women (18%) reported incontinence (stress, 16 of 21, and urge, 5 of 21), while at 6 months post-partum, 16 of 109 (15%) did (stress, 12 of 16, and urge, 4 of 16). Reported incontinence at 6 weeks and 6 months post-partum is related (r = 0.61, p < 0.001). Severe complaints of incontinence was reported in 15 of 115 (13%) and 11 of 109 cases (10%), which were also related (r = 0.691, p < 0.001). Puerperium, pad test and questionnaire Women with self-reported incontinence at 6 weeks and 6 months post-partum had a median pad weight gain of 5.5 (range 0–40) and 6.0 g (range 0–40), respectively; women without self-reported incontinence had a pad test result of 3.0 (range 0–16) and 3.0 g (range 0–9), showing a differences between these groups of women (p = 0.01 and p = 0.045, respectively). The ROC curve for pad tests and self-reported incontinence at 6 weeks post differed from the reference line (area 0.767 vs 0.5, p = 0.001), the optimum cutoff point at 4.5 g and the sensitivity of 0.722 and specificity of 0.742 (Fig. 4). At 6 months post-partum, the ROC curve shows also a significant difference from the reference line (area 0.666 vs 0.5, p = 0.047), the optimum cutoff point at 5.5 g and the sensitivity of 0.629 and specificity of 0.619 (Fig. 4). Again, stratifying for stress incontinence or for severe symptoms of urinary incontinence, the ROC curve did not improve. Fig. 4ROC curve, the diagnostic value of the pad test at different cutoff levels (g/24 h) for measuring self-reported incontinence at 6 weeks (a) and 6 months post-partum (b) Discussion In this study, we focussed on the use of pad testing to investigate its prognostic value for objectively measuring (the severity of) self-reported urinary incontinence during pregnancy and after childbirth. The clinical impact of incontinence in pregnancy and after childbirth is growing because some studies report the efficacy of physiotherapy in pregnancy and because obstetric choices are supposed to have a significant impact on post-reproductive urinary function [1]. This growing impact requires objective measurement. In a meta-analysis, the symptom of stress incontinence was 91% sensitive but only 51% specific for detecting genuine stress urinary incontinence as defined by the International Continence Society, based on history and urodynamic testing [18]. Because of pregnancy and because we are interested in an instrument for screening, it is obvious that urodynamic testing cannot be the instrument of choice. Pad testing is an objective, simple and non-invasive instrument capable of measuring fluid loss in a certain period. First of all, we need to define normal values, data that describe the results of pad testing in a cohort of pregnant women without a history of incontinence before pregnancy. Secondly, we need comparison with the criterion standard in pregnancy and the patients’ history. The median weight gain in the 24-h pad test in pregnancy as reported in our study is in accordance with results reported in a group of non-pregnant, not incontinent, premenopausal women, 2.6–7.0 g, with an upper confidence limit of 5.5–8 g [16, 19]. In post-menopausal continent women, a much lower weight gain of 0.3 g is reported [20]. When compared to the pre-menopausal women, the state of pregnancy does not lead to a higher weight gain in the 24-h pad test nor does the puerperal state. It is remarkable that the group of controls as referred to had similar pad test results but did not report incontinence, whereas in our pregnant group, 30 (28 weeks of pregnancy) and 35% (38 weeks of pregnancy) of the women did. With the same weight gain in pregnancy, women report more incontinence than in the sample of non-pregnant women. It seems therefore that not the amount of weight gain in the pad test but the pregnancy state itself is more discriminating for the chance that a woman qualifies herself as incontinent. In our study, during pregnancy, pad test results had only limited diagnostic value for self-reporting of incontinence. In their review article, Ryhammer et al. [21] stated that “incontinence is a complex condition in which differences in the individual patients’ personal characteristics influence the perception of leakage and the identification of the problem. Pregnancy seems to modulate this perception in such a way that it cannot be measured by pad test. As pad testing did not show to have high sensitivity and specificity for self-reported urinary incontinence in pregnancy and after childbirth, there remains confusion about the accurate diagnosis. This becomes important in deciding on management options such as offering preventive physiotherapy in selected cases or strategies that influence the mode of delivery. After childbirth, the median weight gain is also in accordance with results in non-pregnant continent women. In our study group, 15 to 18% of the women report positive for symptoms of incontinence. Just like in pregnancy, the pad test result has a significant value for testing self-reporting incontinence but again low figures for sensitivity and specificity. Like us, Morkved and Bo [22] reported a discrepancy between self-reported symptoms and stress urinary incontinence assessed by their (short) pad test, 8 weeks after delivery. To assign a women to an intervention, one needs a higher specificity, which according to the curves as shown, will rapidly lead to lower sensitivity. Depending on the chosen intervention, this may or may not be accepted. The calculation of the diagnostic strength of our 24-h pad test was made with the self-reporting of symptoms of urinary incontinence as the gold standard. At 28 weeks, the pad test failed to capture eight subjects who stated they were wet; at 38 weeks, this was higher. Such results are possibly related to the high threshold for definition of incontinence in the women with some leaking less than 9 g describing some leakage. The rationale for using a high cutoff is established in both men and women, but subjects themselves may perceive this a severe incontinence. It is possible that pregnant and post-delivery women perceive leakage differently than their non-pregnant counterparts. When adding severity of symptoms to the gold standard, as reported by VAS, the diagnostic strength of the pad test did not improve. In general practice, questioning about incontinence will provide the clinician with adequate information on the presence, absence or severity of incontinence from a patient perspective, and cumbersome pad tests are unnecessary. In a review article on questionnaires for women with pelvic floor disorders, Barber [23] concludes that measuring symptom severity and quality of life changes in women with pelvic floor disorders is an important part of the evaluation and treatment of women and may be the only practical way to clinically assess symptoms. However, for the purposes of research, pad tests, combined with subjective/qualitative considerations, play a critical role in allowing comparisons across studies, quantifying the amount of urine loss and establishing a measure of severity. Indeed, the International Continence Society standards for research strongly recommend the pad test as one measure in all incontinence research. The fact that the pad test results and patient-reported incontinence were not strongly correlated illustrates the importance of both quantitative and qualitative measures when considered an intervention trial with pelvic floor muscle exercises, for example. From our study, we conclude that pad testing measures fluid loss over a certain period but does not quantify self-reported symptoms of urinary incontinence. Both measurements are of interest but cannot replace each other. Stressing of the pelvic floor by pregnancy and childbirth modulates the sensation of urinary leakage in such a way that women in this state do report symptoms of urinary incontinence more frequently than nulliparous pre-menopausal women do.

Bioinformation-2-5-2241933

Functional gene clustering via gene annotation sentences, MeSH and GO keywords from biomedical literature

Gene function annotation remains a key challenge in modern biology. This is especially true for high-throughput techniques such as gene expression experiments. Vital information about genes is available electronically from biomedical literature in the form of full texts and abstracts. In addition, various publicly available databases (such as GenBank, Gene Ontology and Entrez) provide access to gene-related information at different levels of biological organization, granularity and data format. This information is being used to assess and interpret the results from high-throughput experiments. To improve keyword extraction for annotational clustering and other types of analyses, we have developed a novel text mining approach, which is based on keywords identified at the level of gene annotation sentences (in particular sentences characterizing biological function) instead of entire abstracts. Further, to improve the expressiveness and usefulness of gene annotation terms, we investigated the combination of sentence-level keywords with terms from the Medical Subject Headings (MeSH) and Gene Ontology (GO) resources. We find that sentence-level keywords combined with MeSH terms outperforms the typical ‘baseline’ set-up (term frequencies at the level of abstracts) by a significant margin, whereas the addition of GO terms improves matters only marginally. We validated our approach on the basis of a manually annotated corpus of 200 abstracts generated on the basis of 2 cancer categories and 10 genes per category. We applied the method in the context of three sets of differentially expressed genes obtained from pediatric brain tumor samples. This analysis suggests novel interpretations of discovered gene expression patterns. Background In recent years, increasing amounts of biological data have become available through techniques such as DNA microarrays and other high-throughput gene and protein assays. [1,2] As large numbers of genes can be included in such studies, the task of assigning meaningful biological function to gene patterns or gene clusters is a considerable challenge. Typical analyses using supervised (classification) or unsupervised (clustering) methods require the user to incorporate the necessary background knowledge. [3] This ability to incorporate background knowledge is fundamental to effective and efficient scientific discovery. A substantial amount of biomedical knowledge is captured in free-text form in abstracts and full-text articles and also in specialized biological information systems such as Gene Ontology (GO) [4], Medical Subject Headings (MeSH) [5], Database of Interacting Proteins (DIP) [6] etc. Until only a few years ago, human reasoning was the primary method for the extracting, synthesizing and interpreting the information contained in the biomedical literature and supporting biological information systems. However, in recent years the number of online documents (and other biological information repositories) has grown tremendously. This is both an opportunity and a challenge. On one hand, such resources facilitate automated processing of the knowledge and information contained in these documents. On the other hand, such processing poses considerable algorithmic and computational challenges [7]. For example, the biomedical abstract database MEDLINE [8] currently contains about 15 million citations and about 40 000 citations are added monthly. Text mining is the application of techniques from machine learning, natural language processing (NLP), information extraction and statistical/mathematical approaches to automated extraction of useful knowledge from text [9]. Text mining of biomedical literature has been applied successfully to various biological problems. Many studies focus on protein-protein [10–13] and gene-protein interactions. [14] Other specific relationships between biological entities such as sub-cellular localization of proteins [15,16], molecular binding relationships [17] and interaction between genes and drugs [18] are also explored. Text analysis of biomedical literature has also been applied successfully to incorporate functional information of gene expression data [19–23]. For example, MedMOLE [19] identifies the functions among a group of genes by simple text clustering of entire MEDLINE documents associated with the genes. Blaschke et al. [20] extracted information about the common biological characteristics of gene clusters from MEDLINE using a statistical term weighting approach. This method returns an ordered set of keywords with a high probability of occurrence in abstracts. Liu et al. [24] extended this approach by clustering such keywords to find gene-to-gene relationships. Clustering genes by functional keyword association can provide direct information about the nature of genes and their functional association [25]. However, the quality of the keyword lists extracted from the biomedical literature for each gene significantly affects the clustering results. Commonly, these approaches represent genes by extracting keywords from entire abstracts [25]. These keywords may undergo transformations such as weighting or dimension reduction with the goal of improving clustering quality and efficiency. However, gene clustering using entire abstracts has the following main drawbacks. (a) Abstracts normally contain a large number of irrelevant sentences. These sentences may influence the clustering process and are likely to obscure information useful for gene annotation. (b) The number of unique terms in abstracts is typically very large. This requires the ability to deal with sparse data spaces or methods for dimensionality reduction. (c) Dimension reduction methods such as principal component analysis or signal-to-noise methods increase the computational complexity, may lead to the loss of important keywords and do not guarantee that the reduced dimensionality will yield better clustering/annotation information. Also, the composite features may be hard to interpret. To avoid the above drawbacks and improve the clustering process, we decided to use gene annotation sentences from abstracts instead of using full abstracts to extract the keywords. Current NLP techniques allow such sentence extraction from documents. Using clustering on the basis of sentence-extraction techniques has the advantage of avoiding complex dimensionality reduction and term weighting techniques. Further, this approach is likely to yield more specific terms which are easier to interpret. We first extracted the potential sentences describing gene annotation information from abstracts using a NLP method utilizing gene/protein name dictionaries and pattern-matching-based rules. In addition to this sentence-keywords approach, we carried out two further experiments involving MeSH terms and GO terms as supplementary keywords. Hence, in our method, each gene is represented by set of keywords extracted from sentences, MeSH terms and GO terms. To demonstrate the usefulness of the proposed text mining methods, we performed hierarchical clustering of a gene × keyword matrix to find functionally discrete sub-groups of genes. The overall experimental design and its components are illustrated in the Figure 1. We validated the performance of keywords extracted by our method using a manually annotated corpus of 200 abstracts. We also evaluated the usefulness of our method by sorting differentially expressed genes from a microarray experiment into functional sub-groups. The objective of our gene clustering process using functional keywords is to identify and summarize potential functional gene groups and to complement the conventional gene expression data clustering tasks. Methodology Gene/Protein name and synonym dictionary creation One of the major obstacles in biomedical literature processing is the variety of names each gene or protein is known by. To address this problem in the present study, we developed a gene/protein name dictionary. Essentially, each entry of this dictionary consists of a preferred (or canonical) name for a gene/protein and a list of synonyms used for this gene/protein. The dictionary was created on the basis of the Entrez Gene [26] (previously LocusLink) database, one the most stable and complete sources of information on genes. Since our study is part of a wider investigation in the context of human brain tumor research, we focused specifically on human genes/proteins. We developed PERL scripts to extract and select from the Entrez Gene entries the official symbols as preferred name of the gene and other aliases as known synonyms. In addition, we augmented the dictionary with relevant synonyms from other publicly available databases including GeneCards [27], SwissProt [28], GoldenPath [29] and HUGO [30]. The final dictionary contains 26 731 unique human gene/protein names and 274 845 synonym names. Keywords extraction from biomedical literature In our study each gene is represented by a list of keywords extracted from MEDLINE abstract sentences, MeSH terms and GO terms. The procedure for extracting keywords from each data source is discussed below. MEDLINE abstracts keywords extraction To extract the keywords associated with each abstract, we decided to use gene annotation sentences from the abstracts instead of constructing a large keyword vector based on the entire abstract. The assumption is that the information given on sentence-level is much more specific and therefore useful to characterize the function of the genes. Only sentences that contain one or more genes reference from our gene lists will be considered as gene annotation sentences, all other sentences are discarded from the analysis. We applied the following three steps to extract sentence-level keywords (1) gene-name normalization, (2) sentence filtering, and (3) keyword extraction. Gene-name normalization This process replaces all the gene names in the abstract with its unique canonical identifier (Entrez gene ID) using the gene-synonym dictionary specially constructed for this study. Sentence filtering This process extracts all the gene annotation sentences from abstracts that contain one or more gene names from our gene lists using regular-expression pattern matching rules. We used different regular expressions (which rely on matching of pre-defined patterns or rules such as arrangement of gene/protein names with articles, prepositions and other keywords) to filter sentences containing one to three genes. We defined our regular expressions as nouns describing agents, passive verbs, active verbs and nouns describing actions. Table 1 (in supplementary material) depicts an example for each type of expression. For example, the regular expression ($gene @{0,6} $action (of|with) @{0,2} $gene) extracts sentences that match the structure shown below the expression. The notational construct ‘A → B → ...’ is interpreted as ‘A followed by B followed by ...’. gene name → 0-6 words → action verb → ‘of’ or ‘with’ → 0-2 words → gene name Sentence keyword extraction Sentences containing one or more gene names were parsed using the Brill part-of-speech tagger. [31] This program labels each word in a sentence with its part-of-speech information such as word category like noun, verb, adjective, preposition, etc. This information plays a critical role in identifying corresponding noun and verb phrases. Then, with a simple PERL program, noun phrases containing gene names were filtered out and the remaining noun phrases and verb phrases were extracted as keywords. Initial tests showed that certain keywords were common for most of the genes in the list (e.g., activates, associates, stimulates etc.). We manually removed these common keyword words from the list. The following example illustrates this process: Sentence BRCA1 physically associates with p53 and stimulates its transcriptional activity. Brill-POS-tagged sentence BRCA1/NNP physically/RB associates/VBZ with/IN p53/NN and/CC stimulates/VBZ its/PRP$ transcriptional/JJ activity/NN. /. Sentence keywords associates, stimulates, transcription activity Sentence keywords after manual curation transcription activity MeSH keywords extraction To extract MeSH keywords, we searched for the gene names in our gene lists in the title and abstract of MEDLINE citations related to each gene and extracted the associated MeSH terms for each gene. The extracted gene-MeSH term list was represented by scores indicating the frequency of gene-MeSH term co-occurrence. Initial tests showed that certain MeSH keywords in the list were common biological terms and less useful from the point of view of gene annotation (e.g., human, DNA, animal, Support U.S Govt etc.). A collection of MeSH stop words was created manually and these terms were removed from the gene-MeSH term lists. Finally, from the thus filtered gene-MeSH lists, the 20 highest-frequency MeSH terms associated with each gene were taken as MeSH keywords associated with each gene. For example the MeSH keywords associated with a gene “FOS” in our gene list are oncogene, felypressin, transcription-factor, thermoreceptors, DNA-binding, antibiosis, inflammatory-response, zinc-fingers, gene-regulation, and neuronal-plasticity. GO keyword extraction We used the GO keywords information incorporated in Gene Ontology [Error! Bookmark not defined.] to extract GO keywords associated with each gene. Out of the three GO annotation categories we included only molecular function and biological process as we believe that cellular component (e.g. nucleus, cell membrane etc.) is less important for characterizing genes in the context of this study. Further, due to the hierarchical nature of GO and multiple inheritance in the GO structure, we consider with every ancestor up to level 2 in the GO tree in assigning GO keywords. This enables us to use more generalized GO terms. For example the GO keywords associated with the gene “FOS” in our gene list are protein-dimerization, DNA binding, RNA polymerase, transcription factor, DNA methylation, inflammatory-response, and nucleus. Keyword representation and calculation of numeric vectors After the keyword extraction phase, each gene was described by a list of keywords extracted from MEDLINE abstract sentences, MeSH terms and GO terms. These keyword vectors then served as a basis for clustering (i.e., unsupervised class discovery). To do this, each term vector needed to be represented by a numeric vector representing the relative importance of keywords for each gene. This process is concerned with computing the numeric weight, wij, for each gene-term pair (gi, tj) (i = 1, 2,…n and j = 1, 2, … k) to represent the gene's characteristics in terms of the associated keywords. Common techniques for such numeric encoding includes (1) Binary, the presence or absence of a keyword relative to a gene, (2) Term frequency, he frequency of occurrence of a keyword with a gene (3) Term frequency × inverse document frequency (TF*IDF), the relative frequency of occurrence of a keyword with a gene compared to other genes. As we derived the keywords from gene annotation sentences but not from full abstracts, we found the number of keywords associated with each gene is small. We noticed also that absolute frequency of most keywords tended be one. Therefore, we adopted the binary encoding scheme as illustrated in Table 2 in supplementary material, in which each gene is represented by a vector of ‘normalized’ absolute keywords frequencies. The ‘normalized’ absolute frequency of each vector element (keyword) is either zero or one. Gene clustering Clustering is a data mining technique that groups or clusters data components (typically represented as numeric vectors) according to their similarity or dissimilarity. The goal is to maximize intra-cluster and minimize inter-cluster similarity among the components. [32,33] Clustering is typically used to identify sample groups in data. Unlike supervised learning methods that require explicit class label information, clustering is unsupervised and no information about target groups (classes) is used. Two basic approaches to clustering can be distinguished, hierarchical clustering (e.g., agglomerative and divisive) and non-hierarchical clustering (e.g., k-means/c-means clustering). Agglomerative hierarchical clustering starts with each object representing a cluster and then merges the clusters in sequence. Divisive hierarchical clustering starts with all samples in one cluster and successively split clusters. In hierarchical clustering the distance (similarity) between clusters is measured using different techniques such as single linkage, average linkage or complete linkage [32] and basic distance and similarity metrics (e.g., Euclidean, Minkowski, Hamming distance). K-means clustering requires a priori specification of the desired number of clusters, k. This method clusters data into groups by iteratively optimizing the positions of cluster centers (means) so that the sum of within-cluster similarities (the similarity between data points and their cluster centers) is maximized. Essentially, the sentence-level binary coding scheme adopted in this study consists of numeric row vectors representing genes (via the associated biological function/process terms), and numeric column vectors representing annotation terms (via the associated genes). These two sets of vectors can be independently clustered using available clustering algorithms and tools. This approach can produce useful and specific information about the biological characteristics of sets of genes. In this study, we have used average linkage hierarchical clustering algorithm. [33] Using this algorithm has two advantages for this study. First, clustograms, a visualization of the substructures contained in a gene collection are produced, and second, individual clusters of genes are identified by clustogram splits at different levels. Clustering was performed using Cluto [34] and Cluster/Treeview [35] facilitates visualization of the clustograms. Results and discussion Evaluation To obtain a quantitative measure on the performance of the various keyword encoding schemes, we developed a text corpus of 200 manually annotated abstracts based on two cancer categories brain tumor and breast cancer of our interest (see Table 4 under supplementary material). We used the following procedure to establish the corpus: (1)Determine randomly two cancer categories (brain tumor and breast cancer ), (2) For each cancer category, select randomly 10 genes from Entrez such that species = human and number of associated abstracts ≥ 50, (3)For each gene identified in this way, select randomly 10 abstracts, resulting in a total of 200 abstracts; 10 abstracts for each of the 10 genes associated with each of the two cancer categories, (4) For each of the 200 abstracts, identify manually the keywords characterizing biological function and processes from abstracts, MeSH terms and GO terms. With this text corpus we were able to construct a matrix containing all 20 genes and their associated keywords and keyword frequencies from abstracts, MeSH terms and Go terms. The manually annotated corpus of 200 abstracts and the matrix of 20 annotated genes served as gold standard for our evaluation experiments. We carried our four evaluation experiments: (1) Abstract keywords (baseline). Extracts gene annotation terms based on term frequencies * inverse document frequencies (TF*IDF) within the entire abstract without regard to sentence structure, (2) Sentence keywords. Extracts gene annotation terms based sentence-level keywords, (3) Sentence + MeSH keywords. As in (2) above plus MeSH terms (see Section MeSH keywords extraction), (4) Sentence + MeSH + GO keywords. As in (2) above plus MeSH terms (see Section MeSH keywords extraction) and GO terms (see Section GO keyword extraction). Essentially, in each evaluation experiment the input is the text corpus of 200 abstracts and the output is a list of genes with its predicted annotation terms. Informally, the closer the predicted annotation terms match the manually established annotation terms, the better is the method. Performance is measured via commonly used criteria such a recall (analogous to sensitivity), precision (analogous to positive predictive value) and the F-measure (a score that combines recall and precision). The results we obtained are shown in Table 5 (below in supplementary material). We notice that the baseline method comprising TF*IDF keywords fares worst among all four approaches. We interpret this as evidence for the validity of the methods involving sentence-level processing as this information is likely to carry most specific characterizing terms. The ‘brute-force’ abstract-level processing will have difficulty in extracting these terms correctly and consistently. We further notice that the substantial improvements of precision and recall when we include MeSH terms and GO terms. This may be because these two categories are more specific and MeSH and GO annotations were done using full-papers and these biological functions and process are not described in all abstracts. Clustering of genes resulting from microarray experiment To demonstrate the usefulness of the presented keyword-extraction techniques to microarray data analysis, this method was applied to annotate and cluster gene lists that were found differentially expressed in a microarray experiment investigating the impact of two mitogenic proteins, Epidermal growth factor (EGF) and Sphingosine 1-phosphate (S1P), on glioblastoma cell lines [36]. The microarray data set reveals three sets of differentially expressed genes (p<0.05), namely, genes differentially expressed with response to EGF, G(EGF), genes differentially expressed with respect to S1P, G(S1P) and genes differently expressed in response to both, G(COM). Genes were considered differentially expressed if their p-value is smaller than 0.05. We found that, when compared to the resting state, 19 genes were significantly differentially expressed as a response to EGF, 35 genes as a response to S1P and 30 genes as a response to COM, i.e., combined stimuli of S1P and EGF. The three gene lists are referred to as G(EGF), G(S1P) and G(COM), respectively (see Table 6 in supplementary material). Using these the three gene lists obtained from the microarray experiment (Table 6 shown in supplementary material) as query in MEDLINE returned the three corresponding sets of abstracts A(EGF), A(S1P) and A(COM), respectively. The abstracts were processed with the keyword extraction method involving sentence-level, MeSH and GO terms and the resulting representations were clustered using average linkage hierarchical clustering algorithm. Our gene clustering strategy and clustering algorithms are explained in the Methodology section. The resulting clustograms are presented in Figure 2, Figure 3, and Figure 4, respectively. The clustograms depict associations between genes and biological function/process terms derived from the abstracts obtained with the various gene lists. For the investigating scientist, the clustograms fulfill the following main functions: (1) Squares highlighted in a horizontal line link a gene to one or more biological functions or processes. This is useful to see which genes are associated with which functions/processes and which genes have few or many associations. The interpretation of many and few is very much dependent on the associated biological function/process categories, the particular scientific question under investigation, and also on how extensively a particular gene has been researched and reported in the literature. (2) Users may visually delineate clusters, i.e., rectangular areas with many highlighted squares in them and few highlighted squares around them. Any cluster, small or large, is potentially very useful to have discovered. Each cluster identified in this way relates a set of genes to a group of biological functions and processes. In a sense, each gene in the clustered is characterized by the same set of biological function and process concepts, a kind of ‘guilt by association’. This information is extremely useful as it provides clues as to the roles genes may play collectively in pathways and functions, processes, and possible phenotypes, that are associated with these pathways. Summary of analysis of EGF cluster, G(EGF) The clustograms in Figure 2 show the results obtained from extracting the sentence-level function/process keywords (plus MeSH and GO terms) from 28,913 abstracts (for the 19 genes detected in response to EGF stimulus) and the subsequent clustering. In Figure 2a several individual genes with very many (e.g., CALD1, CLU, FOS) and very few (e.g., HRY, DUSP6) associations stand out. Another interesting feature is the large cluster at the lower left corner of Figure 2a (reproduced in more detail in Figure 2b) containing the genes DUSP, ID1, KLF2, CALD1, ABCA, CLU, FOS, JUN and SLC5A3. Many genes in this cluster are associated with the same set of keywords (transcription factor, cell death and secretion). Summary of analysis of S1P cluster, G(S1P) The clustograms in Figure 3 show the results obtained from extracting the sentence-level function/process keywords (plus MeSH and GO terms) from 19,705 abstracts (for the 30 genes detected in response to S1P stimulus) and the subsequent clustering. In Figure 3a several individual genes with very many (e.g., CCL3, IL6, IL8, F3) and very few (e.g., HERB2, DOC1) associations stand out. Another interesting feature is the large cluster at the upper left corner of Figure 3a (reproduced in more detail in Figure 3b) containing the genes TNAIP, KLF5, BCL6, NAB1, BTG1, NFKBIA, NR4A1, SOCS5, CITED2, NRG1, JAG1, PLAU, CCL2, IL8, IL6, GLIPR1, F3, MAP2K3, and EHD1. Many genes in this cluster are associated with the same set of keywords (atherogenesis, mitogenesis, assemble, inflammation, focal-contact, …, and protein-binding). Summary of analysis of the common gene cluster, G(COM) The clustograms in Figure 4 show the results obtained from extracting the sentence-level function/process keywords (plus MeSH and GO terms) from 39,890 abstracts (for the 30 genes detected in response to EFG and S1P stimuli) and the subsequent clustering. In Figure 4a several individual genes with very many (e.g., MYC, MAFF, ATF3) and very few (e.g., DIPA, UGCG, SNARK) associations stand out. Another interesting feature is the large cluster at the upper left corner of Figure 4a (reproduced in more detail in Figure 4b) containing the genes SPRY2, GEM, ZYX, NEDD9, MYC, LIF, SERPINE1, DTR, MUCL1, C8FW, MAFF, ATF3, RTP801, EGR1, JUNB, FOSL1, CEPED, TIEG, EGR2, EGR3, and ZFP36. Many genes in this cluster are associated with the same set of keywords (DNA binding, zinc fingers, repressor proteins, …, and mitosis). An important aim in microarray data mining is to bind transcriptionally modulated genes to functional pathways or to understand how transcriptional modulation can be associated with specific biological events such as genetic disease phenotype, molecular mechanism of drug action, cell differentiation etc. However, the amount of functional annotation available with each transcriptionaly modulated genes is still a limiting factor because not all genes are well annotated. Our functional clustering/grouping will enable to select literally informative genes (Figure 2b, Figure 3b, and Figure 4b) for further investigations in the above data mining and knowledge discovery pipeline. Our evaluation suggests that this approach will provide more specific and useful information than typical approaches using abstract-level information. This is particularly the case when the sentence-level terms are augmented by MeSH and GO keywords. Conclusion The sequencing of whole genomes and the introduction high throughput analysis (e.g., oligonucleotide and cDNA chips, MALDI/SELDI-TOF MS) provides biomedical research with a global perspective, which necessitates the development of novel mining tools to explore and interpret data in timely manner. This paper presents a novel approach to combine sentence-level keywords with GO and MeSH terms. In our evaluation experiment, this approach has shown promising results. The present evaluation suggests that this approach will provide more specific information than typical approaches using abstract-level information. This is particularly the case when the sentence-level terms are complemented by MeSH and GO terms. Further, clustering of genes into different functional groups based on literature keywords has the potential to help biologists identify and characterize literally informative genes of interest for further investigations. Future work Future enhancements of the system will include additional data resources (OMIM. DIP, KEGG) and the generation of association rules to identify correlations among genes in the same cluster. Association rules between the genes in the same cluster seem particularly interesting because it allows one to find the presence of regularities between gene groups. Finally, abstracts were used in this study as they are readily and easily available but they are limited in content. As full-text contains large number of irrelevent sentences compared to abstracts this approach may be useful for full-text analysis too, as it performs filtering of irrelevant sentences before clustering. The plan to perform the current study with full-text articles and compare the results with that of abstracts is on the way. Supplementary material Data 1

Qual_Life_Res-3-1-1915653

Small-cell lung cancer patients are just ‘a little bit’ tired: response shift and self-presentation in the measurement of fatigue

Background Response shift has gained increasing attention in the measurement of health-related quality of life (QoL) as it may explain counter-intuitive findings as a result of adaptation to deteriorating health. Introduction Quality of life (QoL) is considered an important treatment outcome when the treatment intent is not curative but palliative. However, the expected deterioration in QoL often does not occur, even in cases of serious illness. For example, Groen et al. studied patients with inoperable non-small cell lung cancer treated by radiation with and without chemotherapy [7]. QoL was measured with The European Organization for Research and Treatment of Cancer Core Quality of Life Questionnaire (EORTC QLQ-C30) [1], which has been designed specifically for use in clinical trials focusing on cancer patients. However, contrary to expectation, they did not find significant deterioration in the scale scores over the treatment period of 6 weeks. Although cancer patients are willing to undergo risky and toxic treatments [11], it seemed implausible that the side-effects of treatment had not affected their QoL. Other studies also reported counter-intuitive results. For example, patients with a life-threatening disease or disability were found to report stable QoL, and patients with a severe chronic illness reported QoL levels that were not inferior to that of patients with a less severe illness or to healthy patients [2, 3, 5]. Such counter-intuitive findings, labelled by Breetvelt and Van Dam as ‘underreporting of problems’ suggest that patients report less distress and dissatisfaction than they actually feel [5]. In recent years, response shift theory has gained increasing attention in explaining paradoxical and counter-intuitive findings. Response shift refers to a change in internal standards, values and conceptualization of QoL and is recognized as an important mediator in adaptation to changing health [17]. However, despite the explanatory power of response shift theory, our current understanding of phenomena that can complicate the interpretation of QoL scores is still limited. Therefore, we investigated QoL measurement in small-cell lung cancer (SCLC) patients during 1st line chemotherapy. We were quite surprised when we noticed discrepancies between levels of fatigue measured with the questionnaire and answers spontaneously reported during the interview. We therefore investigated these ‘conflicting’ findings in the measurement of QoL in more depth. This paper reports the results of an exploratory longitudinal multiple-case study, in which we focused on how patients responded to the EORTC QLQ-C30 question ‘were you tired’ at different points in their treatment trajectory. We aimed to describe the patients’ explanations when answering the question, and to search for explanations of counter-intuitive findings. Methods Procedures and study sample Between March 2001 and September 2003, we recruited newly-diagnosed patients with SCLC who were evaluated for 1st line chemotherapy. The patients were attending one of five outpatient clinics for chest diseases in the Netherlands. To maximize the likelihood that we would interview patients from the beginning of their treatment, we were informed about new patients immediately after diagnosis. No restrictions were made with regard to age or treatment (chemotherapy or a combination of chemotherapy and radiotherapy). Participating patients gave written consent and were interviewed at equivalent points in the treatment trajectory. The first interview (T1) was carried out at the start of chemotherapy. In the original plan the second interview was planned after completion of the course of chemotherapy. However, after inclusion and first interviews of 3 patients we made a decision to interview the patients during treatment as well. Therefore the second interview (T2) was conducted 4 weeks after T1 and the third (T3) 7–10 days after completion of the treatment with chemotherapy and the fourth (T4) 6 weeks later. Approval for this study was obtained from the Medical Ethics Committees of the research site and the participating hospitals. During the course of the study, 41 eligible patients were invited to the study. Four patients were unwilling to participate, 3 died before informed consent could be obtained, and 3 were not interviewed because of imminent death. Of the 31 respondents who were interviewed, 8 were excluded from further analysis because their data were incomplete, i.e. they were only interviewed once (six died within a month after T1 and two were too sick at T2 and died before the end of the planned chemotherapy). Consequently, the final study sample consisted of 23 SCLC patients, of whom 12 had limited (3 male and 9 female, mean age 55, range 42–69) and 11 had extended disease (8 male and 3 female, mean age 64, range 39–72). All patients received standard chemotherapy, except for 7 patients whose chemotherapy was combined with local radiation of the tumour. The majority of the patients were married (19, 83%), and had children (17, 74%). Of the 23 patients in our study sample, 15 were interviewed four times, 7 were interviewed three times, and one patient was only interviewed twice resulting in a total of 83 interviews. The interviews were conducted by MW in the homes of the patients. In three cases the 1st interview was held in the hospital. Interviews averaged 80–110 min. Materials and qualitative method In this exploratory, longitudinal multiple-case study, QoL was assessed with EORTC QLQ-C30 (version 3.0) [1] and the lung cancer module QLQ-CL13 [4]. The EORTC QLQ-C30 is the most widely used cancer-specific QoL instrument in European clinical trials. The questionnaire is composed of several scales, which measure among others physical function, mental health, general health and global QoL. Furthermore, it measures different symptoms such as pain, dyspnoea, nausea and fatigue. The Fatigue Scale consists of 3 items: ‘did you need to rest’, ‘have you felt weak’ and ‘were you tired’ (respectively questions 10, 12 and 18), for which there are 4 response categories: ‘not at all’, ‘a little’, ‘quite a bit’ and ‘very much’ (respectively scores 1, 2, 3 and 4). The EORTC QLQ-C30 and CL13 were completed in combination with the Three-Step Test-Interview (TSTI) to investigate how respondents interpreted the items and how they responded to them. The TSTI consists of the following steps [8]: (1) concurrent think aloud, aimed at collecting observational data on how a respondent completes the questionnaire, expressing his thoughts aloud; (2) focused interview, aimed at clarifying respondents’ previous expressions while completing the questionnaire; (3) semi-structured interview, aimed at eliciting respondents’ experiences and opinions with regard to the questionnaire. Interview protocol Each interview was conducted in an identical format. At T1, QoL was measured with the EORTC QLQ-C30, followed by the lung cancer module QLQ-CL13. The questionnaire was conducted in a concurrent think aloud manner and after completion, respondents were asked to clarify previous hesitations, expressions when rating certain items and experiences (i.e. second and third step of the TSTI). Individual QoL was then measured with the Schedule for the Evaluation of Individual Quality of Life—Direct Weighting (SEIQoL-DW) [13, 14]. Finally, we encouraged patients to talk freely about the impact of diagnosis and treatment. At follow-up (i.e. T2, T3 and T4), after the SEIQoL-DW assessment, the EORTC questionnaires were administered a second time as a so-called ‘then-test’ [15, 16, 18] (i.e. the patients filled out the questionnaire in reference to how they perceived themselves as they were in the previous interview). In these 2nd and following interviews, EORTC assessments were conducted in a concurrent think aloud manner and with the second step of TSTI integrated in the assessment. In fact, we encouraged patients to think aloud and we probed for clarification after each item in the case of extra information was considered useful to understand patients’ answer. We used a flexible approach in order not to interrupt the natural flow of both the assessment as well as the patient-interviewer communication. The interviews were audio-taped and transcribed verbatim. In this article we focus on the EORTC QLQ-C30 question ‘were you tired’. Analysis Three types of data were collected and used for analysis [12]: (1) completed EORTC questionnaires (T1–T4), (2) observed respondent behaviour recorded in field notes, and (3) transcriptions of the interviews, including ‘think aloud’. The analysis was aimed at identifying discrepancies, response strategies and explanations of response behaviour. We used the qualitative computer package Kwalitan 5.0 (http://www.kwalitan.net) to extract relevant parts of the transcriptions (1) ‘think aloud’ of the question ‘were you tired’ and of other items that were useful in understanding the response behaviour to the question on fatigue (i.e. two other items of the fatigue scale, the general health and the global QoL question of the QLQ-C30), and (2) ‘comments’ related to fatigue symptoms, impact of treatment on perceived QoL, and attitudes towards life. In order to deal with the still remaining large amount of extracted data, two of the authors (MW, AT) condensed extracted transcripts of the ‘comments’ into core texts. For each patient, the data (i.e. think aloud combined with scores and core texts of comments) were organized per interview in one mind map (see example in Fig. 1) by means of the computer package Mindjet Mindmanager Pro 6 (http://www.mindjet.com). Fig. 1Analysis by means of a mind map. Branch EORTC: scores of GH/QOL and the fatigue scale are organized per interview and complemented by the think aloud data. Branch Comments: core texts of relevant parts of transcripts are organized per code and per interview. Note: (+) not all branches of the different interview moments are shown Furthermore, a different mind map was made to organize think aloud data related to the question ‘were you tired’ for all patients per response category per assessment, including then-test. For the analysis, three authors (MW, AT, TH) each independently read the mind maps of each patient. They studied patients’ scores, their think aloud responses, and examined whether response shift type explanations would be provided: recalibration (i.e., using different standards of comparison to assess fatigue over time), reprioritization (i.e., changes in the importance attached to fatigue over time) and reconceptualization (i.e., changes in the meaning of fatigue over time). Two researchers (MW, AT) searched for additional explanations in the core texts to account for the response behaviour and the discrepancies. The research team (MW, AT, TH, MS) discussed critically the different response strategies used by the patients and the robustness of the interpretations of response shift. Results Patients with or without discrepancies Of the 23 patients, 15 (5 male and 10 female, age 46–72) showed discrepancies at least at one measurement point, i.e. differences between their answer to the EORTC question ‘were you tired’ and their level of fatigue spontaneously reported during the interview. In their answers to the EORTC questionnaire they all presented themselves positively and said that they were not tired. Mary, for example, was 60 years old at the time of the first interview. During all her interviews she reported that she was tired, but she consistently answered ‘not at all’ to the EORTC question (Box 1). Only once she scored ‘a little bit’. This was at T2 when filling in her questionnaire as a then-test for T1 and she commented “a little, more than at the moment”. Box 1Example of a patient with discrepancies in reported level of fatiguePatient MaryMary was 60 years old and married. She had two sons and two grandchildren. Her answer to the EORTC question was consistently ‘not at all’, except for the then-test concerning the interview T1. This suggests that she did not suffer from fatigue in the week prior to the interviews and that fatigue due to chemotherapy did not have any impact at all on her energy level. But, during the interview she spontaneously provided information that indicated that chemotherapy had an impact on her life and that she regularly suffered from fatigue.T1 EORTC score ‘not at all’Think aloud: Were you tired... no not at all, no, no difference compared to the pastInterview: I’m getting tired at the least little thingT2 EORTC score ‘not at all’Think aloud T2: Not tired, last week, not at allThink aloud then-test T1: a little, more than at the moment.Interview: According to the doctors, the X-rays were very good. I’m very optimistic, sometimes I’m tired but that’s my own fault. I don’t have as much energy as I did before I became ill.T3 EORTC score ‘not at all’Think aloud: Last week I wasn’t tired, it’s the second week after my chemo, not at all tiredThink aloud then-test for T2:not at allInterview: Yes, it was my last cycle of the chemo, I was afraid that I wasn’t going to be able to carry on through the treatment. But I managed, okay; I’m tired but apart from that... nothing at all.T4 EORTC score ‘not at all’Think aloud: No I wasn’t tired last week, not at allThink aloud then-test for T3:not at allInterview: Now and then, I’m tired. It’s different to before my chemotherapy. Sometimes I’m so tired, so tired, more than in the past. It comes suddenly...in the middle of the day. For many of our respondents, every new cycle of chemotherapy had a more severe impact on their energy level. Therefore, the highest level of fatigue was to be expected after the 5th and last cycle of chemotherapy, at T3. However, the 15 patients with discrepancies all answered the question ‘were you tired’ with ‘not at all’ or ‘a little’. For example, during her interviews at T2 and T3, Ann reported the growing impact of every cycle of chemotherapy, but her answer to the EORTC question at both interviews was ‘a little’ (Box 2). Box 2Example of a patient using different response strategies, comparison with more sick patients, response shift and self-presentationPatient AnnAnn was 47 years old and living with a partner. She didn’t have any children.Her scores suggest that the chemotherapy had a slight impact on her energy level during her treatment and a greater impact 6 weeks after completion of the treatment. But, the interview and the think aloud provided information that indicated that the chemotherapy had a growing impact on her life, and the score ‘quite a bit’ at T4 was the result of bad news (i.e. a recurrence of the tumour). Furthermore, her data show examples of different response strategies, comparison with more sick patients and self-presentation.T1 EORTC score ‘not at all’Think aloud: Were you tired.. in principle I wasn’t tired. I was mentally tired, it costs me a lot of energy to talk with my relatives. I think you mean physically tired. You’re tired in the sense that you can hardly put one foot in front of the other. That’s being tired. No I’m not tired.Interview: I want to be realistic, think positively. I try not to worry; there is nothing I can do. I don’t know when I will die. I get angry when people are surprised...if I say I’m doing fine...it’s my decision how I’m feeling.T2 EORTC score ‘a little bit’Think aloud: I feel it a little bit, compared to other people who are very sick. So, if I have pain or when I’m tired I say to myself don’t complain, so everything I feel, I only feel a little. I’m doing fine.Think aloud then-test for T1:a littleInterview: Yesterday, I worked for three hours and I was exhausted. I went to bed in the middle of the day in order to be able to show my friend that I’m doing fine. She has trouble in coping...me... having cancer.T3 EORTC score ‘a little bit’Think aloud:A little, yes because I have the feeling that I was able to get over itThink aloud then-test for T2: I think that I’ve said a little last time, it felt a little, but I shift my limit. Interview: It’s a kind of tiredness, I don’t know. I’ve never been like that. To allow yourself to be tired. I think that the story they all tell, that the last cycle of chemo has the most impact, I think that’s very very true.T4 EORTC score ‘quite a bit’Think aloud: Yes, I was quite a bit tired, but only mentally tired. I have to adjust to the idea of a new course of treatment and radiation. My health is excellent. I’m able to do everything I like, better than 6 weeks ago. But, with all the medicine I’m taking to suppress the epileptic fits... I’m scared, just like after the start of chemo.Think aloud then-test for T3:a little but I am not really sure, the pain which I had in my ankels made me tired.Interview: The radiation will make me tired, but it has not started yet, so I am not tired yet. Patients with discrepancies were identified in both stages of disease and with both treatment regimens, i.e. 12 patients (LD n = 5, ED n = 7) treated with chemotherapy and 3 LD patients treated with chemotherapy and radiotherapy. During the course of the treatment (T1–T4, 55 interviews) they answered the question “were you tired” 20 times with ‘not at all, 29 times with ‘a little’, 4 times with ‘quite a bit’ and twice with ‘very much’ (see individual scores per interview in Table 1) Table 1Individual (then-test) scores of patients answering the EORTC QLQ-C30 question ‘were you tired’. Response categories 1, 2, 3 and 4 are representing respectively the category ‘not at all’, ‘a little’, ‘quite a bit’ and ‘very much’. Small-cell lung cancer patients (n = 23), limited (LD) and extended (ED) disease receiving 1st line chemotherapy were interviewed at equivalent points in treatment: at start of chemotherapy (T1), 4 weeks later (T2), at end of chemotherapy (T3), and 6 weeks later (T4). T1t, T2t and T3t are representing then-test scores obtained at respectively T2, T3 and T4, when patients are asked to provide a renewed evaluation of their fatigue at the previous assessment. Two groups were identified: patients with (n = 15) and without (n = 8) discrepancies between their questionnaire answer and fatigue spontaneously reported in the interviewPatients’ characteristicsNr.M/FAgeLD/EDT1T1tT2T2tT3T3tT4DiscrepanciesP 02Male57ED33––1––P 04Female50LD1223232P 08Female69ED2212122P 09Male66ED3222222P 10Male46LD2222222P 12Female47LD1222223P 15Female69LD1223232P 17Female64ED22222––P 18Male72ED112–1–3P 21Male69ED2223222P 22Male55LD411–2––P 24Female56LD1112121P 26Female59LD42122––P 32Female60LD1211111P 34Female51LD1121211No discrepanciesP 01Female42LD22––443P 03Female64ED31––334P 13Male72ED332332P 14Male39LD43223––P 16Male68LD1133332P 20Female44LD1–4–4–4P 27Male69LD322––42P 29Male63ED2332332 No discrepancies were identified in 8 of the 23 patients (LD n = 4, ED n = 4, age 39–72). They answered the question ‘were you tired’ at the end of chemotherapy (T3) with ‘quite a bit’ or ‘very much’. During the course of treatment (T1–T4, 28 interviews) these patients answered the question twice with ‘not at all’, 9 times with ‘a little’, 11 times with ‘quite a bit’ and 6 times with ‘very much’ (Table 1). Then-test scores were dissimilar with scores of the previous assessment in 25 out of 52 cases, with higher then-test scores in 16 cases. Transcripts showed that patients had difficulty remembering either the previous measurement point and/or their fatigue at that time. Response strategies for the four response options The think aloud texts for the response categories ‘quite a bit’ and ‘very much’ were minimal in the entire study population. Only a few patients reacted briefly during the think aloud, e.g. “quite a bit, too tired to keep my eyes open”, “yes quite a bit, very tired” and “next week it will be better...very much”. The same pattern was found in all patients for all four response options when filling in the questionnaire as a then-test, e.g. “a little bit, I think”, “tired then, no” and “very much”. However, patients in the group with discrepancies had much more to say in the conventional QoL measurement when choosing the options ‘not at all’ and ‘a little’. During ‘think aloud’, they seemed to justify the chosen response category. They used various strategies to moderate the impact of fatigue on their life. We summarized their strategies in four categories: (1) I am not tired all the time (e.g. “only in the afternoon”); (2) I am not really tired, it’s something else (e.g. “ it’s the flu”); (3) I have no problems with it (e.g. “I can still cope with it; I don’t want to exaggerate”); (4) I am a little bit tired but it is due to something else (e.g. “I didn’t have a proper meal”). Many of our respondents said that they had expected to become very tired as a result of the treatment, but that they were not as sick as they had expected. They were very happy that they were able to cope with the treatment, and had adjusted to the situation. A male patient, for example, had a score of ‘not at all’ at T2, and explained: “I’m currently doing nothing, so I’m not tired”. The respondents indicated that they wanted to be honest when filling in the questionnaire and did not want to lie or to exaggerate their fatigue and, because many patients were not tired all the time they considered a score of ‘not at all’ or ‘a little’ to be a suitable score (Box 3). Box 3Examples of different response strategies used by patients with discrepancies (n = 15)Think aloud about the question ‘were you tired’I’m not tired all the timeI’m only tired in the afternoonNo not at all, I’m not tired at the moment, it comes suddenlyI have to be honest, sometimes I’m tired, I can’t say not at all, otherwise I would be lyingI’m not really tired, it’s something elseNo not tired, it’s the flue, that’s why I’m tiredIt’s not being tired you know, it’s more like being restlessI’m not physically tired, I’m mentally tiredActually, I can’t be tired because the Hb level in my blood is okayI’ve no problems with itI’m currently doing nothing, so I’ve no problems, I’m not tiredOf course, you can make yourself tired, but I’ don’tI can still cope with it; I don’t want to exaggerateI’m a little bit tired but it’s due to something elseA little, but it was my own fault, I did too muchI didn’t have a proper meal, that’s why I was tiredI didn’t have my lady working for me in the house, she went on holidayIt’s because I’ve got problems with my voice caused by the radiation Optimism Most patients told the interviewer regularly that they had adapted to the situation and had changed their attitude towards a more optimistic perspective. They were not hopeful immediately after diagnosis, but optimism about recovery increased when the tumour was shrinking. Of the 23 patients 17 reported spontaneously that they were optimistic: e.g. “I’ve got good news, I’m as optimistic as can be”, “I’m full of hope because I was diagnosed in an early stage, so I’m good in time”. Furthermore, they said that they felt better off than expected, compared to patients who were worse off: e.g. “I’m lucky not to be very sick, compared to the patients I saw at the hospital”. Although they experienced the impact of every new cycle of treatment as more severe, they said to be able to cope with the treatment and to accept the side-effects: e.g. “it’s part of the package, I’m willing to put up with, knowing the chemo is doing the job properly”. Some patients were actually feeling better after each cycle and happy that they were still alive: e.g. “It was much worse than I wanted to admit last time, I’m feeling much better”, “I’ve already a couple of months extra”. In contrast, 5 of the 23 patients expressed pessimistic feelings: e.g. “I’m a broken man, hard work all my life and now...I don’t think I’ve much time left”, “I’m a bit depressed, when does it stop, if it doesn’t stop it would be better if my life was over”, “No plans for the future, you never know when the tumour will come back”. These patients all reported high levels of fatigue during the course of the treatment. One patient did not provide specific comments about optimism or pessimism. Response shift and self presentation In both groups—with and without discrepancies—we found patients who had reported to have changed their reference point after T1 (i.e. recalibration in contrast to T1). They compared their fatigue at the second and following interviews with that of other patients e.g. “I was tired, yes, but compared to the patients I’ve seen in the hospital, I’m just a little bit tired” or, with the period in which they were more tired (e.g. “Compared to the first week after chemo, it’s the second now ... I’m not tired”). In one case, a patient spontaneously re-evaluated her previous measurement: “I told you that I was really tired then, but compared to how I’m feeling now, it was then just a piece of a cake”. Another one spoke about a shift of limits which also suggests recalibration: “I already told you that I would change my standards”. We did not find indications of reconceptualization and reprioritization of fatigue. The only exception was Ann who made a distinction between being physically and mentally tired (see Box 2). At T1, she said that she was mentally tired but not physically and her answer was ‘not at all’: “Tired means that you can hardly put one foot in front of the other”. At T4, she said that she had recovered from chemotherapy and was physically able to do anything she wanted but, unfortunately, suffered from sudden epileptic attacks caused by metastases. She was feeling anxious in the same way as at the start of her chemotherapy, and had to consider further treatment options. Just like in the interview at T1, she said that she was mentally tired but not physically. However, this time her score was ‘quite a bit’ instead of ‘not at all’: “Purely, because I was mentally tired last week. I have to adjust to the idea of a new course of treatment. Actually, I have to admit that I really am the cancer patient I never wanted to be”. This response pattern might be interpreted as reprioritization (i.e., changes in the importance attached to mental fatigue over time). Box 4Examples of coping strategies used by patients with discrepancies (n = 15)Spontaneously reported coping behaviourProtective behaviourI’m trying to avoid or minimize pessimistic thoughtsI don’t think about it, otherwise I can’t cope with itWe don’t talk about it, just follow my every day routineI’m building a wall around myselfAssertive behaviour/power displayI’ll show others that I’m managing all rightYou have to be positiveYou have to believe in yourself, otherwise you can’t manage it anymoreFighting the stigmaI’m not the cancer patient my neighbour thinks I amThey think I’m lying on my bed all dayPeople look at me, and give advice that I don’t wantI have to admit that I really am a cancer patient... I didn’t want to be (see patient Ann, Box 2). Because the above mentioned response shift type explanations could not adequately explain our conflicting findings in the discrepancy group we questioned: “Why are patients presenting themselves in the questionnaire more positively than in the informal interview”. In our search for an other explanation, we found that 13 of the 15 patients with discrepancies had spontaneously reported how they dealt with having cancer and the perspective of a short life-expectancy. We summarized their comments in three categories (see examples of coping strategies in Box 4): (1) Protective behaviour (e.g. protecting themselves from harmful thoughts); (2) Assertive behaviour/power display (e.g. projecting the image of being positive and managing all right); (3) Fighting the stigma (e.g. fighting against being stigmatized). Taking these strategies into account, we concluded that a possible mechanism underlying the discrepancies in this group was ‘self-presentation’. As the questionnaires are explicitly related to cancer and since these patients want to distance themselves from being reduced to only a cancer patient, they want to present themselves as a person who just happened to have cancer. Therefore, they applied various strategies to respond to the question on fatigue in order to produce a score that was as favourable as possible and presented themselves as positive and managing their fatigue. Discussion Two third of the patients showed discrepancies in their reported level of fatigue. They reported a gradual decrease in energy at the end of chemotherapy, but they were ‘not at all’ or just ‘a little bit’ tired according to their answer to the EORTC questionnaire, with underreporting as a result. They presented a positive image of themselves and used various strategies to explain their choice of response category. A predominant finding was that patients adopted a more optimistic perspective on the treatment. Interestingly, this was not exclusively found in the discrepancy group. The same was true for recalibration and for the only indication of reprioritization. These response shift type explanations did not sufficiently account for the conflicting findings in our discrepancy group. Self-presentation was found to be an additional (coping) mechanism underlying the discrepancies. Our results suggest that patients are not only concerned about the impression they make on others. They try to protect themselves from negative thoughts and they also feel the need to be positive and to distance themselves from the stereotypical cancer patient. With this strategy they are more capable of coping with a situation that they cannot change. The suggestion that self-presentation is an underlying mechanism is supported, for example in the case of Ann. After a recurrence of the tumour she adopted the realistic perspective by admitting that she really was ‘the cancer patient’, which she did not want to be before. It seems that she had given up her attitude of showing others that everything was all right, and for the first time she did not present her self as more positive than she actually was as she did before. Self-presentation (also called impression management [6, 9]) is a phenomenon described by Leary et al. in relation to health behaviour [10]. They discussed its implications for research in health psychology. Our study shows that, in addition to response shift, self-presentation may explain unexpected results, at least in SCLC patients. The question ‘were you tired’ in the EORTC-QLQ-C30 does not unequivocally measure the impact of chemotherapy on the energy level of patients; in fact, with their responses, patients seem to show how that they are managing the situation. From our results we cannot conclude that in the group without discrepancies self-presentation is not present at all, or that whenever self-presentation occurs discrepancies will also be present. However, our study does show that self-presentation affects QoL measurement. These findings must be taken into account when investigating and interpreting QoL data, also in other study populations. Especially after diagnosis and in the initial phase of treatment, self-presentation might be an important coping strategy. In fact, during each phase in which a new equilibrium and a new identity has to be found (e.g. after a recurrence of the tumour, or metastases) self-presentation might affect QoL measurement.

Virchows_Arch-3-1-1888716

Expression microarray analysis of papillary thyroid carcinoma and benign thyroid tissue: emphasis on the follicular variant and potential markers of malignancy

The most common sub-variant of papillary thyroid carcinoma (PTC) is the so-called follicular variant (FVPTC), which is a particularly problematic lesion and can be challenging from a diagnostic viewpoint even in resected lesions. Although fine needle aspiration cytology is very useful in the diagnosis of PTC, its accuracy and utility would be greatly facilitated by the development of specific markers for PTC and its common variants. We used the recently developed Applied Biosystems 1700 microarray system to interrogate a series of 11 benign thyroid lesions and conditions and 14 samples of PTC (six with classic morphology and eight with follicular variant morphology). TaqMan® reverse transcriptase-polymerase chain reaction was used to validate the expression portfolios of 50 selected transcripts. Our data corroborates potential biomarkers previously identified in the literature, such as LGALS3, S100A11, LYN, BAX, and cluster of differentiation 44 (CD44). However, we have also identified numerous transcripts never previously implicated in thyroid carcinogenesis, and many of which are not represented on other microarray platforms. Diminished expression of metallothioneins featured strongly among these and suggests a possible role for this family as tumour suppressors in PTC. Fifteen transcripts were significantly associated with FVPTC morphology. Surprisingly, these genes were associated with an extremely narrow repertoire of functions, including the major histocompatibility complex and cathepsin families. Introduction Papillary thyroid carcinoma (PTC) is the most common endocrine malignancy and encompasses a variety of morphological/architectural variants, all of which are characterized by a distinctive nuclear appearance. In recent years, PTC has become an important paradigm of solid tumour molecular pathogenesis principally arising from intensive investigation prompted by the Chernobyl accident. The discovery of ret rearrangements [12, 13, 35] and their association with radiation [42] was followed by the demonstration of the BRAF V600E mutation [21, 30] found more commonly associated with sporadic PTC in non-radiation exposed populations [22, 32]. In the past, our group and others have noted an association between classic morphology and the BRAF V600E mutation and between variant morphology and ret rearrangements particularly ret/PTC-3 [11, 39]. Similarly ret/PTC-3 appears to strongly correlate with the solid/follicular variant seen commonly in children exposed to the Chernobyl fallout [42]. Nevertheless, the utility of these genetic lesions to diagnostic pathology and clinical practice has remained negligible. Recently, gene expression microarray technology has been used to attempt to identify clinically relevant biomarkers of malignancy related to the thyroid [2, 6, 9, 10, 20, 28]. The discovery of such a biomarker or panel of biomarkers allied to the gold standard triage method of fine needle aspiration cytology (FNAC) would represent a significant advancement in the treatment of the solitary thyroid nodule. An intriguing but commonly occurring variant of PTC is known as follicular variant (FVPTC). This lesion, which by definition retains the classic nuclear features of PTC, shows no evidence of the architectural papillae. FVPTC may be a controversial lesion due to interobserver variation in its pathological diagnosis [25]. Further, the occurrence of follicular patterned lesions with poorly or incompletely developed nuclear features may occur, which are easily dismissed as benign thyroid nodules. This has led to the controversial designation “well-differentiated tumour of uncertain malignant potential” for tumours of this type [45]. It is clear that this is a complex and contentious area, and that further work needs to be done to ascertain the underlying molecular biology of this particular variant. Recently, inroads into elucidation of molecular pathways underpinning PTC have been carried out using microarray studies. The overriding objective of these investigations was to identify clinically useful biomarkers. However, the majority of these studies have analysed PTC as though it were a homogenous singular entity without deference in a detailed manner to sub-variants and, in particular, the most common variant (FVPTC). The identification of specific biomarkers of FVPTC and a deeper understanding of its origins are clearly warranted. The aim of this expression microarray study using a novel microarray platform was twofold: to identify markers that distinguish PTC from benign thyroid tissue and lesions and, secondly, to identify potential markers and further explore the molecular pathology of FVPTC. Materials and methods Patients and tissue samples Tissue from 25 thyroid resections was collected prospectively from patients undergoing partial or total thyroidectomy for a variety of reasons at St. James’s Hospital, Dublin. The study had approval of the local ethics committee and informed consent was obtained from each patient by the clinical team before surgery. Small samples (<1 cm) were divided and immediately snap-frozen in liquid nitrogen for storage at −80°C until use. Histopathological examination of formalin-fixed paraffin-embedded sections was performed by two pathologists (SF and MT) for diagnostic categorisation. Classification of neoplastic tissue was made according to a recognised system [24]. The cohort comprised 11 benign lesions or conditions including follicular adenoma, nodular goitre, normal thyroid tissue, and Graves’s thyroiditis. The remaining 14 samples were diagnosed as PTC and comprised six classical morphology PTC and eight FVPTC (see Table 1). Immediately before RNA extraction, frozen sections were cut and stained to confirm the presence of representative lesional tissue with morphology corresponding to that noted in the diagnostic formalin-fixed paraffin-embedded sections. Table 1Sample cohortIdentifierDiagnosisN1Normal thyroid tissueN2Normal thyroid tissueN3Lymphocytic thyroiditisN4Nodular hyperplasiaN5Follicular adenomaN6Nodular hyperplasia with focal lymphocytic thyroiditisN7Nodular hyperplasiaN8Follicular adenomaN9Nodular hyperplasiaN10Follicular adenomaN11Grave’s thyroiditisT1Solid/FVPTCT2FVPTCT3PTC classic morphologyT4FVPTC-oxyphilT5FVPTCT6FVPTCT7PTC classic morphologyT8FVPTCT9PTC classic morphologyT10FVPTCT11PTC classic morphologyT12FVPTCT13PTC classic morphologyT14PTC classic morphologyList of the 11 benign and 14 malignant lesions that were used in this study.FV follicular variant; PTC papillary thyroid carcinoma RNA isolation and characterization Samples were ground in liquid nitrogen and homogenised in RLT buffer (Qiagen, UK). RNA was then extracted using the RNeasy Mini Kit with optional on-column RNase-free DNase digestion (Qiagen) according to the manufacturer’s instructions. RNA quantity was determined using UV spectrophotometry. RNA quality was assessed using the RNA 6000 Nano LabChip® Kit in conjunction with the Agilent 2100 Bioanalyzer (Agilent Technologies, Waldbronn, Germany). Microarray analysis Applied Biosystems Human Genome Survey Arrays were used to analyse the transcriptional profiles of thyroid RNA samples in this study. Digoxigenin-UTP-labelled cRNA was generated and linearly amplified from 5 μg of total RNA using Applied Biosystems Chemiluminescent RT-IVT Labelling Kit v 2.0 using manufacturer’s protocol. 10 μg of labelled cRNA were hybridized to each pre-hybed microarray in a 1.5-ml volume at 55°C for 16 h. Array hybridization and chemiluminescence detection were performed using Applied Biosystems Chemiluminescence Detection Kit following manufacturer’s protocol. Images were collected for each microarray using the 1700 analyser. Images were auto-gridded and the chemiluminescent signals were quantified, corrected for background and spot, and spatially normalized. TaqMan® PCR validation Sufficient RNA remained from 20 of the initial 25 samples for TaqMan® polymerase chain reaction (PCR) validation in a series of 50 targets. RNA was reverse transcribed using a High Capacity cDNA Archive Kit (Applied Biosystems, CA, USA). Primers and probes for TaqMan® PCR were obtained by using Applied Biosystems’ pre-designed TaqMan® Gene Expression Assays. PCR was carried out using an ABI PRISM 7900 Sequence Detection System (Applied Biosystems). Analysis of relative gene expression data was performed using the ΔΔCT method [23] with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as an endogenous control/reference assay. Statistical analysis Microarrays were analysed using Spotfire DecisionSite™ for Functional Genomics (Spotfire AB, Goteborg, Sweden) and R version 1.9.1, a free language and environment for statistical computing and graphics (R Development Core Team, 2004). Arrays were initially normalized, and genes were deemed undetectable and, therefore, excluded from final gene lists if they had a signal-to-noise ratio of less than three (S/N < 3) in greater than 18 of the 25 arrays. An ANOVA test was used to generate p values for statistical differences between groups. Their p values were then adjusted for multiple comparisons using the technique described by Benjamini and Yekutieli [3]. Genes were deemed statistically different between groups if they had an adjusted p < 0.05 and an average fold-change difference of greater than 2. Hierarchical clustering was performed based on the statistically different genes to determine whether samples grouped appropriately. Gene ontology analysis was performed using an online database known as the Panther classification system (http://www.pantherdb.org). Correlations between microarray and TaqMan® expression data were measured using the Pearson coefficient. Results Unsupervised clustering of all 25 samples demonstrated clustering into two major groups (data not shown) comprising the N group and the T group (see Table 1). There was no tendency for FVPTC to cluster independently with classic morphology PTC, confirming the close relationship of these variants of PTC. To identify potential markers of malignancy, ANOVA with false discovery rate correction was used to compare the benign and malignant thyroid cohorts. A p value cut-off of <0.05 and fold-change difference of ≥2 yielded 236 statistically significant probes. Of these, 172 corresponded to fully annotated probes and are listed in Table 2. Table 2Genes differentially expressed in malignant vs benign thyroid tissueGene NameGene SymbolAdjusted p value1700 probe IDGenes up-regulated in malignant vs benign Active BCR-related geneABR0.014482154399 Adaptor-related protein complex 2, alpha 1 subunitAP2A10.0312115368 Apoptosis, caspase activation inhibitorAVEN0.030998203738 BCL2-associated X proteinBAX0.009782146510 BH3 interacting domain death agonistBID0.021424131216 Brain abundant, membrane attached signal protein 1BASP10.024214198318 Brain acyl-CoA hydrolaseBACH0.014566133876 Bromodomain adjacent to zinc finger domain, 1ABAZ1A0.03538209809 Calcium/calmodulin-dependent protein kinase ICAMK10.041887157712 Cathepsin SCTSS0.046544105790 CD44 antigen (homing function and Indian blood group system)CD440.044181133604 Chemokine (C-X-C motif) ligand 16CXCL160.043234199059 Chromosome 1 open reading frame 38C1orf380.049072202924 CLIP-170-related proteinCLIPR-590.023629102205 Docking protein 1, 62 kDa (downstream of tyrosine kinase 1)DOK10.041898204989 Epidermodysplasia verruciformis 1EVER10.003348175569 FXYD domain containing ion transport regulator 5FXYD50.01444154607 FXYD domain containing ion transport regulator 5FXYD50.023629112771 Galactose-4-epimerase, UDPGALE0.047363141143 Genethonin 1GENX-34140.016836124360 Hypothetical gene BC008967BC0089670.015683108526 Hypothetical protein FLJ10849FLJ108490.013822224983 Hypothetical protein FLJ22531FLJ225310.024391145918 Hypothetical protein MGC4607MGC46070.006507211836 Intercellular adhesion molecule 1 (CD54), human rhinovirus receptorICAM10.028746109070 Jun dimerization protein p21SNFTSNFT0.043301144215 Lectin, galactoside-binding, soluble, 3 (galectin 3)LGALS30.034491179836 Major vault proteinMVP0.0312212354 Matrix metalloproteinase 14 (membrane-inserted)MMP140.038682152076 Milk fat globule-EGF factor 8 proteinMFGE80.02392144588 Mst3 and SOK1-related kinaseMST40.042028112198 neuronal cell adhesion moleculeNRCAM0.011178106462 Phospholipase D3PLD30.034491143388 Promyelocytic leukemiaPML0.016018217558 Protein inhibitor of activated STAT protein PIASyPIASY0.00536153434 Protein tyrosine phosphatase, receptor type, EPTPRE0.048653221568 Rho GDP dissociation inhibitor (GDI) betaARHGDIB0.043853143589 S100 calcium binding protein A11 (calgizzarin)S100A110.019933145550 Similar to rat tricarboxylate carrier-like proteinBA108L7.20.025387179870 SP110 nuclear body proteinSP1100.0312113484 Stimulated by retinoic acid gene 6FLJ125410.043234193986 Syndecan 3 (N-syndecan)SDC30.048804143980 Tax interaction protein 1TIP-10.006673119665 TBC1 domain family, member 2TBC1D20.029907205982 Tenascin C (hexabrachion)TNC0.032355143831 Thymosin, beta 4, Y chromosomeTMSB4Y0.040937193911 Tissue inhibitor of metalloproteinase 1 (erythroid potentiating activity, collagenase inhibitor)TIMP10.032306134692 Topoisomerase (DNA) II alpha 170 kDaTOP2A0.040937135302 Transforming growth factor, beta 1TGFB10.016836170749 Transgelin 2TAGLN20.031971172296 Tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, eta polypeptideYWHAH0.039009188379 v-yes-1 Yamaguchi sarcoma viral related oncogene homologLYN0.016175194134Genes down-regulated in malignant vs benign Aldehyde oxidase 1AOX10.003227106573 Ankyrin 2, neuronalANK20.035357155780 Aspartate beta-hydroxylaseASPH0.002064221656 Aspartate beta-hydroxylaseASPH0.019666114180 ATPase, Cu++ transporting, beta polypeptideATP7B0.044511198852 Brain-specific protein p25 alphap250.023629120622 Casein kinaseLOC1494200.022382149347 Cellular retinoic acid binding protein 1CRABP10.008315100295 Centromere protein JCENPJ0.0312164563 Ceroid-lipofuscinosis, neuronal 5CLN50.011021205999 Chloride channel KbCLCNKB0.040418176266 Chondroitin beta1,4 N-acetylgalactosaminyltransferaseChGn0.013148101140 Chromosome 11 open reading frame 8C11orf80.001977174025 Chromosome 11 open reading frame 8C11orf80.004148108279 Chromosome 21 open reading frame 4C21orf40.0042156895 Clusterin-like 1 (retinal)CLUL10.019631186062 Component of oligomeric golgi complex 3COG30.003664129212 Coxsackie virus and adenovirus receptorCXADR0.004648108284 Crystallin, alpha BCRYAB0.030418190274 Cytosolic sialic acid 9-O-acetylesterase homologCSE-C0.040993213856 Dicarbonyl/l-xylulose reductaseDCXR0.001977103350 DnaJ (Hsp40) homolog, subfamily B, member 4DNAJB40.043853103618 ERO1-like beta (S. cerevisiae)ERO1LB0.013962207998 Extracellular link domain containing 1XLKD10.039738195865 Family with sequence similarity 13, member A1FAM13A10.019631116936 Fatty acid binding protein 4, adipocyteFABP40.014832150137 Fc fragment of IgG binding proteinFCGBP0.001965118361 Fibroblast growth factor receptor 2FGFR20.0042110548 FLJ35740 proteinFLJ357400.020224101102 Friedreich ataxia region gene X123X1230.032602133505 Glutamate-ammonia ligase (glutamine synthase)GLUL0.014649175147 Glycine amidinotransferase (l-arginine/glycine amidinotransferase)GATM0.013962111904 Glycoprotein M6AGPM6A0.011739215326 Growth hormone receptorGHR0.017721190306 HLA complex group 4HCG40.025807191199 Hypothetical protein BC009561LOC1197100.003227211319 Hypothetical protein BC019238LOC1203790.013438201200 Hypothetical protein FLJ13204FLJ132040.003227145066 Hypothetical protein FLJ13842FLJ138420.016448208504 Hypothetical protein FLJ14054FLJ140540.049072202017 Hypothetical protein FLJ20154FLJ201540.014428143310 Hypothetical protein FLJ20513FLJ205130.019493154130 Hypothetical protein FLJ32110FLJ321100.015507229492 Hypothetical protein FLJ32343FLJ323430.012208116902 Hypothetical protein FLJ33516FLJ335160.03965224600 Hypothetical protein FLJ37549FLJ375490.001956218577 Hypothetical protein FLJ39378FLJ393780.005853163223 Hypothetical protein FLJ40021FLJ400210.023629174198 Hypothetical protein LOC134285LOC1342850.018694163671 Hypothetical protein MGC10946MGC109460.022382195982 Hypothetical protein MGC14425MGC144250.015445161569 Hypothetical protein MGC17299MGC172990.026062168452 Hypothetical protein MGC17943MGC179430.0042147296 Hypothetical protein MGC23980MGC239800.018694224619 Hypothetical protein MGC24047MGC240470.001956138122 Hypothetical protein MGC33607MGC336070.033547100645 Ionized calcium binding adapter molecule 2IBA20.0042179489 KIAA0390 gene productKIAA03900.014832119936 KIAA0703 gene productKIAA07030.032602146652 Lectin, mannose-binding, 1LMAN10.031092179632 Leiomodin 1 (smooth muscle)LMOD10.022352120404 Likely ortholog of rat SNF1/AMP-activated protein kinaseSNARK0.044605157942 LIM domain kinase 2LIMK20.002409151439 Low density lipoprotein-related protein 1B (deleted in tumors)LRP1B0.00536209464 Low density lipoprotein-related protein 2LRP20.040937114919 Matrilin 2MATN20.0042167316 Metallothionein 1A (functional)MT1A0.013822204773 Metallothionein 1A (functional)|metallothionein 1E (functional)|metallothionein 1K| metallothionein 2AMT1A|MT2A|MT1E|MT1K0.027037146368 Metallothionein 1A (functional)|metallothionein 1E (functional)|metallothionein 2A| metallothionein 1KMT1A|MT2A|MT1K|MT1E0.043841182305 Metallothionein 1A (functional)|metallothionein 2A|metallothionein 1K| metallothionein 1E (functional)MT1A|MT1K|MT1E|MT2A0.011739223856 Metallothionein 1B (functional)MT1B0.019631174119 Metallothionein 1F (functional)MT1F0.024726144569 Metallothionein 1GMT1G0.0192164525 Metallothionein 1GMT1G0.03965171539 Metallothionein 1JMT1J0.008315227956 Metallothionein 1XMT1X0.008335119685 Metallothionein 1XMT1X0.010748173072 Metallothionein IVMT40.007447223241 Methionine adenosyltransferase II, alphaMAT2A0.014428158350 Mitogen-activated protein kinase 4MAPK40.042306131252 Myc-induced nuclear antigen, 53 kDaMINA530.011959130284 NIMA (never in mitosis gene a)- related kinase 11NEK110.001965194628 OtospiralinLOC1506770.018694182360 PDZ/coiled-coil domain binding partner for the rho-family GTPase TC10PIST0.013822103651 Phospholipase A2 receptor 1, 180 kDaPLA2R10.004029134379 Phospholipase C-like 1PLCL10.022657206894 Phosphotidylinositol transfer protein, betaPITPNB0.014428122698 Polycystic kidney and hepatic disease 1 (autosomal recessive)-like 1PKHD1L10.0042199896 Polymerase (DNA directed) iotaPOLI0.003227167492 Potassium channel, subfamily K, member 9KCNK90.000849108648 Potassium channel-interacting protein 4KCNIP40.011447147058 Potassium inwardly-rectifying channel, subfamily J, member 13KCNJ130.008972124187 pp21 HomologLOC511860.004326127636 Pre-B cell leukemia transcription factor 4PBX40.030311199118 Protein kinase, cAMP-dependent, catalytic, betaPRKACB0.023863198878 Protein phosphatase 4, regulatory subunit 2|hypothetical protein LOC151987PPP4R2|LOC1519870.011338200919 RAB23, member RAS oncogene familyRAB230.000659122394 Ras association (RalGDS/AF-6) domain family 6RASSF60.048804119072 Sarcoglycan, delta (35 kDa dystrophin-associated glycoprotein)SGCD0.011178120415 Serum deprivation response (phosphatidylserine binding protein)SDPR0.011021156433 SH3 and multiple ankyrin repeat domains 2SHANK20.043996193906 Solute carrier family 26, member 7SLC26A70.0042225067 Solute carrier family 26, member 7SLC26A70.005853213530 Solute carrier family 5 (iodide transporter), member 8SLC5A80.031284231731 SPARC related modular calcium binding 2SMOC20.021505135930 Syndecan 2 (heparan sulfate proteoglycan 1, cell surface-associated, fibroglycan)SDC20.001258209676 Syntaxin 12STX120.01117199949 T-box 22TBX220.024297177517 Thioredoxin-like, 32 kDaTXNL0.001102192552 Thyroid stimulating hormone receptorTSHR0.02176108606 Tissue inhibitor of metalloproteinase 4TIMP40.023629184795 Trefoil factor 3 (intestinal)TFF30.004648114445 Trefoil factor 3 (intestinal)TFF30.014428100949 UDP-N-acetyl-alpha-d-galactosamine/polypeptide N-acetylgalactosaminyltransferase 9 (GalNAc-T9)GALNT90.031284161042 WEE1 homolog (S. pombe)WEE10.024101123533 WW domain containing oxidoreductaseWWOX0.012208224298 WW domain containing oxidoreductaseWWOX0.024101135080 Zinc finger protein 258ZNF2580.013962225961 Zinc finger protein 36, C3H type-like 2ZFP36L20.018837210469Two-tail ANOVA with p value correction yielded 173 probes (52 up-, 121 down-regulated in PTC) significantly different (p < 0.05) between the malignant and benign thyroid tissues. Supervised hierarchical clustering was performed on the 25 arrays based on the 236 statistically significant probes to determine whether the samples would segregate appropriately. The resulting heat map can be observed in Fig. 1. Benign lesions and tumours clustered together with the exception of one tumour sample that clustered with the benign group (T5). Fig. 1Hierarchical clustering of samples. This heat map shows the clustering of the 25 samples based on the 236 probes found to be differentially regulated in benign vs malignant thyroid tissue. Clustering was performed using the unweighted pair group method with arithmetic mean, with Euclidian distance as the similarity measure. Average value was used as the ordering function A binomial statistics tool was used to compare classifications of multiple clusters of lists to a reference list (i.e. the complete human genome) to statistically determine over- or under-representation of Panther classification categories. Biological processes over-represented in the up-regulated PTC cohort included tumour suppressor, oncogenesis, DNA replication, cell cycle, and cell adhesion (p < 0.0001). Genes involved in homeostasis and other homeostasis activities were highly over-represented in the down-regulated cohort (p < 0.000001). ANOVA tests were used to determine which genes were differentially regulated in the FVPTC cohort only. Fifteen genes were identified, including cluster of differentiation 14 (CD14), CD74, CTSC, CTSH, CTSS, DPP6, ETHE1, human leucocyte antigen A (HLA-A), HLA-DMA, HLA-DPB1, HLA-DQB1, HLA-DRA, osteoclast stimulating factor 1 (OSTF1), TDO2, and a previously uncharacterized gene (noname). Microarray results were validated using a reverse transcription reaction followed by TaqMan® PCR for 50 gene targets. The ΔΔCT method [23] was used to analyse relative gene expression data. GAPDH was used as an endogenous control, and T12 was chosen as an arbitrary calibrator sample. Gene expression profiles for TaqMan® PCR were plotted in conjunction with those for microarray results in Fig. 2. Pearson co-efficient was used to directly compare data from microarray analysis and TaqMan® RT-PCR. Table 3 depicts genes differentially expressed in both benign vs malignant and FVPTC vs classic morphology PTC. Fig. 2TaqMan® PCR validation of microarray experiments. Profile charts of gene expression levels comparing results obtained by microarray analysis (n = 25) to TaqMan® PCR analysis (n = 20) for six genes. Plots for those genes with multiple probes are also displayed where appropriateTable 3Correlation between TaqMan® and microarray dataGenePearson’s r coefficienttwo-tailed pGenes differentially expressed in FVPTC vs classic PTCCD140.83<0.0001CD74a0.87<0.00010.740.0002CTSC0.530.0170CTSH0.710.0005CTSS0.620.0037DPP60.760.0001ETHE10.300.2008HLA-A0.77<0.0001HLA-DMA0.750.0001HLA-DPB10.96<0.0001HLA-DQB1a0.700.00060.630.0031HLA-DRA0.82<0.0001NONAME−0.020.9323OSTF10.150.5171TDO20.680.0009Genes differentially expressed in benign vs malignantBAX0.200.3997CAMK10.170.4771CD440.570.0094CTSS0.620.0037CXADR0.360.1152FGFR20.84<0.0001GALE0.540.0138ICAM10.620.0038LYN0.450.0483MAPK40.76<0.0001MMP140.360.1175MT1F0.92<0.0001MT1Ka0.680.00110.690.00070.700.0006MT1Xa0.88<0.00010.88<0.0001RAB230.700.0006S100A110.610.0041SDC20.340.1415SDC30.570.0087TFF3a0.97<0.00010.96<0.0001TGFB10.560.0098TIMP10.730.0003TIMP40.80<0.0001TOP2A0.070.7769TSHR0.180.4440Gene expression profiles for TaqMan® PCR and microarray results were compared using the Pearson coefficient.aGenes have more than one probe ID on microarray Discussion The primary aim of this study was to generate an overview of molecular markers of malignancy in PTC with a view to identifying discriminators between common sub-types (classic PTC and FVPTC), using genome-wide expression microarray technology validated by TaqMan® RT-PCR. To this end, lesions that were well characterized histologically were selected. The application of microarray analysis designed to identify transcripts strongly associated with each group of interest yielded a gene list of 173 genes that were differentially expressed between cohorts. Significant down-regulation of Coxsackie virus receptor was recorded in the malignant cohort of thyroid lesions. The Coxsackievirus B and adenovirus receptor (CAR) plays a dual role as a homotypic junctional adhesion protein and as a viral receptor. It is biologically plausible that altered expression may impact on the morphology peculiar to PTC given its association with cellular adhesion. CAR has been shown to be differentially expressed in various human adenocarcinomas, and differential expression may represent a new factor in thyroid tumourigenesisigenesis [27]. Rab 23 expression was also down regulated in the malignant cohort. The Rab small G protein family is composed of approximately 40 members. Many of them are ubiquitous and are expressed and participate in transport processes, such as endocytosis and exocytosis [26]. Other gene targets demonstrating significant down-regulation in the malignant group were syndecans 2 and 3. Syndecans are transmembrane proteoglycans expressed on adherent cells. Changes in syndecan expression have been postulated to influence cell adhesion, migration, and the structure of focal contacts and the cytoskeleton [8]. The abundance of metallothionein genes in the cohort of genes down-regulated in PTC is interesting partly because of the sheer number of isoforms detected (Table 2). There have been many studies showing increased metallothionein expression in a plethora of cancer types but few showing decreased expression [4, 7, 33]. Although previous microarray experiments showed metallothionein genes to be down-regulated in thyroid tumours to a certain extent [2, 9, 16], none have detected so many as the current study. Apart from microarray experiments, there has been little reported in the literature regarding metallothioneins in thyroid cancer. An early report by Nartey et al. [31] showed metallothioneins to be expressed actively in certain human thyroid neoplastic tissues but not in normal thyroid tissue, which would seem to contradict the current authors’ findings. In contrast to this, a later immunohistochemical study showed an absence of metallothionein expression in 13 of 20 PTCs [36]. Interestingly, in three of the seven positive PTCs, metallothionein positivity was restricted to areas of follicular differentiation. In one of the only recent studies, Huang et al. [17] followed up their initial microarray experiment by showing that MT1G is down regulated in PTC via hypermethylation. The biological significance of low metallothionein expression in thyroid tumours is therefore still poorly understood; however, it is interesting to speculate that metallothioneins may have roles as tumour suppressors in thyroid carcinoma. Many genes identified, such as LGALS3 [9, 16, 20], LYN [46], TFF3 [2, 9, 16], CRABP1 [9, 16], BAX [2], MAPK4 [28], CD44 [16], TIMP1 [20], FGFR2 [9], and S100A11 [20, 40], have been previously reported in both microarray and conventional experiments in thyroid cancer. The data generated in this study corroborates the importance of several biological processes in the progression of thyroid neoplasia. For example, S100A11 expression was up regulated in the PTC cohort compared with benign lesions, paralleling the increased protein expression of this gene target identified at the protein level using immunohistochemistry [29]. S100A11 has also been suggested as a biomarker of malignancy in the context of colorectal carcinoma as long ago as 1995 [41]. Correlation of highlighted features with the current state of knowledge of the molecular pathology of thyroid neoplasia goes some way towards providing an external validation of the data obtained from the AB1700 system. However, additional validation using TaqMan® RT-PCR was performed. In general, microarray and TaqMan® data correlated well with approx. 80% of comparisons having p < 0.05. Candidate genes were selected contingent on results identified as over-represented biological processes (oncogenesis, cell cycling, DNA replication, and homeostasis) using the Panther binomial statistics tool as opposed to the more traditional method of selecting the most highly disregulated genes. This may account for the poor correlation observed with certain genes. Some genes, such as TFF3 and FGFR2, had excellent correlation between microarray and TaqMan® results, whereas others, such as BAX and TSHR, showed poor correlation despite previous studies implicating them in thyroid cancer [15, 38]. Where there was discordance in the data, it could be accounted for by differences in the sequences targeted by the TaqMan® target sequences and the microarray probe. In those cases, the array and pre-designed TaqMan assays interrogated different exons or alternative splice variants. This finding highlights the importance of matching the targets to be validated from microarray data sets. Analysis of differentially expressed transcripts in the FVPTC revealed many transcripts showing similar expression level patterns in both FVPTC and classic morphology PTC. Examples of these transcripts are included in Table 2. In addition, somewhat surprisingly, unsupervised clustering of all samples showed no tendency for FVPTC to cluster independently of classic morphology PTC, emphasizing the very close relationship of these PTC variants (data not shown). However, analysis of differentially expressed genes in FVPTC exclusive of those identified in classic morphology PTC revealed 15 genes exhibiting differential expression in FVPTC compared with benign lesions and conditions outwith of classic morphology PTC. These genes displayed a narrow gamut of function represented by the transcripts involved (Table 3). TaqMan® RT-PCR was performed for all of these targets to confirm the array findings and correlation with the array data was strong (see Table 3). Aberrant expression of two major groups of transcripts was noted in FVPTC. Relatively increased expression of class 1 major histocompatibility complex (MHC) genes (HLA-A) and aberrant expression of class 2 MHC genes (HLA-DMA, HLA-DPB-1, HLA-DQB-1, HLA DRA) and associated genes (e.g. CD74 represents the invariant membrane bound moiety of class II HLA molecules and regulates the biology and functions of MHC class II molecules and CD14 is a surface marker of monocytes/macrophages) were the most significant findings. Additionally, relatively up-regulated expression of members of the cathepsin family (cathepsin C, cathepsin H, cathepsin S, and TDO2) was striking in the FVPTC group. HLA expression is generally associated with immune functions such as T cell interaction and antigen presentation. The presence of prominent HLA transcript expression, especially among class 2 in FVPTC, is intriguing. One potential cause of this was that tumour-infiltrating leucocytes were responsible for this finding. However, haematoxylin–eosin (H&E)-stained slides of each case were reviewed to specifically identify the degree of tumour infiltration by leucocytes. Although a mild lymphocytic infiltrate was noted in some cases, there was no apparent over representation of lymphocytes in the follicular variant compared to benign lesions and classic morphology PTC. This raises the clear possibility that the findings represent aberrant increased expression of class 2 HLA transcripts by the epithelium of FVPTC. This is an unexpected finding, as over-expression of MHC class 2 molecules would be expected to increase tumour immunogenicity. A similar aberrant expression of HLA transcripts has been recently described in ovarian neoplasms [34]. Rangel et al. [34] concluded over-expression of HLA-DRA might represent a novel biomarker for malignancy, and this also seems biologically plausible in the FVPTC setting. A recent paper has also described HLA-DRA expression in ret/PTC-activated papillary thyroid carcinoma but not in surrounding normal thyroid follicles [19]. Yu et al. [47] showed discordant expression of (CD74Ii) and HLA-DR in Hashimoto thyroiditis, an autoimmune condition associated with increased incidence of PTC and sharing molecular features such as ret/PTC expression [18, 37]. Hwang et al. [19] have drawn attention to aberrant expression of HLA-DRA in ret/PTC-activated PTC. Expression of HLA-DRA may in some way explain the propensity for PTC to metastasize to lymph nodes and often apparently reside there without markedly worsening prognosis. Cathepsins C (dipeptidyl-peptidase I), H, and S showed up-regulation in FVPTC compared to benign lesions. Cathepsins are a family of proteases that play an important role in protein degradation. They are key players in the proliferative, invasive, and metastatic potential of malignant tumour cells. Their expression in the relatively biologically indolent FVPTC is intriguing, and it remains possible that cathepsins have cellular roles outside of those involved in invasion and dissemination of tumour cells as indeed has been suggested by others [43]. For example, cathepsin L has recently been shown to play a role in nuclear transcriptional activation, and cathepsins are now recognized to play a role in MHC class II antigen presentation [44]. OSTF1 has no defined role in carcinogenesis, although outside of its role in ossification, it is also known to have roles in signal transduction and protein binding, which may be relevant to carcinogenesis and, particularly, FVPTC. A recent paper highlights the role of bone mineralization proteins including osteopontin and osteoclast stimulating factors as potential biomarkers of malignant tumours in general [1]. Indeed, in addition to elevated OSTF1 expression, increased expression of osteopontin was seen in PTC (data not shown); however, in our data, osteopontin does not appear to be specifically up-regulated in FVPTC, and this finding has also been noted by other researchers [14]. In any case, osteopontin is known to be a downstream effector of ret/PTC [5] and mutated BRAF [14], where it acts in association with CD44, another transcript showing increased expression in both classic morphology PTC and FVPTC. A particular focus of this study was to compare transcriptome profiles for PTCs with classic morphology and FVPTC given the propensity for FVPTC lesions to prove problematic from a diagnostic perspective. Although the study confirms the close relationship between the two most common variants of PTC, a narrow portfolio of genes and, in particular, gene functions was elucidated in the FVPTC cohort. The targets identified are easily amenable to analysis by more established techniques such as TaqMan® RT-PCR, with associated potential as additional markers for application in the FNAC setting. Clearly, the potential biomarkers identified in this study will require prospective evaluation in the context of real clinical diagnostic situations in the future to consolidate their merit as adjunctive tests in the diagnostic setting and to validate their altered expression states in the pathobiology of PTC development.

Eur_J_Epidemiol-3-1-2071965

Acute myocardial infarction incidence and hospital mortality: routinely collected national data versus linkage of national registers

Background and Objective To compare levels of and trends in incidence and hospital mortality of first acute myocardial infarction (AMI) based on routinely collected hospital morbidity data and on linked registers. Cases taken from routine hospital data are a mix of patients with recurrent and first events, and double counting occurs when cases are admitted for an event several times during 1 year. By linkage of registers, recurrent events and double counts can be excluded. Introduction Mortality from coronary heart disease, in particular from acute myocardial infarction (AMI), has decreased in many Western countries during the last decades [1]. A decrease in age- and gender-adjusted AMI mortality, assuming a constant quality of diagnosis, is a consequence of either a decrease in incidence, case-fatality or recurrence risk, or a combination of these. Hospital-based registers are often used for surveillance of the morbidity and hospital mortality of AMI [2]. In the Netherlands, the national hospital discharge register has traditionally been used to provide estimates of (trends in) incidence and hospital mortality of AMI [3]. However, in this register, like in many others, a new record is created for each hospital admission. As a consequence, admissions taken from the hospital register will include double counts from patients if they are transferred to a second hospital or if they are admitted for the same event several times during 1 year. Furthermore, patients taken from the hospital register from 1 year include a mix of patients with recurrent events (presence of an event in preceding years) and first events (absence of events in preceding years). Tracking individuals over time based on information from the hospital register only is difficult when a unique personal identifier is absent in the hospital register. The effect of both double counting and admixture of first and recurrent events in nationwide registers on (trends) in incidence or hospital mortality has not been well assessed [4, 5]. For the Netherlands, this effect could only be estimated from comparison with results from regional cohort studies, as nationwide estimates of incidence and hospital mortality of first AMI were not available. Furthermore, it has been argued that statistics from routine data could not be used for providing reliable information on (trends in) incidence and hospital mortality. After we recently showed that hospitalized patients in the Netherlands can be followed longitudinally within the national hospital discharge register in a valid way by using information from the Dutch population register [6], we set out to compare the nationwide (trends in) incidence and hospital mortality of first hospitalized AMI based on routinely collected data in the hospital register (double counts and recurrent events included) and based on linkage of the hospital register with the population register (double counts and recurrent events excluded). Methods Data sources Data on hospital admissions were retrieved from the national hospital discharge register. Since 1986, all general and academic hospitals and most single specialty hospitals participate in the hospital register. There are no private hospitals in the Netherlands that treat patients with AMI. For each hospital admission a new record is created in the hospital register, including the following information: date of birth, gender, numeric part of postal code (since 1991), hospital-specific patient identification code, type of hospital, admission date and principal diagnosis of the admission. The principal diagnosis is determined at discharge and coded using the ninth revision of the International Classification of Diseases (ICD-9-CM) [7]. As the hospital register does not contain a unique personal identifier, we tracked individuals over time within the hospital register by using information from the Dutch population register. This database contains information on all registered persons living in the Netherlands, including date of birth, gender, current address, postal code and nationality. Patients registered in the hospital register were identified in the population register using linkage variables ‘date of birth’, ‘gender’ and ‘numeric part of postal code’. When patients moved, their hospital admissions were recognized by using the new postal code registered in the population register. Recently, the validity of the registries and linkage methods was studied. In a random sample of hospital admissions, 99% of the personal, admission and discharge data and 84% of the principal diagnoses (validated through medical record review by medical specialists) were correctly registered [8]. In a random sample of the population register, over 97% of the addresses were shown to be correctly registered [9]. Furthermore, over 97% of the uniquely linked hospital admissions resulting from linkage of the hospital register with the population register were shown to be correctly linked [9]. These results are similar to most of the studies that reported on the validity of AMI events in hospital and population based registries [10–13]. All analyses were performed at Statistics Netherlands in agreement with privacy legislation in the Netherlands [6]. Cohort enrolment from the hospital register The hospital register comprises information based on all admissions in the Netherlands of the entire Dutch population, including double counts, first and recurrent admissions for AMI, and including AMI admissions of non-residents. In the hospital register, 28,733 and 25,864 hospital admissions with a principal diagnosis AMI (ICD-9-CM [7] code 410 and subcategories) were registered in 1995 and 2000. Cohort enrolment from linked registers After linkage with the population register using linkage variables ‘date of birth’, ‘gender’ and ‘numerical part of postal code’, 25,142 and 22,470 admissions came from patients with a unique combination of linkage variables in the population register (88% and 87%, respectively). Thus, each remaining admission linked to only one unique individual in the population register (one unique individual in the Netherlands). Admissions linking with more than one person (e.g., administrative twins; two persons with the same date of birth, gender and numeric part of postal code registered in the population register) or with no person at all (e.g., non-residents or administrative errors) in the population register were excluded. Selection of the first admission per person of all subsequent admissions of a person occurring during 1995 and 2000 yielded 23,172 patients in 1995 and 20,414 patients in 2000. Thus, 1,970 double counts had occurred in 1995 (8%) and 2,056 in 2000 (9%). Information on admissions in previous years of the patients in 1995 was obtained by selecting all hospital admissions registered in the hospital register with principal diagnosis AMI in the period 1991–1995. These admissions were linked to the cohort of 23,172 patients with linkage variables ‘date of birth’, ‘gender’ and one or both of the variables ‘numerical part of postal code’ and ‘hospital-specific patient identification code’. Linkage with the population register was not possible, since this register started in October 1994. Subjects who linked in this process were patients with previous hospital admissions for AMI (recurrent AMI patients) and were excluded (1,607 patients (7%)). This resulted in the final cohort of 1995 consisting of 21,565 patients. Information on hospital admissions in previous years of the patients in 2000 was obtained by linking of the hospital register of 1995–2000 to the population register with linkage variables ‘date of birth’, ‘gender’ and ‘numerical part of postal code’. All uniquely linked admissions with principal diagnosis AMI were selected and linked to the cohort of 20,414 patients. Patients with previous hospital admissions for AMI (recurrent AMI patients) were excluded (1,356 patients (7%)). This resulted in the final cohort of 2000 consisting of 19,058 patients. Thus the linked register comprises information for only part of the Dutch population (i.e., those that were unique on date of birth, gender, and postal code), and does not include double counts, and recurrent AMI admissions. Data analysis The incidence and hospital mortality of AMI (with 95% confidence interval (95% CI)) based on the hospital register (hospital-based) and on linked registers (linkage-based) was computed by year, age and gender. We compared the hospital-based incidence and hospital mortality to the linkage-based incidence and hospital mortality by calculating incidence rate differences or ratios and risk differences or ratios (with 95% CIs) by age and gender. Trends in incidence and hospital mortality were obtained by calculating incidence rate or risk differences and incidence rate or risk ratios (with 95% CIs) by age and gender. Incidence rate differences and ratios were based on the Poisson model, while risk differences and ratios were based on the binomial model [14]. Pooled age-adjusted incidence rate differences or ratios and risk differences or ratios (with 95% CIs) were calculated according to the Mantel Haenszel method [15]. Results In both 1995 and 2000, the gender and age distribution of the cohort based on the hospital register was comparable to the cohort based on linked registers. In 1995 and 2000, two-thirds comprised men. The mean age in 1995 was 63.7 years in men and 71.5 years in women based on the hospital register. This was 64.3 and 71.9 years, respectively, based on linked registers. In 2000, the mean age based on the hospital register was 63.6 years in men and 70.9 years in women. Based on linked registers, this was 64.2 and 71.6 years, respectively. In men, the (adjusted) hospital-based incidence was 47 per 100,000 person-years or 22% (95% CI 19–25%) higher than the (adjusted) linkage-based incidence in 1995 and 43 per 100,000 person-years or 23% (95% CI 20–26%) higher in 2000 (Table 1). Age-specific (≥30 years) absolute and relative differences ranged from 9–217 per 100,000 person-years or 20–28% in 1995 and from 7–220 per 100,000 person-years or 22–25% in 2000. The (adjusted) hospital-based incidence was also higher than the (adjusted) linkage-based incidence in women in 1995 (19 per 100,000 person-years or 18%; 95% CI 15–22% higher) and 2000 (18 per 100,000 person-years or 20%; 95% CI 16–24% higher). Age-specific (≥30 years) absolute and relative differences varied from 3–103 per 100,000 person-years or 13–20% in 1995 and from 1–116 per 100,000 person-years or 14–33% in 2000. Table 1Incidence (per 100,000 persons per year) of hospitalized acute myocardial infarction in 1995 and 2000 based on the national hospital discharge register and based on linked registers19952000Hospital registerLinked registersHospital registerLinked registersAgeIncidenceIncidenceID1IR2IncidenceIncidenceID1IR2Men<30211 (0 to 1)1.56 (0.99 to 2.48)210 (0 to 1)1.33 (0.88 to 2.01)30–3941329 (4 to 14)1.28 (1.12 to 1.47)40327 (3 to 12)1.23 (1.08 to 1.41)40–4920717334 (22 to 45)1.20 (1.12 to 1.27)18014336 (26 to 47)1.25 (1.17 to 1.34)50–59509409100 (79 to 121)1.24 (1.19 to 1.30)40733276 (58 to 93)1.23 (1.17 to 1.29)60–69887733154 (122 to 186)1.21 (1.16 to 1.26)675551124 (97 to 151)1.23 (1.17 to 1.28)70–791,2261,009217 (169 to 265)1.21 (1.16 to 1.27)1,008824184 (143 to 225)1.22 (1.17 to 1.28)80–891,2541,045209 (126 to 291)1.20 (1.12 to 1.29)1,144925220 (144 to 295)1.24 (1.15 to 1.33)≥90843660184 (−20 to 387)1.28 (0.97 to 1.68)568464104 (−59 to 267)1.22 (0.89 to 1.68)TotalCrude25522134 (29 to 39)1.15 (1.13 to 1.18)22219033 (28 to 37)1.17 (1.15 to 1.20)Adjusted347 (42 to 52)1.22 (1.19 to 1.25)43 (38 to 47)1.23 (1.20 to 1.26)Women<30100 (0 to 1)3.59 (1.21 to 10.67)110 (0 to 1)1.31 (0.65 to 2.63)30–39981 (−1 to 4)1.17 (0.88 to 1.56)1183 (0 to 5)1.33 (1.02 to 1.73)40–4938326 (0 to 11)1.17 (1.01 to 1.36)46388 (2 to 13)1.21 (1.06 to 1.38)50–591139519 (9 to 29)1.20 (1.09 to 1.32)958016 (7 to 24)1.20 (1.08 to 1.32)60–6930626244 (26 to 62)1.17 (1.10 to 1.25)24319647 (31 to 63)1.24 (1.15 to 1.33)70–7958148695 (67 to 123)1.20 (1.13 to 1.26)47239676 (51 to 100)1.19 (1.13 to 1.26)80–89750634116 (73 to 160)1.18 (1.11 to 1.26)643540103 (64 to 143)1.19 (1.11 to 1.27)≥9049143655 (−32 to 142)1.13 (0.93 to 1.36)45039456 (−21 to 133)1.14 (0.95 to 1.37)TotalCrude11810613 (9 to 16)1.12 (1.09 to 1.16)1049113 (10 to 16)1.14 (1.11 to 1.18)Adjusted319 (16 to 22)1.18 (1.15 to 1.22)18 (15 to 21)1.20 (1.16 to 1.24)1 Incidence rate difference (with 95% confidence interval) 2Incidence rate ratio (with 95% confidence interval)3 Pooled age-adjusted incidence rate difference or ratio (with 95% confidence interval) The hospital-based hospital mortality was similar to the linkage-based hospital mortality in men in 1995 (adjusted risk ratio (RR) 1.01; 95% CI 0.95–1.07) and 2000 (adjusted RR 1.00; 95% CI 0.94–1.07) and in women in 1995 (adjusted RR 0.98; 95% CI 0.92–1.05) and 2000 (adjusted RR 0.99; 95% CI 0.93–1.06) (Table 2). Also within the age groups, no significant differences between the hospital-based and the linkage-based hospital mortality were revealed. Table 2Hospital mortality (%) of acute myocardial infarction in 1995 and in 2000 based on the national hospital discharge register and based on linked registers19952000Hospital registerLinked registersHospital registerLinked registersAgeMortality MortalityRD1RR2MortalityMortalityRD1RR2Men<301.93.6−1.6 (−9.5 to 6.2)0.54 (0.03 to 8.29)6.910.8−3.9 (−15.9 to 8.0)0.64 (0.17 to 2.40)30–393.22.40.9 (−1.4 to 3.1)1.36 (0.59 to 3.11)2.83.1−0.3 (−2.6 to 2.0)0.90 (0.42 to 1.95)40–493.33.7−0.4 (−1.5 to 0.7)0.89 (0.65 to 1.22)3.22.60.7 (−0.4 to 1.8)1.26 (0.85 to 1.86)50–594.44.00.4 (−0.5 to 1.3)1.10 (0.88 to 1.37)3.73.50.2 (−0.6 to 1.1)1.07 (0.84 to 1.37)60–698.48.20.2 (−0.9 to 1.3)1.02 (0.89 to 1.16)7.27.10.1 (−1.0 to 1.3)1.02 (0.87 to 1.20)70–7917.117.00.0 (−1.6 to 1.6)1.00 (0.91 to 1.10)15.415.30.1 (−1.6 to 1.7)1.00 (0.90 to 1.12)80–8928.128.5−0.3 (−3.6 to 2.9)0.99 (0.88 to 1.11)24.425.5−1.1 (−4.3 to 2.1)0.96 (0.84 to 1.09)≥9034.732.62.1 (−10.7 to 15.0)1.07 (0.73 to 1.56)47.150.7−3.7 (−19.5 to −2.2)0.93 (0.67 to 1.28)TotalCrude10.710.9−0.3 (−0.9 to 0.4)0.98 (0.92 to 1.04)9.59.8−0.3 (−1.0 to 0.4)0.97 (0.91 to 1.04)Adjusted30.1 (−0.6 to 0.7)1.01 (0.95 to 1.07)0.0 (−0.6 to 0.7)1.00 (0.94 to 1.07)Women<305.925.0−19.1 (−63.0 to 24.8)0.24 (0.02 to 3.01)5.07.7−2.7 (−20.0 to 14.7)0.65 (0.04 to 9.50)30–396.96.10.8 (−6.1 to 7.7)1.13 (0.38 to 3.33)3.43.30.1 (−4.6 to 4.9)1.03 (0.25 to 4.22)40–494.85.1−0.3 (−3.4 to 2.9)0.95 (0.50 to 1.79)6.26.7−0.5 (−3.8 to 2.7)0.92 (0.56 to 1.53)50–594.95.3−0.4 (−2.6 to 1.8)0.92 (0.60 to 1.41)4.95.1−0.1 (−2.3 to 30.)0.97 (0.63 to 1.49)60–699.68.90.7 (−1.2 to 2.5)1.07 (0.88 to 1.31)8.88.10.7 (−1.3 to 2.8)1.09 (0.86 to 1.39)70–7918.718.30.4 (−1.6 to 2.5)1.02 (0.92 to 1.14)16.616.40.2 (−1.9 to 2.3)1.01 (0.89 to 1.15)80–8930.431.8−1.4 (−4.3 to 1.5)0.96 (0.87 to 1.05)28.429.3−0.9 (−4.0 to 2.1)0.97 (0.87 to 1.08)≥9040.540.6−0.1 (−9.3 to 9.2)1.00 (0.79 to 1.25)31.534.4−2.9 (−11.5 to 5.6)0.91 (0.71 to 1.19)TotalCrude17.718.1−0.4 (−1.6 to 0.8)0.98 (0.92 to 1.05)15.916.5−0.5 (−1.8 to 0.7)0.97 (0.90 to 1.04)Adjusted3−0.1 (−1.3 to 1.0)0.99 (0.93 to 1.06)−0.2 (−1.3 to 1.0)0.99 (0.92 to 1.06)1 Risk difference (with 95% confidence interval) 2 Risk ratio (with 95% confidence interval)3 Pooled age-adjusted risk difference or ratio (with 95% confidence interval) From 1995 to 2000, the hospital-based decline in incidence was similar to the linkage-based decline (Table 3). In men, the (adjusted) hospital-based incidence declined by 48 per 100,000 person-years or 18% (95% CI 17–20%) and the linkage-based incidence declined by 46 per 100,000 person-years or 20% (95% CI 18–22%). In women, the (adjusted) hospital-based incidence declined by 18 per 100,000 person-years or 15% (95% CI 13–18%) and the (adjusted) linkage-based incidence declined by 18 per 100,000 persons per year or 17% (95% CI 14–19%). The age-specific relative changes in hospital-based and linkage-based incidence were largely comparable. Table 3Trends in the incidence (per 100,000 persons per year) and hospital mortality (%) of hospitalized acute myocardial infarction from 1995 to 2000 based on the national hospital discharge register and based on linked registersTrends in incidence Trends in hospital mortality Risk difference Relative risk Risk differenceRelative risk AgeHospital registerLinked registersHospital registerLinked registersHospital registerLinked registersHospital registerLinked registersMen<300 (0 to 1)0 (0 to 1)1.16 (0.80 to 1.68)1.36 (0.83 to 2.22)5.0 (−2.5 to 12.5)7.2 (−4.9 to 19.4)3.59 (0.41 to 31.07)3.03 (0.36 to 25.62)30–39−1 (−6 to 4)1 (−4 to 5)0.98 (0.87 to 1.10)1.02 (0.88 to 1.18)−0.4 (−2.5 to 1.6) 0.7 (−1.7 to 3.1)0.87 (0.44 to 1.72)1.30 (0.53 to 3.20)40–49−28 (−39 to −17)−30 (−41 to −19)0.87 (0.82 to 0.92)0.83 (0.77 to 0.89)−0.1 (−1.1 to 0.9)−1.2 (−2.4 to 0.0)0.97 (0.71 to 1.33)0.69 (0.46 to 1.02)50–59−102 (−122 to −83)−78 (−97 to −59)0.80 (0.77 to 0.83)0.81 (0.77 to 0.85)−0.7 (−1.5 to 0.1)−0.6 (−1.5 to 0.4)0.84 (0.68 to 1.04)0.86 (0.67 to 1.12)60–69−211 (−242 to −181)−181 (−210 to −152)0.76 (0.73 to 0.79)0.75 (0.72 to 0.79)−1.2 (−2.3 to −0.2)−1.2 (−2.4 to 0.0)0.86 (0.75 to 0.98)0.86 (0.73 to 1.00)70–79−218 (−265 to −172)−186 (−229 to −143)0.82 (0.79 to 0.86)0.82 (0.78 to 0.86)−1.7 (−3.2 to −0.2)−1.7 (−3.5 to 0.0)0.90 (0.82 to 0.99)0.90 (0.81 to 1.00)80–89−110 (−192 to −28)−121 (−196 to −45)0.91 (0.85 to 0.98)0.88 (0.82 to 0.95)−3.8 (−6.8 to −0.8)−3.0 (−6.4 to 0.4)0.87 (0.77 to 0.97)0.89 (0.79 to 1.01)≥90−276 (−470 to −81)−196 (−370 to −23)0.67 (0.51 to 0.89)0.70 (0.51 to 0.96)12.3 (−1.3 to 26.0)18.1 (2.9 to 33.3)1.35 (0.97 to 1.89)1.56 (1.07 to 2.26)TotalCrude−32 (−37 to −27)−31 (−36 to −27)0.87 (0.86 to 0.89)0.86 (0.84 to 0.88)−1.1 (−1.8 to −0.5)−1.1 (−1.8 to −0.4)0.89 (0.84 to 0.95)0.90 (0.84 to 0.96)Adj1−48 (−53 to −43)−46 (−51 to −41)0.82 (0.80 to 0.83)0.80 (0.78 to 0.82)−1.2 (−1.8 to −0.6)−1.2 (−1.9 to −0.5)0.89 (0.84 to 0.94)0.89 (0.83 to 0.96)Women<300 (0 to 1)0 (0 to 1)1.21 (0.64 to 2.32)3.34 (1.09 to 10.23)−0.9 (−15.6 to 13.8)−17.3 (−62.1 to 28)0.85 (0.06 to 12.59)0.31 (0.02 to 3.88)30–392 (−1 to 4)0 (−2 to 3)1.20 (0.94 to 1.53)1.05 (0.78 to 1.42)−3.5 (−8.9 to 2.0)−2.8 (−9.1 to 3.6)0.50 (0.17 to 1.49)0.55 (0.13 to 2.22)40–498 (3 to 13)6 (0 to 11)1.22 (1.07 to 1.38)1.18 (1.01 to 1.37)1.3 (−1.5 to 4.2)1.6 (−1.9 to 5.1)1.27 (0.75 to 2.17)1.31 (0.71 to 2.40)50–59−18 (−27 to −8)−15 (−24 to −6)0.84 (0.77 to 0.92)0.84 (0.76 to 0.94)0.0 (−1.9 to 2.0) −0.2 (−2.6 to 2.1)1.01 (0.68 to 1.04)0.96 (0.61 to 1.51)60–69−64 (−81 to −46)−66 (−83 to −50)0.79 (0.74 to 0.84)0.75 (0.69 to 0.80)−0.8 (−2.6 to 1.1)−0.9 (−2.9 to 1.2)0.92 (0.75 to 0.98)0.90 (0.71 to 1.15)70–79−109 (−136 to −83)−90 (−116 to −64)0.81 (0.77 to 0.85)0.82 (0.77 to 0.86)−2.1 (−4.0 to −0.2)−1.9 (−4.1 to 0.3)0.89 (0.82 to 0.99)0.90 (0.79 to 1.02)80–89−107 (−149 to −64)−94 (−134 to −54)0.86 (0.81 to 0.91)0.85 (0.80 to 0.91)−2.0 (−4.8 to 0.7)−2.5 (−5.7 to 0.6)0.93 (0.77 to 0.97)0.92 (0.83 to 1.02)≥90−40 (−125 to 44)−41 (−121 to 39)0.92 (0.77 to 1.10)0.90 (0.75 to 1.10)−9.0 (−17.6 to −0.5)−6.2 (−15.5 to 3.1)0.78 (0.97 to 1.89)0.85 (0.66 to 1.09)TotalCrude−15 (−18 to −12)−15 (−18 to −12)0.87 (0.85 to 0.90)0.86 (0.84 to 0.88)−1.8 (−2.9 to −0.7)−1.6 (−2.9 to −0.3)0.90 (0.84 to 0.95)0.91 (0.85 to 0.98)Adj1−18 (−22 to −15)−18 (−21 to −14)0.85 (0.82 to 0.87)0.83 (0.81 to 0.86)−1.6 (−2.7 to −0.5)−1.6 (−2.9 to −0.4)0.91 (0.85 to 0.97)0.91 (0.85 to 0.98)1 Pooled age-adjusted risk difference or ratio (with 95% confidence interval) The (adjusted) hospital-based decline in hospital mortality from 1995 to 2000 was similar to the (adjusted) linkage-based decline (Table 3). In men, hospital mortality declined absolutely by 1% and relatively by 11% based on both the hospital register and linked registers. In women, the absolute and relative decline was 2% and 9%, respectively, based on both the hospital register and linked registers. The age-specific relative changes in hospital-based and linkage-based hospital mortality were largely similar. Discussion We combined data from the national hospital discharge register with data from the population register to determine the (trends in) incidence and hospital mortality of first hospitalized AMI (double counts and recurrent AMI cases excluded) and compared the outcomes with the incidence and hospital mortality based on routinely collected data in the hospital register (double counts and recurrent AMI cases included). The incidence based on the hospital register was considerably and significantly higher than the incidence based on linked registers, whereas hospital mortality and trends in incidence and hospital mortality were identical using either approach. Although we were able to estimate the incidence and hospital mortality of first AMI by linkage of the hospital register with the population register, some aspects of this method should be discussed. First, non-unique persons in the population (register) were excluded from the study population in the linked registry data. If this exclusion produced systematic differences between the linked registry population and the clinically relevant population (i.e., the total Dutch population), it might have affected the incidence estimate in the linked registry to some extent (e.g., an overestimation of incidence resulting from a higher mean age of the study population). A pilot study suggested that non-uniqueness relates to large cities, foreign origin and age [6]. The differences between unique and non-unique persons, however, were small [9] and apply to both 1995 and 2000. Second, information on previous admissions was limited to maximal 5 years for the patients (as the numeric part of the postal code is registered in the hospital register since 1991). Therefore, it seems likely that some “first” AMI patients actually were recurrent AMI patients. However, it has been estimated that most (95%) of recurrent events occur within 5 years. [4, 16] Third, the outcome measures in the present study were incidence and hospital mortality. Mortality after discharge from hospital was not considered, since this outcome is not registered in the hospital register. Differences in mortality after discharge between patients with a first or a recurrent AMI can only be studied by linkage of national registers (i.e., the hospital register with the population register and the cause of death statistics). A final aspect that needs to be addressed is the generalizability of our findings. The results might differ if a change over time occurs in double-count or readmission fractions. Results might also differ for other diseases than AMI or for specific groups of patients (e.g., non-native patients), hospitals or regions. Such differences will not be apparent from routinely collected data. Since trends in incidence and hospital mortality are often based on national routinely collected data, generalization of our findings would be of great relevance. It has been be argued that routine statistics can not be used for providing information on (trends in) incidence and hospital mortality, because of double counting of cases and admixture of first and recurrent events. In order to prevent erroneous inclusion of prevalent cases (recurrent events) that have had a previous hospitalization for AMI prior to the study period generally an clearance period is employed to overcome overestimation of the incidence [16]. For myocardial infarction, it has been shown that a clearance period of 13 years should be taken into account to completely overcome inclusion of prevalent cases. In our study we used a 5-year clearance period because of logistical reasons. This however would suggest that in our incidence estimate around 5% of the subjects should be considered as recurrent AMI-patient rather that first ever AMI patient [16]. Although this does affect the absolute incidence estimate, it may not affect trends in incidence and case fatality, assuming that the erroneous inclusion of prevalent cases occurs at both time windows Indeed this has been shown in a Danish study where the incidence based on the number of AMI-patients without an admission for AMI in the previous year overestimated the incidence based on the number of AMI-patients without (an admission for) AMI in the previous 14 years by 27% in men and 16% in women, but trends reflected trends in true incidence with reasonable accuracy [4]. With respect to double count, in eight states of the USA, it was estimated that double counting of patients resulted in an overestimation of the true incidence of hospitalized AMI and an underestimation of the true hospital mortality. In this study, double counts were defined as readmissions for AMI within 7 days and thought to result from transfer to a second hospital. Correction for double counting revealed a 10–15% lower incidence and a 12% higher hospital mortality [5]. Despite aspects regarding double counts and recurrent events, there have been several consistent reports from different countries using national registries to study trends in case fatality, incidence and survival [17–21]. These time trends indicated a decline in incidence of myocardial infarction and in case fatality after AMI. In the present study, we found a significant decline in both incidence and hospital mortality of first AMI between 1995 and 2000. These declines in incidence and hospital mortality appear to mainly reflect declines in first events, as trends were not altered when recurrent cases were excluded from the data, and thus are best explained by advances in primary prevention and acute management of AMI. Overall, our results based on deterministic linkages of sources using gender, date of birth and postal code, are in line with earlier reports from other studies, where linkage was performed using unique identification numbers. In conclusion, our study shows that the incidence based on routinely collected data in the national hospital discharge register overestimates the actual incidence of first AMI based on linked national registers by least 22% in men and 18% in women. Yet, the hospital mortality based on the hospital register accurately reflects the actual hospital mortality of first AMI. Furthermore, trends in incidence and hospital mortality based on the hospital register are not changed when double counts and recurrent cases were excluded. Since trends in incidence and hospital mortality of AMI are often based on national routinely collected data, it is reassuring that our results indicate that findings from such studies are indeed valid and not biased because of recurrent events and double counts. Electronic supplementary material Below is the link to the electronic supplementary material. (DOC 193 kb)

Eur_J_Epidemiol-4-1-2413078

Low migrant mortality in Germany for men aged 65 and older: fact or artifact?

Migrant mortality in Europe was found to be lower than mortality of host populations. In Germany, residents with migrant background constitute nearly one tenth of the population aged 65+ with about 40% of them being foreigners. The German Pension Scheme follows vital status of pensioners very accurately. Mortality re-estimation reveals two-fold underestimation of mortality of foreigners due to biased death numerator and population denominator. Introduction Migrant mortality has attracted substantial research interest. First, it provides the opportunity to look at the combined influence of earlier life conditions in the country of origin and the more recent life conditions in the receiving country. Second, it allows one to see how cultural and behavioral patterns imported from the country of origin and their further transformation affect migrant health. Third, it enables the researcher to gain insight into selective migration of healthy people (the healthy migrant effect). Indeed, at the very moment of moving, most migrants are in good health because immigration and assimilation to the foreign society are hardly compatible with serious health problems. In addition, migrants often have to pass obligatory medical examinations before they immigrate. In Germany and other European countries, migrant mortality was found to be much lower than the mortality of the host population [1–6]. Prior studies have tended to consider the healthy migrant effect as a central explanation of very low immigrant mortality. In addition, the possible role played by behavioral and psychological factors has been emphasized [1, 2, 7]. At the same time, there has also been serious concern about the reliability of demographic data on the mortality of migrants. The main problem is related to migrant moves between the home and host country. If departures from the host country and deaths in the country of origin are under-recorded, the moves produce a numerator-denominator bias leading to the under-estimation of migrant mortality. Deaths of migrants in the country of origin unregistered by the population statistics in the receiving countries cause “statistical immortality.” The results of under-registered return migrations and deaths abroad may be particularly problematic at advanced ages (90+), i.e., at ages when the share of “immortals” cumulated over time can become significant compared to the relatively small remaining population [4, 8, 9]. Biases in the data can be substantial even when micro-data from population registers or cohort follow-up studies are used [3, 10]. This is so because a segment of migrants (the less-well-integrated and less healthy) may tend to be at higher risk of observation censoring due to departure [3]. In Germany in 2005, 15.3 million people (or 18.6% of the total population) had a migrant background1 and 7.3 million (8.9%) had foreign citizenship [11]. Among the population aged 65+, 7.6% had a migrant background (mostly first-generation immigrants) and 3.4% of them were foreigners [11, 12]. Many labor migrants (Gastarbeiter) arrived in Germany in the 1950 and 1970s, at a time when the rapidly growing German economy experienced a labor shortage. They came mainly from Southern Europe and took up jobs in the coal, iron, steel, and automobile industries of Germany. Many of them have been granted permanent residence in Germany but not German citizenship. The so-called Aussiedler, i.e., ethnic Germans, mostly coming from the former Soviet Union, were the exception. They were granted the legal right to receive German citizenship upon arrival [13]. On average, foreigners have higher unemployment rates than Germans, their educational level is lower, and they have jobs that require relatively low qualifications [14, 15]. German population statistics are not free of problems, possibly resulting in an overestimation of the foreign population (see also [4]). The last West German and East German censuses took place in the 1980s. Since then, the population has been updated on the basis of vital event registration. Foreigners are also registered by the Central Registry of Aliens (Ausländerzentralregister), but a recent update of this source has not been translated into German population statistics [14]. Departures from Germany to stay abroad for long periods legally should be notified. Nevertheless, allegedly many fail to do so since an eventual return to Germany is facilitated by maintaining a German address. In studies of migrant mortality, data problems are usually not addressed by direct checks since it is difficult to find high-quality data sources alternative to the official population statistics. We are aware of only one study that re-estimated migrant mortality. This was done for Sweden on the basis of income and taxation data [10]. The authors calculated death rates for foreign migrants aged 20–64 who had some registered income (either earnings or social benefits) and excluded those who had no registered income. The latter category was considered as the one that included a high proportion of people who (de facto) resided abroad. After the exclusion, the immigrant death rates substantially increased compared to the initial rates calculated from the whole migrant population. Correspondingly, the migrants’ mortality advantage diminished and became statistically insignificant in most immigrant groups. Our study is based on data of the scientific use files of the Federal German Statutory Pension Scheme (DRV Bund-Deutsche Rentenversicherung Bund) for the re-estimation of the mortality of the retired German and foreign populations. These data were provided by the DRV Research Data Centre in Würzburg [16]. The data cover all public old-age pensions resulting from work careers in Germany and all of the three former branches of the pension insurance: workers, salaried employees, and miners. The data also reflect pension terminations due to the death of pensioners. We consider only men because many women who are now retired have never participated in the labor market. The DRV data cover about 90% of the German male population aged 65+. The remaining people are those whose working careers were entirely related to self-employment or to the German Civil Service. The DRV data allow to identify the narrowly defined migrant population, i.e. pensioners with foreign citizenship. In 1995–2004, about 190,000 (or 3.5%) of the 5.4 million male pensioners aged 65+ had foreign citizenship. The DRV information about pensioners’ nationality is considered to be exact [17]. The DRV data on the insurants’ vital status are also highly reliable. As the function of pension insurers is to provide pension payments, they follow the survival of pensioners very carefully, using several sources: information from the pay-out service of the German Post (Postrentendienst), which receives death notices from the registrar’s office or from the undertaker; information provided by surviving dependents; or information about pensions not delivered. Those receivers of German pensions who live abroad must provide annual confirmation of being alive. Nevertheless, we excluded pensioners with current residence abroad in order to achieve greater data homogeneity. The official population figures, population and deaths by 5-year age groups, split by German and non-German nationality, were obtained from the Federal Statistical Office [12]. On the basis of the DRV and official population data, we computed two sets of deaths and population exposures at ages 65–69, 70–74, …, 90+ for the years 1995, 1998, 2001, and 2004. To estimate life expectancies at age 65, abridged life tables were constructed using Chiang’s method [18, 19]. Table 1 presents population figures and deaths from official population statistics and DRV pension data. For German men, the death and population ratios of population statistics to DRV statistics are the same. Both official population and official death figures exceed the figures provided by DRV by 9% over the whole period, indicating that DRV data cover 91% of the German population. For foreigners, however, there is a pronounced disagreement between population and death figures. The national figure for the former exceeds the DRV figure by 33%. For the latter, by contrast, the national figure is lower by 33% than that given by the DRV. This gap suggests a major bias in population statistics on foreigners, induced by an over-stated population and under-stated deaths. Table 1 suggests that mortality of foreigners in the official statistics is about two-fold higher than its true value based on the DRV data.Table 1Comparison of male population and death counts (age 65+) between German official population statistics and DRV data for the years 1995–2004Nationality and yearPopulationDeathsIn thousandsRatioIn thousandsRatioPopulation statisticsDRV statisticsPopulation statistics/DRV statisticsPopulation statisticsDRV statisticsPopulation statistics/DRV statisticsGermans    19954408.04023.11.10277.7255.11.09    19984704.24311.31.09270.7252.41.07    20015208.74760.21.09268.5232.51.15    20045879.45356.61.10277.1264.11.09    1995–200420,200.218,451.31.091,093.91,004.11.09Foreigners    1995109.776.71.432.73.70.74    1998144.7104.21.393.34.60.71    2001190.0144.21.324.05.50.73    2004251.0197.31.274.98.40.58    1995–2004695.3522.31.3314.922.10.67Calculated from: official population statistics from the German Federal Statistical Office [14]; pension statistics from scientific use file SUFRTBNRTWF94-04TDemoKibele [5] Age-specific death rates (displayed in Fig. 1) accordingly show good agreement for Germans between the official population statistics and the pension statistics except for very old ages, i.e. 90+ at which the official population statistics understate the male mortality [8]. For foreigners, there is a very large disagreement in age-specific death rates. Official population statistics on the death rates for foreigners are extremely low at all ages and do not show the expected increase over age. The corresponding figures of DRV data are similar to those for Germans.Fig. 1Comparison of death rates between German official population statistics and DRV data: pooled data for the years 1995, 1998, 2001, and 2004. Calculated from: official population statistics from the German Federal Statistical Office [14]; pension statistics from scientific use file SUFRTBNRTWF94-04TDemoKibele [5] Table 2 shows nearly the same life expectancy figures for Germans at age 65 in both data sources (15.3 years in the population statistics and 15.6 years in the DRV statistics for the whole observation period). For foreigners, the population statistics’ figures are unrealistically high (e.g. 30.2 years) and very different from the figures provided by the DRV data (15.0 years). Importantly, German life expectancy in the DRV data is slightly higher than that of foreigners. Finally, both in the population statistics and the DRV data, overall life expectancy based on pooled data on Germans and foreigners hardly deviates from the life expectancy of Germans because the impact of the foreign population figure is small.Table 2Life expectancy at age 65 for men in Germany, for the years 1995, 1998, 2001, and 2004, in yearsYearGermansForeignersGermans and foreignersPopulation statisticsDRV statisticsPopulation statisticsDRV statisticsPopulation statisticsDRV statistics199514.714.625.914.314.714.6199815.215.229.214.815.415.2200116.016.434.115.716.216.4200416.616.237.815.016.816.11995–200415.315.630.215.015.415.6Calculated from: official population statistics from the German Federal Statistical Office [14]; pension statistics from scientific use file SUFRTBNRTWF94-04TDemoKibele [5] The life expectancy figures of the order of 30 years at age 65 calculated from German official population statistics are obviously implausible. They are much higher than the levels of longevity experienced by Japanese women (23.2 years in 2005), the world’s lowest mortality population. The bias is produced by the inflated population denominator and the under-stated death numerator in German population statistics. By contrast, DRV data (which is based on the rigorous follow-up of people eligible to receive German old-age pensions) show that the mortality of foreigners aged 65+ is actually slightly higher than the mortality of German men. At least among men aged 65+ the seemingly large mortality advantage of foreigners compared to Germans is a statistical artifact. Our study makes clear that the inaccuracies in estimation of the population denominator and the death numerator can result in serious underestimation of immigrant mortality. However, a significant mortality advantage of immigrants could be found even in studies where the researchers were well aware of the possible bias [5]. Even in the Swedish study mentioned earlier that revealed a general underestimation of mortality among immigrants, a statistically significant mortality advantage could be observed in some immigrant groups, i.e., women from Southern Europe, men from Latin America, Africa and Asia [10]. The underestimated mortality of foreigners should result in the underestimation of mortality in the broader group of German residents who have a migrant background. As mentioned earlier, foreigners constitute about 40% of all people with migrant background at ages 65+. Assuming for simplicity the same level of true mortality among all German residents with migrant background and precisely reported mortality of German nationals with migrant background, it is easy to see that the two-fold underestimation of mortality of foreigners leads to a 25% underestimation of mortality of all residents with migrant background. How can the mortality advantage of the migrant population found elsewhere [1, 2, 4, 5] be reconciled with the complete absence of such an advantage in our analysis? The answer may be found by considering the peculiarities of our data compared to the previous studies. First, our study considers “foreigners” or pensioners with current foreign citizenship. Mostly, these are people from Southern Europe who entered Germany in the 1950–1970s and to a much lesser extent more recent immigrants from the former Soviet Union and Eastern Europe. Second, our study includes only people aged 65 and over, while the other studies looked mostly or exclusively at people of working age. Close to the age of their migration, the health advantage of immigrants can be attributed to the selection effect of healthy people, to healthier behaviors and dietary habits, and to positive psychological expectations [1, 2, 7]. For many reasons, the initial advantage does not translate into a mortality advantage at ages of 65 and over. Certain factors affecting immigrants in the country of their new residence can lead to a steeper (compared to the host population) health decline with age. In particular, immigrants may gradually adopt less healthy “western” lifestyles associated with higher coronary risk. In old age they may also experience higher mortality from conditions related to childhood infections and deprivation, such as stroke and stomach cancer [7, 20, 21]. Most importantly, immigrants face the health consequences of lower socioeconomic status such as poorer life conditions, restricted life chances, and corresponding psychological stress [22]. According to DRV data, 86% of foreigners but only 45% of Germans belong to the lower 40% of the lifetime income distribution. Our earlier analysis shows that this category of pensioners experience approximately 50% mortality elevation compared to those belonging to the upper lifetime earnings’ quintile [23]. Foreigners are also characterized by a higher share of former manual workers: 80% vs. 58% among Germans. This status is associated with a 35% mortality excess compared to the white-collar occupations [23]. Speculation about potential reasons for a steeper health decline with age among the immigrants is supported by empirical evidence that confirms that the initial health advantage of foreigners “wears off” over time [3, 24, 25]. This pattern can lead to mortality convergence and even mortality crossover between the host and migrant populations. In general, our results support the concern about the reliability of very low estimates of migrant mortality at ages 65 and over. It is possible that, despite the initial health advantage, the slightly higher mortality of foreigners in the 65+ population is a result of socioeconomic health inequalities. This is a serious public health concern that deserves further investigation.

Semin_Immunopathol-4-1-2315689

Mechanisms of leukocyte migration across the blood–retina barrier

Immune-mediated inflammation in the retina is regulated by a combination of anatomical, physiological and immuno-regulatory mechanisms, referred to as the blood–retina barrier (BRB). The BRB is thought to be part of the specialised ocular microenvironment that confers protection or “immune privilege” by deviating or suppressing destructive inflammation. The barrier between the blood circulation and the retina is maintained at two separate anatomical sites. These are the endothelial cells of the inner retinal vasculature and the retinal pigment epithelial cells on Bruch’s membrane between the fenestrated choroidal vessels and the outer retina. The structure and regulation of the tight junctions forming the physical barrier are described. For leukocyte migration across the BRB to occur, changes are needed in both the leukocytes themselves and the cells forming the barrier. We review how the blood–retina barrier is compromised in various inflammatory diseases and discuss the mechanisms controlling leukocyte subset migration into the retina in uveoretinitis in more detail. In particular, we examine the relative roles of selectins and integrins in leukocyte interactions with the vascular endothelium and the pivotal role of chemokines in selective recruitment of leukocyte subsets, triggering adhesion, diapedesis and migration of inflammatory cells into the retinal tissue. Introduction Immune-mediated inflammation in the retina is regulated by a combination of anatomical, physiological and immuno-regulatory mechanisms, referred to as the blood–retina barrier (BRB). The BRB is thought to be part of the specialised ocular microenvironment that confers protection or “immune privilege” by deviating or suppressing destructive inflammation [120, 121]. These mechanisms are designed to prevent normal immune surveillance and delete or inactivate cells migrating across the BRB to mitigate the effects of deleterious immune responses [15, 99]. Nevertheless, retinal inflammation does occur, and in addition to well-defined inflammatory diseases such as uveoretinitis, immune mechanisms affecting the integrity of the BRB and leukocyte infiltration of the retina are implicated in other ocular diseases such as diabetic retinopathy [76] and age-related macular degeneration (ARMD) [72]. The barrier between the blood circulation and the retina is maintained at two separate anatomical sites. These are the endothelial cells of the inner retinal vasculature and the retinal pigment epithelial cells (RPE) on Bruch’s membrane between the fenestrated choroidal vessels and the outer retina. For leukocyte migration across the BRB to occur, changes are needed in both the leukocytes themselves and the cells forming the barrier [9, 137]. The retina is an extension of the central nervous system (CNS), and as such, many of the molecular structures making up the inner BRB are similar to those found in the vascular blood–brain barrier (BBB) [1]. However, there are also important differences between the tissues, including vascular heterogeneity in the brain and numbers of microglia in the retina. Much research has been carried out into mechanisms controlling inflammatory cell adhesion and extravasation of leukocytes from the circulation into neural tissues, and a paradigm invoking sequential, separate events involving interacting pairs of selectins and their ligands, chemokines and cell adhesion molecules (CAM) has evolved to be one of the central tenets in immunology (Fig. 1) [12]. Less is understood about the process of extravasation, and there is evidence for leukocyte extravasation in the CNS both through intercellular endothelial tight junctions (paracellular route) and for transcellular migration through the endothelial cell itself [45]. Fig. 1Mechanisms involved in leukocyte trafficking across the blood-retina barrier at the endothelium. Initial interactions between leukocyte and endothelium are usually mediated by selectins that induce tethering and rolling (1). If G-protein-coupled receptors on the leukocyte engage appropriate chemokines on surface of endothelium, then the leukocyte may become activated (2), leading to conformational changes in integrin molecules allowing firm adhesion to the endothelium and leukocyte spreading (3). Diapedesis across the endothelium and into the retina can then take place (4). This is triggered by additional chemokine and cytokine signals and gradients and mediated by matrix metalloproteinases secreted by the leukocyte and alterations in the signalling and regulatory molecules of the TJ that control the interaction between the membranous component and the cytoskeleton of the endothelial cell, leading to breakdown of the BRB (5). How inflammatory responses in the retina are initiated or resolved is still controversial. There is evidence that resident and/or infiltrating cells may influence the progression (+Ve signal?) or resolution (−Ve signal?) of the inflammatory response To date, many studies have been based on neutrophil migration, or undefined mixed leukocyte populations, but recent research, including our own, has revealed that different leukocyte subsets have distinct requirements for endothelial cell interactions and subsequent migration. These requirements may also vary depending on the type of tissue, whether normal surveillance or inflammatory trafficking is involved and if inflammatory on the type of stimulus. Mononuclear cells including monocytes and T lymphocytes are primarily involved in the pathologies observed in various retinal inflammations and will be the focus for this review. Understanding these processes and whether mechanisms really differ between peripheral tissues and ocular or other CNS sites such as the brain is important as every stage of the process, from initial engagement to migration into the tissue, is a potential target for immunotherapy to control retinal damage [53]. Leukocyte adhesion cascade Adhesion molecules in the leukocyte adhesion cascade The process of the inflammatory response and how blood leukocytes are recruited from the blood to the tissues has been described as the leukocyte adhesion cascade. Rolling of leukocytes on the luminal endothelial cell wall is usually the first step in the adhesion cascade that culminates in leukocyte extravasation, and the mechanisms underlying the various stages involved has been recently reviewed [82]. Selectins, such as L selectin, expressed on microvilli of most leukocytes and P and E selectin, which may be expressed by activated or inflamed endothelium, mediate both tethering and rolling [119]. These molecules interact with ligands such as P-selectin glycoprotein ligand-1 (PSGL-1), expressed on other leukocytes or endothelium only when correctly glycosylated [89]. This initial tethering and fast rolling allows adherence under flow, and shear stress forces support the adhesive and signalling interactions taking place that then mediate slow rolling. Integrins are also involved in rolling, and α4β1-integrin (VLA4) dependent rolling and LFA-1/ICAM-1 interactions have been shown to be important for firm adhesion and lymphocyte trafficking in the CNS [44, 59]. Leukocyte rolling via low-affinity interactions can then be converted to firm adhesion as chemokine-receptor G-protein dependent activation takes place, resulting in conformational changes in the binding domains of LFA-1 to high-affinity state and in membrane clustering of the integrins giving increased avidity [6, 78]. This leads to firm adhesion and is the prelude to spreading, crawling and migration through the endothelium, either by the paracellular route involving release of endothelial adherens junctional proteins [87, 113, 126] or the transcellular route via small continuous membrane associated structures or vesiculo-vacuolar organelles. Chemokines in the leukocyte adhesion cascade Leukocyte migration is controlled to a large extent by members of a family of chemoattractants, the chemokines [109]. The panel of chemokine receptors that a leukocyte carries enables it to respond to chemokine signals either from normal tissue thus controlling immune surveillance or from an inflammatory site. Forty-three human chemokines have now been officially named, and these do not include isoforms, polymorphisms, splice variants and those chemokines encoded by viruses. There are currently 18 chemokine-specific G-protein-coupled receptors [92]. As these chemokines can be allocated to four groups based on the position of 2 N-terminal conserved cysteines, this has been used as the basis for their systematic nomenclature [7, 143]. The two main groups of chemokines are now designated either CCL, where there is no amino acid separating the cysteines, or CXCL, where there is one amino acid between the cysteine molecules. Chemokines can bind to more than one receptor, and receptors can bind to more than one chemokine. Receptor affinity varies between the ligands but does not necessarily determine the functional potency of the chemokine [32, 130]. In response to ligand binding, the receptors trigger numerous secondary mediators, which initiate functional responses such as cell migration. The resulting functions may differ depending on the cell type and the prevailing microenvironment [77, 111]. The multiple different combinations of chemokines and receptors, expressed either simultaneously or sequentially [13], plus the cell type and the local microenvironment ensure that the chemokine communication system can deliver a precise and accurate communication message to the required cell based on local circumstances, which may be rapidly changing. Chemokines add additional specificity to the adhesive interactions initiated by the adhesion molecules during the recruitment of leukocytes from the circulation. Chemokines produced within the tissue are transported by transcytosis to the luminal surface of the endothelium. The transport across the endothelium is active and may involve the chemokine binding protein DARC [110]. Once at the luminal surface, the chemokines are displayed by glycosaminoglycans (GAGs) [124] at the tips of endothelial processes [95, 96]. This prevents the desensitisation of leukocyte chemokine receptors by chemokines in the blood stream and focusses the action of the chemokines on leukocyte adhesion and emigration. Chemokines have a GAG-binding site as well as specific receptor binding site [5], and GAG binding has been shown to be important in vivo [104]. Chemokines presented on the endothelium stimulate firm adhesion by increasing integrin affinity and avidity [6, 20, 26, 74] and then the spreading of cells, their migration to the endothelial cell junction [129] and the initiation of leukocyte transmigration [23, 24]. More than one chemokine is likely to be needed for the complete process of leukocyte extravasation with specific chemokines at distinct steps [142]. It is thought that chemokines do not need to form a gradient across the endothelium to enable adherent cells to move from the luminal surface into the tissue as surface immobilised chemokines and a shear force may be sufficient for transendothelial migration [24]. Normal retina and immune surveillance In normal physiology, one function of the immune system is to recognise and destroy cells that are abnormal (e.g. transformed tumour cells) or that display foreign or non-self antigens (e.g. virus-infected cells). This entails recirculation of relatively long-lived lymphocytes through the blood stream and lymphatics, moving from one lymph node to another and to peripheral inflammatory sites. In contrast, myeloid monocytes and dendritic cells, generated in the bone marrow, recirculate in the blood for only a few days before migrating into the tissues, forming extensive networks of phagocytic antigen presenting cells (APC). Some of these cells differentiate into long lived, highly specialised cells such as the osteoclasts of bone and microglia of the nervous system. In the tissues, APC may be stimulated by ingesting apoptotic tissue cells or by contact with microbes. They then migrate via the lymphatics to local draining lymph nodes where interaction with T and B lymphocytes can occur, and either tolerance to self-antigens or immunity to altered-self or foreign antigens is generated. Immune surveillance must be regulated in vulnerable tissues such as the brain and retina where tissue repair cannot take place, so structures that anatomically separate the central nervous system have evolved. Thus, the BRB may protect the delicate neural tissue from local, damaging inflammation but at the cost of defective peripheral tolerance to retinal antigens. Consequently, the retina remains vulnerable to autoimmune inflammation, increasing the importance of the BRB in excluding, eliminating or suppressing infiltrating inflammatory cells [15, 99]. Structure and function of the blood–retina barrier The inner BRB formed at the endothelium is supported by perivascular cells such as smooth muscle cells, pericytes or retinal macrophages [53]. Astrocyte foot process contact is known to be essential for formation of intercellular tight junctions (TJ) and maintenance of structural integrity [1]. Other complex interactions between other constituents of the microenvironment surrounding the endothelial cells, such as the basement membrane, nerve endings, microglial cells and the extracellular fluid, are all required for the proper functioning of the barrier. The outer BRB formed at the RPE cell layer is simpler, the barrier being maintained by tight junctions between polarised epithelial cells on Bruch’s membrane. Barrier function develops during embryogenesis, and once formed, is highly selective excluding molecules above 300 kDa. It is also effective in excluding most migrating cells, repopulation of the retina with bone-marrow-derived cells such as microglia taking from 6 to 12 months [4, 132], the rate of haematopoetic-derived cell turnover probably being dependent upon integrity of the RPE cell layer [134]. In normal retina, two distinct molecular structures form the physical BRB; these are the TJ and the adherens junctions (Fig. 1). These closely resemble TJ found in neural tissue in the brain [46, 102]. The TJ are highly organised pericellular structures that appear as multilammellar, usually continuous strands containing several integral membrane proteins that seal adjacent endothelial or epithelial cells, creating distinct membrane domains that restrict soluble and ion transport to apical and basal surfaces and block paracellular passage of macromolecular fluids and cells. The two major components are the claudins and the occludins, which are linked via zonula occludin (ZO) proteins to signalling proteins and the actin cytoskeleton. The junctional adhesion molecules (JAMs) are also located in the TJ, JAM A and C being particularly highly expressed in the RPE consistent with a role in establishing and maintaining cell polarity [33, 90]. Adherens junctions are formed mainly by cadherins (mainly VE cadherin), which interact with catenins to bind the cytoskeleton. PECAM/CD31 expressed by endothelial cells in the junction also binds to the cytoskeleton via catenins and has been implicated in the migration of monocytic myeloid cells [98]. The general structure and regulation of neural tight junctions in retina and brain has been recently reviewed [46, 63]. Leukocyte infiltration of normal retina Despite an apparently intact BRB, lymphocytes have been shown to infiltrate the normal retina. If both the leukocytes and the endothelium are normal, leukocytes do not cross the blood–retina barrier [135, 137], but if lymphocytes are activated, they are able to initiate a transient breakdown in the BRB, enabling sampling of the retinal environment and possibly further recruitment of inflammatory cells [103, 139]. The mechanisms controlling this sampling or immune surveillance of the retina in the absence of retinal inflammation are not clear, and similar observations have been made in the CNS [65, 66]. Using scanning laser ophthalmoscopy, we have shown that circulating, activated T cells induce changes in the retinal vasculature that allow T cell firm adhesion to venule endothelium without rolling (secondary tethering), inferring a role for chemokine or integrin signalling in T cell capture and subsequent diapedesis in the absence of selectin-mediated rolling. Furthermore, recirculation of as few as 1 × 105 activated cells for 8–16 h is sufficient to allow a permissive endothelium to develop, allowing T cell adhesion and diapedesis. This compares with a minimum of 1 h for diapedesis when both leukocytes and endothelium are activated [138]. Interestingly, retinal antigens have been identified as having chemotactic properties for T cells via chemokine receptors, providing a mechanism for recruitment of lymphocytes to the retina in the absence of inflammation [68, 101]. Whether this occurs in vivo and contributes to induction of autoimmune inflammation remains to be proven but is consistent with the observation that migration of activated retinal antigen-specific T cells into the retina is not enhanced compared with non-specifically activated T cells [103]. Leukocyte infiltration of the retina during inflammation Various inflammatory conditions in the eye involve dysfunction of the BRB, and in addition to overt inflammation of the retina in diseases such as uveoretinitis, immune dysfunction leading to vascular permeability and angiogenic changes is linked to pathologies as diverse as retinopathy of prematurity, ARMD and diabetic retinopathy. In premature infants, where the retinal vasculature is underdeveloped, changes in oxygen level in the retina can signal release of vascular endothelial growth factor (VEGF) and development of abnormal vessels [21], and anti-VEGF therapies for ARMD, although possibly harmful to normal circulation, are widely used [88]. VEGF mediates increased vaso-permeability through p38 MAP kinase and ERK1/ERK2-dependent mechanisms that alter occludin expression. VEGF activity, in turn, is regulated by eNOS-dependent dephosphorylation of the VEGF receptor-2, indicating close regulation of these pathways in normal retina [11, 14]. Decreased localisation of occludin at both endothelial and RPE cell borders has been shown to be linked to increased serine and threonine phosphorylation in occludin, which in turn reduces binding to ZO-1 and connection with the cytoskeleton [46]. Angiogenesis and inflammation VEGF can be secreted by many retinal cell types under stress, including pericytes, astrocytes and RPE, and although not normally considered an inflammatory mediator, VEGF is chemotactic for monocytes, adhesion and migration through endothelium being mediated by increased integrin expression at sites of angiogenesis [64]. VEGF164, in particular, has been shown to induce inflammation and cellular immunity during pathological but not physiological ischaemia-induced sites of retinal angiogenesis [70]. Thus, pathologies involving VEGF are increasingly considered to have an inflammatory component. However, in the absence of adaptive immune responses to antigen in these non-infectious conditions, the inflammatory response is limited to the immediate vicinity by the naturally immunosuppressive ocular microenvironment, avoiding the mononuclear cell influx that characterises infectious or autoimmune inflammation [15, 120]. Diabetic retinopathy caused by systemic changes in metabolic and cardiovascular control is a well-described complication of diabetes and also linked to ischaemic release of VEGF, leukostasis and BRB breakdown [69, 105]. Hyperglycemia and advanced glycation end products also promote pathology in a number of cell types including endothelium, neurons and glia. Thus, retinopathy may be considered a chronic inflammation involving all major cell types within the retina [52]. What first triggers retinal microvascular changes and breakdown of the BRB in diabetic retinopathy is unclear. In a rat model of diabetes, retinal vascular endothelial cells and RPE showed elevated levels of the matrix metalloproteinases (MMP) MMP-2, MMP-9 and MMP-14 with specific degradation of occludin [56]. Advanced glycation end products also increased leukostasis through NFκB upregulation of integrins as early events [97]. There is also a direct relationship between the VEGF and prostaglandin-cyclo-oxygenase (COX-2) pathways. These pathways have vasoactive effects and interact with the nitric oxide synthase pathway (NOS), which in turn also increases vasodilation [57, 128]. Endothelial NOS (eNOS) is constitutively expressed by endothelium and is important for maintaining normal vascular physiology, in part by maintaining smooth muscle tone, but the inducible isoform (iNOS) has also been shown to have a predominant role in leukostasis and BRB breakdown in diabetic iNOS knock-out mice [80]. Expression of ICAM-1 and protein levels of all isoforms of NOS were increased in diabetic mouse retinas whilst occludin and ZO-1 protein decreased. These effects were prevented by the NOS inhibitor N(G)-nitro-l-arginine methyl ester (l-NAME) and absent in the iNOS knock-out diabetic mouse. Leukocyte infiltration of the retina in uveoretinitis Endogenous posterior uveoretinitis (EPU) or more correctly, posterior intraocular inflammation, can take many clinical forms such as multifocal choroiditis, sympathetic ophthalmia and pars planitis. These clinical forms have features in common, including retinal vasculitis, macular edema, focal chorioretinal infiltrates and inflammatory cells and exudates in the vitreous, which develop as a result of BRB breakdown. EPU appears to be autoimmune- or immune-mediated rather than infectious in nature. However, it is becoming clear that infectious agents may be linked to autoimmunity through molecular mimicry, polyclonal activation and failure of bystander suppression [49]. Experimental autoimmune uveoretinitis (EAU) is a Th1-type organ-specific autoimmune disease induced by immunisation with retinal antigens such as S antigen and interphotorecepter retinoid-binding protein (IRBP), at distant sites, in susceptible strains of rats or mice. It serves as an animal model of posterior intraocular inflammation, closely mimicking effects seen in the human situation [17, 50]. The BRB is breached in EAU, allowing leukocytes, both CD4+ T lymphocytes and macrophages to move into the retina, resulting in extensive tissue damage [38, 83]. Macrophages are thought to be the main effector cell responsible for tissue damage. EAU can also be induced by adoptive transfer of 5 × 105 retinal antigen-specific T cells [16, 17, 50, 103, 114, 122], and in particular, Th1 cells [18]. Mechanisms of leukocyte migration across the blood–retina barrier in uveoretinitis Adhesion molecules involved in leukocyte–endothelial cell interactions Appropriate adhesion molecule expression on both leukocytes and cells of the BRB is clearly important in enabling the migration of leukocytes into the inflamed retina. During the initial stages of the inflammatory reaction, the endothelium is activated by locally produced cytokines, TNFα and IL-1β in particular. The cytokines up-regulate adhesion molecules, particularly the selectins [108] and chemokines on the luminal surface of the endothelium. Leukocytes in the circulation then respond to the activated endothelium. In vitro studies have shown clear differences in adhesion molecule expression between human RPE and retinal endothelial cells [85]. ICAM is constitutively expressed at high levels on RPE cells, whereas endothelial cells expressed ICAM-1 only after induction with IFNγ. Binding of CD4+ lymphocytes to RPE was dependent on ICAM-1, although after maximal stimulation, there was also an ICAM-1-independent component. Migration of T lymphocytes across retinal endothelial cells in the rat used both LFA-1-dependent and LFA-1-independent mechanisms [58, 61, 94]. Migration of T lymphocytes across unstimulated RPE was dependent on ICAM-1 and LFA-1, whereas after activation of RPE with IFNγ, migration was dependent on both ICAM-1 and VCAM-1 [37]. Receptors for various integrins and adhesion molecules are up-regulated on human RPE and vascular endothelial cells in vivo during ocular inflammation such as sympathetic ophthalmia. These include enhanced or de novo expression of ICAM-1, VCAM-1, CD62E and CD44, which in vitro, may be induced by the cytokines IL-1 and TNF-α. Kuppner et al. [79] compared adhesion molecule expression in acute and fibrotic sympathetic ophthalmia with normal human eyes. Expression of several integrins of the VLA family, which form receptors for extracellular matrix proteins, were found to be expressed by the RPE. However, VLA-4, the ligand for VCAM-1 and fibronectin, VLA-5 (fibronectin) and VLA-6 (laminin), which were increased on endothelium in the inflamed eyes, were not up-regulated on RPE cells. In contrast, ICAM-1 and CD44 expressions were greatly enhanced. Electron microscopy studies have also shown differences in the manner in which leukocytes cross the retinal endothelium and the RPE [40]. Inflammatory cell diapedesis through Bruch’s membrane involved separation of its constitutive layers, migration through the pores in the membrane and between the RPE cells without apparent significant physical disruption of the RPE cell layer. Retinal vascular diapedesis, however, involved changes in post-capillary venules to high endothelial cell morphology, indicating endothelial cell activation [93], with migration appearing to be via the paracellular route [60]. As RPE cells fail to express iNOS during inflammation whilst endothelium does, [19, 67] it has been proposed that the RPE maintains a predominantly immunosuppressive role during inflammation [84]. Mechanisms of leukocyte subset migration into the retina To study the role of adhesion molecules and leukocyte induced breakdown of the BRB, we have developed scanning laser ophthalmoscopy (SLO) methods for use in rodent EAU models. In this technique, syngeneic fluorescently labelled (eGFP CFSE or calcein AM) splenocytes, T helper cell subsets or monocytes are injected into the tail vein of normal or transgenic mice at various disease stages. These transferred cells can then be tracked in vivo in real-time and their interaction with the retinal vascular endothelium quantified in terms of rolling efficiency, rolling velocity and adhesion. These images can then be analysed to investigate the role of various molecules involved in leukocyte recruitment at the BRB [131]. This can be followed by confocal microscopy of retinal wholemounts, which allows both the quantification of cells entering the retina and further immunostaining to reveal changes in phenotype of endothelial cells and trafficking leukocytes. The earliest changes observed are 24 h prior to detectable leukocyte infiltration. Vascular changes occur, which reduce shear stress in retinal veins and venules from approximately 30 to 20 dyn/cm2. Although still considerably higher than in normal fenestrated endothelium, as shear stress falls, rolling and sticking efficiencies of leukocytes increase in the post-capillary venules, correlating with selective upregulation of P and E selectin and ICAM-1 on these vessels, particularly at sites of extravasation [137]. This effect is not observed in arteries or arterioles, although ICAM, VCAM, PECAM, E selectin, P selectin and CD44 are also up-regulated on other retinal endothelium as EAU progresses [135, 140]. As CD44 is widely expressed on most leukocyte subsets, it is required for leukocyte rolling under flow and is involved in early stages of extravasation in non-neuronal tissues [34, 35]. We investigated the role of CD44 in retinal inflammation in our model. Treatment of a mixed leukocyte population with neutralizing antibody against CD44 significantly suppressed the rolling of these cells on inflamed retinal venules and reduced infiltration into the retina. The importance of CD44 in leukocyte recruitment in EAU was also supported by the fact that EAU severity was diminished by administration of the anti-CD44 antibody at the stage of disease in which leukocytes are first seen to infiltrate the retina [140]. Both Th1 cells and Th2 subsets of T lymphocyte are implicated in the pathogenesis of uveitis [10, 27]. So to investigate the role of adhesion molecules in the trafficking of different T cell subsets across the BRB in Th1-mediated EAU, naive CD4+ T lymphocytes were isolated from lymph node cells and polarised in vitro into Th1- and Th2-like cells. Surface PSGL-1 and LFA-1 were up-regulated on both populations but expressed at higher levels on Th1-like cells, whereas CD44 expression was up-regulated to a greater extent on Th2 cells. Pretreatment of the polarised T lymphocyte populations with anti-PSGL-1 inhibited rolling and infiltration of Th1-like cells but not Th2, providing direct in vivo evidence for the inability of Th2 to respond to P/E selectin despite increased expression of PSGL-1. Anti-LFA-1 pretreatment inhibited infiltration of both Th1- and Th2-like cells, but this inhibition was more substantial for the Th-1 cells. Thus, preferential recruitment of Th1 cells in EAU is mediated by PSGL-1:P/E selectin, whereas non-selective trafficking of activated T cells (both Th1 and Th2) across the BRB is mediated by CD44:CD44r and LFA-1:ICAM-1 [137]. Monocytes are also necessary for full expression of disease in EAU. We have established a model for studying bone-marrow-derived monocyte trafficking in vivo, and found that, whereas T cells, whether antigen specific or not, roll on inflamed endothelium and rapidly migrate into the retina, monocytes can only traffic into the inflammatory site once they have acquired a specific phenotype [136]. This phenotype was consistent with the CCR2+ phenotype described by others but was also found to be a constitutive time-limited property of the transferred monocytes and independent of an existing inflammation. Monocytes purified from bone marrow failed to roll on inflamed endothelium until they had undergone in vivo recirculation for at least 24 h, with maximum retinal infiltration not occurring until 48-h post-transfer. Of note was the ability of transferred monocytes to undergo further differentiation during trafficking into the retina. Transferred cells matured into classical tissue macrophages or CD11c+ dendritic cells that persisted within the retina for several days. Several striking contrasts between inflammatory monocyte and Th1 cell trafficking were observed. Monocyte rolling in inflamed venules was faster and CD62-L dependent rather than LFA-1 dependent. While PSGL-1 was found to play a role in regulating diapedesis to the inflammatory site, CD62-L was shown to have a key role in regulating recruitment of monocytes to the lymphoid tissue from the circulation during inflammation (manuscript submitted for publication). These observations reveal that different molecular mechanisms are involved in leukocyte subset adhesion to endothelium and migration into the tissues. Mechanisms of blood–retina barrier breakdown in EAU Whether BRB breakdown is necessary before lymphocytes and monocytes can infiltrate or whether infiltration results in BRB breakdown and free movement of leukocytes between the circulation and the retina is controversial [60, 86]. Blood flow within the CNS and, by extension the retina, is regulated by interactions between neurons, glia and the microvasculature, contact between the microvessel wall and perivascular endfeet of astrocytes, mediated by agrin, being important for maintaining barrier function in the brain [1]. Other cellular components include pericytes and perivascular macrophages, contributing to what has been termed the “neurovascular unit”. Astrocytes can secrete a range of cytokines and chemical mediators that up-regulate TJ and polarised expression of transporters and other enzyme systems that control transendothelial transport of molecules and ions to maintain a metabolic barrier [62]. In BBB disruption, agrin is lost from the abluminal surface of the endothelial cells adjacent to the astrocytic endfeet and contact lost. In the brain, bradykinin activation of astrocytes has been implicated in BBB breakdown in stroke or trauma, but the molecular signals responsible for BRB breakdown in autoimmune inflammation are still not understood [2]. Using our in vivo cell trafficking model, we have studied the relationship between changes in vascular permeability, adhesion molecule expression and leukocyte migration into the retina. As described above, circulation of activated lymphocytes induces changes in the endothelium of the retinal venules, allowing transient breakdown of the BRB. Where T cells are non-antigen specific, this occurs in the absence of cell rolling, without any reduction in shear stress and without generalised inflammation of the retina occurring. In contrast, in EAU where retinal autoreactive T cells are present, cell rolling occurs, and extravasation of T cells in the venules (but not arterioles) occurs together with BRB breakdown in these venules as evidenced by dye leakage and an influx of inflammatory cells [135]. BRB breakdown involves changes in levels of expression and localisation of TJ proteins occludin-1, claudins and ZO-1 [116]. To elucidate mechanisms of BRB and its regulation by inflammatory cells and mediators in vivo, we examined wholemount retinas and choroids from normal and EAU mice. Using confocal microscopy, we showed that in normal retina, TJ proteins were evenly distributed along endothelial cell margins with increased expression in tri-cellular corners, hypothesised to be a “weak spot” in the TJ network. In EAU, disruption of TJ proteins only occurred in the venules where leukocyte infiltration was occurring. Western blot analysis confirmed that claudin 1/3 and occludin proteins were reduced. ZO-1 protein expression was not reduced but redistributed from the cell membrane and the TJs to the cytoplasm [133]. Examination of venules in retinas in early EAU (with few infiltrating cells) showed that loss of occludin-1 from venule TJ occurred at the point of contact with adherent leukocytes undergoing transendothelial migration. In the absence of other migrating inflammatory cells, TJ appeared to reform. In contrast, in areas of the retina where there was significant inflammatory cell infiltration, confocal microscopy revealed that occludin-1 was lost from venules but not capillaries and arterioles. This correlated with loss of astrocyte foot process contact with the endothelium of the venule [133]. These studies are consistent with a major role for venule endothelium in leukocyte recruitment to the retina and imply that breakdown of the BRB in uveitis is an active event triggered by adherent leukocytes rather than a passive event driven solely by local cytokine microenvironment. The role of chemokines in leukocyte migration across the BRB in EAU Chemokines are vital for the development of a focussed immune response, and certain chemokines are associated with inflammatory responses and the attraction of inflammatory cells to the site of infection or injury [91, 111]. Chemokines can be produced by both resident cells at the inflammatory site, such as epithelial cells, endothelial cells, fibroblasts and resident macrophages and dendritic cells as well as early infiltrating macrophages and neutrophils and then at later stages, infiltrating T cells. Chemokine production may be via toll-like receptors [36, 71, 115] or via inflammatory cytokines such as IL-1, TNFα and IFNγ, in the case of a Th1-type response or IL-4, IL-5 and IL-13 in the case of a Th2-type response [8, 107, 109]. Thus, the inflammatory stimulus will generate a chemokine “fingerprint”, which will be the basis for determining the tailored leukocyte response [22, 111]. Production of chemokines is up-regulated in the murine eye in EAU and generally shows, as expected, a Th1-like pattern of expression; for example, CCL2, CCL3 and CCL5 [28]. CCL3 and CCL2 were associated with retinal and choroidal vessels as well as with infiltrating cells, whereas CCL5 was associated predominantly with infiltrating cells [28]. RT-PCR analysis found expression of mRNA transcripts for these and several other chemokines including CXCL9, CXCL10 CCL6, CCL9, CCL19 and CCL22 is also up-regulated, although some of these studies were carried out with whole eyes [31, 51]. Recipients receiving Th1 cells had a similar pattern of expression [51]. Laser capture microdissection showed that both RPE and infiltrating leukocytes expressed chemokine transcripts in distinct but overlapping patterns [51]. In vitro, human cells of the BRB, both RPE and retinal microvascular endothelial cells respond to pro-inflammatory cytokines IL-1β, TNFα and IFNγ by producing substantial levels of CCL2, CCL5, CXCL8, CXCL10 and CXCL1 [29, 41, 42]. Receptors for these chemokines are also up-regulated in EAU. Expression of mRNA transcripts for CCR1, CCR5, CXCR3, CXCR2, CCR6, CCR8 and CCR2 was detected in whole eyes [51], and cell surface expression of CCR5, CXCR3 and CCR2 on infiltrating cells was detected by immunohistochemistry (pers. comm. IJC). We have investigated the importance of CCL3 in recruitment of leukocytes in vivo at the inflamed BRB using SLO. We showed that CCL3, a ligand for CCR5, was involved in leukocyte recruitment at the BRB and was linked to the inflammatory process and disease pathogenesis in EAU [30]. The effect of short-term anti-CCL3 treatment was examined by injecting anti-CCL3 antibody into mice with EAU 1 h prior to tracking a labelled population of activated leukocytes (splenocytes from mice with EAU) in real-time using SLO. This treatment inhibited leukocyte slowing and accumulation and subsequent extravasation of leukocytes at the blood–retina barrier. This was effective predominantly in the post-capillary venules, which have been shown to be the main site of passage of leukocytes across the BRB. Long-term anti-CCL3 treatment also prevented decreased leukocyte velocity and reduced disease severity as measured clinically, histologically and in terms of BRB breakdown [30]. To define this further, we examined the trafficking of Th1-like cells, polarised in vitro. These cells, which expressed high levels of CCR5, were labelled and adoptively transferred and their trafficking monitored in vivo at an early disease stage in EAU using SLO. Treatment of the cells with antibody against CCR5 prior to transfer resulted in a reduction in their infiltration into the retina. However, rolling velocity, rolling efficiency and adherence of the cells to retinal endothelium was not reduced. CCR5 is clearly important for Th1 cell recruitment at the BRB and acts at the level of transendothelial migration rather than at the earlier stage of rolling on the endothelium [31]. This is consistent with other work on the arrest of monocytes or Th-1-like T cells by CCL5 (RANTES) immobilised to activated endothelium under flow conditions. Arrest was mediated predominantly by CCR1, whereas CCR5 was responsible for spreading. This was irrespective of the degree of expression of the receptors on the different cell types [127]. Interestingly, studies in which CCR1/CCR5 was blocked in EAU have emphasised the fact that the action of a chemokine and its receptor in cell recruitment at the BRB may differ dramatically depending on the antigen used to induce uveitis and the stage of the disease and the resulting microenvironment. In Lewis rats in which EAU was induced with S-antigen peptide, treatment with the CCR1/CCR5 receptor antagonist Met-RANTES was effective in reducing uveitis in the efferent phase, possibly inhibiting migration of activated T cells or monocytes into the retina [39]. When EAU was induced with IRBP peptide, however, Met-RANTES appeared to affect activation of T cells, reducing intraocular inflammation if administered early in the initiation phase of the response but moderately enhancing uveitis if given during the efferent phase [39]. Antibody against RANTES given during the efferent phase in C57BL/6 mice immunised with IRBP peptide also led to exacerbation of EAU, and this was attributed to a change in the ratio of T cell subsets recruited favouring CD4 over CD8 T cells [117]. Monocyte trafficking across the BRB has also been examined, and in particular, the role of two chemokine receptors which have been shown to be involved in monocyte recruitment, CCR2 and CX3CR1. This type of experiment is constrained by the number of monocytes, which can be isolated from the blood. However, when in vitro-cultured bone marrow monocytes, non-activated peritoneal monocytes and freshly isolated bone marrow monocyte precursors were compared, only the latter precursors continued to circulate and trafficked efficiently to the inflamed retina in EAU [136]. These were, therefore, labelled and used to investigate trafficking in EAU, and interestingly, it was found that these cells could not initially migrate across the BRB and required 24–48 h in vivo before they started to infiltrate the inflamed retina as described above. This coincided with an increase in expression of CCR2 on these cells. This was only a transient increase, and CCR2 expression and the migratory phenotype were lost 72 h after adoptive transfer [136]. However, evidence from further studies indicates that although CCR2 may improve the efficiency of monocyte transendothelial migration at the inflamed BRB, it is not essential for monocyte emigration at this site (manuscript in preparation). This would confirm findings from other inflammatory sites in which it has been shown that although CCR2-positive monocytes may be recruited preferentially to an inflammatory site, CCR2-negative monocytes are also able to traffic to these sites [106, 112, 123]. CCR2 also has a role in facilitating the egress of monocytes from the bone marrow, with more monocytes present in the bone marrow in CCR2−/− mice and fewer circulating even before infection [112], but its role in monocyte recruitment to an inflammatory site is, in addition to this, and may be mediated by CCL7 (MCP-3) and CCL2 (MCP-1) [125]. In some inflammatory situations, CCR2 is clearly more critical for monocyte recruitment than we show at the BRB; for example, for recruitment into the peritoneum in response to thioglycolate [125]. Thus, reliance on CCR2 for recruitment of monocytes may vary depending on the particular inflammatory stimulus and the microenvironment that is generated. CX3CR1 has also been implicated in the trafficking of monocytes, and although it has been proposed to be important for immune surveillance [55], there are also reports that it has a role in recruitment in an inflammatory situation; for example, in atherosclerosis [25, 81], crescentic glomerulonephritis [47] and in cerebral ischaemia [118]. It is possible that this receptor is important for the recruitment of a subpopulation of inflammatory CCR2-negative monocytes. However, recently, it has been shown in an atherosclerosis model that although CX3CR1-positive monocytes can enter atherosclerotic plaques, CX3CR1 was not necessary for this entry. In contrast, CCR2-positive monocytes, which also express substantial levels of CX3CR1 [55], required CX3CR1 in addition to CCR2 for the accumulation of monocytes in these lesions [123]. In EAU, our studies to date indicate that CX3CR1 is not essential for the trafficking of monocytes into the inflamed retina (manuscript in preparation). Conclusions and implications for design of novel therapies Current treatment for uveitis is still dependent in the most part on systemic, non-specific immunosuppression. Additional therapies such as cyclosporin A, tacrolimus and rapamycin may act by interfering with the production and action of cytokines, particularly interleukin-2, targeting T cell function. Mycophenolate mofetil, an antimetabolite affecting purine synthesis, is also valuable. Other immunosuppressants such as azathioprine, cyclophosphamide and chlorambucil have severe side effects restricting their use. Blockade of TNF-α is also effective at least in the short term and now used clinically in different forms. Despite the usefulness of these therapies, they are all relatively non-specific, and in recent years, the quest for specific, tailored therapies for uveitis has been paramount. The ability of leukocytes to migrate across the BRB depends on many different factors, the importance of which will vary depending on the leukocyte subset, the specific site and the microenvironment created as a result of the inflammatory stimulus and genetic background. Therapies designed to block the passage of cells into the inflamed retina will, therefore, need to be tailored to the situation to increase their potential for success. Understanding in depth the stages in the process of leukocyte migration across the BRB will be fundamental for this. It is clear that there is a degree of redundancy in adhesion molecule use as well as in cytokine, chemokine and integrin signalling, and different inflammatory cell subsets will require different strategies to block their recruitment and effector functions. However, some therapies targeting key intracellular signalling pathways may be beneficial. For instance, lymphocyte trafficking is dependent upon endothelial cell G-protein signalling, and recently, lovastatin has been shown to reduce retinal disease in the mouse model of EAU. Lovastatin inhibited the synthesis of precursors required for prenylation and post-translational activation of endothelial Rho GTPase, an essential step in ICAM-1-mediated leukocyte migration [3, 54]. Although different statins may have different effects in different diseases, they are proven safe drugs and provide attractive therapeutic options. Some therapies that target specific aspects of leukocyte migration are already in use and in clinical trials in some inflammatory diseases. Although targeting some adhesion molecules has not always proved successful [100, 141], there are some promising results. These include natalizumab, directed against VLA-4, which is in use for MS and Crohn’s despite a low risk of progressive multifocal leukoencephalopathy, a viral infection of the CNS [73]; and efalizumab, a recombinant humanised antibody, which binds to LFA-1 preventing binding to ICAM-1, is in use for the treatment of psoriasis [75]. The possibility of targetting chemokines and their receptors to prevent recruitment of inflammatory cells has also received much attention from pharmaceutical companies. For example, a number of compounds with anti-CCR2 activity have been developed and patented with use planned in multiple sclerosis and atherosclerosis [48]. An antagonist of CCR5 (maraviroc) is also available, which has been designed for use in HIV [43]. These drugs may have potential use in inflammatory conditions involving the BRB.

Cancer_Immunol_Immunother-3-1-2082063

Association of antigen processing machinery and HLA class I defects with clinicopathological outcome in cervical carcinoma

HLA class I loss is a significant mechanism of immune evasion by cervical carcinoma, interfering with the development of immunotherapies and cancer vaccines. We report the systematic investigation of HLA class I and antigen processing machinery component expression and association with clinical outcome. A tissue microarray containing carcinoma lesions from 109 cervical carcinoma patients was stained for HLA class I heavy chains, β2-microglobulin, LMP2, LMP7, LMP10, TAP1, TAP2, ERAP1, tapasin, calreticulin, calnexin and ERp57. A novel staining evaluation method was used to ensure optimal accuracy and reliability of expression data, which were correlated with known clinicopathological parameters. Partial HLA class I loss was significantly associated with decreased 5-years overall survival (61% vs. 83% for normal expression; P < 0.05) and was associated with decreased 5-years disease-free survival (DFS) (65% vs. 82% for normal expression; P = 0.05). All APM components except LMP10, calnexin and calreticulin were down-regulated in a substantial number of cases and, except ERAP1, correlated significantly with HLA class I down-regulation. LMP7, TAP1 and ERAP1 loss was significantly associated with decreased overall and (except LMP7) DFS (P < 0.05 and 0.005, respectively). ERAP1 down-regulation was an independent predictor for worse overall and DFS in multivariate analysis (HR 3.08; P < 0.05 and HR 2.84; P < 0.05, respectively). HLA class I and APM component down-regulation occur frequently in cervical carcinoma, while peptide repertoire alterations due to ERAP1 loss are a major contributing factor to tumour progression and mortality. Introduction Tumour cells utilize several mechanisms to escape immune-mediated recognition and destruction [1]. Loss of surface-expressed human leukocyte antigen (HLA) class I molecules is particularly important, as this enables tumour cells to evade recognition and lysis by cytotoxic T-lymphocytes [2–5]. These molecules consist of a glycoprotein heavy chain (encoded by genes within the HLA regions of chromosome 6p) and a β2-microglobulin (β2-m) light chain (encoded on chromosome 15q) [6, 7]. Defects in HLA class I expression are caused by multiple mechanisms, including loss of heterozygosity at chromosome 6p; β2-m or HLA class I mutations; and defective expression and/or function of components of the antigen processing machinery (APM) [8–10]. The APM is the combination of cellular processes responsible for the presentation of endogenous peptides by HLA class I molecules. These peptides are generated by specialized immunoproteasomes (composed of LMP2, LMP7 and LMP10 subunits); and transported from the cytoplasm to the endoplasmic reticulum by TAP (transporter associated with antigen presentation) which is composed of TAP1 and TAP2 subunits. Transported peptides undergo length-specific trimming by ERAP1 (endoplasmic reticulum aminopeptidase associated with antigen presentation 1) and are loaded onto empty HLA class I molecules (with involvement of tapasin and the chaperone molecules calnexin, calreticulin and ERp57) [11–14]. Although various in vitro studies have demonstrated partial or total APM component loss in human malignancies, the extent to which this occurs in vivo is largely unknown. Two recent reports describe down-regulation of several APM components in head and neck squamous cell carcinoma and astrocytoma, and demonstrate an association of these defects with prognosis [15, 16]. As HLA class I-restricted processing and presentation of endogenous peptides is crucial to the generation of a virus-specific cytotoxic T-lymphocyte-mediated reaction, APM defects are assumed to contribute significantly to virus-mediated carcinogenesis. Several studies have confirmed interference in various APM processes by viral peptides [17]. However, the extent to which defective expression of APM components contributes to HPV-induced cervical carcinogenesis remains to be established. One study has assessed the expression of various APM components, although only at the mRNA level [18]. Other studies of APM component expression in cervical carcinoma were largely limited to TAP1 and TAP2 and encompassed relatively small sample sizes. Moreover, the expression data yielded by these studies are conflicting, with TAP1 loss being reported in 0–50% of cases [18–20]. Though TAP1 and HLA class I down-regulation has occasionally been shown to be associated with progressive malignant transformation [21–24], a comprehensive study of the association of APM defects with prognostic parameters has yet to be performed. Since information regarding defective APM component expression in cervical cancer can contribute to the development of T cell-based immunotherapies, we have investigated the expression of all known APM components in a large number of cervical cancer specimens using a tissue microarray (TMA)-based approach [25, 26]. Additionally, we demonstrate for the first time a negative association of APM defects with prognosis and other clinicopathological variables. Materials and methods Clinical characteristics and tissue samples From 109 patients with cervical carcinoma who underwent radical hysterectomy with bilateral pelvic lymphadenectomy (by the same surgical team) between 1985 and 1999, formalin-fixed, paraffin-embedded tissue blocks were retrieved from the archives of the Department of Pathology, Leiden University Medical Centre, The Netherlands. Twenty-six adenocarcinomas and 83 squamous carcinomas were obtained. These cases were not consecutive, but chosen on the basis of availability of sufficient material. The use of clinical material was approved by the institutional review board according to the guidelines of the Dutch Federation of Medical Research Associations. All patients were inhabitants of The Netherlands and had not received pre-operative radiotherapy or chemotherapy. Mean age was 48.5 years, the youngest patient being 24 years and the oldest 87 years at the time of surgery. Follow-up of these patients until 2005 provided information concerning recurrence rate and performance state. Tissue microarray construction The archival slides for all the cases were reviewed; a slide containing representative tumour was selected, and an area of tumour was encircled on the slide. The corresponding tissue blocks were recovered from the archives and the selected area on the slide was circled on the block for tissue microarray construction. Using a manual tissue microarrayer (Beecher Instruments, Silver Spring, MD, USA), the area of interest in the donor block was cored with a 0.6-mm diameter needle and transferred to a recipient paraffin block. The microarray was constructed with a threefold redundancy (three spots for each patient) to increase accuracy. The finalized arrays were then cut into 3 μm-thick sections and mounted on glass slides using an adhesive tape-transfer system (Instrumedics Inc., Hackensack, NJ, USA) with ultraviolet cross-linking. Antibodies The mouse monoclonal antibody (mAb) HCA2 which recognizes a determinant expressed on β2-m-free HLA-A (excluding HLA-A24), HLA-B7301 and HLA-G heavy chains [27, 28], the mAb HC10 which recognizes a determinant on all β2-m-free HLA-B and HLA-C heavy chains and on β2-m-free HLA-A10, -A28, -A29, -A30, -A31, -A32 and -A33 heavy chains [27, 29], the LMP2-specific mAb SY1 [30], the LMP7-specific mAb HB-2 [30], the LMP10-specific mAb TO-7 [30], the TAP1-specific mAb NOB-1 [31], the TAP2-specific mAb NOB-2 [31], the calnexin-specific mAb TO-5 [32], the calreticulin-specific mAb TO-11 [32], the Erp57-specific mAb TO-2 [32] and the tapasin-specific mAb TO-3 [32] were developed and characterized as described. The rabbit polyclonal anti-β2-m-antibody A-139 was purchased from DAKO, Copenhagen, Denmark. The rabbit polyclonal anti-ERAP1 antibody has been described previously and was kindly provided by Dr. M. Tsujimoto, RIKEN, Wako, Saitama, Japan [33–35]. Immunohistochemical staining The slides were dried for 16 h at 56°C before being dewaxed with standard xylene and rehydrated through graded alcohols into water. Standard immunohistochemical staining was performed as described previously [36]. Stromal cells, including infiltrating leukocytes, were used as internal positive controls. Negative control sections of each specimen were processed with omission of the primary antibodies. Evaluation of TMA immunostaining Two researchers (A.M. and E.J.) evaluated all patient cores, and scores were averaged to produce a single evaluation with the scorers being blind to the source of the samples. When slides were scored differently, which occurred rarely (<5%), a consensus was reached by simultaneous evaluation. Staining was scored semi-quantitatively by the quality control system proposed by Ruiter et al. [37]. The intensity and percentage of positive cells in the tumour and the stromal cells surrounding the tumour were determined. The intensity of staining was scored as 0, 1, 2, or 3 indicating absent, weak, clear, or strong expression, respectively. The percentage of positive cells was scored as 0 for 0%; 1 for 1–5%; 2 for 5–25%; 3 for 25–50%; 4 for 50–75% and 5 for 75–100%. The sum of both scores was used to identify three categories of expression: normal expression (total score 7–8), partial loss (3–6) and total loss (0–2). Immunohistochemical staining demonstrated strong positive expression of all examined markers in stromal tissue and tumour-infiltrating inflammatory cells, thereby providing an internal positive control. Statistical analysis The Chi-square test or, where appropriate, Fisher’s exact test was used for evaluation of associations between expression and clinicopathological parameters. All statistical analyses were performed with the SPSS Version 12 software package. Five-year survival rates were calculated and survival rates were compared according to the Kaplan–Meier method using the log rank test, while multivariate analysis of survival was performed according to Cox proportional hazard models. Overall survival (OS) was defined as survival till death due to cervical carcinoma (the two patients that dyed from other causes were excluded), i.e. disease-specific survival, while disease-free survival (DFS) was defined as time to disease recurrence, metastasis, or disease-specific death. All tests were two-sided and the significance level was set to 5%. Results Clinical features Of the total group of 109 patients, 52 were diagnosed as FIGO stage IB1, 30 as IB2, 21 as FIGO IIa and 2 as FIGO IIb. For four patients no FIGO staging was available. Fifty-five patients (50%) received post-operative radiotherapy due to narrow tumour-free margins of excision, lymph node metastases, vaso-invasive growth, parametrial infiltration, depth of infiltration exceeding 15 mm and/or tumour size exceeding 4 cm. Twenty-seven patients had lymph node metastases. Fifty-one patients (58%) were HPV16-positive, 21 (24%) were HPV18-positive, while the remaining patients were positive for HPV31, HPV33, HPV45, HPV52 or HPV68. At the end of follow-up, 80 patients were alive, 2 suffered from recurrent disease, 3 had a metastasis, 22 had died of the disease and 2 had died of causes unrelated to the primary disease. The median follow-up time was 45 months. HLA class I and APM component expression Representative staining patterns for the HLA class I heavy chains and various APM components are shown in Figs. 1 and 2. Expression of the HLA class I heavy chains and β2-m is summarized in Table 1. Of the 109 cases, eight exhibited total β2-m loss; correspondingly, these cases were also negative for HLA heavy chain cell surface expression. As HLA class I expression, and therefore tumour immunogenicity, is not affected by the APM in a β2-m-negative background these eight cases were excluded from further analysis of APM effects. Furthermore, as HLA class I expression is regulated by the APM in a β2-m-independent manner, the overall HLA class I expression score was based on expression of the heavy chains. If either or all the heavy chains were expressed normally, overall HLA class I expression was considered to be normal. Total loss of all heavy chains was counted as total HLA class I loss, and all other cases were counted as partial loss. Fig. 1Immunohistochemical analysis. Staining patterns of a–c HLA-A, d–f HLA-B/-C, g–i LMP7 and j–l TAP1 showing a, d, g, j normal expression, b, e, h, k partial loss of expression and c, f, i, l total loss of expressionFig. 2Immunohistochemical staining patterns of ERAP1. a Normal expression. b Partial loss of expressionTable 1Expression of HLA class I heavy chains, β2-microglobulin and overall HLA class IHLA-AHLA-B/-Cβ2-microglobulinHLA class ITotal lossPartial lossNormal expressionTotal lossPartial lossNormal expressionTotal lossPartial lossNormal expressionTotal lossPartial lossNormal expressionOverall (n = 109)36 (33)22 (20)51 (47)24 (22)29 (27)56 (51)8 (7)28 (26)73 (67)21 (19)24 (22)63 (59)Tumour type AC (n = 26)6 (23)5 (19)15 (58)4 (15)8 (31)14 (54)1 (4)7 (27)18 (69)4 (15)5 (19)17 (66) SCC (n = 83)30 (36)17 (21)36 (43)20 (24)21 (25)42 (51)7 (8)21 (25)55 (67)17 (21)19 (23)46 (56)Data shown as n (%) As shown in Table 2, expression of all APM components except LMP10, calnexin, and calreticulin was lost in some of the cases. This down-regulation was significantly associated with HLA class I down-regulation for all components, except ERAP1. Moreover, defects in overall TAP expression (defined as partial or total loss of either or both of the TAP subunits) were significantly associated with HLA class I down-regulation (P < 0.001). Down-regulation of any APM component was significantly associated with HLA class I down-regulation (P < 0.001). Table 2Overall and histological type-specific expression of APM componentsOverall (n = 101)AC (n = 25)SCC (n = 76)Total lossPartial lossNormal expressionPTotal lossPartial lossNormal expressionPTotal lossPartial lossNormal expressionPLMP214 (14)30 (30)57 (56)0.0145 (20)11 (44)9 (36)0.7729 (12)19 (25)48 (63)0.001LMP73 (3)16 (16)82 (81)<0.0012 (8)9 (36)14 (56)0.1001 (1)7 (9)68 (90)<0.001LMP1000101 (100)TAP14 (4)19 (19)78 (77)<0.0010 (0)4 (16)21 (84)0.2304 (5)15 (20)57 (75)0.001TAP211 (11)26 (26)64 (63)<0.0014 (16)6 (24)15 (60)0.0057 (9)20 (26)49 (65)<0.001ERAP1014 (15)80 (85)0.3140 (0)2 (9)20 (91)0.582012 (16)60 (84)0.307Tapasin1 (1)24 (24)76 (75)<0.0010 (0)10 (40)15 (60)0.2841 (1)14 (18)61 (81)<0.001Calnexin00101 (100)ERp5712 (12)40 (40)49 (48)0.0171 (4)6 (24)18 (72)0.61511 (15)34 (45)31 (40)0.002Calreticulin00101 (100)Data shown as n (%). P values are shown for association with overall HLA class I down-regulation Association with histopathological parameters Depth of invasion exceeding 15 mm and vaso-invasive growth are unfavourable prognostic indicators. Of the cases examined, 51 (45%) demonstrated depth of invasion greater than 15 mm and 38 (33%) demonstrated vaso-invasive growth. Total loss of overall HLA class I expression and of any APM component was found to be significantly associated with depth of invasion exceeding 15 mm (P = 0.034 and 0.045, respectively), while no significant association was found for defective expression of any marker and vaso-invasion. None of the markers was found to be significantly associated with FIGO stage, corresponding to the lack of a significant association with histopathological parameters. No association of any of the markers with specific HPV status of the patients was detected. Interestingly, partial LMP2 and/or LMP7 loss was significantly associated with an absence of detectable lymph node metastases (P = 0.023 and 0.021, respectively). Association with overall and disease-free survival As shown in Table 3 and Figs. 3 and 4, partial loss of overall HLA class I expression was significantly associated with decreased OS, and exhibited a trend towards a significant association with DFS, whereas total loss was not significantly associated with survival. Moreover, partial HLA-A loss (evaluated by HCA2 reactivity) was significantly associated with decreased OS and DFS, whereas total loss was not significantly associated with survival. Partial, but not total HLA-B/-C loss (assessed by HC10 reactivity) showed a trend towards significant association with decreased survival. Table 3HLA class I and APM component expression and 5-years survival ratesOverall survivalDisease-free survivalNormal expressionPartial lossTotal lossNormal expressionPartial lossTotal lossHLA-A42 (83)12 (55)29 (81)42 (83)13 (60)28 (77)HLA-B/-C45 (81)19 (64)18 (75)46 (82)19 (67)18 (73)HLA class I52 (83)15 (61)16 (77)52 (82)16 (65)15 (75)LMP724 (74)17 (87)1 (33)62 (76)13 (80)2 (50)TAP161 (78)8 (70)1 (33)63 (81)13 (67)1 (25)ERAP166 (82)5 (38)–65 (81)7 (50)–Data shown as n (%)Fig. 3Kaplan–Meier curves for overall survival. Patients with normal expression, partial loss, and total loss of a HLA-A, b HLA-B/-C, c overall HLA class I, d LMP7, e TAP1 and f ERAP1 expression. P values are shown for comparison with normal expressionFig. 4Kaplan–Meier curves for disease-free survival. Patients with normal expression, partial loss, and total loss of a HLA-A, b HLA-B/-C, c overall HLA class I, d LMP7, e TAP1 and f ERAP1 expression. P values are shown for comparison with normal expression Decreased OS and DFS was significantly associated with total TAP1 loss, but not with normal expression and partial loss. A similar pattern was observed for partial LMP7 loss, though this was not significantly associated with DFS. ERAP1 loss was significantly associated with decreased OS and DFS as compared to normal ERAP1 expression. Multivariate analysis As expected, well known prognostic parameters such as depth of infiltration exceeding 15 mm, presence of vaso-invasive growth, and presence of lymph node metastases were significantly associated with a shorter survival, while FIGO-stages of IB2 or more showed a significant association with shorter DFS (Table 4). HPV type was not significantly associated with survival (data not shown). Subsequently, multivariate analysis was performed on overall HLA class I, TAP1, LMP7 (for OS only) and ERAP1 expression, combined with the aforementioned prognostic factors. ERAP1 down-regulation was demonstrated to be an independent prognostic parameter for shorter OS (hazard ratio (HR) 3.08; 95% CI 1.07–8.90; P = 0.037) and shorter DFS (HR 2.84; 95% CI 1.01–8.03; P = 0.049), as was presence of lymph node metastases (HR 7.02; 95% CI 2.55–19.32; P < 0.001 for OS and HR 6.64; 95% CI 2.53–17.44; P < 0.001 for DFS). Total TAP1 loss was also an independent predictor for shorter DFS (HR 6.72; 95% CI 1.29–35.06; P = 0.024). Table 4Histopathological parameters and 5-years survival ratesOverall survivalDisease-free survival%P%PDepth of infiltration <15 mm (n = 59)8890 ≥15 mm (n = 51)60<0.00157<0.001Vaso-invasion Absent (n = 68)8185 Present (n = 38)590.031540.002Lymph nodes Negative (n = 80)8686 Positive (n = 27)43<0.00139<0.001Histopathological class SCC (n = 84)7576 AC (n = 26)770.924740.662FIGO stage ≤IB1 (n = 52)8291 ≥IB2 (n = 53)690.060570.002 Discussion In the present study we have investigated HLA class I and APM component expression in cervical carcinoma patients and have evaluated the relationship of these factors with survival and other clinicopathological parameters. Our results show that HLA class I down-regulation occurs frequently (∼40% of cases) in cervical carcinoma, corresponding to previous reports [19, 21]. Synchronous partial or total loss of various APM components was found in a substantial number of cases and was demonstrated to be significantly associated with HLA class I down-regulation. We observed partial HLA class I loss to be significantly associated with decreased OS. Of the APM components, TAP1, LMP7 and ERAP1 down-regulation exhibited significant associations with decreased survival. ERAP1 down-regulation was shown to be an independent, significant predictor of shorter survival. Though defective APM component expression is hypothesized to be an important mechanism of HLA class I down-regulation and therefore immune evasion by tumour cells, information regarding APM down-regulation in carcinomas is limited. This is partly due to the scarcity of reliable antibodies for most APM components other than the TAP subunits. Furthermore, most previous studies of APM expression in cervical cancer were limited by small sample sizes. Lastly, the use of various scoring methods for evaluation of immunohistochemical staining has rendered proper interpretation of findings difficult, leading to conflicting data, e.g. regarding the frequency of TAP1 loss [9, 18–20]. In this study, a panel of recently reported monoclonal antibodies was used to investigate expression of all known APM components in a large set of cervical carcinoma lesions using a TMA-based approach [30–33]. Staining was evaluated according to a previously reported quality control system [37], which is more stringent in defining total and partial expression loss than other evaluation methods, which consider ≤25% positive cells as total loss. The system used here defines total loss as complete or near-complete negativity and is therefore more reliable. Moreover, by excluding β2-m-negative cases from the analysis, all outcome parameters can be reliably correlated to expression levels of the various APM components, as any influence of defective APM expression will only be detectable in a β2-m-positive setting. As an APM component, ERAP1 has been shown to be an important determinant of the repertoire of HLA class I-presented peptides [14, 34, 38–41]. However, to date no systematic investigation of its potential role in HLA class I down-regulation in tumours has been performed. To our knowledge, this is the first description of ERAP1 down-regulation in a human carcinoma. Additionally, we have identified ERAP1 loss as an independent predictor of shorter survival, in the absence of associated HLA class I down-regulation. A possible explanation for this observation is the role of ERAP1 in shaping the antigen repertoire: ERAP1 down-regulation may lead to the preferential loading and presentation of non-tumour-associated peptides, thereby yielding a less immunogenic phenotype and facilitating tumour growth and progression. This hypothesis is supported by previous reports on the disrupted nature of presented peptide repertoire in ERAP1-deficient mice [41, 42]. Also, the effect of down-regulation of several APM components on the presentation of specific HPV 16 E6-derived epitopes has been firmly established [43–45]. Although we found that total TAP1 loss was an independent predictor for shorter DFS, this finding should be interpreted with caution as only four patients exhibited this loss. The decreased significance of ERAP1 as an independent predictor for shorter DFS can be explained by the interdependency of ERAP1-mediated peptide trimming and TAP-mediated peptide transport. An interesting finding in this study is the observed association of LMP2 and LMP7 down-regulation with an absence of lymph node metastasis; this finding is seemingly corroborated by reports that certain immunogenic peptides in melanoma and renal-cell carcinoma are not generated by immunoproteasomes and that presentation of these peptides is reduced upon up-regulation of the immunoproteasomal subunits [46]. Though altered proteasomal cleavage of endogenous proteins can lead to an alternative, less immunogenic antigen repertoire, an adequate mechanism for the association of LMP down-regulation with an absence of lymph node metastases remains to be postulated. Interestingly, we have recently observed that genetic variation in various APM components is significantly associated with increased cervical carcinoma risk. Although the underlying mechanisms of APM component loss are largely unknown, the occurrence of single nucleotide polymorphisms in the various APM genes could form a genetic basis predisposing to either mutations, or regulatory down-regulation [47, 48]. Partial HLA class I loss was observed to be significantly associated with decreased OS; correspondingly, reports in colon cancer have shown partial, but not total HLA class I loss to be associated with worse OS [49]. This finding can be explained by the ability of low levels of HLA class I molecules to provide tumour cells with an escape not only from cytotoxic T lymphocyte-mediated, but also from natural killer (NK) cell-mediated cytotoxicity [2]. This hypothesis is further supported by the association of decreased survival with total TAP1 loss, which has been shown previously to lead to partial HLA class I loss [2, 50–52]. In the present study, an association of loss or down-regulation of expression of the various APM molecules with aberrant HLA expression was observed. Moreover, the partial HLA loss observed is most likely an underestimation of the extent of HLA alterations present as immunohistochemical staining on paraffin sections does not reveal loss of specific HLA alleles [8]. However, there are also other aberrations known to lead to HLA class I alterations and it is plausible to hypothesize that all these mechanisms can coexist in the same tumour sample. In conclusion, we have shown here that defective expression of HLA class I and of various APM components occurs frequently in cervical carcinoma. Partial HLA class I loss and total TAP1 loss are associated with decreased survival, while down-regulation of immunoproteasomal subunits is associated with decreased risk of lymphogenic metastasis. Moreover, we report the first description of ERAP1 down-regulation in a human carcinoma, and have found that this is an independent prognostic parameter for decreased survival. A complete understanding of the mechanisms and relevance of HLA class I and APM component down-regulation and immune evasion may contribute to the rational design of tumour vaccines and T-cell-based immunotherapies.

Eur_J_Appl_Physiol-3-1-1914232

Reproducibility of onset and recovery oxygen uptake kinetics in moderately impaired patients with chronic heart failure

Oxygen (O2) kinetics reflect the ability to adapt to or recover from exercise that is indicative of daily life. In patients with chronic heart failure (CHF), parameters of O2 kinetics have shown to be useful for clinical purposes like grading of functional impairment and assessment of prognosis. This study compared the goodness of fit and reproducibility of previously described methods to assess O2 kinetics in these patients. Nineteen CHF patients, New York Heart Association class II–III, performed two constant-load tests on a cycle ergometer at 50% of the maximum workload. Time constants of O2 onset- and recovery kinetics (τ) were calculated by mono-exponential modeling with four different sampling intervals (5 and 10 s, 5 and 8 breaths). The goodness of fit was expressed as the coefficient of determination (R2). Onset kinetics were also evaluated by the mean response time (MRT). Considering O2 onset kinetics, τ showed a significant inverse correlation with peak- (R = −0.88, using 10 s sampling intervals). The limits of agreement of both τ and MRT, however, were not clinically acceptable. O2 recovery kinetics yielded better reproducibility and goodness of fit. Using the most optimal sampling interval (5 breaths), a change of at least 13 s in τ is needed to exceed normal test-to-test variations. In conclusion, O2 recovery kinetics are more reproducible for clinical purposes than O2 onset kinetics in moderately impaired patients with CHF. It should be recognized that this observation cannot be assumed to be generalizable to more severely impaired CHF patients. Introduction Oxygen (O2) kinetics describe the rate change of oxygen uptake during onset or recovery of exercise and reflect changes in cardiac output and tissue oxygen extraction. Compared to healthy individuals, patients with chronic heart failure (CHF) have slower O2 onset and recovery kinetics, resulting in early fatigue and slow recovery after exertion due to a greater reliance on anaerobic metabolism (Wasserman et al. 1996; Koike et al. 1995). Although peak is widely accepted as a reliable indicator of maximal aerobic capacity in CHF patients (Weber et al. 1982; Janicki et al. 1990), O2 kinetics provide additional objective information on the ability to adapt to and recover from exercise that is indicative of daily life (Riley et al. 1994; Koike et al. 1995). Furthermore, O2 kinetics are potentially useful for risk stratification of CHF patients (Brunner-La Rocca et al. 1999; Schalcher et al. 2003) and for measuring the effects of exercise training, which has already been demonstrated in healthy individuals (Carter et al. 2000) and patients with chronic obstructive pulmonary disease (COPD) (Puente-Maestu et al. 2000). In order to use O2 kinetics for these clinical purposes it is necessary to know more about the applicability and reproducibility of these exercise parameters in this specific patient group. Until now there has been no uniformity in the assessment of O2 kinetics in patients with CHF (Arena et al. 2001). In addition, the reproducibility of O2 kinetics at submaximal exercise in CHF patients has not been studied extensively. Two studies that assessed O2 onset kinetics by different modeling techniques, suggest an acceptable reproducibility of nonlinear regression and an algebraic method (Belardinelli et al. 1998; Sietsema et al. 1994). In both studies, however, intra-class correlations and limits of agreement were not mentioned. The purpose of this study is to evaluate the goodness of fit and reproducibility of previously described clinically applicable methods to characterize O2 onset and recovery kinetics in moderately impaired patients with CHF. Furthermore, we aim to define interventional changes that are required to distinguish from the normal test-to-test variations. Methods Subjects Nineteen patients (15 men, 4 women) with stable CHF (New York Heart Association class II–III and echocardiographical ejection fraction ≤40%) attributed to idiopathic dilated cardiomyopathy (n = 4) or ischemic heart disease due to myocardial infarction (n = 15) were selected at the cardiology outdoor clinic of the Máxima Medical Centre (Veldhoven, The Netherlands). Fifteen patients were in NYHA functional class II and four in class III. Subject characteristics are listed in Table 1. Patients with recent myocardial infarction (<3 months), angina pectoris at rest, atrial fibrillation, or atrial flutter were not included. All patients performed a pulmonary function test using a spirometer (Masterlab, Jaeger, Würzburg, Germany) including measurement of forced expiratory volume in 1 s (FEV1) and forced vital capacity (FVC) during a maximal forced expiratory effort. Patients with chronic airways obstruction, defined as FEV1/FVC < 60% were excluded. Table 1Characteristics of included patients with CHF (N = 19)VariablesMean ± SDRangeAge (years)62 ± 843–78Height (cm)172 ± 8155–184Weight (kg)85 ± 1054–97Body mass index (kg/m2)29 ± 422–37Fat mass (%)a30 ± 720–43Time since diagnosis (months)20 ± 246–96Left ventricular ejection fraction (%)33 ± 719–40a Fat mass was assessed by skinfold measurements (biceps, triceps, subscapular, and suprailiacal) according to standard procedures (WHO 1995) Fifteen patients used beta-blockers and angiotensin-converting enzyme inhibitors, three patients used an angiotensin-converting enzyme inhibitor only, and one patient used a beta-blocker only. Sixteen patients used diuretics. The average duration that patients were using beta-blockers was 34 ± 33 months (range 7–112 months) and 32 ± 29 months for ACE inhibitors (range 7–118 months). Patients who did not use beta-blockers were not different from the other patients with respect to age, peak- or left ventricular ejection fraction. The research protocol was approved by the local Research Ethics Committee of the Máxima Medical Centre, and all patients provided written informed consent. Exercise testing Subjects performed a symptom-limited, incremental exercise test, and on a separate day (at least 3 days later), a constant-load test at 50% of the maximum workload achieved at the first test. This test was repeated at the same time on another day within 2 weeks (mean difference between tests 6.7 ± 3.9 days). All subjects took their medication at the usual time and were instructed not to perform any extra physical activity on testing days. During the testing period all patients were on sinus rhythm. Furthermore, none of the patients reported changes in symptoms, functional status or medication use. Therefore, they could be considered to be in a stable physical condition during the study period. All exercise tests were performed in an upright seated position on an electromagnetically braked cycle ergometer (Corival, Lode, Groningen, The Netherlands). Measurements of and respired fractions of and were obtained breath by breath (Oxycon α, Jaeger, Germany). Volumes and gas analysers were calibrated before each test. The incremental exercise test was performed using an individualized ramp protocol with a total test duration of 8–12 min (Working Group on Cardiac Rehabilitation & Excercise Physiology and Working Group on Heart Failure of the European Society of Cardiology 2001). During the tests all patients were instructed to maintain a pedaling frequency of 70 per minute. A 12-lead electrocardiogram was registered continuously, and blood pressure was measured every 2 min (Korotkoff). The test was ended when the patient was not able to maintain the required pedaling frequency. Maximal workload was defined as the final registered workload, peak- as the average of the last 30 s of the test. Predicted value of peak- was calculated with use of the Wasserman equation, normalizing peak- for age, gender, weight and height (Hansen et al. 1984). The ventilatory threshold was determined by two independent observers using the V-slope method as described by Beaver et al. (1986). The constant-load tests included 2 min of rest, 2 min of unloaded pedaling, 6 min at 50% of the maximum workload and 5 min of rest. Data analysis Mono-exponential model Time constants (τ) were calculated by fitting the data of the constant-load tests to a first-order (mono-exponential) model using the non-linear least squares method (Whipp and Wasserman 1972). Calculations were performed with breath-by-breath data averaged into four different sampling intervals that were used in previous studies: 5 s (5 s) (Matsumoto et al. 1999), 10 s (10 s) (Arena et al. 2002), 5 breaths (5b) (Koike et al. 1995) and 8 breaths (8b) (Pavia et al. 1999). The following formulas were used: A=B=Td=time delayτ=time constant (s)Baseline- was defined as the average of the last minute of the unloaded-cycling stage and steady-state- as the average of the last minute of exercise. The time delay (Td) is a parameter allowed to vary in order to optimize the fit, representing the time between onset of exercise and the start of the mono-exponential increase of One of the determinants of this time delay is the lag time between the computer signal to deliver the work rate and the actual response of the ergometer, which amounted to 2.2 ± 0.6 s in this study. Occasional errant breaths (e.g., due to coughing, swallowing or talking) were deleted from the data set when exceeded three standard deviations of the mean, defined as the average of two following and two preceding sampling intervals (Lamarra et al. 1987). In total, about 1% of the breaths had to be deleted. Algebraic method O2 onset kinetics were also evaluated by an algebraic method calculating mean response time (MRT) (Sietsema et al. 1994), using the following formulas:MRTO2 deficitThe expected amount of O2-uptake was calculated by multiplying the O2-amplitude with exercise duration (6 min). Oxygen deficit was calculated by subtracting the summed above the baseline-value from the expected Statistical analysis All data (presented as mean ± SD) were analyzed using a statistical software program (SPSS 11.0). The ‘goodness of fit’ for mono-exponential modeling was evaluated by the coefficient of determination (R2). The fitting procedure was considered acceptable when R2 ≥ 0.85, as previously described by de Groote et al. (1996). Differences between calculation methods were evaluated by one-way ANOVA with repeated measures and Bonferroni post hoc analyses. In order to assess differences between kinetic parameters of the two tests the paired Student’s t test was used. Linear regression was used to define correlations between variables. Agreement between the kinetic parameters was assessed by intra-class correlation coefficients, limits of agreement (mean difference ±1.96 × SD) (Bland and Altman 1986) and coefficients of variation (SD of difference as a percentage of the mean value). Probability values <0.05 were considered statistically significant. Results Symptom-limited exercise tests All subjects completed the exercise tests. The maximum workload was 109 ± 32 W, peak- was 20.0 ± 4.0 ml min−1 kg−1 (73 ± 9% of predicted peak- ) and the maximal respiratory exchange ratio was 1.13 ± 0.13. The ventilatory threshold could not be determined in three patients (16%) because of excessive ventilatory oscillations. In the remaining 16 patients the independent observers agreed on the determination of the ventilatory threshold (mean :16.4 ± 3.2 ml min−1 kg−1, 60 ± 11% of predicted peak- ). Constant-load exercise tests The mean value of during the second minute of unloaded pedaling was 655 ± 78 ml min−1 (30 ± 5% of predicted peak- ), and the steady-state value at 50% of the maximal work load was 1,185 ± 228 ml min−1 (53 ± 7% of predicted peak- ). Figure 1 shows changes in during a constant-load test in a representative subject. Fig. 1-response to steady-state exercise at 50% of the maximal workload (50 W) in a representative subject. The solid line represents 10 s averages of The curved dashed line is the computer-derived representation of the best fit of the mono-exponential model to the -response. The first dashed vertical line indicates onset of exercise and the second vertical line the end of exercise In 16 subjects, in whom the ventilatory threshold could be determined reliably, steady-state- was below the ventilatory threshold. None of the other three subjects demonstrated a significant rise of defined as in increase from the third to the sixth minute of exercise of more than two times the SD of the mean in the fourth minute. This indicates that these three patients also exercised below the ventilatory threshold (Whipp 1994). Comparison of calculation methods Concerning the onset phase, there were no significant differences between mono-exponential modeling and the algebraic method (differences between τ and MRT −6.6–1 s, SD 11–17 s). In the recovery phase, the use of different sampling intervals yielded significantly different time constants (P < 0.001). Paired comparisons showed that in two cases the time constants were not significantly different (5 s vs. 5 breaths, P = 0.14 and 10 s vs. 8 breaths, P = 0.31). When comparing the applicability of the mono-exponential model to the -response during onset and recovery of exercise, the results indicate a better ‘goodness of fit’ during recovery (P < 0.001) (Table 2). Parameters of O2 onset kinetics showed significant correlations with peak- when calculating τ with sampling intervals of 10 s, 5 breaths and 8 breaths (Table 3). However, mainly due to the limited number of patients, the differences between these correlation coefficients were not statistically significant. During recovery of exercise τ was only correlated with peak- when 5 breath sampling intervals were used. Table 2O2 Kinetics using an algebraic method and a mono-exponential model with different sampling intervals (n = 19)Kinetic parametersMean ± SD (s)Goodness of fit (R2) Mean ± SDUseful measurements (%)aO2 Onset kinetics MRT71 ± 19–– τ-5 s67 ± 200.69 ± 0.1526 τ-10 s74 ± 22 0.81 ± 0.1242 τ-5b75 ± 300.87 ± 0.0879 τ-8b74 ± 220.91 ± 0.0784O2 Recovery kinetics τ-5 s60 ± 130.89 ± 0.0684 τ-10 s66 ± 140.94 ± 0.04100 τ-5b61 ± 130.96 ± 0.03100 τ-8b64 ± 130.98 ± 0.02100Average duration of 5 breath sampling interval 15.9 ± 3.6 s at baseline, 13.0 ± 3.0 s at steady-state. Average duration of 8 breath sampling interval 25.0 ± 5.2 s at baseline, 20.9 ± 4.7 s at steady stateMRT Mean response time, τ time constant calculated with a mono-exponential model, 5s 5 s, 10 s 10 s, 5b 5 breaths, 8b 8 breathsa The assessment of τ was considered useful when R2 ≥ 0.85Table 3Correlation coefficients between averaged kinetic parameters and peak Kinetic parametersCorrelation with peak 95% Confidence intervalO2 Onset kineticsMRT−0.33−0.680.15τ−5 s–τ-10 s−0.88*−0.99−0.24τ-5b−0.67*−0.90−0.16τ-8b−0.57 *−0.83−0.10O2 Recovery kineticsτ-5 s−0.43−0.760.08τ-10 s−0.45−0.750.01τ-5b−0.47*−0.76−0.02τ-8b−0.39−0.720.08Only data with R2 > 0.85 when using a mono-exponential model were included; differences between correlation coefficients were not statistically significant* P < 0.05MRT Mean response time, τ = time constant calculated with a mono-exponential model, 5s 5 s, 10s 10 s, 5b 5 breaths, 8b 8 breaths Reproducibility In the two constant-load tests there were no statistically significant differences between during unloaded pedaling (655 ± 78 vs. 633 ± 98 ml min−1), at steady-state exercise (1,185 ± 228 vs. 1,202 ± 235 ml min−1), and respiratory exchange ratios at steady-state exercise (0.94 ± 0.04 vs. 0.94 ± 0.06). In addition, there were no statistically significant differences between the kinetic parameters of both tests. Considering limits of agreement, coefficients of variation and intra-class correlation coefficients, recovery kinetics show better reproducibility than onset kinetics (Table 4). Table 4Comparison of kinetic parameters in two consecutive constant-load testsKinetic parameterNDifference mean ± SDLimits of agreementCoefficient of variation (%)Intraclass correlation coefficientO2 Onset kinetics  (ml min−1 kg−1)190.1 ± 1.2−2.32.58.50.97 MRT (s)192.8 ± 12.9−232918.40.86τ−5 s (s)0––––τ-10 s (s)69.3 ± 11.8−143317.10.79τ-5b (s)125.9 ± 19.8−344628.40.63τ-8b (s)165.6 ± 16.7−283923.00.77O2 Recovery kineticsτ-5 s (s)16−0.4 ± 7.2−151412.20.91τ-10 s (s)19−0.9 ± 10.7−222116.30.86τ-5 b (s)19−0.9 ± 6.4−141210.40.94τ-8 b (s)19−1.4 ± 8.6−191613.40.91Only data with R2 > 0.85 when using a mono-exponential model were included Oxygen uptake at steady-state exercise, MRT mean response time; τ time constant calculated with a mono-exponential model, 5 s 5 s; 10 s 10 s; 5b 5 breaths, 8b 8 breaths Figure 2 shows Bland Altman plots of the kinetic parameters during onset and recovery of exercise with the highest intra-class correlation coefficients. Fig. 2Bland Altman plots showing the difference of O2 kinetics between two constant-load tests during onset (upper graph) and recovery of exercise (lower graph), using an algebraic method and mono-exponential modeling with 5 breath sampling intervals, respectively. The solid lines represent the mean difference between the two tests, the dashed lines indicate the 95% confidence intervals of the difference. MRT mean response time. τ -Rec = time constant of during recovery with 5-breath sampling intervals Discussion The principal finding of this study is that, using the applied exercise protocol, O2 recovery kinetics were more reproducible than O2 onset kinetics in moderately impaired patients with CHF. It should be recognized, however, that this observation cannot be assumed to be generalizable to more severely impaired CHF patients. O2 onset kinetics Because of the lack of standardized protocols, previous authors used both time constants (τ, non-linear regression) (Belardinelli et al. 1998; Koike et al. 1995; Matsumoto et al. 1999) and mean response times (MRT, algebraic method) (Sietsema et al. 1994) to assess O2 onset kinetics in CHF patients. Considering non-linear regression, the use of 10 s sampling intervals yielded the best results in terms of reproducibility in this study. However, the goodness of fit, which was comparable with the study of Arena et al. (2002) (0.81 vs. 0.78, respectively), was insufficient in 58% of the patients. Moreover, the wide limits of agreement restrict its use for clinical applications. While the calculation of τ with larger sampling intervals resulted in a better goodness of fit, their reproducibility was even lower. To our knowledge, only one study previously addressed reproducibility of O2 onset kinetics using mono-exponential modeling in CHF patients. In that study five patients with a peak- that was comparable to our study performed three constant-load tests at a workload of 50 W, starting from unloaded cycling. Although the low mean difference between the tests (2 s) suggests good reproducibility, the actual variability between the tests was not mentioned, making it impossible to compare these results to our study (Belardinelli et al. 1998). In spite of the fact that almost all of the onset procedures showed a significant correlation with peak- (Table 3), differences between these correlations were not statistically significant. Therefore, one could not suggest a preferential use of one of the procedures. Although showing a slightly higher intra-class correlation coefficient (Table 4), the wide limits of agreement of the MRT also indicate low reproducibility of this parameter. In addition, this parameter did not correlate significantly with peak Sietsema et al. (1994) previously addressed reproducibility of the MRT in 18 CHF patients, reporting small mean differences between two tests. Again these results cannot be compared to our study, because intra-class correlations and limits of agreement were not mentioned. When comparing reproducibility of O2 onset kinetics in CHF patients with healthy subjects (Kilding et al. 2005) and patients with COPD (Puente-Maestu et al. 2001) we found a lower reproducibility of τ (coefficients of variation 17.1, 6.2 and 8.7%, respectively). There are several physiological and methodological factors that may explain this discrepancy. One significant physiological factor that may cause the large variability of O2 onset kinetics in CHF patients is the influence of typical ventilatory oscillations, especially when small sampling intervals are used (Francis et al. 2002). Although we did not quantify the effect of oscillations, we did observe ventilatory oscillations more clearly in patients with low coefficients of determination. Furthermore, it is postulated that ventilatory oscillations in patients with CHF increase in the transition from rest to exercise (Kremser et al. 1987) and diminish with increasing exercise (Francis et al. 2002). This suggests that these oscillations have a greater influence on O2 onset kinetics than O2 recovery kinetics. Based on our study, variations in τ during onset of at least 24 s are needed to exceed the limits of the 95% confidence interval, whereas a change of only 13 s in τ during recovery is sufficient to exceed the normal test-to-test variations. This latter variability is also observed during the on-transient response in patients with COPD (Puente-Maestu et al. 2001), which supports our view that the relatively large variability of the on-kinetic parameters was mainly caused by ventilatory oscillations. Another explanation for the limited reproducibility of O2 onset kinetics in CHF patients is their low exercise capacity, which reduces the amplitude of O2 uptake, and consequently the reliability of the determination of the time constant (Lamarra et al. 1987). Furthermore, the period between the tests (6.7 ± 3.9 days) could be a long enough period of time to induce changes in the cardio circulatory condition of the patients. However, all patients were in a stable clinical or functional condition during the study period. Considering methodological factors, the fact that all patients started the constant-load test with unloaded pedaling in stead of rest resulted in a relatively low -amplitude. This could be a major factor contributing to the relatively low reproducibility of the on-kinetic parameters found in this study. The reasons for the authors, however, to apply this exercise protocol were threefold. First, starting exercise from unloaded pedaling results in a reduction of the duration of the early rapid increase of oxygen uptake, representing an initial increase of pulmonary blood flow (cardiodynamic phase). Since in this phase the change of is functionally distinct from the subsequent mono-exponential increase of (phase II), reducing the relative contribution of this phase might result in a better fit of the mono-exponential model to the data. Second, starting exercise from unloaded cycling might reduce ventilatory oscillations at the onset of exercise because of a reduction in variation of pedaling frequency and upper limb muscle activity. Third, in daily life exercise will be frequently started from an active state. Given the results of this study, however, the authors feel that additional research is required to re-assess the reproducibility of O2 onset kinetics using an exercise protocol starting from rest in stead of unloaded cycling. Starting exercise from rest might result in a better reproducibility of O2 onset kinetics than observed in this study, which may in particular be of importance for more severely impaired CHF patients than used in this study. Another factor that may have influenced reproducibility of O2 onset kinetics is the choice of the workload (i.e. 50% of the maximum workload). The approach to maximize -amplitude by relating the workload to the ventilatory threshold (e.g. 90% of ventilatory threshold) was not used in this study, because it was postulated that such a fixed threshold cannot be determined reliably in a substantial number of CHF patients (Meyer et al. 1996). When looking at the patients of this study in whom determination of this threshold was successful, at steady state amounted to 86% of at the ventilatory threshold, with none of the patients exercising above this threshold. This suggests that the applied constant-load exercise protocol was adequate, because the exercise intensity remained below the ventilatory threshold. O2 recovery kinetics O2 recovery kinetics were more reproducible in moderately impaired patients with CHF than O2 onset kinetics. This difference in reproducibility is at least partly explained by the larger -amplitude in recovery due to the fact that subjects were not cycling in the recovery phase. From the results of this study it is difficult to conclude to what extent this difference between onset and recovery kinetics is caused by a smaller influence of the cardiodynamic phase during recovery or a more stable breathing pattern during recovery causing less ventilatory oscillations. Nevertheless, these observations are in line with previous studies in healthy individuals, which also reported a better reproducibility of O2 recovery kinetics than O2 onset kinetics (Kilding et al. 2005; Ozyener et al. 2001). In terms of reproducibility, the most optimal method to characterize O2 recovery kinetics was mono-exponential modeling with sampling intervals of 5 breaths (intra-class correlation 0.94). Using this method a change of at least 13 s in τ is needed to exceed the normal test-to-test variations. In addition, this method yielded an excellent goodness of fit (R2 0.96 ± 0.03). To our knowledge, other data on the reproducibility of the recovery kinetics after submaximal exercise in CHF patients are not available. Cohen-Solal et al. (1995) studied reproducibility of O2 recovery kinetics after maximal exercise in 10 patients with CHF (NYHA II-III) using a mono-exponential model with sampling intervals of 7 breaths. They found a coefficient of variation comparable to our study with 5 breath sampling intervals (12.3 vs. 10.4%, respectively). Conclusion This study shows that the reproducibility of O2 onset kinetics assessed by mono-exponential modeling is too low to warrant their use for measuring effects of therapeutic interventions in moderately impaired patients with CHF. This might be mainly attributable to physiological factors like ventilatory oscillations and the applied exercise protocol. Future studies should address the effect of different exercise protocols. The determination of time constants of O2 recovery kinetics has been shown to be feasible and reproducible when using mono-exponential modeling with 5 breath sampling intervals. Since this variable represents recovery from exercise that is indicative of daily life, it is potentially useful for clinical purposes like grading of functional impairment in patients with CHF and measuring effects of therapeutic interventions.

Planta-3-1-2039816

Relationship between petal abscission and programmed cell death in Prunus yedoensis and Delphinium belladonna

Depending on the species, the end of flower life span is characterized by petal wilting or by abscission of petals that are still fully turgid. Wilting at the end of petal life is due to programmed cell death (PCD). It is not known whether the abscission of turgid petals is preceded by PCD. We studied some parameters that indicate PCD: chromatin condensation, a decrease in nuclear diameter, DNA fragmentation, and DNA content per nucleus, using Prunus yedoensis and Delphiniumbelladonna which both show abscission of turgid petals at the end of floral life. No DNA degradation, no chromatin condensation, and no change in nuclear volume was observed in P. yedoensis petals, prior to abscission. In abscising D.belladonna petals, in contrast, considerable DNA degradation was found, chromatin was condensed and the nuclear volume considerably reduced. Following abscission, the nuclear area in both species drastically increased, and the chromatin became unevenly distributed. Similar chromatin changes were observed after dehydration (24 h at 60°C) of petals severed at the time of flower opening, and in dehydrated petals of Ipomoea nil and Petunia hybrida, severed at the time of flower opening. In these flowers the petal life span is terminated by wilting rather than abscission. It is concluded that the abscission of turgid petals in D. belladonna was preceded by a number of PCD indicators, whereas no such evidence for PCD was found at the time of P. yedoensis petal abscission. Dehydration of the petal cells, after abscission, was associated with a remarkable nuclear morphology which was also found in younger petals subjected to dehydration. This nuclear morphology has apparently not been described previously, for any organism. Introduction Floral life span depends on the fate of the petals. In several species the end of floral life span is due to petal withering or wilting. The petals in most of these species desiccate, and finally abscise or fall due to growth of the fruit. Only in a few species the desiccated petals remain attached to the fruit. In another large group of species the end of floral life span is determined by abscission of petals that are mostly or fully turgid. Wilting in petals, at the end of their lifespan, is due to a series of highly regulated processes that lead to cell death, thus is a form of programmed cell death (PCD; van Doorn et al. 2003). Prior to visible PCD, many macromolecules in the petal cells are degraded to molecules that are suitable for phloem transport. These mobile molecules are transported to other parts of the plant (Winkenbach et al. 1970a, b). Typical for PCD in wilting petals is the disappearance of most of the organelles and accumulation of degrading enzymes in the vacuole. This is followed by rupture of the vacuolar membrane, whereby several hydrolases are released. Tonoplast rupture in numerous cells results in their collapse, which is visible as wilting (Winkenbach 1970a, b; van Doorn et al. 2003). PCD in several plant parts (Gunawardena et al. 2001; Wojciechowska and Olszewska 2003), including petals, is also usually accompanied by DNA degradation (van der Kop et al. 2003; Wagstaff et al. 2003; Yamada et al. 2006a, b) and chromatin condensation (Yamada et al. 2006a, b). It is not known if the abscission of turgid petals is preceded by cellular changes that are indicative of PCD. We therefore investigated the time line of some of these changes, in Prunus yedoensis and Delphiniumbelladonna. In both species the petals abscise without any visible sign of turgor loss. We used DNA degradation (measured both on agarose gels and by using flow cytometry), chromatin condensation, and a decrease in nuclear diameter as parameters that indicate PCD. According to our hypothesis these parameters would not detectably change, prior to petal abscission. Materials and methods Plant material Plants were grown in the garden of National Institute of Floricultural Science of Japan (P. yedoensis cv. Someiyoshino) or the greenhouse of the institute (D. belladonna cv. Bellamosum, Ipomoea nil cv. Violet, and Petunia hybrida cv. Double Duo Pink). Just before the onset of flower opening, cut branches (P. yedoensis) or potted plants (the other species) were transferred into a growth chamber at 24 ± 1°C, about 70% relative humidity (RH), and 12 h of daily illumination from cool-white fluorescent lights (100 μmol m−2 s−1). The time to petal abscission and petal desiccation was observed daily, on plant material in the growth chamber. The timing of these symptoms was noted in six flowers from two plants. All determinations were made using petals that were collected from plant material in the chamber. DNA degradation DNA breakdown was detected by agarose gel analysis. Total DNA was extracted according to Yamada et al. (2003). The amount of DNA was estimated by molecular absorption spectrophotometry. Equal amounts of total DNA (3 μg) were immediately electrophoresed in a gel containing 3% agarose and stained with SYBR Gold nucleic acid gel stain (Molecular Probes, Eugene, OR, USA). The gel pattern was photographed, using an electronic UV transilluminator system (model FAS-III mini + DS-30; Toyobo, Tokyo, Japan). The DNA degradation index was calculated according to Yamada et al. (2006b). Two biological replications, each containing several petals, were used for each stage of development. Depending on the species, two to several petals were used in each sample. Flow cytometry The petal was chopped in nuclear extraction buffer, part of the kit (High Resolution kit) for plant DNA analysis (Partec, Münster, Germany), a reagent set provided by the manufacturer of the flow cytometer used (Ploidy Analyzer, Partec). The extraction buffer in this kit was a low pH solution containing Triton X-100. The extract was filtered through 50 μm nylon mesh. The medium with the isolated DNA masses (nuclei) was collected, the buffer with the fluorescent DNA stain 4, 6-diamidino-2-phenylindole (DAPI) from a standard reagent set (Partec) was added, and the solution was vortex mixed. The DNA content of the isolated nuclei was analysed using the flow cytometer. The flow cytometer apparatus can be set to count all nuclei in a sample, and can be set to analyse a fixed number of nuclei. We here set it to detect 5,000 nuclei. DNA levels were obtained in a total of 5,000 nuclei. It should be noted that the large histogram peak was adjusted to 10 on the fluorescence scale. This means that nuclear condensation was not taken into account in the flow cytometry data. Two biological replications, each containing several petals, were used for each stage of development. Depending on the species, three to several petals were used in each sample. In each species, the number of petals was the same at all stages of development. Nuclear morphology; chromatin condensation Nuclei were isolated from petals in the same way as for flow cytometry. Isolated nuclei were stained with DAPI, which stains DNA. The nuclei were observed under a fluorescence microscope (model PROVIS AX70, Olympus, Tokyo, Japan) using U-excitation (330–385 nm). Digital photography of nuclei was obtained for each stain using low light cool CCD camera (model DP30BW, Olympus). Dehydration and ethylene treatment Flowers were excised when the petals were fully open. Cut flowers were placed in air in the growth chamber (24°C, about 70% RH) or in a desiccator (at 60°C). For ethylene treatments, stems of cut flowers were placed in sterile distilled water, in a closed 70 l clear plastic box containing 2 μl/l ethylene. Petal abscission was checked at hourly intervals and abscised petals were collected immediately. The petals were then used instantly for the measurement of the various parameters mentioned. Statistics The number of replications is mentioned in the text and/or in the figure legends. Statistical analysis used one way analysis of variance, at P < 0.05. All experiments were repeated at least once. Results Symptoms of petal life span cessation P. yedoensis flowers have five pale pink petals. After flower opening the petals remained attached to the flower for about 4 days, and then fell without visible wilting (Fig. 1a, P2). During the first day following abscission, petals that were held at 24°C and 70% RH changed colour from pale pink to darker pink, and subsequently became infiltrated with liquid (results not shown). By the end of the first day following abscission, the petals started to shrivel at their margins (Fig. 1a, P3). By the end of the second day after abscission the petals had much more decreased in size (Fig. 1a, P4). The stages 2–4 (pictures P2–P4) in Fig. 1a use the same scale thus show the decrease in petal size. Fig. 1Petal abscission and petal dehydration in P. yedoensis, indicated by a P (a) and D.belladonna, indicated D (b). Cut branches of (P. yedoensis) and potted plants (D.belladonna) stood in a growth chamber (24°C, about 70% RH). The time to abscission and to dehydration symptoms (in days from full flower opening) is shown in the left lower corner of the pictures (n = 6). Stages P1 and D1: full flower opening; P2 and D2, petal abscission; P3 and D3, petals visibly dehydrated; P4 and D4, petals desiccated. All pictures have the same scale (bars = 10 mm) Flowers of Delphinium species contain five petals one of which has a large spur. The petals of D. belladonna fell about 7 days after flower opening (Fig. 1b, D2). They did not show any visible wilting symptom by the time of abscission. Two days after abscission the petals had slightly shrivelled (Fig. 1b, D3) but by day-3 after abscission the shrivelling had become very clear (Fig. 1b, D4). The stages 2–4 (pictures D2–D4) in Fig. 1b have the same scale. DNA degradation Figure 2 shows the petal DNA agarose gels of one of two repeat experiments. The pixel ratios under the lanes are the means of two experiments. In both species, the agarose gels showed no DNA fragments at the time of flower opening (Fig. 2, lanes P1 and D1). By the time of abscission, the petals of P. yedoensis did not show a statistically significant (P < 0.05) increase in DNA fragmentation, compared with the time of floral opening (Fig. 2, lane P2). In contrast, the petals of D. belladonna showed a considerable increase in DNA fragmentation by the time of abscission (Fig. 2b, lane D2, pixel intensity had increased from 4.3 to 17.4). Following petal abscission the degree of DNA fragmentation, as observed on agarose gels, increased in P. yedoensis (Fig. 2a, compare lane P3 with lane P2) and further increased in D. belladonna (Fig. 2b, compare lane D3 with lane D2). By the time of petal desiccation (P4 and D4) the amount of fragmented DNA had become considerably lower than at the previous stage investigated, when the abscised petal showed visible wilting and dehydration (Fig. 2). Fig. 2Agarose gel analysis of total DNA isolated from the petals of P. yedoensis, annotated as P (a) and D. belladonna, annotated as D (b). 3 μg of total DNA was extracted from the petals and electrophoresed in a 3% agarose gel. Lanes P1–P4 and SD1–SD4 refer to DNA isolated from petals at stages P1–P4 and D1–D4 as shown in Fig. 1. Stages P1 and D1: full flower opening; P2 and D2, petal abscission; P3 and D3, petals visibly dehydrated; P4 and D4, petals desiccated. Pixel ratios below the lanes are means of two repeat experiments Nuclear morphology In P. yedoensis petals the nuclear morphology had not clearly changed by the time of petal abscission (Fig. 3a). This is in contrast with D. belladonna where visual inspection showed that the nucleus was much smaller by the time of abscission, compared with the size observed at flower opening. The small nuclei of D. belladonna showed brighter DAPI fluorescence than the nuclei at flower opening. This indicates a high concentration of DNA per unit volume, thus chromatin condensation (Fig. 3b, D2). Nuclear morphology was studied using light microscopy, giving a two-dimensional impression only. For this reason nuclear size will be referred to as nuclear area. Fig. 3Morphology of DAPI-stained nuclei in the petals of P. yedoensis, indicated by a P (a) and D.belladonna, indicated D (b). Two representative nuclei (-1 and -2) are shown at each stage of development. Nuclei were isolated from the petals and observed using a fluorescence microscope under U-excitation. Bars: 5 μm (a) or 15 μm (b). Stages as in Fig 1. P1 and D1 1: full flower opening; P2 and D2, petal abscission; P3 and D3, petals visibly dehydrated; P4 and D4, petals desiccated After abscission, during the period of dehydration, the nuclei of both species showed an increase in area (Fig. 3) In P. yedoensis the nuclear area at stage P4 seemed even considerably larger than that at floral opening (Fig. 3a). In D. belladonna the nuclear area was apparently similar at D4 compared to the stage of floral opening (D1), but it was ostensibly much larger compared with the area at the stage of petal abscission (P2; Fig. 3b). In both species the nuclear periphery was no longer smooth at stage P4 or D4 (Fig. 3). In both species investigated, numerous nuclei showed the morphology shown in Fig. 3, stages P4 and D4. The area of the remaining nuclei did not increase as much, whilst a few nuclei had not changed in nuclear area with respect to controls at flower opening. Flow cytometry of nuclei at various stages prior to and after abscission Flow cytometry of nuclei stained for DNA (with DAPI) shows the distribution of nuclei in fluorescence classes. The X-axis in Fig. 4 is divided in numerous classes of increasing fluorescence, and the Y-axis shows the number of nuclei per fluorescence class. Flow cytometry of 5.000 petal nuclei at the time of P. yedoensis flower opening showed a peak of nuclei at phase G0/G1 (which was set at 101 on the arbitrary logarithmic scale of the X-axis), and no mitotic nuclei (G2/M), which have a fluorescence about two times that of normal G0/G1 nuclei. By the time of petal abscission, no change had occurred in the DNA fluorescence histogram (Fig. 4a, P2). Two days after abscission, the height of the G0/G1 peak had diminished but the peak had not shifted. This is shown by mixing nuclei from day-0 to -6 (Fig. 4a, P1 + P4). The data suggest that a small amount of DNA had become degraded by day-2 after abscission. Fig. 4Fluorescence of DAPI-stained nuclei. Histograms were obtained by flow cytometry of 5,000 nuclei isolated from the petals of P. yedoensis (indicated by a P) as shown in a, and petals of D.belladonna (D) shown in b. Petals were severed from cut twigs placed in water (Prunus) or potted plants (Delphinium), which stood both in a growth chamber (24°C, about 70% RH). Stages as in Fig. 1. P1 and D1: full flower opening; P2 and D2, petal abscission; P3 and D3, petals visibly dehydrated; P4 and D4, petals desiccated By the time of petal abscission in D. belladonna, flow cytometry revealed a decrease in the height of the G0/G1 phase nuclei, and several nuclei with a fluorescence less than that of G0/G1 phase nuclei (Fig. 4b, D2). During the 3 days following abscission the G0/G1 peak height further decreased and the number of nuclei with fluorescence lower than G0/G1 further increased. No clear shift in the G0/G1 peak was observed: a mix of nuclei at day-0 and -10 did not result in a bimodal curve (Fig. 4b, D1 + D4). The results indicate extensive DNA degradation, before, during, and after abscission. Effects of ethylene Ethylene treatment greatly hastened the time to petal abscission, in both species (Fig. 5a, compared with Fig. 1a). In both species, the rate of petal dehydration after abscission was not much affected by the ethylene treatment (Fig. 5a) compared with the control without ethylene (Fig. 1a). For example, 3 days after petal fall in D. belladonna the petals had extensively shrivelled, independent of the time of petal fall (day-7 in controls and day-2 in ethylene-treated flowers). Fig. 5Effect of ethylene on petal abscission and DNA fluorescence of nuclei in the petals of P. yedoensis and D.belladonna. Time to petal abscission and petal dehydration symptoms (a) and flow cytometric determination of 5,000 DAPI-stained nuclei (b) of ethylene-treated branches of P. yedoensis and potted plants of D.belladonna. Ethylene treatment is annotated by E. Compare stages of development with untreated controls shown in Fig. 1, and with fluorescence data of controls in Fig.4. The left lower corner of the pictures series a show the times to abscission (EP2 and ED2) and the time to dehydration symptoms (both expressed in days from full flower opening; n = 6). Stages P1 and D1: full flower opening; EP2 and ED2, petal abscission; EP3 and ED3, onset of petal dehydration; EP4 and ED4, petals desiccated. All pictures have the same scale (bars = 10 mm) In both species, flow cytometry revealed no effect of ethylene treatment on DNA fluorescence at the time of petal fall, compared to the fluorescence at the time of flower opening (Fig. 5b). After petal fall, a clear shift to lower DNA fluorescence of the G0/G1 peak of fluorescence was observed in P. yedoensis (Fig. 5b, upper sequence), which was not found in controls without ethylene treatment (Fig. 4, upper sequence). A small shift to lower DNA fluorescence (Fig. 5b, lower sequence), also not found in controls (Fig. 4 lower sequence), was also observed after ethylene treatment of D. belladonna. Superimposition of the stages P1 and EP4 (control and ethylene-treated stage 4 of P. yedoensis) showed two peaks of DNA fluorescence (Fig. 5b, upper sequence). A similar tendency was observed by superimposing the stages D1 and ED3 (Fig. 5b, lower sequence). Dehydration of petals cut at flower opening Flowers were harvested just after opening and subjected to a dehydration treatment at 60°C for 24 h. The treatment resulted in a petal FW of about 20% of that of controls. Preliminary experiments showed that such a treatment resulted in a decrease of FW, and in changes of other parameters such as DNA degradation and nuclear morphology, that were very similar to dehydration on the laboratory bench (about 20°C and 70% RH) for about 96 h. Treatment of P. yedoensis flowers at 60°C for 24 h (Fig. 6a, DP) resulted in a slight increase in petal DNA degradation, as shown on agarose gel (Fig. 6b, DP). In contrast, this treatment induced a large increase in DNA degradation in D. belladonna petals (Fig. 6b, DD). Fig. 6Effect of dehydration on DNA degradation, DNA fluorescence of nuclei, and nuclear morphology in the petals of P. yedoensis (P) and D.belladonna (D). Dehydration in the two species is indicated as DP and DD, respectively. a Flower morphology before (P1 and D1) and after (DP and DD) dehydration at 60°C for 24 h. Photographs show small branches with fully opened flowers and desiccated flowers of P. yedoensis (P1 and DP) and the same for individual flowers of D.belladonna (D1 and DD). All pictures have the same scale (bars = 10 mm). b Agarose gel analysis of total DNA isolated from the petals. c Flow cytometric determination of nuclei isolated from the petals. d Morphology of nuclei in the petals. Two representative nuclei (-1 and -2) are shown for each species, after the dehydration treatment (DP and DD). Bars = 5 μm (P1 and DP) or 15 μm (D1 and DD) The dehydration treatment also had a small effect on DNA fluorescence of nuclei from P. yedoensis petals (Fig. 6c). Superimposition of the nuclei prior to and after treatment showed no shift to lower DNA fluorescence (Fig. 6c, P1 + DP). In D. belladonna, the dehydration treatment resulted in lower G0/G1 DNA fluorescence, and many nuclei with a lower than G0/G1 fluorescence (Fig. 6c). Superposition of the treatment and the control showed no shift of the G0/G1 peak of DNA fluorescence (Fig. 6c, D1 + DD). Following the dehydration treatment, the nuclei of P. yedoensis had apparently somewhat increased in area, but no clear increase in nuclear area was discernable in D. belladonna (Fig. 6d). The chromatin distribution in the nuclei, after the dehydration treatment (Fig. 6d) was similar to that after dehydration due to abscission (Fig. 3). Visual inspection indicated that about 80% of all nuclei had the morphology shown in Fig. 6c, d. The area of the remaining nuclei had not as much increased. Dehydration of petals in species that show petal wilting (Ipomoea and Petunia) Similar dehydration treatments (60°C for 24 h) were carried out with I. nil and Petunia hybrida, species in which the petals wilt by the end of flower life span. Wilting is shown as inrolling in I. nil, which occurs at about 12 h after flower opening, in plants held at 24°C (Fig. 7a). Wilting is observed after 7 days at 24°C in flowers attached to P. hybrida plants (Fig. 7a). Fig. 7Effect of dehydration on fresh weight, DNA degradation, and DNA fluorescence of nuclei in the petals of I. nil and P. hybrida. Potted plants were grown in a greenhouse and then placed in a climate-controlled chamber (24°C, about 70% RH). Petals were detached from flowers that had just opened, and were placed at 60°C for 24 h to dehydrate. In a the time until about half of each petal had wilted (in days from full flower opening; n = 6) is shown in the left lower corner of the pictures. a Morphology of the flowers, at opening (left) and at petal wilting (right). All pictures have the same scale (bars = 10 mm). b Agarose gel analysis of total DNA isolated from the petals. c, d Flow cytometric determination of 5,000 DAPI-stained nuclei isolated from the petals of I. nil (c) and P. hybrida (d). Ip, Ipomoea; Pe, Petunia, D, dehydration Flowers were detached from potted plants, just after full opening. Fig. 7b shows considerable DNA fragmentation on agarose gels, after the dehydration treatment (60°C for 24 h), in both species tested. Flow cytometry of Ipomoea petals held at 60°C for 24 h showed a large decrease in DNA fluorescence, compared to controls of Ipomoea (Ip) before desiccation (Fig. 7c, compare DIp with Ip, whereby D means dehydrated). Fig. 7d shows that DNA degradation was also found in Petunia (Pe). Superimposition of the fluorescence in controls and the treated samples showed two peaks (Fig. 7c, d). The morphology of DAPI-stained nuclei of I. nil and P. hybrida, after petal desiccation at 60°C for 24 h, is shown in Fig. 8. Compared to nuclei of petals in flowers that had just fully opened (two panels upper left), relatively large parts not stained with DAPI were observed (two panels upper right). In I. nil the nuclear area had not become clearly increased, compared to those in petals that had not been desiccated (Fig. 8a), but inflated nuclei were observed in P. hybrida (Fig. 8b). Visual inspection indicated that about 80% of all nuclei showed the morphology shown in Fig. 8b. The area of the remaining nuclei had not as much increased. Fig. 8Effect of dehydration on nuclear morphology in the petals of I. nil and P. hybrida. Nuclei were isolated from the petals of I. nil (a) and P. hybrida (b) after the indicated treatments, then observed using a fluorescence microscope under U-excitation. For comparison, the nuclei are shown during PCD of petals that remained attached to the plant (for data of Ipomoea and Petunia see also Yamada et al. 2006a, b). Stage S1: full flower opening; S2: onset of petal wilting; S3: full petal wilting; and S4: petals desiccated but still attached to the flower. Bars = 10 (μm Both in I. nil (Fig. 8a, S1–S4) and in P. hybrida (Fig. 8b, S1–S4) the nuclear morphology after the dehydration treatment was quite different from the one observed during normal PCD. During PCD of petals that remained attached to the plant the area of the nuclei became increasingly smaller, whereas the brightness of DNA fluorescence remained high at least until the nuclei had become quite small. Discussion In reports on animal cells the suggestion that PCD occurs is usually backed up by showing changes such as phosphatidylserine externalization to the outer leaflet of the plasma membrane, caspase activation, and DNA degradation. Only the addition of a number of such parameters seems indicative of PCD. In the present study we took massive DNA degradation (measured in two ways), chromatin condensation, and a decrease in nuclear diameter as indicators of PCD. In other studies on petal PCD, it was shown that these parameters were each correlated with other indicators of PCD and also correlated with visible PCD symptoms such as petal wilting (Yamada et al. 2003, 2006a, b). We therefore suggest that these parameters, which separately represent only circumstantial evidence for PCD, are, when occurring together, adequate to indicate PCD. We hypothesized that PCD parameters will not show in petal cells, prior to the abscission of turgid petals. The results of D. belladonna petals contradict the hypothesis, while those of P. yedoensis are in agreement with it. The data suggest that the degree of PCD, prior to abscission of turgid petals, depends on the species. The present data indicate that abscission of fully turgid petals may coincide with an advanced stage of PCD in many petal cells. The absence of visible wilting in abscising petals, therefore, does not seem to show absence of PCD. This is so because the wilting that is often associated with PCD is quite a late stage in the PCD process. In petals the timing of PCD and that of abscission may be regulated independently. Independent regulation would explain the large variation of their relative timing, among the Angiosperms. In many species PCD comes much earlier than abscission (species showing petal wilting at the end of floral life), whereas in other species abscission comes first. In most species in which floral life span is ended by petal wilting, the petals eventually abscise, even when most of the petal surface has become desiccated. In these species the petal base, where abscission takes place, remains alive. Only in a few species fully desiccated petals are torn-off by the growing fruit, or (in still other species) remain attached to the fruit (van Doorn and Stead 1997; van Doorn 2001). The agarose gel data (Fig. 2) showed substantial DNA degradation prior to and during abscission in D. belladonna and less DNA degradation prior to and during petal fall in P. yedoensis. Flow cytometry data (Fig. 4) confirmed the agarose gel data. DNA degradation seems a relatively late PCD process. Mass protein degradation, a late process, may serve as a yardstick. In Ipomoea petals, bulk degradation of proteins and DNA occurred at about the same time (Winkenbach 1970a, b). In Iris petals, 40% of total DNA had become degraded (van der Kop et al. 2003) by the time of the onset of mass protein degradation (Pak and van Doorn 2005). In Alstroemeria petals DNA breakdown (Wagstaff et al. 2003) started slightly before the decrease in protein content (Wagstaff et al. 2002). When using agarose gels to show DNA breakdown during PCD, a ladder of DNA fragments is often observed, at least in association with apoptosis in animal cells. DNA laddering on gels is the result of oligonucleosomal cleavage, resulting in DNA sequences of multiples of about 180 bp. In contrast to apoptosis in animal cells, the DNA breakdown pattern during plant PCD is quite variable and depends on the system. For example, during petal PCD clear DNA laddering without a background of DNA fragments of various lengths is rarely found. Most reports show DNA laddering against a background of fragments of various lengths (Orzáez and Granell 1997; Wagstaff et al. 2003; Yamada et al. 2003) or only fragments of various lengths (van der Kop et al. 2003). The present results fall in the last category. Nuclear morphology at the time of petal abscission also suggested advanced PCD in the petal cells of D. belladonna and little PCD in those of P. yedoensis. The nuclei of D. belladonna were much smaller and showed bright DAPI fluorescence, indicating chromatin condensation. No chromatin condensation was observed in P. yedoensis petals, but here it might possibly have occurred between the stages P2 and P3 and therefore may have been missed. We found that most but not all nuclei underwent the changes shown at the stages 2–4 of Fig. 3. The difference might relate to differences in PCD progress, within the petal. It is known that cells associated with the vascular bundle undergo PCD at a later stage than the other petal cells. In some species the mesophyll cells also show earlier PCD than the epidermis cells (van Doorn et al. 2003). Little is known about the physiology of petal cells, after abscission. The cells of abscised petals, as long as they remain alive, will be subject to a considerable decrease in water potential. Using young detached corollas (petals) of Ipomoea flowers, we observed that FW was about 65% of initial after 24 h on the laboratory bench (at about 20°C and 70% RH) and 50% of initial after 48 h (unpublished data). This indicated that some physiological processes would occur at least during day-1 and -2, but that these processes would be influenced by lower water potential. Both agarose gel (Fig. 2, P3 and D3) and flow cytometry analysis (Fig. 4) showed DNA degradation after the petals had abscised. The data indicate that the PCD process, once started, goes on even under conditions of increasing water stress. However, by the time of total desiccation of the petals, the amount of degraded DNA on gels was smaller than at the time of visible petal wilting of the abscised petals (Fig. 2, P4 and D4). This indicated that the DNA breakdown became halted by that time. This could be due to the fact that there was no more DNA, or might relate to the very low water potential in the tissue. After petal abscission, the petal nuclei unexpectedly increased in size. In both species the chromatin became unevenly distributed. The chromatin became also much less condensed, as indicated by decreasing DAPI brightness. The increase in nuclear size and uneven and reduced DAPI brightness were also observed after dehydration of young D. belladonna and P. yedoensis petals. In these experiments flowers were cut from the plant just after floral opening and then subjected to a desiccation treatment (24 h at 60°C). This treatment, apart from inducing dehydration, also involved an increase in temperature. The effects did not seem to be due to high temperature though, as preliminary experiments had shown that 72–96 h desiccation at room temperature had the same effects as 24 h treatment at 60°C. This indicates that the effect was due to dehydration rather than high temperature, and that it was not only taking place in cells that were relatively old, but also in much younger cells. I. nil and P. hybrida flowers both exhibit petal wilting rather than abscission at the end of floral life span. An increase in nuclear volume and a decrease in DAPI brightness were also found in the nuclei, if the flowers of these species were severed when just fully open and then subjection to a 24 h 60°C desiccation treatment. Preliminary experiments with these species had also shown that 72–96 h desiccation at room temperature had the same effects as 24 h treatment at 60°C. These results indicate that the particular nuclear morphology observed after dehydration in P. yedoensis and D. belladonna is also found in the petals of other species, which show petal wilting rather than petal abscission at the end of flower life. The particular nuclear morphology, therefore, seemed related to the low water potential as a result of the dehydration treatment. After the dehydration treatment in I. nil or P. hybrida the nuclear morphology was quite different from the one observed during wilting as the final phase of PCD. At the time of the normal senescence-related petal wilting in Ipomoea (Yamada et al. 2006b) and Petunia (Yamada et al. 2006a) flowers, the nuclei had become much smaller than at previous phases of development, and had a rather bright DAPI fluorescence. The desiccation treatment, therefore, induced changes quite different from those observed during PCD: a large nucleus with faint DAPI fluorescence rather than a small one with bright fluorescence. It is concluded that the tested parameters indicated advanced stages of PCD prior to petal abscission in D. belladonna whereas PCD was much less obvious by the time of petal abscission in P. yedoensis. Dehydration of the petal cells, after abscission, was associated with a remarkable change in nuclear morphology and chromatin distribution. Similar changes in the nucleus were found after dehydration of younger petal cells, and after dehydration of the petal cells in two other species. This nuclear morphology has apparently not been described previously.

J_Abnorm_Child_Psychol-3-1-1915605

Children’s Feedback Preferences in Response to an Experimentally Manipulated Peer Evaluation Outcome: The Role of Depressive Symptoms

The present study examined the linkage between pre-adolescent children’s depressive symptoms and their preferences for receiving positive vs. negative feedback subsequent to being faced with an experimentally manipulated peer evaluation outcome in real time. Participants (n = 142) ages 10 to 13, played a computer contest based on the television show Survivor and were randomized to either a peer rejection (i.e., receiving the lowest total ‘likeability’ score from a group of peer-judges), a peer success (i.e., receiving the highest score), or a control peer evaluation condition. Children’s self-reported feedback preferences were then assessed. Results revealed that participants assigned to the negative evaluation outcome, relative to either the success or the control outcome, showed a significantly higher subsequent preference for negatively tuned feedback. Contrary to previous work and predictions derived from self-verification theory, children higher in depressive symptoms were only more likely to prefer negative feedback in response to the negative peer evaluation outcome. These effects for depression were not accounted for by either state mood at baseline or mood change in response to the feedback manipulation.   Depression is a leading cause of disability across the life span (Murray & Lopez, 1996). Whereas the point prevalence of major depressive disorder (MDD) in school-age children is relatively modest (i.e., between 1 and 3 percent; Cohen et al., 1993), epidemiological community surveys of childhood disorders suggest that 10 to 20 percent of school-aged children in the general population experience periods of elevated depressive symptoms, including depressed mood (e.g., Compas, 1997; Hammen & Rudolph, 1996). Several studies have shown that moderate levels of depression are associated with significant impairment in school and peer functioning (e.g., Nolen-Hoeksema, Girgus, & Seligman, 1992; Petersen, Sarigiani, & Kennedy, 1991), and may persist for years in some children (Nolen-Hoeksema et al., 1992). Moreover, children displaying moderate levels of depressive symptoms are at high risk to suffer from severe, recurrent major depression later in life (e.g., Costello, Angold, & Keeler, 1999; Lewinsohn, Rohde, Klein, & Seeley, 1999). Taken together, although moderate levels of depressive symptoms may not meet DSM-IV-TR (APA, 2000) criteria for MDD, they are of significant concern. Several theorists have argued that feedback seeking may be implicated in the maintenance and/or exacerbation of dysphoric states, including the development of full-blown depression. For instance, Coyne (1976) has asserted that people high in depression systematically alienate their relationship partners through excessive reassurance seeking. This behavior is particularly likely to elicit negative reactions in others—including rejection, raising the risk of becoming even more depressed—when people’s feedback seeking reflects their suspicion that they will not receive such reassurance (e.g., constantly asking their partners if they still love them). Whereas Coyne suggested that depressed people merely inadvertently create the social conditions that make them suffer, the more recently advanced self-verification theory (e.g., Swann, 1990; Swann, Rentfrow, & Guinn, 2002) posits that people high in depression seek out negative social feedback more actively, as a means of verifying their negative self-conceptions. Specifically, according to self-verification theory (e.g., Swann, 1983; Swann, Pelham, & Krull, 1989; Swann, Rentfrow, & Guinn, 2002) individuals are drawn to feedback that confirms their self-conceptions because it tends to buttress their feelings of existential security (“epistemic” concerns) and/or because they want their social interactions to unfold smoothly, and thus work to ensure that others do not form appraisals of them that are overly positive or overly negative (“pragmatic” concerns). Negative self-views including worthlessness, repulsiveness, and excessive self-criticism figure prominently in depression (e.g., Beck, 1967; Ellis, 1977). Consistent with self-verification theory, several studies among adults have provided evidence to suggest that people higher in depression are more inclined to actively seek out (or prefer) negative social feedback, as well as interaction partners likely to provide them with such unfavorable feedback. For instance, Swann and colleagues observed that depressed college students, relative to controls, preferred friends who evaluated them negatively, were more inclined to seek negative feedback from their roommates, and preferentially solicited unfavorable feedback from their dating partners (Swann, Wenzlaff, Krull, & Pelham, 1992). Moreover, when presented with self-discrepant (i.e., positive) feedback, adults high on depressive symptoms appeared to attempt to restore their sense of worthlessness by subsequently seeking feedback about their weaknesses from a different evaluator; whereas adults low on depressive symptoms were more likely to solicit feedback about their strengths (Swann, Wenzlaff, & Tafarodi, 1992). Finally, participants higher on depressive symptoms, relative to controls, were more likely to choose interacting with someone who appraised them negatively over the opportunity to avoid the encounter (Swann et al., 1992). Contrary to research with adults, the investigation of how depression in children is associated with feedback seeking preferences is still in its infancy. Joiner and colleagues (Joiner, Katz, & Lew, 1997) examined the linkage between depression and feedback seeking in a sample of inpatient youths. They found that children who were more depressed showed a greater preference for receiving feedback of a negative nature, based on responses to a questionnaire assessing feedback preferences in four distinct self-relevant domains (i.e., social, intellectual, athletic, and physical attractiveness). As noted by the authors, an important limitation of this study was the unknown predictive validity of their feedback preference questionnaire. Hence, it is unclear to what extent scores on this measure converge with participants’ actual preference for negative feedback in the face of actually occurring events (see Joiner et al., 1997). In the only study to examine feedback seeking in children using a community sample, Cassidy and colleagues used a bogus feedback paradigm to examine children’s actual feedback preferences (Cassidy, Ziv, Mehta, & Feeney, 2003). Specifically, participants provided personal information about themselves by completing a series of questionnaires. They were led to believe that their responses would be shown to peers from a different school who would use this information to answer specific questions of an evaluative nature about each participant. Next, participants were provided the opportunity to select some of the alleged answers to these questions, knowing that some answers would include negative feedback and others would include positive feedback. In line with findings for adults and children diagnosed with depression, children displaying higher depressive symptoms were more likely to display a relative preference for negative information about themselves. A similar pattern of findings consistent with self-verification theory was observed for the linkage between feedback preferences and global self-representations, as indexed by level of general self-esteem. An important contribution of the Cassidy et al. study is that it provides evidence to suggest that the tendency for depressed individuals to prefer rather negative feedback extends to moderately depressed children and adolescents, as indexed by level of self-reported depressive symptoms. However, the study is limited by its relatively narrow scope of inquiry with regard to the role of depressive symptoms. First, the potential role of state mood was not examined. Whereas self-verification theory claims that interest in negative feedback is motivated by negative self-views, an alternative explanation for the linkage between elevated depressive symptoms and more negative feedback preferences is that due to high levels of depressed state mood, people displaying elevated depressive symptoms are prone to be drawn to all that is negative, including unfavorable appraisals from others (e.g., Beck, 1967; Beck, Rush, Shaw, & Emery, 1979). Moreover, enhancing state mood with adults by providing them with experimentally manipulated favorable evaluations has been shown to increase levels of subsequent positive feedback seeking (Mischel, Ebbesen, & Zeiss, 1973). A second limitation of the Casssidy et al. study is that the valence of the evaluative feedback was not manipulated. Beck has asserted that depressed people seek unfavourable feedback only insofar as their negative self-views have been activated by a stressful negative event. Work conducted by Swann and colleagues (Swann et al., 1992) has shown that adults high in depressive symptoms respond quite differently with respect to their feedback preferences in the face of positive, negative, or neutral evaluations. Specifically, these individuals responded to positive—i.e., self-discrepant—evaluations with a more pronounced preference for negatively tuned feedback, relative to a control condition; whereas they were not more inclined to seek negative feedback subsequent to receiving unfavorable—i.e., self-congruent—evaluations. The present study examined the linkage in pre-adolescent children between depressive symptoms and feedback preferences, in response to a salient and ego-involving emotion-eliciting event. Specifically, we examined these linkages in response to both a positive and a negative peer evaluation outcome. We chose peer evaluation as the emotion provocation stimulus because peer praise and peer rejection are common emotion-eliciting events in childhood (Coie, 1990). Moreover, ample evidence suggests that rejection ranks among the most aversive of human experiences and is associated with marked negative affect (e.g., Baumeister & Tice, 1990; Leary, Tambor, Terdal, & Downs, 1995). Finally, peer rejection figures prominently in the development and/or maintenance of several forms of psychopathology, including depression (e.g., Nolan, Flynn, & Garber, 2003). The reason for our focus on pre-adolescent children is twofold. First, the most salient content of self-representations among pre-adolescents is one’s social appeal and social skills/attributes that influence interactions with others (Harter, 1998). Second, in this age group approximtely 50% of children’s social activities involve peers (Grusec & Lytton, 1988; Rubin, Bukowski, & Parker, 1998), and peers as a reference group are of central importance in these children’s lives (e.g., Hay, Payne, & Chadwick, 2004; Hartup, 1996). Toward this aim, we devised an experimental peer evaluation manipulation based on the television show Survivor. In brief, participants were led to believe that they were playing an Internet version of Survivor against four same-sex contestants of comparable age (all of them were computerized bogus contestants) from four different schools in the same area. They were informed that a team of 16 same-age peer judges consisting of eight boys and eight girls would evaluate all participants. Specifically, each juror would give them a score between 0 and 100, with higher scores reflecting higher levels of perceived likeability. Moreover, the jurors would indicate what they liked most and what they disliked most about each contestant. Participants were randomized to one of the following three peer evaluation feedback conditions: (a) success feedback (i.e., having obtained the highest total score), (b) failure feedback (i.e., having obtained the lowest total score), or (c) control feedback (i.e., having obtained neither the highest nor the lowest score). These feedback conditions were designed to induce mood improvement, mood worsening, and no mood change, respectively. Subsequent to receiving the bogus peer evaluation feedback outcome, children’s feedback preferences were assessed. The assessment of state mood at pre- and post-feedback afforded the possibility to examine the linkages between participants’ level of depressive symptoms, feedback valence, (changes in) state mood, and subsequent feedback preferences. To increase the specificity of our findings for depressive symptoms, we also examined the potential role of children’s peer-nominated social standing in affecting their feedback preferences. In sum, the present study aimed to address the following specific research questions: (a) Does the type (valence) of peer feedback influence children’s subsequent feedback preferences? (b) Are the effects of feedback valence moderated by children’s level of depressive symptoms? (c) Do differences in state mood at baseline account for these potential effects of depression? and (d) Are the effects of feedback valence and depressive symptoms on subsequent feedback seeking mediated by changes in mood from pre to post-feedback? We hypothesized that children assigned to receive a positive peer evaluation outcome would be more likely to prefer subsequent favorable feedback than those assigned to the neutral peer feedback condition; whereas children assigned to receive a negative peer evaluation outcome would not be more likely to prefer subsequent negative feedback. This set of predictions was based on findings with adults suggesting that positively valenced feedback leads to a subsequent preference for positively tuned feedback (Mischel et al., 1973), whereas Swann and colleagues (Swann et al., 1992) found that negative feedback exerted no effect on adults’ subsequent preference for negative feedback. We also hypothesized that level of depressive symptoms would qualify the relationship between our peer feedback manipulation and children’s subsequent feedback preferences, such that children higher on depressive symptoms would be more likely to respond to positive—i.e., self-discrepant—evaluations with a preference for negatively tuned feedback, relative to the control condition. Finally, we hypothesized that the effects of the peer feedback manipulation on subsequent feedback preferences would be mediated by changes in state mood from pre- to post-feedback. Method Participants Participants were 142 children (73 boys, 69 girls) enrolled in 5th and 6th grade classes from two public elementary schools in the Netherlands. They were predominantly Caucasian (92.9%), ranged in age from 10 to 13 years (M = 11.2, SD = .66) and were predominantly from a middle-class SES background. For the initial sample of 185 children, classroom teachers sent parent permission letters home with children. Of the 165 letters returned, 142 parents (86.1%) gave their consent for their children to participate in the study, and 23 (13.9%) declined. We also obtained verbal permission to perform the study from the principal of the school and each child’s teacher. Children were informed that they could discontinue their participation at any time. Procedure In the first of two sessions, approximately one week apart, participants were administered the Children Depression Inventory (CDI, Kovacs, 1981) in their regular classrooms during school hours. Moreover, participants’ social standing in their peer group was assessed via a commonly used procedure, which asks children to indicate whom among their classmates they like most and whom they like least (e.g., Newcomb, Bukowski, & Pattee, 1993). During administration of the measures the classroom teacher remained in the room. A research assistant read the directions aloud and children were encouraged to ask for help if they had questions or encountered problems completing the questionnaires. At the end of the first session, which lasted approximately 30 min, children were informed that later that week they would participate in a computer-contest. The second session was carried out in a quiet room on the school grounds. Participants were told that their class was selected to take part in an Internet computer-contest called ‘Survivor.’ In reality, the contest was a computer program written in Visual Basic designed to present the illusion of playing on-line with four other children. Survivor contest Upon arrival, the participant was seated in front of a laptop computer equipped with a web-cam to have their photo taken. Participants were told that their picture would allow all the contestants to see what each of the other children looked like. Prior to beginning (Time 1), participants completed a computer-administered baseline mood measure, i.e., a one-item mood scale ranging from 1 (extremely negative) to 10 (extremely positive). In an attempt to add both to the credibility and the attractiveness of the contest, the opening bars of the hit ‘Survivor’ (produced by the band “Destiny’s Child”) were played at the start of the game. In addition, an eye-catching logo of the American TV-show appeared on the computer display. The objective and rules of the contest were presented on screen. Participants were encouraged to read the information, which was pre-tested on comprehensibility for children in this age-range, carefully in their own pace and click “continue” to progress to the next screen. Participants were informed that they would be playing against four same-sex contestants of comparable age (all of them were computerized fictitious co-players) from four different schools in the same area, and that all participants would be evaluated by a jury consisting of 16 members, eight boys and eight girls. Specifically, participants were explained that each judge would give them a score between 0 and 100, with higher scores reflecting higher levels of perceived likeability. After receiving this information, the computer displayed a screen announcing that in a moment pictures and names of all 16 judges would be presented one-at-a-time. The children whose pictures appeared were child actors from two different modeling agencies in the Netherlands. In a pilot study, 31 adults rated the attractiveness of 43 children’s faces, (20 boys, 23 girls) on a scale ranging from 1 (extremely unattractive) to 10 (extremely attractive). For the purposes of the present study, we selected the eight children rated highest (four boys, M = 7.9, and four girls, M = 8.1) and the eight children rated lowest (four boys, M = 5.2, and four girls, M = 5.1). Following the viewing of these children, participants were directed through a series of screens in which they were asked to answer a series of questions that would give the members of the jury and the other contestants information about them. Participants responded to questions about their favorite musical group, hobbies, future occupation, things they liked and disliked about themselves, a number of character traits (e.g., sense of humor, agreeableness, intelligence, trustworthiness), how they got along with other children, and their academic performance. Most of the questions were in a multiple-choice format but some (e.g., “what is your favorite musical group?”) required an open-ended response. In view of the potential risk that children would respond to personal questions in a socially desirable fashion, instructions emphasized the importance of responding to questions honestly. To further minimize response bias, most questions were worded such that the “best” or “optimal” answer was unclear (e.g., “what is your favorite leisure activity”?). Participants were informed on screen that their picture (previously taken by a web camera) along with the biographical information from their answers to the personal questions would be transmitted over the Internet and viewed by the judges who would then give them a ‘likeability’ score ranging from 0 to 100. Moreover, children were informed that the judges would also indicate what they liked most and what they disliked most of each participant (e.g., “(s)he doesn’t seem to be a nice person,” “(s)he is witty,” “(s)he is unattractive,” “(s)he seems fun to hang out with”). Subsequent to answering all the biographical questions, participants were informed that pictures and descriptions of each of the other contestants would be presented one-at-a-time for review. Upon clicking “continue,” the picture of the first bogus co-player was displayed together with his or her self-description. The latter consisted of the alleged answers to the same questions that the participant had answered earlier. To enhance credibility of the bogus co-players, actual self-descriptions were taken from those of same-age children participating in another study. These participants gave their explicit consent to have this information viewed by other children, provided that the alleged self-description profiles would contain randomly combined personal information from at least three different children. Participants progressed through the game examining each of the profiles at their own pace. Following the participant scrutinizing the last profile, a message appeared on the screen indicating that the computer would now for every player add the scores from the judges to determine which player had received the highest total score and which player had received the lowest total score. After a 5 s waiting period, the names of the players with the highest and the lowest score appeared in capital letters on the screen. In the success condition, the name of the participant was displayed as having obtained the highest total score; one randomly chosen alleged co-player’s name appeared as having obtained the lowest total score. Conversely, in the failure condition the name of the participant was displayed as having obtained the lowest total score, while one alleged co-player’s name appeared as having obtained the highest total score. In the control condition, the participant received neither the highest nor the lowest score. Five seconds after receiving feedback (Time 2), participants were re-administered the mood measure via computer. Instructions emphasized the importance of rating how they felt ‘right now.’ Subsequent to completing this measure, participants were administered the feedback measure which assessed preferences for obtaining feedback from jurors that had evaluated them favorably versus jurors that had evaluated them unfavorably. Subsequently, a computer screen appeared announcing that children would now participate in another study. The participant was then accompanied to an adjacent room where a female research assistant debriefed him or her thoroughly. Debriefing Each child was thoroughly debriefed with the aim of removing any lingering effects of the false rejection feedback while participating in the Survivor contest. During the debriefing, the child was informed that the judges, the co-players and the received feedback were entirely fictitious and that this deception was a necessary part of the procedure. At this point it should be noted that in our previous work, more than 100 participants were assigned to receive Survivor-administered peer rejection feedback (Reijntjes et al., 2006), including children with elevated depressive symptoms. Interviews with all participants, both immediately post-feedback as well as at one-week follow-up, indicated that the peer failure experience was not too emotionally upsetting. For instance, when asked, none of the participants made mention of any feelings of regret with regard to participation and none reported any objections to the procedure. On the contrary, several children—spontaneously—reported to consider the rejection experience an unnerving but useful experience, for themselves and/or their peers. Still, during the investigation a registered clinical psychologist was available when needed. Moreover, before being dismissed children were interviewed at length about a recent positive social experience in which they felt positive and/or were successful. In so doing, we aimed to maximize the remedy for children showing a strong emotional response to the peer feedback. Towards the end of the debriefing, participants were encouraged to ask questions or voice their concerns. All children reported that they understood the purposes of the research, as well as the necessity of having been deceived. The credibility of the deception manipulation was also assessed during the debriefing by asking each participant whether they had believed that they were playing against other children. With no exception, participants indicated that they believed that the game was genuine. Finally, all participants reported that prior to participating they had not talked with classmates about Survivor. At the conclusion of the debriefing, participants were urged to observe complete secrecy by not talking with their classmates about Survivor until all the other children had participated. To increase adherence to this instruction, children were asked to sign a non-disclosure agreement and were then provided a choice of one of several possible small gifts for playing the game (e.g., a small tape recorder, a gift certificate worth about 3 dollars). Measures State mood Similar to previous work among children (Reijntjes, Stegge, Meerum Terwogt, Kamphuis, & Telch, 2006) and adults (e.g., Gross, 1998), children rated their current feeling state using a one-item global mood index, ranging from 1 (extremely negative) to 10 (extremely positive). This Likert-scale was administered to assess participants’ state mood at baseline (pre-feedback), and their changes in state mood from pre- to post-feedback. The mean score at baseline was 7.96 (SD = 1.36). Scores did not differ as a function of age, gender, or their interaction. Children Depression inventory (CDI; Kovacs, 1981) The CDI is a 27-item self-report measure designed to assess the social, behavioral, and affective symptoms of depression in children. Each item consists of three sentences that describe a symptom of depression in increasing degrees of severity. The respondent chooses the sentence that best describes him or her during the past week. Each item set is scored from 0 (symptom absent) to 2 (symptom is present always or most of the time). The CDI has adequate discriminant and convergent validity, test-retest reliability, and internal consistency (Saylor, Finch, Spirito, & Bennett, 1984). Coefficient alpha in the present sample, using the Dutch version of the instrument (Braet & Timbremont, 2002), was .80. Total scores ranged from 0 to 23; the median is 6. The top quartile (i.e., 24.6% of the participants) obtained a score of 12 or higher. Eleven participants (7.7%) obtained a score of 20 or higher, which is recommended as a cut-off point for general screening (Kovacs, 1992). Mean scores of this sample (M = 8.03, SD = 6.59) were similar to those previously reported (Reijntjes et al., 2006; Smucker, Craighead, Craighead, & Green, 1986) and did not differ as a function of age, gender, or their interaction. Table 1Means and standard deviations of baseline measures by conditionFeedback conditionSuccess (n = 47)Neutral (n = 47)Failure (n = 45)MeasureMSDMSDMSDDepressive symptoms 7.837.18 8.346.78 8.116.00Social acceptance 3.7227.47 3.9027.63 3.0226.75Age (months)134.628.22134.178.11132.387.52 Social standing in the peer group Participants completed a widely used nomination-based sociometric questionnaire, in which they identified the three classroom peers whom they liked most and the three classroom peers they disliked most (Newcomb et al., 1993). Children who declined participation in the present study were included in the classroom lists, however the data collected on these children were not used. From these nomination data, two continuous scores for each participating child were computed. These scores included: (a) a measure of preference, by dividing the number of times each participant was nominated for the ‘like most’ question by the total number of participating children in the class, and (b) a measure of rejection, by dividing the number of times each participant was nominated for the ‘like least’ question by the total number of participating children in the class. Our measure of social acceptance (i.e., social standing in the peer group) was computed by subtracting the measure of rejection from the measure of preference. This figure was then multiplied by 100, yielding scores ranging from—84.6 to 61.5 (M = 3.56; SD = 27.10). These scores did not differ as a function of gender, age, or their interaction. For the majority of participants (58%), a positive social acceptance score was observed. Based on the widely used taxonomy for sociometric classification (Newcomb et al., 1993), the number of children that were classified as “popular,” “rejected,” “neglected,” “controversial,” and “average” amounted to 26, 26, 24, 8, and 58, respectively. Self- reported feedback preference Subsequent to receiving the bogus peer evaluation outcome, participants responded to the measure assessing the valence of feedback preferences (i.e., positive versus negative). The specific dichotomous probe included: ‘Which judges would you prefer to get feedback from, those that hold a positive view on you or those that hold a negative view on you’? Results Preliminary analyses Exploratory analyses revealed a departure from symmetry for the distribution of the CDI, as evidenced by a skewness value of 5.84. We therefore performed a square root transformation on the scores, which was successful in producing a distribution that was no longer skewed (skewness value is .73). This transformed variable (CDIt) was used in all analyses reported below. Participants reporting higher CDIt scores displayed lower levels of state mood at baseline (r = −.26, p < .01). Moreover, higher CDIt scores were negatively associated with peer nominated social acceptance ratings (r = −.21, p < .02). Equivalence of the experimental groups Children were matched on age and gender, and then randomly assigned to one of the three experimental conditions. Means and standard deviations for Time 1 measures are presented in Table 1. To confirm that the randomization procedure resulted in comparable groups, baseline differences on continuous measures were examined using one-way ANOVAs. Results revealed no significant between-group differences. Change in state mood: Effects of the feedback manipulation Means and standard deviations for state mood scores at Time 1 (baseline) and Time 2 (immediately post-feedback) are presented in Table 2. Two a priori repeated measures ANOVA’s were performed to examine the magnitude and direction of emotional reactivity elicited by success feedback and failure feedback, each relative to the control (neutral) feedback condition. In the first planned contrast, state mood scores at Time 1 and Time 2 served as the dependent variables. Condition (Success vs. Neutral feedback) served as the between subjects factor, and Assessment phase (Pre-feedback vs. Post-feedback) served as the within subject repeated measures factor. Interaction effects were followed up with simple main effects analyses, comparing Time 1 vs. Time 2 scores for each feedback condition separately. In the second analysis, the planned contrast comparing Failure vs. Neutral feedback was tested using the same analytic approach as that outlined above for the Success vs. Neutral feedback contrast. Table 2Means and standard deviations for state mood at baseline (Time 1) and immediately post feedback (Time 2) by conditionFeedback conditionSuccess (n = 46)Neutral (n = 45)Failure (n = 42)MSDMSDMSDTime 17.831.528.031.088.021.47Time 28.831.258.191.196.842.05Note. Due to technical failures, data were not collected for 6 participants (1 randomized to the success condition, 2 to the neutral condition, and 3 to the failure condition). Effects for success feedback The analysis comparing Time 1 vs. Time 2 scores yielded a significant effect for Time, F(1, 89) = 34.70, p < .001, which was qualified by a significant Time by Condition interaction, F(1, 89) = 18.53, p < .001, η2 = .17. Subsequent simple effect analyses for both conditions separately revealed a significant increase in state mood in the success condition, F(1, 45) = 61.79, p < .001, η2 = .58, but no significant change in state mood in the neutral condition, p > .30. Taken together, these findings show that the success feedback condition elicited change in state mood in the expected positive direction, whereas the control feedback condition elicited no emotional reactivity. Effects for failure feedback The analysis comparing Time 1 vs. Time 2 scores yielded a main effect for Time, F(1, 85) = 12.84, p < .001, which was qualified by a Time by Condition interaction, F(1, 85) = 21.08, p < .001, η2 = .20. Subsequent simple effect analyses for both conditions separately revealed a significant decrease in state mood in the failure condition, F(1, 42) = 20.83, p < .001, η2 = .34, but no significant change in state mood in the neutral condition, p > .30 (see above). These findings indicate that the failure feedback condition elicited change in state mood in the expected negative direction. Do depressive symptoms, social acceptance, or gender moderate change in state mood? The potential moderating effects of depressive symptoms, social acceptance, and gender on affective change were examined using the regression approach outlined by Aiken and West (1991). Separate hierarchical regression analyses were performed for the success feedback condition and the failure feedback condition, each relative to the control (neutral) feedback condition. In Step 1 of the first analysis, state mood score at Time 2 was regressed on state mood score at Time 1. In Step 2, feedback condition, gender (both dummy coded as 0 or 1), social acceptance score (centered), and CDIt score (centered) were entered. In Step 3, all three two-way interactions were entered, including the CDIt by Condition interaction term. Finally, the three-way interactions (e.g., Condition by Gender by CDIt) were entered in Step 4. Moderator effects for success feedback Change in mood from pre to post-feedback was significantly predicted by feedback condition (see above). However, neither depressive symptoms, social acceptance score, nor gender moderated the relationship between feedback condition and mood change, as evidenced by non-significant interaction effects (p’s > .25). Moderator effects for failure feedback A similar analytic strategy as reported above for success vs. neutral feedback was employed. Results revealed that change in mood from pre to post-feedback was significantly predicted by feedback condition (see above). Again, none of the three within-person variables examined moderated the relationship between feedback condition and mood change, as evidenced by non-significant interaction effects (p’s > .25). Self-reported feedback preferences: Effects of the feedback manipulation Self-reported feedback preferences for each of the three conditions are presented in Table 3. Results are reported for the total sample and for the top and bottom third as indexed by children’s score on the CDI. Similar to the analyses reported above, we examined feedback preferences in response to success feedback and failure feedback, each relative to the control (neutral) feedback condition.Table 3Feedback preferences across conditions broken down by CDI statusFeedback conditionSuccess (n = 47)Neutral (n = 47)Failure (n = 45)PreferredPreferredPreferredPreferredPreferredPreferredCDI statuspositive (%)negative (%)positive (%)negative (%)positive (%)negative (%)Low (Bottom third) CDI<594.55.582.417.683.316.7High (Top third) CDI>1088.211.893.76.335.764.3Total sample91.58.591.58.566.733.3 Effects of success feedback As displayed in Table 3, there was no significant effect for Condition. Binominal tests showed that in both the success and the neutral condition children preferred significantly more positive feedback than expected by chance (p’s < .001). Effects of failure feedback A Fisher-Exact test revealed a significant effect for condition, p < .004. This finding indicates that children randomized to receive a negative peer evaluation outcome, relative to controls, were significantly more inclined to subsequently prefer negatively tuned feedback. However, a binominal test revealed that children in the failure condition still preferred significantly more positive feedback than expected by chance, p < .04. Do depressive symptoms, social acceptance, or gender moderate feedback preferences? The potential moderating effects of depressive symptoms, social acceptance, and gender on feedback preferences were examined using logistic regression analyses. In these analyses feedback preference (positive vs. negative) served as the binary outcome measure. In Step 1 of the first analysis, state mood at baseline (centered) was entered. In Step 2, feedback condition (success vs. control), gender (both dummy coded as 0 or 1), social acceptance score (centered), and CDIt score (centered) were entered. In Step 3, the two-way interaction terms were entered, including the CDIt by Condition interaction term. Finally, the three-way interaction terms (e.g., Condition by Gender by CDIt) were entered in Step 4. An identical analytic strategy as reported above was used for the failure vs. neutral contrast. Moderator effects for success feedback Feedback preferences were not predicted by either state mood at baseline, gender, condition (see above), CDIt, or social acceptance score. Moreover, all interaction terms, including the interaction between CDIt and condition were not significant (p’s > .20). Moderator effects for failure feedback Feedback preferences were not predicted by state mood at baseline, gender, or social acceptance score. However, we observed a significant effect for condition (see above). Moreover, our findings yielded a significant effect for the interaction between CDIt and Condition (p < .05). Subsequent analyses revealed that children higher in depressive symptoms were significantly more likely to prefer negative feedback subsequent to the negative peer evaluation outcome (relative odds = 2.44, 95% CI = 1.14 to 5.22, p < .02). In contrast, no significant linkage between depressive symptoms and feedback preferences emerged in the control condition (p > .30).1 Are the effects of feedback valence on feedback preferences mediated by changes in state mood from pre- to post-feedback? The proposed mediation model depicted below was examined following Baron and Kenny (1986), who asserted that a variable functions as a mediator when it meets the following three conditions: (a) variations in the level of the independent variable account for variations in the presumed mediator (i.e., change in mood; path A); (b) variation in the level of the presumed mediator account for variations in the dependent variable (i.e., feedback preferences, path B); and (c) after controlling for paths A and B, a previously significant relation between the independent and dependent variables (i.e., path C) is no longer significant, with the strongest demonstration of mediation when path C becomes zero (Fig. 1).Fig. 1Mediational model of feedback preferences Our analyses revealed that the second condition for mediation was not met, as evidenced by a non-significant relation between changes in state mood and feedback preferences. Taken together, our findings show that neither state mood at baseline (see above), nor changes in affect from pre- to post-feedback were significantly associated with children’s feedback preferences. Discussion The present study sought to advance our knowledge on how sub-clinical depression in pre-adolescent children is associated with self-reported feedback preferences. To our knowledge, this is the first study to investigate this linkage in response to an experimentally manipulated, ecologically relevant emotion-eliciting event in real time. In so doing, we included both a positive and a negative peer feedback manipulation, along with the assessment of the roles of state mood at baseline, changes in affect from pre- to post-feedback, and actual standing in the peer group. Data on participants’ immediate changes in state mood in response to the feedback manipulation revealed that the Survivor contest was successful in eliciting differential affective reactions in the expected direction as a function of feedback valence. Moreover, our debriefing interviews revealed that children were involved in the activity and that none of the participants reported being aware that the feedback they received was bogus. Taken together, these data suggest that the Survivor contest was successful in achieving its major objective of providing a credible and ecologically relevant emotion-eliciting event. Several interesting findings emerged with respect to the examination of feedback preferences subsequent to receiving a positive (success) and a negative (failure) peer evaluation outcome, each relative to the control feedback condition. For instance, contrary to findings reported by Mischel and colleagues (1973), feedback preferences did not differ as a function of receiving a favorable versus a neutral control peer evaluation outcome. In contrast, relative to the control condition, the negative peer evaluation outcome lead to a significant increase in negatively tuned feedback preferences. However, it should be noted that the negative evaluation outcome merely induced a relative preference for negative feedback, as evidenced by the fact that in all three conditions children sought significantly more positive feedback than expected by chance. What might account for the finding that the effects of manipulated feedback on subsequent feedback preferences in the present study are at odds with those observed in prior work? One possible explanation may involve the (single and joint) effects of differences in the domain of feedback seeking, the age of the participants, and the specific measures that were employed. For instance, Mischel et al. (1973) provided adult participants with feedback on an intellectual task, and observed that a favorable feedback outcome increased subsequent positive feedback seeking; whereas in the present study pre-adolescent children received feedback on a social task. Another explanation may be that our manipulated peer evaluation outcome differs from the types of manipulations used in other studies examining feedback preferences. It may be that the manipulation employed in the present study may have been more personally relevant (i.e., ego-involving), and hence elicited a different pattern of feedback preferences relative to exposure to less personally relevant feedback (e.g., impressions evaluators had allegedly formed on the basis of participants’ displayed nonverbal behavior while delivering a speech; see Swann et al., 1992). Specifically, it seems likely that participants in the present study were more concerned and/or distressed after being rejected based on their personal and physical attributes, relative to participants receiving a negative evaluation based on their non-verbal behavior. Consequently, they may have displayed an increased interest in (preference for) negative feedback, which might be considered more diagnostic and informative on learning why the judges viewed them negatively. Our findings on the linkages between the valence of the manipulated feedback outcome and subsequent feedback preferences were not consistent with previous work among adults. In a similar vein, our results on the relationship between depressive symptoms and feedback preferences qualify the findings reported by Cassidy et al. (2003). Specifically, in the present study the predictive role of depressive symptoms on subsequent feedback preferences emerged only for children in the negative feedback condition, such that children with higher levels of depressive symptoms displayed a significantly stronger preference for negative feedback. This depression effect was not observed in either the success or control conditions. Interestingly, across conditions we found that children’s actual standing in the peer group was not associated with feedback preferences. In this context, it should be noted that in the Joiner et al. (1997) study it was also observed that interest in negative feedback evidenced symptom specificity to depression. What might account for the observed effects of depressive symptoms on children’s feedback preferences in response to the negative peer evaluation outcome? We considered the possibility that the observed effects were due to the higher levels of depressed mood observed at baseline. However, our findings indicated that this was not the case. Moreover, the observed linkage between depressive symptoms and feedback preferences cannot be explained by change in affect, given that change in affect from pre- to post-feedback did not mediate the single or joint effects of feedback valence and depressive symptoms on feedback preferences. These findings converge with the Joiner et al. (1997) study showing that interest in negative feedback among depressed inpatient children was not associated with the emotional features of depression. We also considered the possibility that self-verification processes might account for the observed linkage between depressive symptoms and feedback preferences. However, our finding that the linkage was present in the failure condition but not in the success or neutral conditions is not in line with self-verification theory, which would predict depression effects across all three conditions. Rather, our findings are more consistent with the formulation put forth by Beck (1967), asserting that depressed (dysphoric) people seek negative feedback only insofar as their negative self-views have been activated by a potent stressful negative event. Possible implications of our findings deserve mention. The greater tendency for children high in depressive symptoms to seek out negative peer feedback in response to negative peer experiences might contribute to further negative evaluations from peers, and thus serve to maintain or exacerbate depressive symptoms in a cycle similar to that described by Coyne (1976). In line with this reasoning, considerable work has shown both concurrent and prospective linkages between heightened depressive symptoms (depression) in children and peer rejection (Panak & Garber, 1992), decreased peer interactions (Axelson et al., 2003), and less secure peer attachment (Armsden & Greenberg, 1987; Armsden, McCauley, Greenberg, Burke, & Mitchell, 1990). Moreover, in a recent longitudinal study Borelli and Prinstein (2006) observed that a more negative feedback seeking style at baseline was positively associated with more negative perceptions of best friendship criticism at one-year follow-up, and positively associated with increases in depressive symptoms over time. However, this linkage was only found for girls. Several features of the present study deserve further comment. First, only self-reported feedback preferences were assessed, with no attention given to feedback seeking behavior. Future work should examine children’s feedback seeking when they are free to choose how to divide their time between different sources of feedback information. Second, because all significant results are based on children’s self-report, it is possible that our findings are partly due to shared method variance. Third, our findings are based on a community sample of children, rather than a clinical sample with a diagnosed mood disorder. Children’s mean scores on the CDI suggest that depressive symptoms were relatively modest in magnitude. It is therefore an empirical question to what extent our findings can be generalized to children who meet criteria for major depressive disorder. Finally, this investigation focused only on children’s feedback preferences in response to peer evaluation. While peer evaluation has the benefit of being both a potent elicitor of positive and negative affect as well as being ecologically sound, we should not assume that our findings are generalizeable to other relevant domains such as academic failure, interpersonal conflict, and loss or separation. Notwithstanding these limitations, the results of the present study extend our limited knowledge base on children’s feedback preferences in several ways. First, our findings suggest that children respond quite differently with respect to their feedback preferences subsequent to positive or negative evaluations. Second, in line with the Joiner et al. (1997) study, our findings contradict the view that the baseline feeling state of children high in depressive symptoms or manipulations of affect may be involved in governing their preference for more unfavorable feedback. Third, our findings provide evidence to suggest that children displaying elevated depressive symptoms may prefer favorable feedback so long as their negative self-views are not accessed by a potent negative stressful event.

Eur_Radiol-2-2-1766022

MR imaging of osteochondral grafts and autologous chondrocyte implantation

Surgical articular cartilage repair therapies for cartilage defects such as osteochondral autograft transfer, autologous chondrocyte implantation (ACI) or matrix associated autologous chondrocyte transplantation (MACT) are becoming more common. MRI has become the method of choice for non-invasive follow-up of patients after cartilage repair surgery. It should be performed with cartilage sensitive sequences, including fat-suppressed proton density-weighted T2 fast spin-echo (PD/T2-FSE) and three-dimensional gradient-echo (3D GRE) sequences, which provide good signal-to-noise and contrast-to-noise ratios. A thorough magnetic resonance (MR)-based assessment of cartilage repair tissue includes evaluations of defect filling, the surface and structure of repair tissue, the signal intensity of repair tissue and the subchondral bone status. Furthermore, in osteochondral autografts surface congruity, osseous incorporation and the donor site should be assessed. High spatial resolution is mandatory and can be achieved either by using a surface coil with a 1.5-T scanner or with a knee coil at 3 T; it is particularly important for assessing graft morphology and integration. Moreover, MR imaging facilitates assessment of complications including periosteal hypertrophy, delamination, adhesions, surface incongruence and reactive changes such as effusions and synovitis. Ongoing developments include isotropic 3D sequences, for improved morphological analysis, and in vivo biochemical imaging such as dGEMRIC, T2 mapping and diffusion-weighted imaging, which make functional analysis of cartilage possible. Introduction Articular cartilage injuries are one of the most common types of injuries seen in orthopaedic practice. In a retrospective review of 31,510 knee arthroscopies, the incidence of chondral lesions was 63% [1]. Full-thickness articular cartilage lesions with exposed bone were found in 20% of patients, with 5% of these occurring in patients less than 40 years old [1]. The treatment of articular cartilage damage remains a challenge because cartilage has a limited capacity for spontaneous repair after a traumatic insult or degenerative joint disease [2]. Joint surface defects that exceed a critical size heal poorly and usually lead to osteoarthritis. As a result, several therapeutic strategies have been developed to restore articular cartilage and produce a durable repair. These methods may be arthroscopic or open surgical techniques and include marrow-stimulation techniques, osteochondral grafting, and chondrocyte implantation technique [3, 4]. As the use of articular cartilage repair techniques has become more widespread, techniques for imaging articular cartilage have become increasingly important. Arthroscopy is unsuitable for routine follow-up due to its invasive nature and associated risks. Conventional radiography does not allow direct visualisation of cartilage and arthrography combined with conventional radiography or computer tomography only provides information related to the cartilage surface [5]. By contrast, through the use of appropriate magnetic resonance imaging (MRI) techniques it is possible to evaluate the biochemical and biomechanical status of cartilage in addition to cartilage morphology. These benefits make MRI a powerful tool for the initial diagnosis and subsequent post-operative monitoring of cartilage lesions and cartilage repair tissue. Articular cartilage repair techniques Marrow stimulation There are several established marrow stimulation techniques, including abrasion arthroplasty, subchondral drilling and microfracture. The principle of marrow stimulation techniques is to abrade (abrasion arthroplasty) or pierce (drilling and microfracture) the subchondral bone at the base of the cartilage defect, causing controlled bleeding and clot formation in the cartilage defect, which leads to the subsequent formation of fibrocartilage. The newly formed fibrocartilage may fill the defect and give relief of symptoms, but it lacks the structural, biomechanical and biochemical properties to sustain normal joint function in the long term [6]. Therefore, there is an increasing interest in osteochondral autografts and autologous chondrocyte implantation (ACI) therapies. Osteochondral grafting Osteochondral autograft transplantation is currently considered as the only surgical technique that provides and retrains proper hyaline articular cartilage [7]. Cylinders of autologous bone with their associated overlying hyaline cartilage are harvested from relatively non-weight-bearing areas of the knee and transferred into similarly sized holes created within the articular defect to be treated [8]. This technique is used most frequently at the knee and ankle joints [9, 10]. The main clinical indications for autologous osteochondral transplantation are focal cartilage defects, osteochondritis dissecans and osteonecroses. Autologous osteochondral transplantation is recommended for cartilage defects up to 3–4 cm2 [8]. Autologous chondrocyte implantation (ACI) ACI is currently considered the treatment of choice in young patients with a symptomatic full-thickness cartilage defect between 2 and 12 cm2 on the femoral condyle [11]. The classical ACI was first described in the mid-1990s and is a two-stage procedure. In the first stage, chondrocytes are harvested and cultured. In the second stage, a periosteal patch, usually harvested from the tibia, is sewn over the defect and the proliferated and differentiated cells are injected beneath the patch in order to fill the cartilage defect [12–14]. ACI provides significant and long-term benefits for patients in terms of diminished pain and improved function [15]. Most complications are directly related to the periosteal graft and a revision arthroscopy rate between 4.8% and 60% has been described due to the problems with the periosteal flap [15, 16]. Early problems (<6 months) include periosteal graft detachment and delamination, and the most common late complication is periosteal hypertrophy. Delamination has been reported in 14% of patients undergoing ACI [17, 18]. Periosteal hypertrophy, has been reported to cause symptoms in 10–26% of the cases [11, 15, 19, 20]. Matrix-associated ACT (MACT) The necessity of a periosteal flap with classical ACI and the complications associated with this periosteal flap have led to the development of biomaterials as carrier for chondrocyte cells. The use of three-dimensional (3D) scaffolds has been shown to favour the maintenance of a chondrocyte-differentiated phenotype [21]. Thus efforts are now focused toward a tissue-engineering approach, which combines laboratory-grown cells with appropriate 3D biocompatible scaffolds. Second generation ACI The second generation of ACI technique use a bioengineered bi-layer collagen membrane rather than a periosteal flap [22, 23]. The use of a collagen membrane simplifies the surgical procedure and reduces overall surgical morbidity. Furthermore, the problem of periosteal hypertrophy can be avoided. Third generation of ACI Further technological advances have led to the third generation of ACI, which use biomaterials seeded with chondrocytes as carriers and scaffolds for cell growth. These “all-in-one” grafts do not need a periosteal cover or fixing stitches and can be trimmed to exactly fit the cartilage defect. The advantages of these new techniques are their technical simplicity, shorter operating time and the possibility to perform the surgery via a mini-arthrotomy or arthroscopy [23–26]. MR techniques for cartilage and cartilage repair imaging Appropriate combinations of MRI techniques allow objective non-invasive measures of the properties of the grafted regions after biological cartilage repair, facilitating the longitudinal follow-up and evaluation of the repair tissue. MRI is currently the standard method for cartilage evaluation as it allows morphological assessment of the cartilage surface, thickness, volume and subchondral bone [27–33]. Additionally MRI techniques can be used to evaluate the biochemical and biomechanical status of articular cartilage. MRI is therefore ideal for the evaluation of the morphologic status of cartilage defects and the repair tissue throughout the post-operative period [34–37]. MR evaluation of cartilage repair can be performed using the same acquisition techniques as used for native cartilage, as recommended by the International Cartilage Repair Society [29, 35]. The most commonly used MRI techniques are intermediate-weighted fast spin-echo (FSE), and 3D fat-suppressed gradient-echo (GRE) acquisition [27–29, 31–35]. Fast spin-echo (FSE) imaging combines heavy T2 weighting, magnetization transfer effects and relative preservation of high signal intensity in the marrow fat, so that the subchondral bone exhibits high signal intensity. In FSE imaging, cartilage appears dark against bright synovial fluid and there is consecutive high contrast between joint fluid and cartilage, and cartilage and bone marrow [40]. Collagen fibres with a highly regular structure, in particular near the bone-cartilage interface, tend to immobilize water molecules and promote dipolar interactions between their protons, thus accelerating T2 relaxation. T2-weighted FSE sequences are therefore useful for both the detection of surface and matrix damage assessed by intrachondral signal abnormalities. Proton density-weighted T2 FSE is of intermediate signal adjacent to the low signal subcortical bony plate, which allows to better visualise intrachondral abnormalities and lesions near to the cortical bone; whereas with T2-weighted FSE cartilage has a low signal intensity, which gives a high contrast to bright synovial fluid, resulting in better delineation of cartilage surface defects. The T2-weighted FSE sequence is relatively insensitive to magnetic susceptibility artefacts, which is advantageous in patients who have undergone previous surgery of the joint. FSE sequences are advocated because of the possibility of acquiring high-resolution images in a relatively short scan time and are normally included in the standard MRI protocol for the knee joint for detection of meniscal and ligamentous lesions [28, 31]. An exciting new development is a 3D FSE sequence [sampling perfection with application optimised contrasts using different flip angle evolutions (SPACE)] with isotropic voxels, which allows multi-planar reformatting in any plane without loss of resolution. The advantage of fat-suppressed 3D spoiled gradient echo sequences is the relatively high signal intensity of articular cartilage in contrast to low signal intensity from the adjacent fat-suppressed tissue. Three-dimensional acquisitions yield images with higher resolution and contrast-to-noise ratio than 2D acquisitions. Fat-suppressed, 3D spoiled gradient echo imaging is easy to perform, widely available and, contrary to other cartilage imaging techniques, such as magnetization transfer imaging, it requires no post-processing of data and avoids misregistration artefacts [27, 31–33, 35, 38]. The 3D data set can be reformatted in any plane, allowing 3D visualisation and volume measurements [27, 30, 39]. New isotropic 3D gradient echo sequences such as DESS (double-echo steady-state), true FISP (fast imaging in steady state precession), balanced FFE (fast field echo), VIBE (volume interpolated breath-hold examination) and MEDIC (multi-echo data image combination), with a voxel size down to 0.5 mm3, at 3 T, and also at 1.5 T with a high gradient strength, have been developed and seem to be very promising for cartilage imaging. However, the usefulness of these techniques is yet to be validated in clinical studies. Caution should be exercised as a flare phenomenon in the double-echo steady-state sequence has been shown to sometimes cause a spurious wavy appearance of a normal graft or cartilage surface [41, 42]. Furthermore, there are limitations in the sensitivity of the fast low angle shot (FLASH) and double-echo steady-state sequences for depicting cartilage abnormalities and opinions vary regarding which sequence is superior for cartilage imaging [42–44]. Indirect MR arthrography has been advocated in the evaluation of ACI [20, 29]. The indirect technique is performed by intravenous administration of contrast agent followed by a period of joint exercise, which leads to uptake of contrast agent by the synovial tissue and subsequent diffusion into the joint cavity. Indirect MR arthrography can be particularly helpful in differentiating delamination of the base of the graft from normal high signal intensity repair tissue in the immediate postoperative period [20, 29]. Thus, indirect MR arthrography may play a role in the evaluation of cartilage repair techniques in the early post-operative period. Quantitative MRI of articular cartilage Quantitative geometric measurements of cartilage parameters, such as cartilage thickness and volume, have been suggested as sensitive image based biomarkers for detecting and monitoring cartilage degeneration in osteoarthritis [45]. This has been possible due to the development of higher field magnets and stronger gradients systems as well as a parallel improvement in image analysis and processing techniques. New techniques allow rapid fully automated generation of accurate 3D reconstructions of articular cartilage layers [46]. Three-dimensional reconstructions can be generated from pre- and post-operative MR examinations and spatially registered to facilitate longitudinal comparisons of the cartilage repair site. This area holds a lot of promise for cartilage repair surgery planning; with on-going work to co-register biochemical and biomechanical sequences, such as dGEMRIC, with high-resolution 3D sequences. High-resolution MRI Most MRI studies on articular cartilage have tried to optimise pulse sequences that accentuate the contrast-to-noise ratio for cartilage, but less attention has been paid to image resolution. Rubenstein et al. [47] demonstrated that the image resolution of standard MR sequences is insufficient to detect fraying of the articular surface of cartilage. It has also been reported that the smooth articular surface of healthy cartilage is indistinguishable from early superficial degenerative changes [27, 31–33]. Therefore, MRI of cartilage repair tissue must be performed at a sufficiently high resolution to detect early surface changes. In order to obtain a sufficient resolution for cartilage repair imaging a 1.5-T MR scanner with a high performance gradient system and a dedicated extremity coil (quadrature/phased array coil) is a minimum requirement. Three-Tesla clinical MR systems are becoming more widespread and can generate images with both a high signal-to-noise ratio and a high resolution. In our experience with cartilage repair patients, the signal-to-noise ratio issue arising with standard field MR units can be partially resolved by using surface coils and optimising the pulse sequence to increase resolution for a given imaging time. Surface coils provide a high signal-to-noise ratio allowing a small field of view and high matrix size to be used resulting in an in-plane resolution of 0.234 mm2 acquired in a clinically acceptable time. However, the use of a surface coil is limited to the evaluation of one knee compartment; since cartilage repair is usually restricted to one compartment, this rarely poses a problem. Using this approach an appropriate assessment of the cartilage repair tissue can be made. Moreover, this approach formed the basis for the definition of pertinent variables for describing articular cartilage repair tissue following biological cartilage repair [48]. MR evaluation of the biochemical and biomechanical status of cartilage Various tissue parameters which may be evaluated by MRI reflect the biomechanical properties of articular cartilage. The T1 and T2 relaxation times and apparent diffusion constants change during the cultivation of cartilage implants [49], indicating that the biomechanical properties of cartilage implants change as the graft matures. Furthermore, the spatial distribution of the T2 relaxation time can be used for in vivo monitoring of the biomechanical properties, pathological changes or aging of various cartilage layers [50, 51]. Glycosaminoglycans (GAG) are the main source of fixed charge density (FCD) in cartilage and changes in GAG concentration are one of the features of cartilage graft maturation. Intravenously administered gadolinium diethylenetriamine pentaacetate anion (Gd-DTPA2−), equilibrates in inverse relation to the FCD, which is in turn directly related to the GAG concentration. Therefore, T1 which is determined by the Gd-DTPA2− concentration becomes a specific measure of tissue GAG concentration. The delayed gadolinium-enhanced MRI of cartilage (dGEMRIC) technique can be considered the method of choice for visualisation and quantitative evaluation of proteoglycan in articular cartilage [52, 53]. The dGEMRIC technique is feasible at 3 T [54] and is useful for evaluating cartilage repair [55]. In vitro studies show that the dGEMRIC index has a good correlation with cartilage biochemical properties [56]. However, in vivo studies of the dGEMRIC index in ACI grafted tissue are not so clearly correlated with stiffness of the repair tissue. The differences are presumably due to differences in the collagen content and architecture of the repair tissue. Therefore, the both GAG (dGEMRIC) and collagen (T2 mapping) imaging techniques are required when evaluating the biomechanical status of cartilage repair tissue. MR classification systems of cartilage implants For the long-term follow-up, evaluation and classification of cartilage repair tissue clinical scores in addition to morphological and biochemical evaluation of biopsies taken during control arthroscopies have previously been used [37, 57]. For optimal use of MRI in the evaluation of cartilage repair tissue, a simple evaluation and a point scoring system that allows efficient statistical data analysis is necessary. A few different classification systems for the description of articular cartilage repair tissue have been proposed [37, 48] (personal correspondence with Dr C. Winalski, Brigham and Womens Hospital, Boston, Mass., USA); however, early systems have certain limitations and deficiencies that can potentially lead to confusion. Roberts et al. [37] used four parameters to assess cartilage repair on MR images: surface integrity and contour, cartilage signal in graft region, cartilage thickness and changes in underlying bone. A score is obtained by summing the values of the four parameters; scores range from 0, no repair, to a maximum of 4, complete repair. Unfortunately the system only assesses each parameter as normal or abnormal and provides no assessment of the graft integration, degree of defect fill or the presence of adhesions. The magnetic resonance observation of cartilage repair tissue (MOCART) scoring system, defined by our working group and shown in Table 1 [48], was designed to systematically record only those observations that can be most accurately and reproducibly determined and to avoid the use of ambiguous terms [48]. The MOCART system has been shown to be reliable and has excellent inter-observer reproducibility for the defined variables [58]. The system is very helpful for the longitudinal follow-up of cartilage repair patients [58], and facilitates prospective multi-centre studies comparing different cartilage repair techniques. Table 1The MOCART scoring systemVariableClassesDegree of defect repair and defect fillingComplete (on a level with adjacent cartilage)Hypertrophy (over the level of the adjacent cartilage)Incomplete (under the level of the adjacent cartilage; underfilling)        >50% of the adjacent cartilage        <50% of the adjacent cartilageSubchondral bone exposed (complete delamination or dislocation and/or loose body)Integration to border zoneComplete (complete integration with adjacent cartilage)Incomplete (incomplete integration with adjacent cartilage)demarcating border visible (split-like)Defect visible  <50% of the length of the repair tissue        >50% of the length of the repair tissueSurface of the repair tissueSurface intact (lamina splendens intact)Surface damaged (fibrillations, fissures and ulcerations)        <50% of repair tissue depth        >50% of repair tissue depth or total degenerationStructure of the repair tissueHomogeneousInhomogeneous or cleft formationSignal intensity of the repair tissueDual T2-FSE        Isointense        Moderately hyperintense        Markedly hyperintense3D GE-FS        Isointense        Moderately hypointense        Markedly hypointenseSubchondral laminaIntactNot intactSubchondral boneIntactOedema, granulation tissue, cysts, sclerosisAdhesionsNoYesEffusionNo effusionEffusion MR findings following osteochondral autografting Due to the invasive nature of arthroscopy, MRI has become the most important tool for the follow-up of patients with osteochondral autografts. There have been a limited number of studies focusing on the morphologic assessment of osteochondral grafts during clinical follow up. With the exception of the study by Link et al. [59], these studies have evaluated only relatively small numbers of patients [60, 61]. An MRI evaluation of osteochondral grafts should include: the number and size of the grafts; bone and cartilage integration, the cartilage surface contour; the contour of the cartilage bone interface; an assessment of the signal in the graft, the adjacent bone marrow and at the donor site; details of any soft tissue abnormalities and an assessment of the contrast enhancement patterns. Cartilage and bone integration and congruity When assessing integration and surface congruity following osteochondral grafting cartilage and bone should be considered separately. With respct to the cartilage, Link et al. [59] found only 15% of patients had an incongruity of the cartilage-cartilage interface. Furthermore, no substantial defects or irregularities of the cartilage overlying the bony cylinders were seen on MRI (Fig. 1). Lastly, gaps between cartilage plugs and between cartilage plugs and adjacent native cartilage were rarely visualised in this study [59] (Fig. 2). Fig. 1Normal cartilage integration of osteochondral autografts in the weight bearing region of the femoral condyle in a patient 2 years after osteochondral autograftsFig. 2Cystic cavities in the osteochondral graft area with fissure-like gap (arrow) between cartilage caps of the grafts in a patient 2 years after osteochondral autografts. Both signs are associated with a poor prognosis Sanders et al. [61] examined 21 patients between 1 and 22 months following osteochondral transplantation using a series of five dynamic contrast-enhanced sequences at 1-min intervals post contrast. Their findings were very similar to Link et al. [59] in terms of cartilage-cartilage interface incongruity. In a study to evaluate the use of indirect MR arthrography in examinations of osteochondral graft patients Herber et al. [60] found that indirect MR arthrography helped to identify a persistent fissure gap between the implanted cartilage and native cartilage. They showed that the grafts were well seated and the cartilage-cartilage interface demonstrated a smooth contour in the majority of cases. Moreover, the authors felt that indirect MR arthrography was superior to unenhanced imaging in the assessment of the cartilage surface. Their results highlight a potential role for indirect MR arthrography in challenging cases. Regarding the bony integration, Link et al. [59] reported that poor bony integration of the osteochondral cylinder may be suggested by the presence of cystic cavities with fluid-like signal intensity and/or a persistent oedema-like signal within the subchondral bone. Cartilage and bone signal intensity From 105 osteochondral cylinders in 55 patients (51%) Link et al. [59] reported bone marrow signal intensities consistent with oedema (hypointensity on T1-weighted images and hyperintensity on the fat-suppressed T2-weighted or PD-weighted images), during the first 12 months post-operatively (Fig. 3). During the 12–24 month period this dropped to 17%. They found cystic changes in the osseous component of the graft in four of 99 cylinders. Furthermore, eight cylinders showed no or partial enhancement after contrast agent was administered and was reported as necrosis. The T2 signal intensities varied depending on whether the osteonecrosis caused a fibro/sclerotic or cystic degeneration. This limits the usefulness of T2-weighted images for diagnosing graft osteonecrosis. Interestingly, only two of the six patients who showed signs of osteonecrosis of one or more cylinders had associated clinical abnormalities [59]. These osteonecroses of the graft cylinders did not lead to collapse of the bone or pathological changes of the cartilage that could be visualised by the MRI. Since cartilage derives its nutrition almost exclusively from the synovial membrane, this may provide an explanation. By comparison, Sanders et al. [61] and Hangody and Fules [9] reported no cases of osteochondral necrosis in their patient groups. Fig. 3Coronal STIR images 12 weeks (a), 52 weeks (b) and 2 years (c) after osteochondral transplantation show gradual resolution of severe bone marrow oedema in and around the grafts as bony incorporation occurs When the cartilage layer of the osteochondral cylinder was examined, the cartilage signal intensity of the graft was similar to the surrounding cartilage in the vast majority of cases, 86% [59]. These findings are supported by Sanders et al. [61], who found that the graft cartilage signal intensity was similar to adjacent cartilage in most cases. Additionally they found similar rates of graft and perifocal graft oedema. Graft and adjacent bone Subchondral bone marrow oedema is often present in the early post-operative phase but usually resolves as the graft incorporates into the subchondral bone. A normal fatty marrow signal is seen within and around the plugs when solid bony incorporation occurs (Fig. 4). Herber et al. [60] examined ten patients at 3, 6 and 12 months using indirect MR arthrography and noted a high rate of early post-operative subchondral marrow oedema, which settled in most cases by 12 months. Fig. 4a, bSagittal T1-weighted SE images of the normal development of autologous osteochondral transplants. a Marked oedema in and around the osteochondral plugs at the recipient site 12 weeks (arrows) after surgery and b bony incorporation of the grafts with fatty bone marrow in and around the grafts (arrows) and filling of the donor site with cancellous bone after 2 years Normal findings The results of these earlier follow-up studies have helped to define normal and abnormal MR findings following osteochondral autografting and help in the identification of possible complications. “Normal” MR findings associated with after osteochondral autografting include bone marrow oedema in and around the grafts in approximately 50% of the subjects during the first 12 months, with a gradual reduction thereafter. However persistent oedema may be seen in a small number of cases for up to 3 years post-operatively. Joint effusion and synovitis appear to follow a similar trend. Incongruities at the bone-bone interface occur frequently, while incongruities at the cartilage-cartilage interface occur in approximately 15% of grafts. The sometimes substantial incongruities of the bone-bone interface should not be interpreted as an abnormal finding or complication. Differences occur because the cartilage thickness of the donor site commonly differs in thickness from the implant site, but the cylinder is inserted to a depth to ensure a smooth cartilage surface. Abnormal findings Complications which can be determined by MRI include graft loosening or migration, incongruencies of the cartilage-cartilage interface, significant gaps between osteochondral plugs and adjacent native cartilage and partial or complete necroses of the grafts (Figs. 5, 6). However, it should be noted that over time fibrocartilagenous tissue fills the gaps between osteochondral plugs and adjacent native cartilage thereby improving surface congruity. Fig. 5Incongruity at the cartilage-cartilage interface on a medial femoral condyle 12 months after osteochondral autograftingFig. 6Osteonecrosis of the osteochondral autograft 2 years after surgery shown by marked hypointense signal alteration of the graft on T1-weighted SE image in the sagittal plane (arrows) Future studies in this field should provide better understanding of the pathophysiology of transplanted hyaline cartilage and its function, which is important for the long-term prognosis of these patients. MR findings following ACI and MACT Assessment and interpretation of MR examinations for ACI and MACT patients should be performed in a systematic fashion. Careful attention should be paid to the degree of defect filling, the integration of the graft to adjacent cartilage and underlying bone, the graft’s internal structure and surface, its signal intensity and any changes in the subchondral bone. Last but not least, the presence of adhesions to the graft or joint effusion should be evaluated. Defect filling One of the major goals of ACI and MACT is to ensure the graft cartilage has the same thickness as the adjacent native cartilage in order to restore the smooth contour of the articular cartilage surface (Fig. 7). When evaluating defect fill MRI has been shown to be effective in detecting cases of incomplete filling, either focally or globally [20, 29, 34, 36]. Fig. 7A sagittal FSE image of a stable cartilage implant at 2 years after MACT surgery shows complete filling of the defect (arrows mark the borders of the implant) Longitudinal MRI follow-up of ACI patients has shown that a consistent volume of repair tissue can be visualised after 3 months, which remains stable for at least 2 years post-operatively [62]. Tins et al. [63] demonstrated that 63% of patients in their series had normal cartilage thickness compared with native articular cartilage. Roberts et al. [37] also found a similar percentage of grafts to have normal cartilage thickness compared with adjacent native cartilage. In a different study with a 2-year MR follow-up [64], complete filling of the defect by repair tissue was seen in 65.2% of patients. Graft hypertrophy is often asymptomatic, but may produce pain and catching. It usually occurs between 3 and 7 months and has been reported to complicate between 10% and 39% of cases [15, 18, 65]. Graft hypertrophy is seen on MRI as the ACI graft protruding above the level of the native articular cartilage and may involve part or the full width of the graft (Fig. 8). It is important to note that hypertrophy of grafts close to the intercondylar notch may cause impingement on the anterior cruciate ligament. Treatment consists of arthroscopic debridement of the hypertrophied tissue. Fig. 8Severe hypertrophy of cartilage implant 104 weeks after MACT surgery on a sagittal T2-weighted FSE image The reported incidence of graft hypertrophy following MACT is lower than with the classical ACI with periosteal flap technique [22, 66]. We found an incidence of 20% graft hypertrophy following MACT surgery [64], which is similar to the hypertrophy rate following classical ACI surgery [18, 67]. Interestingly, however, in three out of four cases the graft hypertrophy resolved, with the cartilage thickness returning to the level of adjacent cartilage within one year. This was hypothesised to be due to increasing weight bearing during rehabilitation and subsequent remodelling of the repair site. Henderson et al. [19] analysed 81 lesions in 58 knees 12 months after ACI and reported that MRI showed 81.6% of the lesions had normal or nearly normal cartilage at the site of repair. These investigators analysed four categories, including the filling and the signal intensity of the repair site, as well as the presence and severity of bone-marrow oedema and effusion. A complete fill of the repair site was found in 79%, 13.6% had >50% filling, and 2.5% had <50% filling. By comparison, we found filling to the level of adjacent cartilage in 65.2% and underfilling in 17.4% in a 2-year follow-up of MACT patients using Hyalograft C [64]. Integration The interface between ACI and native cartilage should be indiscernible, a fluid-like split, in particular a broad split or one that extends beneath the base of the ACI has been described as pathological [20, 34, 36] (Fig. 9). Poor graft integration can be identified on high-resolution sequences by fluid signal clefts or ill-defined high signal intensity at the interface between the graft tissue and native cartilage [15, 18, 20, 34, 36, 67]. Fig. 9A split-like integration defect, seen on a sagittal FSE image, 52 weeks after surgery. Arrows mark the borders of the implant, the right-hand arrow points to the defect The clinical importance of a full-thickness fissure or a broader gap, suggesting incomplete integration, is still unclear. Potentially, a gap between the native cartilage and the edge of the graft may act as a focal point for cartilage wear [36]. We found integration to be complete in 18 out of 23 knees after 2 years [64]. One patient had incomplete integration with a split-like fissure between the graft and adjacent cartilage. Four patients showed incomplete integration with a slightly broader gap between the graft and adjacent cartilage (17.4%). Poor integration of the ACI repair tissue to the bone or to the adjacent native cartilage may result in delamination of the graft from the underlying bone (Fig. 10). On MRI, a delaminated graft may appear as a loose body in the joint if it has dislocated, or if still in situ at the repair site, a thin rim of fluid between the base of the graft and the subchondral bone plate, resembling a cartilage flap, may be seen [20, 34]. Clinically, patients may complain of pain, swelling or locking. This complication may occur in between 5% and 14% of cases [15, 18]. Fig. 10A sagittal FSE image of incomplete delamination of the repair tissue 24 weeks after MACT surgery. Fluid partially demarcates the bone interface Structure and surface The articular surface of the ACI or MACT site should appear smooth and be continuous with the adjacent native cartilage. Irregularities of the graft surface on MRI have been described previously and seem to be relatively common [19, 20, 34, 68, 69]. This is in keeping with findings during follow-up arthroscopy [70]. In some patients, we have seen a transition from an initially irregular surface to a regular smooth surface over time (Fig. 11). We believe this may represent continuous organisation of the graft as it matures. In contrast, the development of surface defects over time should be considered as abnormal [62]. Fig. 11Gradual integration and normalisation of the graft surface and internal structure after MACT surgery from (a) 12 weeks to (b) 24 weeks on sagittal T2-FSE image The internal structure of normal hyaline cartilage usually has a trilaminar appearance on most sequences [71]. Although the widths of these signal bands seen on MRI did not correlate with the thickness of histological layers and were affected by chemical shift and the magic angle effect, they represent a homogeneous layering present in normal collagen architecture. Regarding the internal structure of the graft we found only eight out of 23 patients had homogeneous repair tissue at their first follow-up scan. In a further eight patients the repair became homogeneous with signal layering during the follow-up period [62]. This most likely represents a normal maturation process, whereas the reverse process with repair tissue going from a homogeneous to inhomogeneous appearance over time may be considered abnormal. Signal intensity In addition to a smooth surface and homogeneous signal layering, the signal intensity of repair tissue should resemble normal hyaline cartilage. The post-operative signal characteristics seen after ACI surgery are clearly affected by the choice of sequence used. In the uncomplicated case, there is a gradual change in the signal characteristics of the repair tissue over time to resemble those of normal articular cartilage [62, 72] (Fig. 12). Typically, when a FSE sequence is used the signal intensity of the ACI/MACT repair tissue steadily decreases with time. By contrast the signal intensity steadily increases over time with fat-suppressed T1-weighted GRE sequences [19, 62, 65]. Fig. 12a–eChanges of implant signal intensity, on PD and T2-FSE images following MACT. a, b Fluid-like signal after 4 weeks; c, d hypointensity at 24 weeks and e isointensity with native hyaline cartilage after 52 weeks In the early post-operative period (4 weeks) the graft may have a fluid-like appearance, which may be misinterpreted as complete delamination. However, on high-resolution imaging the surface of the implant is seen as a thin dark line, this feature is more commonly seen with classical ACI [34]. Several report show that at approximately 12 months post-operatively the signal intensity of the ACI graft on both FSE and fat-suppressed GRE sequences resembles native hyaline cartilage and can no longer be differentiated from it [19, 20, 34, 68, 69]. By contrast, Brown et al. [17] found that a significant proportion of grafts (69%) remained hyperintense on FSE sequences more than 18 months after surgery. In clinical follow-up studies, Henderson et al. [19] reported signals identical to the adjacent articular cartilage in 63% of patients following classical ACI grafting and nearly normal signals in 29.6% at 12 months. In our group of MACT patients, we found all grafts evaluated at 24 months were isointense [64]. These differences may be attributable to the different evaluation times and the further maturation of the graft over the additional 12-month period in our study. Subchondral lamina and bone The subchondral lamina is normally not violated during the ACI or MACT procedure and therefore should be intact. However, cartilage repair for cases of osteochondritis dissecans or involving partial removal of subchondral sclerosis may be associated with defects of the cortical endplate. An intact cortical endplate which becomes damaged during follow-up may be the result of over-use or recurrent trauma [62]. The bone contour underneath the graft may show the presence of central osteophytes, defects, and irregularities. When central osteophyte formation occurs beneath a graft the surface of the overlying cartilage is usually smooth and level with adjacent cartilage. The importance of central osteophytes has still to be determined [63]. Subchondral oedema commonly occurs at the repair site in the early postoperative period [17, 19, 29, 34, 63, 68], and has been described as part of the normal healing process during the first 3 months following surgery [20, 34]. Abnormal loads transmitted to the bone have also been suggested to account for the marrow oedema [73]. The presence of oedema-like marrow signal beyond 12 months or an increase in intensity of oedema should be considered abnormal and requires close clinical follow-up [20, 34]. Possible reasons for persistent or reappearing subchondral bone marrow oedema include: over-use, an abnormality of the leg axis and new trauma. Conversely, it has been reported that the persistence of oedema-like signal intensity is a sign of yet undetermined importance [34]. For the diagnosis of bone marrow oedema, a fat-suppressed sequence such as STIR or fat-suppressed FSE is required. Cystic changes in the subchondral bone underneath the cartilage implant have been described and indicate a problem with the graft which requires close clinical follow-up [19]. Cyst formation has been associated with oedema-like signal intensity in 10% of cases with these cases also demonstrating a fibrocartilage appearance rather than hyaline-like articular cartilage [63]. In our series of 23 patients who underwent MACT surgery, we found 13% had an oedema-like signal in the subchondral bone at 2 years [64]. A comparison of ACI and MACT patient groups [19, 64] showed the rate of bone marrow oedema and effusion was similar, approximately 40%, with both techniques. Adhesions Adhesions are demonstrated on MRI as bands of intermediate to low signal intensity tissue traversing the joint and demonstrating contact to the repair tissue (Fig. 13). Adhesions most commonly connect to the infra-patellar fat pad, suprapatellar pouch and parapatellar recesses [15, 18, 67]. Knee stiffness from intraarticular adhesions requiring arthroscopic release has been reported in up to 10% of ACI patients [15, 67, 73]. Patients who have undergone extensive cartilage repair or have multiple grafts appear to be more at risk [18]. Adhesion to the graft surface may lead to graft tearing or dislocation as the patients activity level increases. Fig. 13An adhesion seen as a thin band-like structure, running from the tibial articular surface to the cartilage implant on a sagittal T2-FSE image. Arrows mark the borders of the implant Effusion Post-operative reactive synovitis should be considered as a possible cause of pain in this patient group. For the evaluation of synovitis intravenous contrast media application is usually required. Synovitis is generally accompanied by effusion and clinical signs of swelling of the affected joint. The majority of cases with synovitis in the early postoperative period resolved during the follow-up, which can be explained by its reactive nature [64]. In addition to the cartilage repair surgery itself other causes of synovitis have to be considered, such as native cartilage defects or meniscal lesions. Maturation of cartilage repair tissue Serial follow-up MRI scans of MACT patients show that the cartilage repair is a dynamic process that can be non-invasively monitored [62]. High-resolution MRI examinations at 4, 12, 24, 52 and 104 weeks post-operatively revealed characteristic changes in the repair tissue over time, which we believe represent the normal maturation process of the repair tissue. The most significant features were: first, early filling defects showed progressive filling by 6–12 months (Fig. 14); second, initial graft hypertrophy seen at 3 months resolved by 6 months; third, small surface defects became smooth over time. Finally, signal intensity gradually changes over time from fluid-like appearance in the early post-operative stage to iso-intensity with surrounding native hyaline cartilage by 6–12 months. One may conclude that inverse developments are associated with a poor prognosis. Fig. 14a, bProgressive defect filling at the repair site. After matrix-based autologous chondrocyte implantation, significant improvement of filling of the defect at the repair site from (a) 4 weeks to (b) 24 weeks post-operatively, depicted on sagittal FSE images Histology In a histological analysis of follow-up biopsies after ACI surgery, 57% of patients after ACI demonstrated articular cartilage [74]. However, it is generally hypothesised that continuous remodelling of the graft occurs with the transplant becoming more like hyaline cartilage. To our knowledge only one study has attempted to correlate MRI findings with ACI graft histological findings [63]. Tins et al. [63] studied 41 patients at 1 year following ACI grafting on the femoral condyle. The histological appearance of the graft was classified into one of the four morphological categories recommended by the ICRS. MRI findings were correlated to the histology findings. The authors found no relationship between any of the MRI features assessed and the histological appearance of the cartilage repair tissue; however, this is unsurprising given the sequences used. Studies of the correlation between histology and the dGEMRIC index and/or T2 maps are a very interesting area for future studies. Functional outcome Marlovits et al. [58] recently described the statistical correlation of clinical outcome scores with the radiological variables of the MOCART scoring system. For the variable “filling of the defect” a statistically significant correlation with all KOOS (knee injury and osteoarthritis outcome score) variables and also with the VAS (visual analogue score) was observed. For the variable “integration to border zone” and “surface of the repair tissue” no statistically significant correlation was found. The variable “structure of the repair tissue” showed a statistically significant correlation with the VAS and nearly all KOOS group except for the variable “symptoms”. For the variables “adhesion”, “subchondral lamina”, and “effusion” no statistically significant correlation with the clinical scores was found. In contrast the variable “subchondral bone” showed a statistically significant correlation with the VAS and nearly all KOOS group except for the variable “symptoms”. For the signal intensities a statistically significant correlation was found with the KOOS variables “symptoms”, “sport”, and Activities of daily Living (ADL) function. Recommendations From reviewing our studies and those of other centres we recommend that follow-up MR studies should be performed at 3 months and 1 year. The initial imaging at 3 months allows the volume and adherence of repair tissue to be assessed. Imaging at 1 year demonstrates the maturation of the graft and allows complications to be identified both non-invasively and at a sufficiently early stage. Conclusion In the near future as the use of clinical high-field (3 Tesla) systems with modern multi-element coil configurations becomes more widespread and new high-resolution isotropic 3D sequences are utilised a further improvement in the morphological analysis of cartilage implants can be expected. Moreover, advanced cartilage imaging techniques which allow the biochemical composition of cartilage to be studied will be possible in vivo (Fig. 15). This is particularly promising for evaluating the maturation of the graft and whether or not hyaline cartilage has developed. The ability to non-invasively assess graft maturity will help to define the optimal postoperative rehabilitation and to detect the early stages of graft failure. Fig. 15a, bA dGEMRIC image of a matrix-associated ACT 2 years after surgery. a The cartilage layer of the graft shows different T1 values, representing proteoglycan concentration, compared with hyaline cartilage. b a 3D-GRE image of the same patient, which shows morphology of cartilage implant with hypointense signal alteration of the cartilage implant in comparison with normal hyaline cartilage

Anal_Bioanal_Chem-4-1-2226005

Selected isotope ratio measurements of light metallic elements (Li, Mg, Ca, and Cu) by multiple collector ICP-MS

The unique capabilities of multiple collector inductively coupled mass spectrometry (MC-ICP-MS) for high precision isotope ratio measurements in light elements as Li, Mg, Ca, and Cu are reviewed in this paper. These elements have been intensively studied at the Geological Survey of Israel (GSI) and other laboratories over the past few years, and the methods used to obtain high precision isotope analyses are discussed in detail. The scientific study of isotopic fractionation of these elements is significant for achieving a better understanding of geochemical and biochemical processes in nature and the environment. Introduction The development of multiple collector inductively coupled mass spectrometry (MC-ICP-MS) about 15 years ago immediately indicated the advantages of this technique, at least in ratio measurements from mass 80 and above. In general, the acquired data showed precision equal or better to the well-established thermal ionization mass spectrometry (TIMS) [1, 2]. The ease of data acquisition was remarkable, the ionization efficiency was significantly larger, and consequently the sensitivity was improved. When MC-ICP-MS was introduced it had to compete with TIMS, which at that time was fully developed and provided the best isotope ratio data for metallic elements in terms of precision and accuracy. Even in the stages of application tests MC-ICP-MS exhibited several advantages: high precision, high reproducibility, high analytical throughput, simple sample preparation, and, as mentioned above, improved sensitivity [3]. MC-ICP-MS requires a more complicated instrument compared to TIMS. It uses an ion source at atmospheric pressure and therefore requires a specially designed ion beam introduction inlet system equipped with efficient pumping systems to reduce the argon gas pressure to 10−7−10−8 mbar. Further pumping systems along the ion flight tube maintain the vacuum at 2 × 10−9 mbar. The ion beam emerging from the interface has a circular profile which requires a tunable quadrupole lens, efficiently changing it to a rectangular shape. This rectangular ion beam possesses ions with an energy spread of up to 30 eV. An electrostatic energy filter is used to separate an almost monoenergetic ion beam acceptable for the magnetic sector mass separator. A further inherent problem in an ICP-MS is the relatively high instability of the plasma, observed as fluctuations and drifts of the ion beam. A single collector system, whether a Faraday cup or an electron multiplier, is too slow to follow this instability; therefore an array of Faraday cups (and also a Daly electron multiplier) is mounted at the end of the instrument, allowing the simultaneous monitoring of two or more ion beam intensities. This technique had already been applied in the multiple collector TIMS instruments. Plasma 54 (P54), the first MC-ICP-MS instrument manufactured by VG Elemental in the UK [1], contained an ion collector equipped with seven Faraday cups and an analog Daly detector. A unique feature of all double focusing instruments is the wide flat-topped ion peak shape, which is of primary importance for high precision data acquisition. Two further modifications were applied to this instrument: a wide-end flight tube to allow simultaneous measurement of ions with large mass differences, such as U–Pb or 6Li–7Li, and the addition of an electrostatic energy filter before the Daly detector to enhance abundance sensitivity measurements. Isotope ratio measurements using laser ablation have also been demonstrated [4]. Several years later VG Elemental-Thermo Elemental introduced the Axiom, a smaller double focusing instrument of vertical configuration, providing high mass resolution capability. This mass spectrometer was aimed for high precision, fast analytical work [5]. Several other MC-ICP-MS instruments were manufactured by other companies applying different approaches for efficient reduction of ionic spectral interferences formed by argon and atmospheric gases, e.g., thermalizing the ions produced by the plasma, introducing a detection system comprising a fixed array of Faraday cups and ion counters, and including an adjustable ion beam dispersion device. Micromass in the UK introduced the IsoProbe MC-ICP-MS where an off-axis hexapole collision cell is mounted in a jacket between the second interface cone and a beam focusing system [5]. The cell is flushed with a low pressure inert gas which reduces the ion energy to less than 1 eV. Also spectral interferences such as ArC+, ArN+, ArO+, ArCl+, and dimers produced in the atmospheric plasma, which interfere with 52Cr+, 54Fe+, 56Fe+, 75As+, and 80Se+, respectively, are removed by collision-induced cleavage of the Ar–X bonds, allowing measurement isotope ratios at masses below 80. The remainder of the instrument comprises a magnetic mass separator and a multiple collector detection system. Nu Instruments in the UK introduced the Nu Plasma MC-ICP-MS. This machine, as with the Plasma 54, was developed by P.A. Freedman who maintained the same basic concepts, but introduced a different design for almost all the modules. Essentially it is a forward geometry (electrostatic sector followed by magnetic sector), double focusing mass spectrometer with a C-shape configuration compared to the S-shape of the P54. The ion beam profile conversion system is of different construction and the electrostatic energy filter and the magnetic separator were changed in shape and size. The profound advance in this design is the multiple collector detection system. The detection module comprises an array of 15 fixed detectors, 12 Faraday cups, and three electron multipliers, in front of which a zoom lens system deflects each separated ion beam into a chosen detector. Also, one of the multipliers is equipped with a filter for cutting off the tail from highly abundant isotopes. A later model from Nu Instruments, the Nu1700, is a large geometry, high resolution MC-ICP-MS [6] that allows one to overcome most of the interferences. Finnigan MAT in Germany introduced the Neptune MC-ICP-MS, a double focusing C-shape instrument with a large magnet providing 16% mass dispersion, movable collector array, and high resolution capability. Flat-topped peaks are achievable at R > 4,000 [7]. When discussing isotope ratio measurements utilizing plasma ionization, several further phenomena should be mentioned. Isotopic fractionation is of utmost importance. It is caused by repulsive forces in the intensive positive ion beam emerging from the plasma and supersonic ion expansion through the sample cone. Both effects yield radial repulsion of the lighter isotope from the beam center, i.e., increasing the heavy mass over light mass ratio. The fractionation effect is inversely mass dependent from several per mils in uranium to more than 10% in lithium. It is constant in time, since fresh sample solution is continuously aspirated into the ion source. This is in contrast to TIMS ionization, where a fixed sample is used, permanently changing in composition as the lighter isotope is preferentially vaporized. Fractionation may be corrected in one of three techniques: (1) internal and (2) external normalization, including double spike method; and (3) “standard-sample-standard bracketing”; consecutive measurements of the same ratio in a sample and standard. Further details will be given when describing the ratio measurements in the elements discussed in this review. ICP mass spectrometry is subject to other interferences originating from various sources. Spectral interferences are products of interactions between the carrier gas and atmospheric gases or the solvent molecules. As mentioned above they obscure the lower mass range. Sodium ions in the sample solution in the case of copper analysis may interfere as ArNa+ at mass 63. Further cases are molecular interferences such as oxides, nitrides, and hydrides of trace elements in the solution and isobaric ion interferences. These cases can be removed by using expensive high resolution MS. Other ways to reduce interferences include using of desolvation nebulizer and chromatographic separation of the analyte from the matrix prior to introduction into the MS. This separation is also necessary to prevent the additional effects of mass fractionation connected with the matrix effect. In the case of the heavy isotope ratio measurements, this fractionation is corrected by internal or external normalization, but in the case of standard-sample-standard bracketing, generally used for light masses correction, the analyte must be carefully and precisely separated from the matrix. As with the stable isotopes of the light elements, the isotopic composition of a sample is given in δ units (‰) relative to standard (I is isotope, x and y are mass numbers). Generally, the isotope variations are derived from bracketing the measured sample ratio with the mean ratios of a standard measured before (std1) and after the sample run (std2), and are presented as deviations in parts per 1,000 of the measured ratio from that of the standard (normalizing) ratio: Results and discussion Lithium Lithium isotopes are of significant importance in a number of fields, such as geochemistry [8–11], astrophysics [12–13], nuclear technology [14], and biomedicine [15]. MC-ICP-MS opened new frontiers in Li isotope measurements by providing high precision data (<0.3‰, 2σ). Lithium isotopes fractionate during hydrothermal processes, and significant variations may be observed in 7Li/6Li ratios in water derived from marine sedimentary rocks and from hydrothermal altered igneous rocks, thereby providing valuable information regarding regional ground-water flow paths. Lithium isotopes are a powerful tracer of recycling processes in the Earth. This is particularly the case for understanding the geochemical evolution of the Earth’s mantle, because (a) Li is a moderately incompatible constituent of minerals in peridotite, (b) Li isotope ratios show large variations in the terrestrial system, caused by low-temperature fractionation and mixing, (c) Li is a fluid-mobile element. Therefore, combined with existing geochemical information, its isotopes could be powerful geochemical tracers, especially for fluid-related (metasomatic) mantle processes. In recent years, almost all lithium isotope studies dealing with geological and related systems have applied the MC-ICP-MS measurement technique. Lithium isotope ratio measurements require careful separation of Li from the matrix in natural samples. For solid samples, digestion and ion exchange chromatography separation are used in general; liquid samples are processed by chromatographic separation [16–21]. Acid leaching [22, 23] and direct resin techniques [24] are also applied. The most rapid method of separation was published by Hall et al. [21] for TIMS and modified at the Geological Survey of Israel for MC-ICP-MS. Any traces of Na, K, Mg, Sr, and Rb were detected using this modified method, so the samples were sufficiently pure for ratio measurements by Nu Plasma MC-ICP-MS. The values of the ratio in standard L-SVEC NBS (NIST, Li2CO3) are [25], and in seawater [26]. A typical analysis sequence consists of blank, standard, blank, sample 1, blank, standard, blank, sample 2, and so on. Lithium ratio measurements exhibit time-dependent drifts and random shifts. Tomascak et al. [16] reported a typical 7Li/6Li ratio drift for the L-SVEC standard from 12.7 to 12.3 during an 8-h period. Nishio and Nakai [17] rejected data when two successive measurements of the bracketing standard ratio shifted by more than 2‰. These authors also reported a stability of (2σ) for an in-house isotope standard measured over an 8-month period. It is good practice to carry out the sample and standard measurements under identical instrumental and solution conditions. The analyte concentrations in both solutions should be as close as possible and the impurity levels identical and as low as possible. It should be noted that sample purity in TIMS analysis relative to MC-ICP-MS is of major importance, since the large isotopic fractionation in Li is highly sensitive to impurities in the loaded sample, especially when Na and K are interfering elements [21]. It was reported that a Na/Li ratio greater than approximately 5 may cause unstable instrumental fractionation [16]. Moriguti and Nakamura [27] developed a four-step chromatographic separation for Li purification, allowing precise TIMS Li ratio analyses. Bouman et al. [28] studied a wide range of Li samples originating from islands in the Pacific, Atlantic, and Indian Oceans. Thirty six samples and three standard materials were analyzed. The precision of these measurement are 0.01–0.5(‰), 0.6–1.0(‰), and 1.2–1.9(‰) for 22, ten, and four samples, respectively. The data for standard materials are presented in Table 1, which also summarizes selected MC-ICP-MS Li isotope ratio measurements in geological samples. Table 1Li isotopic composition in standards and selected natural samples relative to NIST NBS L-SVEC (‰)Sampleδ7Li±2σReferenceJR-2 rock standard3.840.18[24]Magmatic arc lavas (Kurile)[18]8322/3 Onekotan4.2<1.0K33 Keli-Mutu5.1<1.0VB30 granite−1.41.0[19]MG20 granite0.81.0Inorganic calcite CM019−7.60.6[20]Coral Acropora21.00.4Indian Ocean water33.01.2[21]Atlantic Ocean water32.11.2Foraminifera Orbulina universa28.41.6Inorganically grown carbonates[23]Aragonite (salinity 10 psu)−10.90.8Calcite (salinity 50 psu)−1.90.8Seawater29.70.4[28]BHVO-1 rock standard5.01.5Seawater 680 W30.70.4[29]Mediterranean Sea30.590.26[30]Red Sea30.490.12Dead Sea28.780.11Yarkon spring15.140.21SC -1 olivine3.41.0[31]Ia/211 clinopyroxene−2.41.0M 1 saprolite−11.61[32]10 saprolite0.21SH 65 Ocean Island basalt7.0<1[35]KRS9806 xenolith, Japan−7.70.83[36]9708 xenolith, Australia6.00.83Zagami meteorite4.40.5[38]15495 lunar low-Ti mare basalt5.60.214-1 amphibolite0.91[39]12-2 amphibolite−14.21 Bryant et al. [29] studied Li isotope ratio measurements by MC-ICP-MS under “cold plasma” conditions. The results are characterized by fewer baseline interferences and improved reproducibility as compared with conventional hot plasma techniques. The 2σ precisions for 1,200-W, 800-W, and 680-W plasma energy are conservatively estimated as 1.1‰, 0.7‰, and 0.5‰, respectively. The effects of analyte, acid, and matrix concentrations were discussed. Seitz et al. [31] determined the Li isotopic composition of coexisting olivine, clinopyroxene, and orthopyroxene from spinel- and garnet-bearing peridotite xenoliths. The degree of intramineral fractionation correlated negatively with equilibration temperature. The lithium isotopic composition of saprolites developed on a granite and diabase dike from South Carolina was measured to document their behavior during continental weathering. A general trend of decreasing δ7Li with increasing weathering intensity was observed [32, 33]. A similar study was carried out by Kısakürek et al. [34]. The internal precision on 7Li/6Li measurements was usually less than 0.20‰, and the external precision of Nu Plasma MC-ICP-MS, was 0.8‰. Ryan et al. [35] defined lithium isotope variations in mantle sources (volcanic rocks from the Antarctica oceanic islands), and Nishio et al. defined lithium isotope variations in mantle-derived xenoliths [36]. These isotopic measurements have provided much information on the mantle source region. Similarly, Elliott et al. [37] showed that Li isotopes promise to provide significant new constraints on the distribution of recycled material in the mantle and its implications for mantle convection. Lithium isotope compositions and concentrations of lunar samples, including basalts, breccias, and glass, and martian meteorites were measured using Thermo Finnigan MC-ICP-MS [38]. Teng et al. [39] measured lithium isotopic compositions in the country-rock amphibolites and schists using Nu Plasma MC-ICP-MS. δ7Li decreased dramatically with contact distance (along a −10 m traverse from the pegmatite into amphibolite) from +7.6 to −19.9. Pogge von Strandmann et al. [40] investigated the behavior of Li isotopes in glaciated basaltic terrain. The δ7Li value of the suspended load was always lower than that of the bedload due to preferential retention of 6Li in secondary minerals during weathering. In turn, the δ7Li value of the dissolved load was always greater than that of the bedload. The authors suggested that δ7Li decreases with increasing chemical weathering. Similarly, Hathorne and James [41] used Li isotopes in seawater as a tracer for silicate weathering and showed that between 16 and approximately 8 Ma silicate weathering rate has increased, while weathering intensity has decreased. Nishio et al. [42] studied Li isotope ratios in North Atlantic and Indian Ocean waters and its relation to those of Sr and Nd. The Li isotope results support the recent proposal that significant amounts of recycled lower continental crust might produce the radiogenic isotope signatures of the Indian Ocean basalts. Rudnick and Ionov [43] examined Li isotopic disequilibrium in olivine and clinopyroxenes from far-east Russia. Jeffcoate et al. [44] studied Li isotope fractionation in peridotites and mafic melts. Their results highlight the potential of Li isotopes as a high resolution geospeedometer of the final phases of magmatic activity and cooling. Wunder et al. [45] studied Li isotope fractionation between Li-bearing staurolite, Li-mica, and aqueous fluids. Magnesium Magnesium has three naturally occurring isotopes at mass numbers 24, 25, and 26 with relative abundance of 78.99, 10.00, and 11.01%, respectively. Galy et al. [46] briefly summarized the various fields where natural variations in the isotopic composition of this element may arise: (1) stellar nucleosynthesis and incorporation of presolar grains into meteorites, (2) the decay of 26Al to 26Mg, (3) isotopic fractionation in volatilization/condensation reactions, (4) isotopic fractionation during low temperature fluid/rock interactions, and (5) kinetic and thermodynamic isotope effects in biological processes. Galy et al. [46] performed high precision magnesium isotope ratio measurements using MC-ICP-MS for three groups of materials: commercial Mg materials including the NIST isotopic standard SRM 980, natural magnesite, and commercial chlorophyll in spinach. The observed ratio 26Mg/24Mg was 0.139828 ± 0.000037 (2σ), compared to the NIST value of 0.13932 ± 0.000 6 (2σ) [47]. The variations in the Mg materials reported in δ25Mg and δ26Mg units are given in Table 2. A desolvation nebulizer was used to minimize the introduction of H2O, CO2, O2, and N2, therefore reducing the presence in the plasma of molecular interferences such as , C2H+, , CN+, and NaH+. Other possible interferences are doubly charged ions such as 48Ca2+, 48Ti2+, 50Ti2+, 50V2+, 50Cr2+, and 52Cr2+. These elements, if present, should be removed by purification. Table 2Mg isotopic composition in standards and selected natural samples relative to DSM-3 (‰)Sampleδ26Mg±2σδ25Mg±2σReferenceAldrich Mg solution2.600.171.330.08[46]aDead Sea Mg metal3.960.152.030.08AG 177 magnesia2.030.041.030.01OUM 10988 magnesite, Italy1.220.040.600.03AG 27 chlorophill b, spinach1.060.070.540.03North Atlantic Sea water2.590.041.330.08[51]aMixed foraminifera−1.880.16−0.990.10Dolomite1.640.040.880.01Seawater[52]aEPR1 surface1.950.140.980.06WP45N 5800 m2.010.161.000.04Med-T1 surface2.020.071.050.09Hydrothermal 2,650 m−1.010.16−0.510.06River water MB160.760.460.380.22MB60.090.220.040.11PH60.220.060.110.06PH5−1.010.16−0.510.14Ganges−1.390.06−0.70.09[53]Amazon−1.030.07−0.530.04Lena−1.280.08−0.660.02Seawater−0.840.130.430.15[54]MT 66 solute−1.740.07−0.900.07Ett113 bulk silicate rock−0.420.03−0.210.02M 201 biotite−0.070.09−0.010.00MO33 100–110 soil0.020.010.010.01Ace78 travertine−4.010.11−2.060.07Dead Sea water−0.600.08−0.330.10[55]Seawater−0.740.05−0.360.12Sataf spring−2.500.10−1.300.10Carbonaceous chondrites[56]144A, Al-Ti diopside0.110.06−0.090.12144A, Al-Ti diopside−2.740.11−1.740.11Olivines[57]DR9894 Australia−1.430.2−0.660.13ZS56-2 Siberia−1.050.2−0.540.13aRelative to NIST isotopic standard SRM 980 Variations of the Mg isotope ratio in the metallic chips of the NIST SRM 980 magnesium isotopic standard were shown using five different MC-ICP-MS instruments of two types [48]. The chips were 1–50 mg in size. The differences in δ25Mg and δ26Mg of the SRM 980 were up to 4.2 and 8.19‰, respectively, while the long-term repeatability of the δ values was up to 0.09 and 0.16‰ respectively. Because of the heterogeneity of the NIST reference material, two homogeneous isotope standards, DSM-3 (Dead Sea magnesium) and Cambridge1 were prepared and characterized. The heterogeneity of Mg isotopes was also reported by Zhu et al. [49]. Carignan et al. [50] discussed the isotopic homogeneity of existing reference materials and suggested the acceptance of DMS-3 as a new Mg isotope reference standard material. Chang et al. [51] developed a Mg separation technique for low-Mg biogenic carbonates with yields close to 100%. The technique was applied to the determination of Mg isotopes in three natural samples: seawater, foraminifera, and dolomite. De Villiers et al. [52] established the magnesium isotopic composition of seawater and evaluated its constancy as a function of depth and geographic location. The authors demonstrated that the magnesium isotopic composition of ancient oceans can be used to make important inferences about the relative contribution of different lithologies to the global continental weathering flux. Tipper et al. [53, 54] analyzed river water, rock, travertine, and soil and demonstrated that both Ca and Mg isotope ratios are fractionated during weathering. The Mg isotope composition of the rivers is intermediate between limestone and silicate rock. Silicate soil has a δ26Mg of −0.03‰, heavier than that of silicate rock by 0.5‰. This fractionation in the soil creates a complementary groundwater reservoir of light Mg. Seasonal variations in Mg isotope ratios in the dissolved load are small, but define an array which can be modeled as a mixture between a fractionated groundwater reservoir and surface runoff. Mg isotope fractionation during brine evolution in the Dead Sea is presently being studied using the MC-ICP-MS technique [55]. Samples from the solar system were used for 26Al−26Mg dating, by measuring the variations in the Mg ratios using laser ablation combined with MC-ICP-MS [56]. Pearson et al. [57] investigated Mg isotopic variations in the lithospheric mantle by analyzing olivine in mantle-derived peridotite xenoliths and megacrysts using a laser-ablation microprobe and MC-ICP-MS. δ26Mg ranges from −3.01 to +1.03‰ and δ25Mg from −1.59 to +0.51‰, relative to the magnesium isotopic standard DSM-3, were found. The in situ measurement of Mg isotopes thus provides a powerful new method for investigating processes in the mantle. Calcium Calcium has six stable isotopes at mass numbers 40, 42, 43, 44, 46, and 48 with relative abundance of 96.941, 0.647, 0.135, 2.086, 0.004, and 0.187%, respectively. Halicz et al. [58] studied in detail the MC-ICP-MS isotope ratio measurements of calcium. The more important points are discussed here. Under normal instrumental operation conditions it is not possible to use the isotope 40Ca because it is masked by the intense 40Ar+ ion beam. Furthermore, the ion dispersion of Ca isotopes is too large to allow the 48Ca−42Ca mass range to be accommodated on the multicollector array; therefore only 42Ca, 43Ca, and 44Ca are monitored. Ar-scattered interferences were monitored at half masses for elevated background correction. Corrections of doubly charged Sr were made by measuring doubly charged 87Sr2+. A desolvation nebulizer was used to reduce molecular interferences such as , , and . The results for the Ca ratios in the studied sample are given in δ units in Table 3 and were derived using the bracketing technique relative to the NIST SRM 915a Ca standard. Table 3Ca isotopic composition in standards and selected natural samples relative to NIST SRM 915a (‰)Sampleδ44Ca±2σReferenceCaCO3 Merck0.610.28[58]2-8-E3 speleothem0.250.32Acropora coral0.580.16Shell from marine organism[59]USGS EN-10.53a0.07CaCO3 JM 9912−0.630.07CaCO3 JM 4064−6.620.08Inorganic calcite[61]CM0400.090.2CM039−0.070.2Growth solution0.310.2Foraminifera[62]OMEX-12b0.210.10WIND-10b0.280.11Mediterranean Sea water0.980.14Seawater1.090.09[54]MT 66 solute0.540.06Ett113 bulk silicate rock0.310.05MO33 0-10 soil0.510.05MO33 100-110 soil0.540.13Ace78 travertine0.140.05Carbonate fluorapatite[63]OG-50.260.11D4/010.420.08S30/030.230.0132a/010.520.1232/01s0.130.03D49/01−0.230.06aRelative to IAPSO seawater Wieser et al. [59] developed a high precision Ca isotope ratio measurement technique for a Finnigan Neptune magnetic sector ICP-MS. Delta values including δ44Ca/43Ca, δ44Ca/42Ca, and δ48Ca/42Ca were measured with an external reproducibility better than 0.2‰ in seawater and biogenic and non-biogenic marine carbonates. Fietzke et al. [60] developed a new technique for direct measurements of 44Ca/40Ca ratios on an MC-ICP-MS using “cool plasma”. Reducing the plasma energy from the usually applied approximately 1,300 W to about 400 W significantly reduced the 40Ar+ isobaric effect, allowing simultaneous and precise measurements of the two Ca isotopes. Repeated measurements of the 44Ca/40Ca ratios in various Ca standard materials were in good agreement with data reported in the literature. Marriott et al. [61] investigated temperature dependence of Ca isotopes in solution, inorganic calcite, and foraminifera and concluded that they are lighter than in the growth solution, and only weakly dependent on temperature. Sime et al. [62] also found negligible temperature dependence of calcium isotope fractionation in planktonic foraminifera. River water, rock, travertine, and soil were studied by Tipper et al. [54] who demonstrated that Ca, as well as Mg isotope ratios, are fractionated during weathering. Fractionation of Ca during continental weathering is of importance to the global cycle of Ca. The riverine input of Ca to the oceans is controlled not only by the composition of the primary continental crust, but also by the size and composition of the fractionated reservoir on the continents. The impact on the oceanic cycle of Ca depends on the relative residence times of dissolved Ca in the ocean and the storage time of fractionated Ca. Soudry et al. [63] explained fluctuations of Tethyan phosphogenesis through time, and whether or not they reflect long-term changes in ocean circulation or in continental weathering. A δ44Ca increase during the Late Cretaceous–Eocene also reflects a decrease in weathering Ca2+ fluxes together with increased biological removal of isotopically light Ca2+, fostered by increased continental flooding and concomitant carbonate (chalk) sedimentation on shelves. The overall concordant trends between the measured δ44Ca, the eustatic sea level curve, and the sizes of the flooded continental area throughout the Early Cretaceous–Eocene, point to a linkage between the Ca isotopic composition of paleoseawater and long-term paleogeographic and oceanographic changes. Steuber and Buhl [64] analyzed the calcium isotope composition in modern and ancient marine carbonates. No significant temperature dependence of Ca isotope fractionation was found in Cretaceous shells. δ44Ca of Cretaceous seawater was 0.3–0.4‰ lower than that of the modern ocean. Chu et al. [65] developed a procedure for the precise determination of Ca isotope ratios in natural and organic samples, such as bones, milk, and other biological materials. The data demonstrated that geological/environmental conditions do not cause large variability and it was suggested that diet is the major cause for variations in bones; so Ca isotope ratios may serve as a paleodiet tracer. Skeletal carbonates from the Tethyan realm were analyzed by Farkaš et al. [66]. The observed late Mesozoic δ44Casw was simulated using a Ca isotope mass balance model, and the results indicated that the variation in δ44Casw can be explained by changes in oceanic input fluxes of Ca that were independent of the carbonate ion fluxes. Sime et al. [67] interpreted Ca isotope behavior in marine biogenic carbonates. The 18 million year record of planktonic foraminifera in the Atlantic averages and is a good match to Neogene Ca isotope record based on foraminifera, but is not similar to those in bulk carbonates. There are also publications on Ca using seawater as a standard for normalization in the bracketing method. Copper Copper isotope ratio measurements have been used to determine natural isotope variations in ore geology, geochronology, and archaeometry [68, 69]. Three different types of ICP-MS instrument, a quadrupole, a single collector magnetic sector, and a multicollector magnetic sector instrument, were compared by Diemer et al. [70]. Although precision of results significantly differed, excellent agreement was observed between results obtained using all the instruments. The results for the Cu-isotopic composition are given in Table 4. Table 4Cu isotopic composition in standards and selected natural samples relative to SRM NIST 976 (‰)Sampleδ65Cu±2σReferenceNative copper[71]aOUM15126 Michigan, USA0.450.06OUM00061 Yekaterinburg, Russia−0.330.06OUM15127 Cornwall, England0.410.06Minerals OUM23585 malachite, England−0.260.06OUM1616 azurite, USA1.590.06OUM25139 chalcopyrite, England0.070.06Acc. Ser. 25300 chalcopyrite, Canada0.400.06G-4 chalcopyrite, sulfide deposits−0.440.06Ultra-pure Cu standards[75]JMC Cu Axiom0.6190.058JMC Cu IsoProbe0.6410.019IMP Cu IsoProbe0.200.10IMP Cu IsoProbe0.2070.049Copper sulfide precipitates[78]A-Cu- 0 Cu(I)S bulk0.700.06A-Cu- 0 Cu(I)S precipitated−2.520.06A-Cu-48 Cu(I)S bulk0.620.06A-Cu-48 Cu(I)S precipitated−2.320.06C-Cu-0 Cu(OH)2 bulk−0.260.06C-Cu-0 Cu(OH)2 precipitated−0.540.06Chalcopyrite from the Grasberg deposit region[73]aF1-001 Grasberg skarn0.1070.031XC05-001 Ertsberg diorite0.6330.035ES-005 Ertsberg skarn0.1930.054XC22001 pyrite shell0.2180.023XC25-002 bornite-2.69 Dalum−0.2690.031Alexandrinka VHMS sulfides[76]SW-1(a) chalcopyrite0.184<0.07SW-1(b) quartz0.318<0.07HV-1(a) sphalerite0.330<0.07CA-1 pyrite0.054<0.07CB-1(b) covellite−0.300<0.07CB-2(a) silicate−0.058<0.07Supergene copper sulfides[81]Tt-1, El Teniente bornite0.370.16Tt-6, El Teniente chalcopyrite−0.150.16El-1, El Salvador chalcopyrite0.810.16M-1, Mocha chalcopyrite−0.060.16aOriginal results were given in ɛ units (parts per 10,000) and have been converted to δ units MC-ICP-MS ratio determinations of copper were comprehensively studied by Zhu et al. [71]. A sample purification procedure was described, noting that possible isotope fractionation must be avoided. Potential interferences in 65Cu/63Cu ratio measurements are (23Na40Ar)+ and at mass 63 and (25Mg40Ar)+ at mass 65. It has been shown that for samples where Na/Cu and Mg/Cu ratios are below 10−4 and 10−3, respectively, and under instrumental working conditions applied in this study, the abundances of the polyatomic ions relative to Na+ and Mg+ were approximately 10−4, i.e., isobaric interferences were negligible. Two Cu isotope measurement procedures were used: the standard-sample-standard bracketing technique where variations are calculated relative to Cu isotope standard (SRM NIST 976) as in Eq. 2, and the “doping” technique [2] where the sample is doped with Zn at a concentration Cu/Zn ratio close to 1. Early results were given in “ɛ” units, but in the last papers “δ” is accepted. A constant value of the 68Zn/66Zn ratio was used as a normalization factor to correct the measured 65Cu/63Cu ratios. External precision (2σ) better than 0.03‰ and 0.06‰ was achieved by applying the doping and bracketing techniques, respectively [71]. Borrok et al. [72] presented a new method for efficient separation of Cu, Fe, and Zn from the greater concentrations of matrix elements using a single anion-exchange column with hydrochloric acid media. The possibility of laser ablation combined with a MC-ICP-MS for Cu isotope ratio determinations was presented by Graham et al. [73] About 160 solid samples were analyzed with a 2σ uncertainty of approximately 0.04‰. Jackson and Günter [74] examined the influence of various processes of laser ablation on fractionation of the Cu isotope ratio. The data suggest that the dominant source of isotopic fractionation at high laser fluence was the preferential volatilisation of 63Cu during incomplete vaporization and ionization in the ICP of particles greater than approximately 0.5 μm in diameter. Mason et al. [5, 75] performed a detailed study on spectral interferences across the mass range 63Cu to 70Zn and the mass discrimination corrections using two different MC-ICP-MS instruments: the double focusing VG Axiom and the single focusing Micromass IsoProbe. It was observed that the ion types and their intensities depend on the different instrumental configuration. The importance of removing problematic matrix components prior to the Cu and Zn measurements was also emphasized. Based on their method Mason et al. [76] studied Zn and Cu isotopic variability in the Alexandrinka sulfide ore deposit from the Urals, Russia. A further study on mass discrimination correction shows the importance of matrix removal, particularly Fe and Ti, the dependence of mass discrimination on Cu/Zn ratio in the solution and that sample introduction system with a desolvating membrane causes variable behavior of the Cu standard, probably due to variations in Cu oxidation state in the solution [77]. Ehrlich et al. [78] studied copper isotope fractionation between aqueous Cu(II) and CuS, the latter being precipitated from CuSO4 solution with Na2S under anoxic conditions. Using Ni as a doping agent and the standard-sample-standard bracketing technique a 2σ error of 0.06‰ was achieved. The mean fractionation factor at 20 °C was derived: . Additional experiments over the temperature range 2–40 °C showed an inverse dependence of the fractionation factor on temperature. Markl et al. [79] used copper isotopes as monitors of redox processes in hydrothermal mineralization. The authors concluded that copper isotope analyses cannot be easily used as a reliable fingerprint for the source of copper in archaeology and geology because the variation caused by redox processes within a single deposit is usually much larger than between deposits. However, Asael et al. [80] showed that systematically large Cu isotope fractionation occurred during redox processes in sedimentary copper ore deposits. Cu isotope fractionation was also observed in bacterial oxidizing environments [81]. The roles of Cu and Zn isotopes in chondrites and iron meteorites were studied by Luck at al. [82, 83]. A typical error (external precision) for δ65Cu ± 0.4–0.5‰ was reported. Moynier et al. [84] investigated isotopic composition of zinc, copper, and iron in lunar samples. Recently isotopic ratio studies of Cu and Zn were performed in seawater [85]. The very low concentration of these elements and high TDS matrix demanded development of preconcentration and purification, followed by separation processes. In this work accurate data and their uncertainties are given only for Zn. Conclusions We have reviewed a group of more than 60 papers that were published in the period from 2000 up to present. We also quote results from a few unpublished works made at the GSI. We have chosen to focus on the isotope ratio determinations of only four elements, namely lithium, magnesium, calcium, and copper, where high precision ratio measurements were for various reasons impossible to achieve with the early MC-ICP-MS instruments. The second-generation machines have detection systems with sufficient dispersion to accommodate Li ions and the capability to be tuned to partial mass numbers. The high resolution capability allows one in some cases to resolve interferences. New types of desolvation nebulizers allow significant reduction of solvent and atmospheric interferences and the “cold plasma” technique reduces nebulizing gas ion intensities. Consequently, precision (±2σ) for δ7Li, δ25, 26Mg, δ44Ca, and δ65Cu down to the range of (0.1–0.2)‰, (0.01–0.1)‰, (0.05–0.1)‰ and (0.06–0.15)‰, respectively, can now be achieved.

Pediatr_Radiol-4-1-2292491

MRI evaluation of tissue iron burden in patients with β-thalassaemia major

β-Thalassaemia major is a hereditary haemolytic anaemia that is treated with multiple blood transfusions. A major complication of this treatment is iron overload, which leads to cell death and organ dysfunction. Chelation therapy, used for iron elimination, requires effective monitoring of the body burden of iron, for which serum ferritin levels and liver iron content measured in liver biopsies are used as markers, but are not reliable. MRI based on iron-induced T2 relaxation enhancement can be used for the evaluation of tissue siderosis. Various MR protocols using signal intensity ratio and mainstream relaxometry methods have been used, sometimes with discrepant results. Relaxometry methods using multiple echoes achieve better sampling of the time domain in which relaxation mechanisms take place and lead to more precise results. In several studies the MRI parameters of liver siderosis have failed to correlate with those of other affected organs, underlining the necessity for MRI iron evaluation in individual organs. Most studies have included children in the evaluated population, but MRI data on very young children are lacking. Wider application of relaxometry methods is indicated, with the establishment of universally accepted MRI protocols, and further studies, including young children, are needed. Introduction β-Thalassaemia major is a hereditary anaemia characterized by ineffective erythropoiesis and haemolysis [1]. The underlying mechanism is defective production of haemoglobin β-chains, resulting in excess of α-chains, which are unstable and precipitate to form intracellular inclusion bodies [2, 3]. This excessive intracellular deposition of α-chain material is responsible for accelerated apoptosis of the erythroid precursors and for peripheral haemolysis of the erythrocytes [3]. By the age of 3 months, severe anaemia develops leading to increased intestinal iron absorption. To maintain haemoglobin at a level of 10–12 g/dl, patients suffering from β-thalassaemia major need to be given repeated blood transfusions [1]. A major drawback of this treatment is transfusion siderosis, which, in association with the increased intestinal iron absorption, apoptosis of the erythroid precursors and peripheral haemolysis, leads to iron overload [1]. Iron is ubiquitous in eukaryotic organisms. It is essential for cellular survival and proliferation and for haemoglobin synthesis [4, 5]. Human demands for iron are covered partly by intestinal absorption, but mainly from the recycling of iron from old or abnormal erythrocytes phagocytized by macrophages [6]. Iron released from macrophages binds mainly to transferrin, but also to citrate and albumin to form non-transferrin-bound iron (NTBI) which is a toxic form [7–9]. Extracellular iron bound to transferrin enters the cell via transferrin receptors by a process of endocytosis. Once iron is released into the cytoplasm, it enters a poorly defined cellular compartment termed the labile iron pool (LIP). Iron in this compartment is loosely bound and therefore is highly toxic [10]. Intracellular iron that is not needed for immediate use is stored in the form of ferritin which consists of an apoprotein storing up to 4,500 atoms of iron (loading factor 4,500) [5, 11], and is not cytotoxic. When ferritin storage capacity is exceeded the LIP increases and haemosiderin is generated from ferritin denaturation [10]. Iron in the form of haemosiderin is thought to be more cytotoxic [10]. The iron in the LIP is in both the ferrous (Fe2+) and ferric (Fe3+) forms [12]. Fe2+ reacts with hydrogen and lipid peroxides and generates highly toxic hydroxyl and lipid radicals (Fenton reaction) that damage cellular membranes, proteins and nucleic acids [4]. Iron overload, just as lack of cellular iron, may lead to cell death and organ dysfunction. Iron overload is a major cause of morbidity and mortality in β-thalassaemia major [1]. Iron accumulates initially in the reticuloendothelial system (bone marrow, spleen, liver) and then in the hepatocytes, the heart (myocytes) and the endocrine glands [1, 9, 13]. In contrast to the reticuloendothelial cells, the turnover of iron in the hepatocytes, myocytes and endocrine glands is very low [9]. Chelation therapy has been used to eliminate excess iron [1, 7]. Chelatable iron is derived from the catabolism of haemoglobin in macrophages [14]. Chelators remove NTBI from the plasma, but they do not interact with ferritin and haemosiderin at clinically relevant rates [10, 15]. Desferrioxamine (DFO), which is the chelating agent most widely used over the last 30 years, requires parenteral administration. It removes mainly extracellular iron and only a fraction of the intracellular LIP iron [14, 15]. Deferiprone is an oral iron chelator that penetrates the cellular membrane and chelates intracellular species of toxic iron [16]. Cellular uptake of chelators takes place at different rates in different cells [15]. DFO is of low toxicity, mainly affecting the optic, auditory and skeletal systems [7]. Skeletal changes develop mainly due to the toxic effects of DFO on the growth cartilage and are manifested as disproportionate truncal shortening with loss of seated height [7]. Adverse effects of deferiprone are arthralgia, gastrointestinal symptoms, elevated liver enzymes and rarely agranulocytosis [7, 16]. The effective management of patients, and especially of children, with thalassaemia requires optimal monitoring of the toxic effects of both iron overload and excessive chelation therapy. Serum ferritin has been widely used as a surrogate marker and a target ferritin level of 1,000 μg/l is generally recommended [7, 17]. However, serum ferritin represents only 1% of the total iron pool, and as an acute-phase protein it is not specific because the levels can be raised in inflammation (e.g. hepatitis) and liver damage [7, 18, 19]. Chronic liver inflammation is not rare in patients with thalassaemia, since over 40% of them have positive anti-hepatitis C virus (HCV) antibodies and more than 50% have chronic (persistent or active) hepatitis [20, 21]. Liver iron concentration (LIC) measured on needle biopsy is currently considered the gold standard for the evaluation of siderosis [7, 18, 19]. However, needle biopsy is an invasive technique, it is not easily repeatable and the accuracy of the resulting LIC measurement is greatly affected by hepatic inflammation-fibrosis and uneven iron distribution [22]. Furthermore, it appears that cardiac iron overload, which is the leading cause of death in thalassaemia, cannot be predicted from the degree of liver siderosis, probably because of differences in iron kinetics between liver and heart cells [9]. The use of non-invasive techniques for monitoring iron overload in each of the affected organs would be preferable, and to this end MRI has been used increasingly over the last two decades [19, 22–44]. Other non-invasive methodologies are too complex, too costly and not readily available (i.e. magnetic susceptometry), or lack imaging capability (i.e. magnetic resonance spectroscopy) [45, 46]. MRI assessment of tissue iron Relaxation theories MRI evaluation of tissue iron overload is based on T2 relaxation enhancement induced from the interactions between high-molecular-weight iron complexes such as ferritin, haemosiderin and ferrioxamine with water molecules [5, 47, 48]. Ferrioxamine is produced after cellular uptake of DFO, and in extrahepatic tissues, where there is no active excretion and it may remain for days [7]. Ferrioxamine-based agents have been used in the past as contrast media because of their T2* effect [49]. “Inner sphere” and “outer sphere” relaxation theories have been proposed to explain the T2 relaxation enhancement induced in siderotic tissues [50–55]. According to the inner sphere theory, iron electron spins enhance the relaxation of protons of water molecules buried in iron-containing proteins. Enhanced relaxation of the bound water protons is then transferred to the free water protons by water or proton exchange [48, 56, 57]. According to the outer sphere theory, magnetic field gradients induced at the periphery of iron-containing proteins create loss of phase and relaxation enhancement of free water protons diffusing in the neighbourhood [53, 54]. MR methodologies MR methods for assessing tissue iron can be separated into two groups: signal intensity ratio (SIR) methods and relaxometry methods. Various techniques have been described, including: (a) methods measuring SIR based on T2-weighted (spin-echo) or T2*-weighted (gradient-echo) sequences [19, 23–26, 30, 58–62], (b) relaxometry methods measuring absolute T2, (c) relaxometry methods measuring absolute T2*, and (d) hybrid relaxometry methods [30–32, 34–37, 39–43, 58, 63]. SIR methods These have been used for the study not only of the liver but also of other organs such as the spleen, pancreas, pituitary gland, bone marrow and abdominal lymph nodes [19, 23, 25, 27, 30, 58–62]. For SIR assessment the signal intensity of the target organ is divided by the signal intensity of a reference tissue (e.g. fat, muscle) or noise. Signal intensity measurements are performed in the same slice by using the region-of-interest (ROI) method. For large organs such the liver, spleen and pancreas more than one ROI is used, positioned in areas lacking vascular structures and movement artefacts [19, 23, 25, 27, 30, 58, 61, 62]. The mean signal intensity from the different ROIs is then divided by the signal intensity of the reference tissue. One ROI is used for signal intensity measurement of the pituitary gland, bone marrow, abdominal lymph nodes and the reference tissue [58–60]. For SIR evaluation of the liver various reference tissues have been used, of which paraspinal muscle appears to be the best choice because it combines good sensitivity with minimum intersite variability [19, 64]. A disadvantage of the SIR methods is that in most cases they use only one echo time (TE) and thus lose their detection sensitivity in tissues with heavy siderosis, where transverse relaxation is much faster than the TE. This occurs particularly in the liver at the upper range of LIC values where signal intensities are widely dispersed [39]. Gandon et al. [19], by using an algorithm that combined signal intensity ratios from multiple sequences with different TEs, achieved extension of the detection range up to about 21 mg Fe per gram dry liver tissue, with sensitivity and specificity similar to those of biochemical analysis. T2 relaxometry methods These assess T2 relaxation time or R2 (1/T2) by using the Carr-Purcell-Meiboom-Gill (CPMG) spin-echo sequence, which employs multiple (2–32) equidistant refocusing 180° pulses, each followed by an echo [65, 66]. Most scanners, by using a pixel-by-pixel, log-linear fitting model, automatically derive the corresponding T2 maps. Signal intensity measurements in the T2 maps correspond to the mean T2 relaxation time of the included voxels [67]. To avoid motion artefacts from respiration, studies of abdominal organs should be performed with respiratory triggering. To this end a pressure-sensitive pad is attached to the abdomen and depending on the patient’s respiratory cycle a delay time is chosen. Data acquisition takes place at the end-expiration phase [39, 40]. During cardiac MRI, in addition to respiratory triggering, cardiac triggering is necessary. To obtain a minimum repetition time (TR) of 2,000 ms two to three R–R intervals should be allowed between successive excitations [39, 40]. The mid-systole phase and a cardiac trigger delay of around 250–300 ms are usually chosen to image the myocardium with sufficient thickness for the best signal intensity measurements [39, 40]. A pseudo four-chamber or short-axis view of the heart is generally used. A T2 relaxometry method that received FDA approval for clinical liver iron estimation has recently been developed by St. Pierre et al. [44]. This method uses multiple T2-weighted single spin-echo sequences with different TEs acquired in half-Fourier mode to reduce acquisition time. The calculated mean R2 values combined with LIC values, obtained from liver biopsies, are used to create calibration curves. T2* relaxometry methods These evaluate T2* or R2* (1/T2*) by using multiple gradient-echo sequences with different TEs. These methods have been developed to further accelerate acquisition, in order to increase sensitivity and eliminate artefacts related to respiration or cardiac motion. To obtain R2* (1/T2*) values, the signal decay curve is usually fitted with an exponential model: S=S0e−TE/T2*, where S is the net image signal intensity, TE is the echo time and S0 is a constant [68]. T2* relaxometry methods have been used mainly for myocardial iron assessment and cardiac gating is always applied [35, 42, 68–71]. Breath-hold sequences have greatly eliminated motion artefacts [35, 42, 68]. Hybrid relaxometry methods Hybrid approaches have been applied in high fields and measure both R2 and R2* to calculate the inhomogeneity factor R2′=R2*−R2 [34, 72, 73]. These approaches assume that R2′ is more specific to mechanisms of relaxation related to iron than R2 [34, 73]. Comparison of the MRI methodologies SIR versus relaxometry SIR methods require shorter acquisition times but lack a wide range of iron assessment [63]. Relaxometry methods, mainly the T2* method, by using multiple echoes create in- and out-of-phase effects between water and fat transverse magnetization (Fig. 1) [35]. Relaxometry methods, although taking longer, are preferable because they achieve a better sampling of the time domain in which relaxation mechanisms take place and lead to more precise results [63]. Fig. 1A 15-year-old male with β-thalassaemia major. Axial scan with a multiecho gradient-echo sequence (TR/TE 150/1.07, 2.14, 3.21, 4.28, 5.35, 6.42, 7.49, 8.56, 9.63, 10.7, 11.77, 12.84, 13.91, 14.98, 16.05, 17.12, 18.19, 19.26, 20.33, 21.4 ms; flip angle 35°). a Second echo shows pixel annulation (arrows) due to out-of-phase phenomena, at the interfaces of abdominal organs and muscles with fat. b Fourth echo: no pixel annulation is observed because water and fat protons are in phase (courtesy of Dr. M. Douskou) T2, T2* and hybrid methods There is no general consensus on which relaxometry method (spin-echo or gradient-echo) or index (R2, R2*, R2′) is best for tissue iron quantification. Theoretically, R2* and R2′ are more sensitive than R2 to iron-induced field inhomogeneities. Gradient-echo sequences evaluating R2* are preferred for cardiac relaxometry, because of their short acquisition time, but gradient-echo sequences are more prone to artefacts and R2* is more dependent on factors unrelated to iron, such as susceptibility artefacts from the lungs and the blood oxygenation level-dependent (BOLD) effect [35, 74]. According to the BOLD effect, the T2* decreases as the concentration of intracapillary deoxyhaemoglobin increases [74]. A pronounced BOLD effect may be observed in early diastole and results in signal loss [75, 76]. Data collection during mid-systole helps to overcome BOLD effect interference with accurate T2* measurements of the myocardium. In liver and pituitary studies both T2 and T2* have been assessed. T2* is more sensitive to low iron content, but studies in the liver show that these methods suffer from inaccuracies at high iron concentrations [77]. Furthermore, in high fields susceptibility artefacts from the sphenoid sinus are more pronounced with gradient-echo sequences and may lead to inaccurate measurements of pituitary siderosis [43]. MRI studies of individual iron overloaded organs The degree of siderosis, the crystalline structure of ferritin, the rate of iron elimination under chelation therapy and the degree of ferrioxamine formation are all organ-specific [7, 15, 18, 78]. All these parameters may be responsible for differences in the T2 relaxation enhancement induced in the various organs. Individual organs should be considered separately, and the effect of age on iron overload should be taken into account. Higher survival probabilities have been reported in patients with thalassaemia born in the last 30 years [79]. Patient compliance with treatment regimens and effective chelation therapy are thought to be the main factors associated with improved survival [79]. The combination of DFO, deferiprone and the new oral chelators is considered very promising, but will require effective monitoring by non-invasive methods [80]. An increasing number of studies have evaluated iron in the various affected organs by MRI [19, 22–44]. Many of these studies have included paediatric patients, but very few have investigated children younger than 10 years and none has included children younger than 5 years [17, 18, 28, 39, 43, 58, 60–62, 77, 81–84]. In addition, very few studies have evaluated the effect of age on individual organ siderosis and only one has evaluated and compared MRI data from younger and older patients with thalassaemia [28, 81, 82, 84]. Studies have tended to focus on specific organs. Liver For the MRI evaluation of liver siderosis both SIR and relaxometry techniques have been used (Fig. 2) [18, 23, 28, 32, 44, 68, 85]. R2 of the liver demonstrates a significant positive correlation with serum ferritin and LIC determined from liver biopsy material [18, 23, 32, 39, 40, 44, 68]. Comparative evaluation of hepatic R2 and R2* in iron-overloaded patients demonstrates that both parameters correlate closely with LIC [85]. The relationship of SIR with LIC and serum ferritin varies among studies [18, 30, 58]. In most studies R2 and SIR show a better correlation with LIC than with serum ferritin [18, 23, 28, 32, 44, 68, 85]. This can be explained in part because the HCV-positive thalassaemia patients in the studies had higher serum ferritin levels than those who were HCV-negative [77, 86]. Liver R2 shows no association with the hepatic inflammation histological activity index or the type of hepatitis (chronic persistent or chronic active), but is affected by hepatic fibrosis [18, 39, 77]. Fig. 2A 17-year-old male with β-thalassaemia major. Axial scan, fourth echo of a multiecho spin-echo sequence (TR/TE: 2000/20, 40, 60, 80, 100, 120, 140, 160 ms), shows low-signal intensity in the liver, bone marrow of the spinal body and the pancreas suggesting iron overload In iron overload states, over 70% of body iron is found in the liver and LIC has been considered to be the best marker of total body iron burden [7, 18, 19]. Based on the good correlation between hepatic R2 or SIR and LIC a number of recent studies have tested the relationship between siderosis of the liver and other organs [28, 39, 40, 58–61, 87]. No correlation has been found between liver and pituitary siderosis [28, 58]. With regard to the heart, a correlation with liver siderosis has been found only in cases of heavy myocardial iron deposition [39, 40]. This lack of correlation can probably be explained by differences in transferrin receptor concentration, iron kinetics, the crystalline structure of ferritin and the degree of organ inflammation or fibrosis [7, 78, 88–92]. Furthermore, under chelation therapy with DFO intracellular paramagnetic ferrioxamine is formed, which exits slowly from cells unless there is an active excretion pathway as is present in the hepatocytes [88]. Young patients with thalassaemia studied longitudinally have shown absence of substantial improvement in the MR parameters of liver siderosis under different chelation therapy regimens [82]. This may be explained by the fact that liver siderosis progresses very fast in thalassaemia patients, and iron overload develops after only 2 years of transfusion therapy [93]. Therapy with the most widely used chelating agent is started at the age of about 3 years, and until growth is completed DFO should not exceed a dose of 40 mg/kg per day [7]. An early start to monitoring the progress of tissue iron deposition with MRI might be useful in deciding whether to begin chelation therapy at a younger age and when to introduce new chelating agents. Heart Cardiac failure is the leading cause of death from iron overload in patients with thalassaemia [94, 95]. Thalassaemia patients on a regimen of regular blood transfusions who are not receiving chelation therapy, develop heart enlargement by the age of 10 years and heart failure by the age of 16 years [22]. Cardiomyopathy secondary to iron overload is potentially reversible with vigorous chelation therapy [7]. Myocardial iron has been evaluated by MRI using SIR and relaxometry techniques (Fig. 3) [22, 28, 39, 40, 68, 82, 96–98]. In the prechelation era, post-mortem examination of patients with thalassaemia showed a close correlation between cardiac and hepatic iron concentrations. MRI studies have demonstrated discordant results regarding the relationship of myocardial iron with hepatic iron and serum ferritin [28, 39, 40, 68, 97]. Differences in iron kinetics and variations in chelation schemes may be responsible for the lack of correlation of the MRI-determined myocardial iron with that of the liver [88]. More active elimination of iron from the hepatocytes than from the myocytes may play a role in the absence of correlation in patients receiving chelation therapy [28, 39, 68]. Fig. 3A 16-year-old male with β-thalassaemia major. a T2 map of the short axis of the heart shows low values of the left ventricle (arrow) and the septum (arrowhead). b R2* (=1/T2*) map of the short axis of the heart shows high signal of the septum (asterisk). Note the reduction of motion artefacts and better delineation of the heart boundary in the R2* image (courtesy of Dr. M. Douskou) Another important issue is whether myocardial iron evaluated by MRI could predict which patient will develop cardiac failure and arrhythmia. Thalassaemia patients with significantly reduced left ventricular ejection fraction (LVEF) have myocardial T2* values about 20 ms [22, 37, 38, 68], although most patients with T2* values less than 20 ms have normal LVEF [68]. Furthermore, the relationship between myocardial R2 and LVEF is either non-existent or weak [22, 38–40]. Differences in cardiac functional indices among patients with similar MRI measurement of iron might also be explained by genetic factors that predispose to the more rapid development of cardiac failure when a critical level of myocardial siderosis is reached [99, 100]. Heterogeneous iron distribution in the myocardium has been demonstrated in histological studies, and this would suggest the necessity for global MRI estimation of myocardial iron [42, 101]. Pepe et al. [42], using T2* multiecho cardiac MRI, demonstrated that global T2* shows a close correlation with midseptal T2*, suggesting that the latter may be used for a quick assessment of myocardial siderosis. MR studies have shown that cardiac siderosis increases with age, but that siderosis progresses more slowly in the heart than in the liver [28, 82]. Systematic MRI studies in children are necessary in order to evaluate the age at the start of cardiac iron overload. Pituitary gland and brain Increased iron deposition in the anterior pituitary gland is the cause of hypogonadotropic hypogonadism and growth hormone deficiency [102, 103]. Iron accumulates in all five cell types of the adenohypophysis, but preferentially in the gonadotropin-secreting cells [102, 103]. Gonadotropin cell death due to iron toxicity is probably the cause of the decreased pituitary gland height observed in thalassaemia patients with hypogonadotropic hypogonadism (Fig. 4) [104]. MRI has been used to evaluate pituitary siderosis with SIR and relaxometry techniques [28, 43, 83, 105]. SIR and R2 findings show significant correlation with serum ferritin [83]. The very few studies evaluating simultaneously the pituitary and other solid organs have demonstrated no correlation between their MRI parameters of siderosis [28, 43, 81]. Pituitary siderosis appears to increase with age, but MR data are lacking for children under the age of 7 years [28, 81] [84]. Fig. 4The pituitary. a A 12-year-old male with β-thalassaemia major. Midsagittal T1-weighted (TR/TE 500/20 ms) scan shows low-signal intensity of the anterior pituitary lobe (arrow) and the bone marrow (asterisks) suggesting iron overload. The pituitary gland is small measuring 3.5 mm (normal for age 5.3±0.8 mm) [112]. This patient developed hypogonadotropic hypogonadism. b An 11-year-old male with β-thalassaemia major. Mid-sagittal T1-weighted (TR/TE 500/20 ms) scan shows normal signal intensity of the anterior pituitary lobe (arrow) along with normal pituitary gland height (6 mm). The bone marrow (asterisk) shows low-signal intensity suggesting iron overload The single study assessing iron in the brain of thalassaemia patients by MRI [106] showed significantly higher R2 values in the cortex, the putamen and the caudate nucleus in patients than in controls. A lack of correlation between R2 and serum ferritin was found, possibly because serum ferritin, which does not cross the blood–brain barrier, is not a good marker of brain siderosis [106]. Pancreas Iron overload leads to impairment of the endocrine and exocrine functions of the pancreas [107, 108]. Pancreatic siderosis has been evaluated in thalassaemia patients using SIR techniques and relaxometry techniques [61, 84, 109]. Midiri et al. [109], using fat as the reference tissue, found a significant negative correlation of SIR with serum ferritin and trypsin. Papakonstantinou et al. [61], using muscle as the reference tissue, found no relationship of the pancreatic SIR with the hepatic SIR or serum ferritin. The findings of Argyropoulou et al. [84], using T2 relaxometry techniques, were similar. In addition, relaxometry techniques demonstrated that the T2 relaxation time of the pancreas of patients with thalassaemia is significantly lower in children than in adults [84]. Fatty degeneration of the pancreas in adult patients may explain these differences (Fig. 5). Fig. 5Siderosis of the spleen, pancreas and adrenals. a A 14-year-old male with β-thalassaemia major. Axial scan, fourth echo of a multiecho sequence (TR/TE 2,000/20, 40, 60, 80, 100, 120, 140, 160 ms) shows low-signal intensity of the liver, pancreas (arrow) and spleen suggesting increased iron deposition. b A 27-year-old-male with β-thalassaemia major and diabetes. Axial T2-W (TR/TE 1,800/80 ms) scan shows low-signal intensity of the liver and adrenals suggesting siderosis. High signal intensity of the pancreas (arrow) suggests tissue damage and fatty degeneration Adrenals Abnormalities in adrenal function have been reported in patients with thalassaemia [103]. There is only one study evaluating adrenals for iron overload with MRI (Fig. 5), which showed a significant correlation between adrenal and liver siderosis [87]. Spleen, lymph nodes and bone marrow In spite of the fact that the spleen, lymph nodes and bone marrow, which all contain reticuloendothelial cells, are among the first organs to be affected by iron overload [1, 13], there have been very few studies evaluating their iron overload in thalassaemia by MRI, and these used mainly SIR techniques [59, 60, 62, 84]. SIR of the spleen shows a significant correlation with serum ferritin but not with SIR of the liver (Fig. 5) [62]. The absence of correlation between liver and spleen siderosis could be explained by differences in iron kinetics, by differences in the cluster size of iron proteins, by haemochromatosis gene mutations in β-thalassaemia major carriers, and by the presence of extramedullary haemopoietic tissue in the spleen [78, 110, 111]. Intraabdominal lymph nodes in β-thalassaemia have been related to chronic hepatitis C [60]. Lymph node siderosis correlates with liver, but not with spleen siderosis [60]. In the few studies that have been reported, SIR and relaxometry methods have shown discordant results for MR parameters of bone marrow siderosis and serum ferritin [59, 84]. Normal bone marrow signal associated with liver siderosis has been reported in a few patients with thalassaemia and this may be due to differences in genotype or differences in chelation therapy regimens [59]. Conclusion It is evident that MR relaxometry has the potential to become the method of choice for non-invasive, safe and accurate assessment of iron load. Further theoretical research, along with studies monitoring wider age groups of patients are needed before a generally accepted protocol can be established. Until then, extreme caution is needed in its clinical application and interpretation. Experimental artefacts, non-optimized protocols, poor data analysis and unawareness of the inherent limitations of current methodologies in assessing a heavy body iron burden can result in misleading diagnosis and inappropriate management of thalassaemia patients with iron overload. Future experimental developments in relaxometry based on a better theoretical understanding of the contribution of the iron-containing proteins to the MR signal are expected to further strengthen its clinical role in the monitoring of patients with β-thalassaemia major.

Placenta-1-5-1895600

Oxidative Stress and the Induction of Cyclooxygenase Enzymes and Apoptosis in the Murine Placenta

Placental oxidative stress has been implicated in many complications of human pregnancy, including preterm delivery and preeclampsia. It is now appreciated that reactive oxygen species can induce a spectrum of changes, ranging from homeostatic induction of enzymes to apoptotic cell death. Little is known regarding the occurrence of placental oxidative stress in other species. We investigated markers of oxidative stress in the labyrinthine (LZ) and junctional (JZ) zones of the murine placenta across gestational age, and correlated these with expression of the cyclooxygenase enzymes COX-1 and COX-2, and apoptosis. We tested a causal link between the two by subjecting placental explants to hypoxia-reoxygenation (H/R) in vitro, a known stimulus for generation of oxidative stress. Western blotting demonstrated significant increases in the concentrations of hydroxynonenal (HNE), COX-1 and COX-2 with gestational age. Dual-labelling demonstrated co-localisation of HNE, and COX-1 and COX-2 within the trophoblast of the LZ, and glycogen cells of the JZ. An apoptotic index based on TUNEL-positivity demonstrated an increase with gestational age, and dual-labelling showed co-localisation of TUNEL labelling with HNE and active caspase-3 within the trophoblast of the LZ. H/R significantly increased oxidative stress, induction of COX-1 and COX-2, and the apoptotic index. Co-localisation demonstrated the increases in COX to be within the trophoblast of the LZ, and in particular the glycogen cells of the JZ. Apoptosis was restricted to the LZ. We speculate that the induction of COX enzymes is a physiological response to oxidative stress, and may play a role in initiating or augmenting parturition. Generation of oxidative stress may also play a role in influencing the growth trajectory of the placenta, and its component cell types. The mouse may provide an experimental genetic model in which to investigate these phenomena. 1 Introduction Oxidative stress of the placenta has been implicated in the pathogenesis of many complications of human pregnancy, including miscarriage, preeclampsia, and preterm labour [1–3]. These complications all share the common predisposing feature of reduced trophoblast invasion and incomplete conversion of the uterine spiral arteries [4,5]. Therefore, it is generally held that malperfusion of the placenta leads to increased oxidative stress, resulting in placental dysfunction. In the case of preeclampsia it is thought that the stress induces the release of a cocktail of factors, including pro-inflammatory cytokines, anti-angiogenic factors and apoptotic debris [6], into the maternal circulation that causes activation of the peripheral endothelial cells. Placental inflammatory lesions have also been associated with premature onset of labour [7]. By contrast, there are few data relating to placental oxidative stress available for other species. The mouse is potentially a powerful model in which to study the mechanisms of placental and obstetrical pathologies, and is the best-studied mammalian experimental genetic model system. The definitive murine placenta is a discoid haemochorial organ as in the human, and there are three anatomically and physiologically distinct regions: the labyrinth zone (LZ), junctional zone (JZ) and decidua basalis (DB) [8]. The labyrinth zone is the principal area of exchange, and consists of a meshwork of maternal blood spaces lined by trophoblast. The trophoblast comprises three layers; an outer layer (I) composed of cytotrophoblast cells, and two layers (II & III) of syncytiotrophoblast [9]. Deep to the syncytiotrophoblast are the fetal capillaries, embedded in a small amount of connective tissue. The junctional zone, interposed between the labyrinth and the decidua basalis, is of unknown function, but contains spongiotrophoblast cells and trophoblastic glycogen cells (GC) [8]. The decidua basalis is composed principally of maternal uterine tissues, and contains maternal arteries and veins that are continuous with the arterial and venous channels traversing the JZ. Glycogen cells migrate into the DB and surround the maternal arteries that supply the placenta late in gestation. The mechanisms of myometrial stimulation and activation appear to parallel those of the human [10]. For example, prostaglandins (PGs) are central components of labour in both humans and mice [11]. There are fundamental similarities in the changes occurring prior to labour, including an increase in myometrial contractile activity, increased coupling of myometrial cells through the formation of gap junctions, an increase in PG receptors in the myometrium, and enhanced sensitivity to oxytocin at term. There is also a rise in intracellular calcium resulting in an increase in myosin light chain phosphorylation. In this study we determined the degree of oxidative stress present in the murine placenta at different gestational ages in normal pregnancies. We also tested whether there was a temporal and spatial association between oxidative stress and two markers of placental function, one physiological and one pathological. These were the induction of cyclooxygenase (COX) enzymes and apoptosis respectively. Evidence from other systems points to potential links between oxidative stress and COX induction, resulting in the production of PGs [12–16]. PGs are derived from arachidonic acid (AA) through the actions of the COX enzymes, which are rate-limiting. There are two known isoforms of COX, COX-1 and COX-2, which share similar structures but differ in their function. COX-1 is believed to play a constitutive “housekeeping” role, whereas COX-2 is considered to be an inducible isoform. However, these contrasting roles have not been proven in the mouse [17]. In mice induction of COX-2 by administration of lipopolysaccharide leads to preterm labour, which can be blocked using the specific COX-2 inhibitor SC-236 [18]. Apoptosis has been described within the LZ of the normal mouse placenta, and is most frequent towards term and in post-mature placentas [19]. The stimulus for this is not known, but oxidative stress is a potent inducer of trophoblast apoptosis in the human placenta [20]. 2 Materials and methods 2.1 Tissue collection Placentas were collected from C57BL/6J inbred mice, and all experiments were carried out in accordance with the UK Government Home Office licensing procedures. Stages E14, E16, E18, and E19 were studied, where E1 of gestation was the morning when a copulation plug was found. 2.2 Western blotting Placentas from three different animals at each gestational age were homogenized in ice-cold sample buffer, and centrifuged at 15,000 rpm. Protein concentration within the supernatant was determined using a colorimetric assay (Bio-Rad Bradford protein assay). An equal amount of protein from each sample (30 μg) was subjected to electrophoresis in a 10% SDS-polyacrylamide gel under reducing conditions and transferred onto nitrocellulose membrane by a semidry transfer machine. Membranes were incubated in a blocking solution (TBS containing 5% powdered non-fat milk and 0.01% Tween-20) at room temperature for 1 h, and then in the same solution containing primary antibody (Table 1) at 4 °C overnight. After washes in TBS-T, membranes were incubated with the horseradish peroxidase secondary antibody at room temperature for 1 h. The immunoreactive bands were visualized by enhanced chemiluminescence with the ECL system (Amersham Biosciences) according to the manufacturer instructions. The levels of protein expression were quantified densitometrically and normalised against their respective expressions of β-actin. Prestained protein markers were used as molecular weight standards for each analysis. 2.3 Immunohistochemistry At each gestational age three placentas were fixed by immersion in 4% paraformaldehyde overnight, dehydrated and embedded in paraffin wax. Sections (6 μm thick) were dewaxed and immunostained according to our standard protocol [21]. When necessary, antigen retrieval was performed in a pressure cooker for 2 min in 0.01 M Citrate buffer at pH 6. Primary antibodies (Table 1), diluted in goat serum, were applied overnight in a humidified chamber at 4 °C. Binding was detected using Vectastain Elite ABC kits (Vector Laboratories) and SigmaFast DAB (Sigma), according to the manufacturers’ instructions. Sections were then lightly counterstained with haematoxylin. Negative controls were performed by omitting incubation with the primary antibody. For each antibody slides from all the gestational ages were prepared and immunostained in the same batch, ensuring identical conditions for comparisons. In order to localise expression of the antigens to specific cell types staining intensity was graded visually from 0–3, where 0 represented the negative control. Three people scored the sections independently, blinded to the gestational age or experimental protocol, and the mean value was taken. 2.4 Immunofluorescent dual-labelling Dual immunofluorescent labelling was performed in order to test for co-localisation of markers of oxidative stress and COX expression in individual cells. Sections were dewaxed, permeabilised in TBS containing Triton X-100 (0.1%) and Tween 20 (0.1%) (TBS-TT) for 30–60 min, and blocked in 5% bovine serum albumin for 20 min at room temperature. Primary antibodies diluted in TBS-TT were applied overnight at 4 °C. Negative control sections had primary antibodies omitted. After three 10-min washes in TBS-TT, sections were incubated for 1 h at room temperature with a mixture of fluorescent secondary antibodies, containing anti-goat Alexa 488 and anti-rabbit Alexa 568 (both used 1/100; from Molecular Probes) in TBS-TT. Sections were washed in TBS-TT as before and then twice in distilled water for 5 min and subsequently mounted in Vectashield mounting medium containing DAPI (Vector, UK). Images were captured using a Leica confocal microscope (LeicaTCS-NT, Leica Instruments GmbH). Co-localisation was identified in overlay images by a yellow signal caused by the combination of the green signal (COX-1/COX-2) and the red signal (HNE). 2.5 TUNEL assay Staining was performed using In-Situ Cell Death Detection kit-Texas Red (Roche Applied Science) according to the manufacturer's instructions. Conditions were adjusted so that the staining pattern matched morphological assessments of apoptosis based on examination of resin-embedded samples. Once the protocol had been optimised an apoptotic index was calculated as follows: 2.6 Dual-fluorescence TUNEL assay & Active Caspase-3, and HNE co-localisation Staining was performed using In-Situ Cell Death Detection kit-Texas Red (Roche Applied Science) according to the manufacturer's instructions, and using an Anti-Active Caspase-3 antibody (Promega) at 1:50, and Anti-HNE as previously at 1:200. Samples were also DAPI stained to intercalate the DNA and fluoresce cell nuclei. For negative controls, omission of primary antibody and TdT enzyme were performed separately. For positive controls, post-lactational mouse mammary glands were used. For confocal microscopy, all samples were analysed during one session to avoid bias, and multiple (usually 5) fields of view per labyrinth were saved for further analysis. 2.7 Ex-vivo culture experiment In order to induce oxidative stress placental samples were subjected to hypoxia-reoxygenation (H/R) in vitro [22]. Placentas were taken from 3 animals at gestational age E18, halved on an ice block and immersed in ice-cold PBS prior to culture. Three samples, each 5–10 mm3, from each placenta were incubated in culture medium (TCS large vessel endothelial cell basal medium (TCS CellWorks, Milton Keynes, UK)), which had been maintained for several hours in the specified gas composition at 37 °C. Data for intraplacental oxygen concentrations are not yet available for the mouse, and so estimates were made on the basis of results for another small mammal, the guinea-pig. The umbilical venous pO2 towards term is 29.5 mmHg, which is close to that for fetal primates and ungulates [23]. Therefore, for normoxia a gas mix of 5% O2, 90% N2, 5% CO2 was used, whereas for hypoxia it was 1% O2, 94% N2, 5% CO2. Gas composition was monitored and controlled throughout using Biospherix (Redfield, NY) Pro-Ox and Pro-CO2 probes. Control samples were cultured for 4 h in normoxia, whereas samples subjected to H/R experienced 1 h of hypoxia followed by 3 h of normoxia. Following culture the tissues were either frozen or fixed for Western blotting and immunohistochemistry respectively. 2.8 Statistical analyses Statistical significance was calculated by using one-way analysis of variance (ANOVA). Differences between groups were assessed using Fisher's Least Significant Difference Test. Statistical significance was assumed at a P-value <0.05. 3 Results 3.1 Markers of oxidative stress Hydroxynonenal (HNE) is a marker of lipid peroxidation. Western blotting revealed significant changes across gestational age (P = 0.003), with low concentrations of HNE at E14, a significant increase at E16, and a subsequent decline by E18 (Fig. 1a,b). Scoring based on IHC revealed different intensities of immunoreactivity in the different cell populations, and different temporal patterns. At E14 immunoreactivity was low in all cell types (Fig. 1c,d). At E16 there was moderate staining (scored 2) in the syncytiotrophoblast of the LZ, which peaked (3) at E18 and declined to 1 by E19 (Fig. 1e). The cytotrophoblast cells of the LZ stained uniformly at level 1 throughout, and no immunoreactivity was observed in the fetal endothelial cells. In the JZ strong staining was observed in the GCs at E16 (2.5) and E18 (3) (Fig. 1f), and this was maintained at 2.5 at E19. The spongiotrophoblast cells were less immunoreactive, scoring only 0.5 throughout gestation, and no immunoreactivity was observed in the giant cells or the decidua. Nitrotyrosine (NT) staining indicates the formation of the prooxidant peroxynitrite, and of an imbalance in the production of the superoxide ions and nitric oxide. Immunostaining revealed a similar pattern to HNE, with almost undetectable staining in early samples and significant increases from day 16 onwards (data not shown). Therefore, oxidative stress increases during gestation in the murine placenta, specifically in syncytiotrophoblast and glycogen cells. 3.2 COX-1 Western blotting revealed significant changes in concentrations of COX-1 across gestational age (P = 0.013). There was an increase from E14 to E16, followed by a significant decrease by E18 (Fig. 1a,b). At E14 COX-1 immunoreactivity was not observed in any cell type, except for low intensity (level 1) staining in the GCs. At E16 strong reactivity (3) was visible in the GCs, with lower intensity staining (1) being present in the spongiotrophoblast and cytotrophoblast cells (data not shown). By E18 there was a general increase in staining, with many of the cytotrophoblast cells (2) reacting positively, along with the syncytiotrophoblast (2), and all GCs (3) (Fig. 2a,b). This pattern persisted at E19, when the decidua (2) was also immunoreactive. By contrast, the spongiotrophoblast cells displayed low (0.5) immunoreactivity throughout gestation. Towards term COX-1 and HNE immunoreactivity were present within the same cells in both the LZ and JZ, as evidenced by the dual-labelling experiments (Fig. 2c,d). 3.3 COX-2 Western blotting revealed a significant change in concentrations of COX-2 across gestation (P = 0.002), peaking at E19 (Fig. 1). IHC demonstrated that COX-2 expression is low (level 1) within the syncytiotrophoblast, spongiotrophoblast and GCs at E14. With advancing gestational age immunoreactivity increased, principally in the cytotrophoblast cells (2), GCs (2.5) and spongiotrophoblast (1) at E18 and E19 (Fig. 3a,b). The syncytiotrophoblast displayed low immunoreactivity (0.5) throughout gestation. Again there was strong co-localisation between immunoreactivity for COX-2 and HNE (Fig. 3c). 3.4 Detection of apoptotic changes During apoptosis cytokeratins within trophoblast cells are cleaved by active-caspase 3 to yield a specific product detected by the M30 antibody [24]. The labyrinth showed low immunoreactivity for active-caspase 3 until E18 (Fig. 4b), but thereafter it increased in intensity until E19. The staining was localised to both the syncytiotrophoblast and individual cytotrophoblast cells. Immunoreactivity for M30 also increased with gestational age, and was most abundant at E18, declining slightly at E19 (data not shown). In order to estimate the number of apoptotic nuclear profiles TUNEL staining was performed and an apoptotic index calculated. TUNEL-positive nuclei were first observed at E16 when the apoptotic index was 2.7%, and steadily increased in frequency until E19, when the index reached 20.0% (P < 0.001). There was a strong temporal and spatial association between TUNEL staining and immunoreactivity for HNE (Fig. 4a) and active caspase-3 (Fig. 4b). At E16 a number of cells were caspase-positive but TUNEL-negative. Later, almost all TUNEL-positive nuclei were surrounded by cytoplasm immunoreactive for active caspase 3. TUNEL-positive nuclei were rarely observed in the absence of caspase-positivity, when they were considered to be necrotic. To further confirm trophoblast cells were undergoing apoptosis, morphological evidence was sought by examining resin embedded samples. Cells displaying peripheral nuclear chromatin condensation, some in the classical crescent shape, were seen with increasing frequency towards E19 (Fig. 4c). There was occasional evidence of apoptotic body formation, but other classical characteristics, such as plasma membrane blebbing, could not be resolved with light microscopy. The position of the nuclei suggested that the majority was within the syncytiotrophoblast, but the involvement of cytotrophoblast and endothelial cells could not be excluded at the light microscopic level. 3.5 Ex-vivo culture experiments To test the relationships between oxidative stress and the induction of COX enzymes and apoptosis experimentally, placental explants were subjected to hypoxia/reoxygenation (H/R), a known stimulus for the generation of reactive oxygen species [22]. Western blotting confirmed increased concentrations of HNE and COX-1 in the explants following H/R compared to control samples frozen at the start of the experiment (time zero) or cultured under normoxic conditions. For COX-2 a significant increase was only observed compared to the time zero controls (Fig. 5a,b). IHC localised the HNE to individual cytotrophoblast cells and to the syncytiotrophoblast layers of the labyrinth. In the JZ individual spongiotrophoblast cells were immunopositive, but the strongest staining was seen in the GCs within the JZ and decidua. Dual-labelling revealed strong co-localisation for COX-1 and COX-2 in the spongiotrophoblast and GCs in the JZ, and within the syncytiotrophoblast and cytotrophoblast cells in the labyrinth (Fig. 5c). IHC revealed increased immunoreactivity for M30 within the trophoblast following H/R. The apoptotic index based on TUNEL labelling increased from 2.8% and 1.6% in the time zero and normoxic controls respectively, to 20.6% following H/R (P = 0.006). The majority (78%) of TUNEL positive cells were also immunoreactive for HNE on dual-labelling (Fig. 5d). From resin-embedded material it appeared that most of the cells with condensed peripheral chromatin following H/R were within the syncytiotrophoblast layers (data not shown). From these data it can be concluded that there is a significant increase in oxidative stress following H/R, and that this is associated with increased expression of COX-1 and COX-2 expression in the LZ and JZ, and with increased apoptosis in the labyrinth. 4 Discussion These results indicate that oxidative stress increases with gestational age in the murine placenta during normal pregnancies, and that it may play a significant physiological role by inducing higher concentrations of the COX-1 and COX-2 enzymes, and hence increasing prostaglandin synthesis. Our data also indicate a close temporal and spatial association between oxidative stress and trophoblast apoptosis within the labyrinth. Trophoblast apoptosis has been linked to the pathophysiology of preeclampsia, and the mouse might provide a useful genetic model in which to elucidate the mechanisms underlying the shedding of apoptotic debris. We have recently proposed that placental oxidative stress arises through fluctuations in oxygenation [3,25], and its development in the murine placenta towards term is consistent with this view. As fetal and placental oxygen extraction reach their peaks so any transient mismatch between maternal oxygen supply and feto-placental demand will lead to a dip in the intraplacental oxygen concentration. The fact that H/R in vitro is a powerful inducer of oxidative stress, and that the stress localises to the same tissues as in late gestation supports this hypothesis. It is now accepted that oxidative stress induces a spectrum of cellular changes, ranging from the physiologically homeostatic to the frankly pathologic. Induction of enzymes is towards the more physiological end of that spectrum, whereas apoptosis is towards the opposite extreme. Induction of the COX enzymes has been linked to oxidative stress through activation of the p38MAPK and the NF-κB family of transcription factors in other cell types [26–30]. In our study there appeared to be strong co-localisation between the formation of HNE and expression of COX-1 and -2 to individual cytotrophoblast, spongiotrophoblast and glycogen cells, both in vivo and in vitro. However, there was an apparent discrepancy between the Western blot results and our IHC findings, for the former indicated a peak in oxidative stress and COX-1 at E16 while the staining intensity for HNE and COX-1 increased until E19. This difference can be accounted for by the fact that the number of GCs declines by about half between E16 and E18 [31], so reducing their contribution to the overall tissue homogenate. Expression of the COX enzymes appears to be mainly altered in placental rather than decidual tissues. Changes with gestational age were observed in the trophoblast populations within the LZ, but the greatest changes in immunoreactivity were seen within the GCs and the spongiotrophoblast cells in the JZ. Expression of COX-1 and COX-2 has recently been reported in the rat placenta, and that of COX-2 similarly increases with gestational age, particularly in the JZ [32]. The placental cells in this zone have direct access to maternal blood as they line the maternal venous sinuses. Therefore, it is possible that PGs synthesized in this zone in response to oxidative stress are released into the maternal circulation. It is also notable that the GCs migrating into the DB are strongly immunoreactive for COX, and so PGs synthesised by these cells could exert local paracrine effects in the decidua. As PGs have been linked with the induction of parturition [10,11,33] this raises the possibility that placental oxidative stress could play a role in initiating or augmenting uterine contractions should the mismatch in maternal blood supply and feto-placental demands become extreme. Trophoblast apoptosis increased late in gestation as assessed by IHC for M30 and TUNEL, and cell morphology. There was a close temporal and spatial co-localisation with oxidative stress, although on Western blotting evidence of oxidative stress appeared to decline in the last days of gestation. This may again reflect changes in the cell populations within the placenta and associated decidua. The apoptotic index based on TUNEL-positivity showed a continual increase until term, although M30 immunoreactivity suggested a decline after E19. There is strong evidence that following initial cleavage, exposing the M30 epitope, cytokeratin 18 becomes further cleaved to smaller fragments [24]. In these late stages of apoptosis the M30 epitope may be lost, and so the antibody no longer detects all apoptotic trophoblast [34]. A causal link between oxidative stress and apoptosis was strengthened by the fact that apoptosis could be induced by H/R in vitro. As in the human placenta this was associated with increased concentrations of active caspase-3 [20]. The mitochondrial pathway was strongly implicated in the apoptotic process in the human, but further work is required to determine whether the same mechanism operates in the mouse. The increase in oxidative stress and apoptosis in the LZ with gestational age may explain some of the growth dynamics of the murine placenta. Overall placental volume increases with gestational age until E16.5 and then plateaus, coinciding with the rise in oxidative stress. It is notable that within the LZ the volume and surface area of the trophoblast plateaus at E16.5 whereas those of the fetal capillaries continue to increase until E18.5 [35]. The fetal endothelial cells did not display immunoreactivity for HNE or nitrotyrosine, and did not appear to label with TUNEL, suggesting that oxidative stress and its consequences are lower in this cell type. The links between oxidative stress and apoptosis suggest that the mouse may provide a useful genetic model in which to investigate the relative roles of antioxidant enzymes and signalling pathways in regulating trophoblast apoptosis. Apoptosis increases towards term in normal human pregnancies [36], but increased rates have been reported in preeclamptic placentas and in placentas from cases of intrauterine growth restriction [37–39]. It is thought that microparticles of apoptotic, or aponecrotic, debris released from the apical surface of the syncytiotrophoblast into the maternal circulation result in an enhanced maternal inflammatory response in preeclampsia [6]. Whether apoptotic debris is shed from the murine labyrinth as in the human placenta towards the end of gestation remains to be determined.

Int_J_Colorectal_Dis-3-1-2039795

The sentinel node procedure in colon carcinoma: a multi-centre study in The Netherlands

Background Lymph node status is the most important predictive factor in colorectal carcinoma. Recurrences occur in 20% of the patients without lymph node metastases. The sentinel lymph node (SLN) biopsy is a tool to facilitate identification of micrometastatic disease and aberrant lymphatic drainage. We studied the feasibility of in vivo SLN detection in a multi-centre setting and evaluated nodal micro-staging using immunohistochemistry (IHC). Introduction Survival in patients with colon carcinoma is strongly correlated with lymph node status: the 5-year disease-free survival rate is 70–80% for patients with lymph node-negative disease (stage I/II) but only 45–50% for those with node-positive disease (stage III) [1]. The presence of lymph node metastases indicates the use of adjuvant chemotherapy in these patients, which increases the 5-year survival rate with about 10% [2]. Despite the favourable prognosis of patients with localized colon carcinoma without regional lymph node metastasis, 20–30% of these patients will develop recurrent disease after apparently curative resection [3]. It is possible that in this group of patients, small lymph node metastases have been missed, resulting in understaging. This may be due to an inadequate surgical lymphadenectomy or insufficient pathological examination [4]. According to international guidelines, meticulous pathological examination of at least 12 lymph nodes is warranted for adequate staging of patients with colon carcinoma [5]. However, several studies showed that the minimal number of lymph nodes necessary for correct staging varied considerably from 6 to 18 to as many as possible in the study of Goldstein et al. [4, 6–9]. In addition, in-depth pathological examination of lymph nodes by immunohistochemical staining for cytokeratin or reverse transcriptase polymerase chain reaction (RT-PCR) may reveal micrometastases that could have been missed by routine haematoxylin and eosin (H&E) examination. Several authors have reported a decreased survival rate when micrometastases are detected in colon carcinoma [10–13]. The possible benefit of adjuvant therapy in this group of patients is not clear yet. These (ultra-)staging techniques are time consuming, labour intensive and costly. For optimal staging, in depth examination of only the sentinel lymph node (SLN) could be helpful. In colon carcinoma, the SLNs are defined as the first one to four blue-stained nodes with the most direct lymph drainage from the primary tumour. They have the greatest potential to harbour metastatic disease when present, enabling focussed examination with multi-level micro-sectioning of the SLNs to provide a more efficient and cost-effective detection of micrometastases. In addition, patterns of aberrant lymphatic drainage can be visualized with SLN mapping, which may lead to a more extended resection. Several studies have reported varying results of the SLN procedure in colon carcinoma [14–20]. This study presents the results of the SLN procedure in six Dutch hospitals. The primary aim of this study is to test the accuracy and sensitivity of the SLN procedure in a multi-centre setting. Furthermore, we looked at upstaging and possible aberrant lymphatic drainage. Materials and methods This study was performed between May 2002 and May 2005 in five teaching hospitals and one university hospital. All procedures were supervised by one of the coordinating surgeons (Plukker, Braat). Only patients with histologically proven primary colon carcinoma were included in the study. Patients with distant metastases or gross lymph node involvement as shown by pre-operative examinations or palpation during surgery were excluded. The procedure was only performed when one of the study coordinators was available for supervision (Plukker, Braat, Kelder). The study was approved by the local scientific ethics committee, and all patients had given informed consent. Patients with rectal cancer were excluded from the study. SLN mapping was carried out through an open procedure by injection of 1–3 ml Patent Blue with a tuberculin syringe and 29-gauge needle sub-serosally in four quadrants around the tumour. The sub-serosal injection was carried out before vascular ligation. Within 5 to 10 min after the blue dye injection, the SLN could be identified by following the blue-stained lymphatic vessels leading to the blue-stained SLN. These lymph nodes were tagged with a long suture. SLNs were defined as the first one to four blue-stained lymph nodes seen within the regional basin. After marking of the SLNs, routine resection was performed. If the SLN was found outside the normal lymphatic basin, we performed an extended resection. The tumour and all lymph nodes were examined according to standard guidelines [5]. If the SLNs were negative after routine H&E staining, they were sectioned at 150-μm intervals and examined at three levels with H&E as well as immunohistochemistry on cytokeratins (CK7/8 or 18). Metastases between 0.2 and 2 mm were referred to as micrometastases. Metastases smaller than 0.2 mm were referred to as isolated tumour cells [5]. Upstaging was defined as the presence of micrometastases or isolated tumour cells after immunohistochemistry (IHC) in patients with a negative lymph node status after H&E. Definitions See Fig. 1. Identification rate is the number of patients with one or more SLNs identified (b)/the total number of procedures (a) × 100%. Negative SLNs were false negative if one of the other regional lymph nodes (non-SLNs) were tumour positive (d). The accuracy of the SLN procedure suggests a conformity of the SLN status and the regional nodal status, i.e. the total number of patients with a positive SLN (c) + the number of patients with a true-negative SLN (e)/the number of patients with an identified SLN (b) × 100%. Sensitivity is the number of patients with a positive SLN (c)/The total number of node positive patients (c + d) × 100%. Upstaging is the number of patients with positive SLNs by IHC (g)/the number of patients who were node negative by H&E examination (e + g) × 100%. Fig. 1Flowchart. SLN Sentinel lymph node, pts patients Results The SLN procedure was performed in 69 patients. Tumour characteristics are shown in Table 1. Figure 1 shows the total number of patients and SLN results. At pathological examination, a mean of 11 nodes per specimen was found; per hospital, this varied between 9 and 17 (9, 10, 12, 14, 17, 17, respectively). The mean number of SLNs was 2.3 per patient. The SLN was identified in 67 out of 69 patients (97%). One of the two failed procedures was in a patient with a carcinoma in the sigmoid colon surrounded by a concurrent diverticulitis. The other patient had extended lymph node metastases with angio-invasion at pathological examination. In 28 patients, lymph node metastases were identified at the pathological examination; this includes the one case where the SLN procedure failed because of extensive lymph node metastases (28 of 69, 41%). This one case with lymph node metastases and a failed procedure was excluded from further statistical analysis on the SLN procedure, leaving 27 node-positive patients in the final analysis. In 24 patients, the SLN was positive, either with H&E staining or with IHC, resulting in a sensitivity of 89% (24 of 27) in the group of 67 patients with a successful SLN procedure. If we leave out the patients who had a positive sentinel node only after IHC, the sensitivity is 15 of 18 (83%). In 15 of 24 SLN-positive patients, the SLN was the only involved lymph node (63%). In 9 of 27 lymph node-positive patients, metastases were found only after IHC. In four patients, these were micrometastases, whereas in five cases, isolated tumour cells were found. Therefore, without IHC, the number of node-positive cases would have been 27 minus 9, which is 18. This corresponds to a total of 49 node-negative cases by H&E in the group with a successful SLN procedure. With IHC, the upstaging is 9 of 49 or 18%. The SLN was negative in 43 patients. In 40 patients, the non-SLNs were also negative. This results in a negative predictive value of 93% (40 of 43). One of the three patients with a false-negative SLN had lymph node metastases with extra-nodal growth in the non-SLN. In another patient, a small tumour deposit was found in the mesocolon right next to the primary tumour. This was classified as N1 according to the American Joint Committee on Cancer classification, although it is unclear whether this is a true lymph node metastasis or some kind of ‘in transit’ metastasis. The last patient with a positive non-SLN showed micrometastases at H&E examination in a small peritumoural lymph node. Aberrant lymphatic drainage was seen in three patients (4%). In two cases, the SLN was found on the left side of the middle colic artery in patients with a tumour in the ascending colon. In both cases, an extended right hemicolectomy was performed. The third patient had a tumour near the rectosigmoid junction with a high para-aortal SLN. Therefore, we performed an extended left sided resection en-bloc with a partial para-aortal dissection. None of these lymph nodes contained metastases. All other SLNs were found in the mesocolon in close proximity to the tumour. In these cases, the central lymph node as identified by the pathologist was always a non-SLN. The accuracy of the SLN procedure in this study was 96%, as the pathological status of the SLN corresponds with the definitive lymph node status in 64 of the 67 patients. Table 1Tumour characteristicsCharacteristicsValuesTumour location Right colon35 Left colon2 Sigmoid colon32T-stage 11 214 348 46Mean number of lymph nodes11Mean number of SLN2.3 Discussion With an identification rate of 97%, accuracy of 96%, sensitivity of 89% and negative predictive value of 93%, this study shows that it is possible to perform the SLN procedure properly in patients with localized colon carcinoma in a multi-centre setting. Other multi-centre studies showed varying results of this technique (Table 2) [14–21]. Our results correlate with those from other larger studies, which show accuracy and sensitivity rates of 95–98 and 89–93%, respectively [15, 18–20]. Most smaller studies show worse results with low accuracy and success rates and corresponding low sensitivity rates and negative predictive values [14, 17, 21]. In one study, the time between injection of the blue dye and identifying the SLN was too long, leading to a larger number of SLNs [17]. It is very likely that not all of these blue nodes were true SLNs. In the study by Bertagnolli et al. [14], 79 patients were operated on by 25 different surgeons in 13 different hospitals. A mean of three procedures per surgeon seems insufficient to adequately learn this technique. It is known that the learning curve of the SLN in colon carcinoma stabilizes after about five procedures [20]. To minimize technical failures, the procedure in our study was performed by a few surgeons under direct supervision of one of the two surgeons coordinating this study (Braat, Plukker). Apart from too few procedures, the worse results in some studies might be explained by inclusion of patients with advanced disease. Some studies included patients with clinically apparent stage III or stage IV disease [21]. Widespread lymph node metastases could result in obstruction of lymphatic channels, and lymphatic drainage is bypassed to other (non-sentinel) lymph nodes. This phenomenon is called skip-metastasis. It was noted in one of the patients with a false-negative SLN in our study who had advanced lymphatic metastases with extra-nodal growth. Patient selection is therefore important for a reliable SLN procedure in colon carcinoma. In fact, the SLN procedure is not useful in patients with clinically apparent stage III or stage IV disease as false-negativity rates will be higher. Moreover, in these patients, metastases will be easily found at routine pathological examination, and the SLN procedure will not have any additional value. The SLN procedure could be useful in those patients with (micro)metastases that would not be identified with routine pathological examination. Furthermore, the failed procedure in one of our patients with concurrent diverticulitis also suggests the importance of an undisturbed lymphatic drainage for a successful SLN procedure. Table 2 Results of multi-centre studies of the SLN procedure in colon cancerStudyNumber of patientsNumber of centresIdentification rate (%)Accuracy (%)Sensitivity (%)Upstaging (%)Bilchik et al. [15]40310010010010Saha et al. [18]131399979216Bertagnolli et al. [14]72139281420Read et al. [21]3827976253Kelder and Braat69697968913 or 18 We saw aberrant lymphatic drainage in three patients (4%). This percentage correlates with the literature [15, 18, 20]. In this study, none of these aberrant SLNs showed metastases. However, potentially, these aberrant SLNs are the only lymph nodes containing metastasis, as shown in a previous study [22]. In an experimental situation, it seems justified to perform an extended resection in these cases. Further study should be performed to justify an extended resection in the daily practice. Literature not clearly indicates how many nodes should be examined to accurately predict lymph node status [6–8, 23, 24]. One study showed that a colon specimen usually contains about 50 lymph nodes and that more than 70% of the lymph nodes containing metastases are smaller than 5 mm [9]. It is also known that the prognosis in node-negative patients with colon carcinoma is better when more lymph nodes have been examined [24]. Taking this into account, the pathologist takes only a sample of the lymphatic basin of a resected colon specimen, even when international guidelines are followed, which state that at least 12 lymph nodes are needed for adequate staging [5]. The mean number of 11 lymph nodes in our study is not enough to predict lymph node status according to the international guideline. This fact could theoretically lower the chance to detect metastases in non-SLN and thus could lower the false-negative rate. However, we did not find any differences in false negative rates between the two hospitals with a mean number of nine and ten examined nodes (40 cases) and the hospitals with more than 12 examined nodes (29 cases). With regards to upstaging, most studies show an upstaging of 10–16% [15, 18–20]. However, they calculated upstaging by dividing the number of IHC-positive patients by the total number of patients (Fig. 1g/a, 9 of 69, 13% in our group). We think it is better to consider upstaging solely in the H&E node-negative group, as this is the group to be upstaged by IHC. Using this method, we find 18% upstaging in our series. In addition to this true upstaging, patients with a SLN as the only site of metastases could have been ‘possibly upstaged,’ as conventional pathological dissection of the mesentery might have missed this lymph node. The SLN procedure with patent blue might be able to improve adequacy of the lymph node examination by selecting the right lymph nodes, even small nodes less than 5mm, to be examined in depth by the pathologist. We found the SLN to be the single lymph node with metastasis in 15 (21%) of the patients (Fig. 1 f). ‘Possible upstaging’ might play a role here, but we cannot prove this. As we believe that even isolated tumour cells are important for staging, we assigned patients with micrometastases or isolated tumour cells to the group of node-positive patients. It must be remarked, however, that these cases were also used for the calculation of upstaging. Our idea of the biological importance of micrometastases and isolated tumour cells is based on a recent meta-analysis that showed that micrometastases detected retrospectively by RT-PCR correlated better with overall survival than IHC and carried significant prognostic value [12]. Regarding the detection of micrometastases, two studies showed a high reliability of the SLN concept to predict micrometastases and/or isolated tumour cells also in non-SLNs. Therefore, it seems sufficient to perform IHC only on the SLN, while examining the non-SLNs with H&E [25, 26]. Prospective studies are needed to evaluate the potential benefit of systemic chemotherapy in patients with these micrometastases. A reliable SLN procedure might facilitate this intensive pathological examination by allowing focussed examination of only the SLN and thereby aid in a better patient selection for adjuvant therapy in the future.

Acta_Neuropathol_(Berl)-3-1-1824787

Intraneuronal Aβ immunoreactivity is not a predictor of brain amyloidosis-β or neurofibrillary degeneration

Amyloid β (Aβ) immunoreactivity in neurons was examined in brains of 32 control subjects, 31 people with Down syndrome, and 36 patients with sporadic Alzheimer’s disease to determine if intraneuronal Aβ immunoreactivity is an early manifestation of Alzheimer-type pathology leading to fibrillar plaque formation and/or neurofibrillary degeneration. The appearance of Aβ immunoreactivity in neurons in infants and stable neuron-type specific Aβ immunoreactivity in a majority of brain structures during late childhood, adulthood, and normal aging does not support this hypothesis. The absence or detection of only traces of reaction with antibodies against 4–13 aa and 8–17 aa of Aβ in neurons indicated that intraneuronal Aβ was mainly a product of α- and γ-secretases (Aβ17–40/42). The presence of N-terminally truncated Aβ17–40 and Aβ17–42 in the control brains was confirmed by Western blotting and the identity of Aβ17–40 was confirmed by mass spectrometry. The prevalence of products of α- and γ -secretases in neurons and β- and γ-secretases in plaques argues against major contribution of Aβ-immunopositive material detected in neuronal soma to amyloid deposit in plaques. The strongest intraneuronal Aβ17–42 immunoreactivity was observed in structures with low susceptibility to fibrillar Aβ deposition, neurofibrillary degeneration, and neuronal loss compared to areas more vulnerable to Alzheimer-type pathology. These observations indicate that the intraneuronal Aβ immunoreactivity detected in this study is not a predictor of brain amyloidosis or neurofibrillary degeneration. The constant level of Aβ immunoreactivity in structures free from neuronal pathology during essentially the entire life span suggests that intraneuronal amino-terminally truncated Aβ represents a product of normal neuronal metabolism. Introduction Neurofibrillary degeneration and brain amyloidosis with deposition of fibrillar amyloid β (Aβ) in plaques are diagnostic features of Alzheimer’s disease (AD). Intracellular processing of amyloid precursor protein (APP) with β- and γ-secretases generates Aβ1–40 and Aβ1–42 in the endoplasmic reticulum, trans-Golgi network, and endosomal–lysosomal system [10, 19, 25, 62, 66]. An amino-terminally truncated 3-kd peptide (Aβ17–40/42) is the product of APP cleavage with α- and γ-secretase. Cells may produce and secrete several species of Aβ, including Aβ1–40, Aβ1–42/3 and Aβ17–40/42 [8, 23, 24]. Human neurons are Aβ-immunoreactive [11, 17, 22, 38, 39, 57]. The nature, distribution, and role of intraneuronal Aβ are the subject of controversy (see review by Takahashi [58]). Intraneuronal Aβ immunoreactivity has been localized in lipofuscin deposits [4, 65], cathepsin D-positive vesicles of lysosomal origin [11], multivesicular bodies within presynaptic and postsynaptic compartments [57], and intracellular and extracellular neurofibrillary tangles (NFTs) [1, 21, 26, 27, 34, 42]. Several studies of cytoplasmic Aβ immunoreactivity in neurons and glial cells in the human brain have been conducted to determine the properties of Aβ in human neurons and its role in fibrillar plaque formation [11, 17, 22, 39, 57]. The key observation has been the absence of [64] or only minimal intraneuronal Aβ immunoreactivity [22] in normal brain. Therefore, the appearance of or an increase in Aβ immunoreactivity has been suggested as a sign of neuronal pathology [22] leading to fibrillar plaque formation in the brain of people with AD [11, 15, 22, 39, 63]. Overexpression of APP, overproduction of Aβ, and early intracellular accumulation of Aβ [33] have been considered the foundation for early onset of AD pathology and functional deterioration in adults with Down syndrome (DS) in their 40’s [22, 39, 63]. The loss of Aβ immunoreactivity in areas of plaque formation has led to the conclusion that neurons release intracellular Aβ, which initiates a seeding process leading to plaque formation [39]. According to Bahr et al. [2], the death of neurons, a prominent feature of AD, is associated with the release of oligomerized intracellular Aβ42 into the surrounding milieu, which may stimulate the production of amyloidogenic fragments of APP, amplify the levels of intracellular Aβ in neighboring cells, and act as a nidus for the deposition of secreted Aβ. The association of intracellular Aβ42 with intraneuronal tangles has been considered an indication that the growing concentration of Aβ42 may contribute to NFT formation [40]. The presence of altered aspartyl residues in intracellular NFTs has been interpreted as indicating that racemized Aβ peptides are involved in neurofibrillary degeneration [41, 52]. The pattern of intraneuronal Aβ immunoreactivity observed in the control cohort in our previous study [61] was in conflict with the hypothesis that Aβ immunoreactivity in neurons is the key and an early event in the cascade of pathology leading to AD. Therefore, the aim of this study was to reexamine the hypothesis that intraneuronal Aβ immunoreactivity is an early manifestation of Alzheimer-type pathology leading to the formation of fibrillar plaque and/or neurofibrillary degeneration. To test this hypothesis, we examined patterns of Aβ accumulation in the human brain during development, adulthood, and aging, and patterns of changes in the amount and distribution of Aβ in neurons in people with DS and DS/AD, and sporadic AD. Materials and methods Human tissue The brains of 99 individuals were examined. Control brains included 32 cases, from 3 months to 102 years of age (3-, 6-, 11-, and 13-month-old infants; 4-, 8-, 14-year-old children; and adults from 23 to 102 years of age; 19 males and 13 females). DS brains included 31 cases from 3 weeks to 72 years old (infants that were 3 weeks, and 3, 6, and 9 months old and adults from 28 to 72 years of age; 15 males and 16 females). In the AD cohort, the brains of 36 subjects from 65 to 97 years of age (17 males and 19 females) were studied. Diagnostically, three subjects had a mild cognitive impairment (MCI) corresponding to global deterioration scale (GDS) stage 3 [45, 46], a frequent clinical precursor of AD. GDS stage 4 (mild AD) was reported for three subjects; stage 5 (moderate AD) for three subjects; stage 6 (moderately severe AD) for 11 subjects, and stage 7 (severe AD) for 16 subjects. One brain hemisphere was fixed in 10% buffered formalin for 1.5 to several months and then dissected into 1 cm thick slabs. The tissue blocks were dehydrated for 5 days in 70% ethanol, then 2 days in 80% ethanol, and finally 1 week in 96% ethanol. The samples were infiltrated with polyethylene glycol (PEG) 400 (Merck #807,485) for 6 days (two changes of 3 days each, at room temperature) and with PEG 1000 for another 6 days (two changes of 3 days each, at 42°C). The slabs were embedded in fresh PEG 1000 [28]. The tissue blocks were cut serially at 50 μm thick sections. The immunoproperties of intracellular Aβ in tissues fixed in formalin for 1.5 to several months were also compared with immunostainings of sections from five brains (two DS and three AD) prefixed in 4% paraformaldehyde for 2 h, dissected into 1 cm thick slabs, fixed in 4% paraformaldehyde for the next 24 h, cryoprotected with 15 and 30% solutions of sucrose (4 days), frozen, and cut serially into 50 μm thick sections. The methods selected for this study were approved by the New York State Institute for Basic Research (IBR) Institutional Review Board. The tissue samples were provided by the Silberstein Institute for Aging and Dementia at New York University, the Brain Bank at IBR, and the University of Miami Brain and Tissue Bank. In postmortem examination, all samples were identified only by an anonymous case number, and tissue was examined blind to clinical and demographic information. Immunostaining Several antibodies were applied to serial sections to evaluate Aβ immunoreactivity in cortical and subcortical structures, the cerebellum, and brainstem. Monoclonal antibodies (mAbs) 6E10 and 6F/3D were used for characterization of the amino-terminal portion of Aβ (Table 1). mAb 6E10 recognizes an epitope in residues 4–13 of Aβ (Signet Laboratories, 1:10,000) [30, 36, 59]. mAb 6F/3D recognizes an epitope in residues 8–17 of Aβ (Novocastra Laboratories Ltd; NCL-β-amyloid). The central portion of Aβ was detected with mAb 4G8, which recognizes an epitope in residues 17–24 of Aβ [29]. The carboxyl terminus of Aβ was characterized with pAbs purified from rabbit serum by epitope-specific affinity chromatography. These antibodies react with Aβ residues 32–40 (Catalog nr. 44–348) and 32–42 (Catalog nr. 44–344; BioSource International, Inc., CA, USA). The reactivity towards other species of Aβ peptides was eliminated through a series of preabsorption steps. Purified rabbit polyclonal antibodies, R164 and 165, specific for the carboxyl terminus of Aβ1–42 (residues 35–42) were also used [37]. Fibrillar Aβ was detected with rabbit polyclonal antibody R262 produced by immunization of rabbits with fibrillar Aβ1–42. Rabbit antibodies were purified according to the protocol described by Miller et al. [36]. Table 1Mouse monoclonal and rabbit polyclonal antibodies used for immunocytochemistryTypeNameEpitopeDilutionSourcemAb6E104–13 aa Aβ1:10,000Signet Laboratories (developed at IBR)mAb6F/3D8–17 aa Aβ1:50NovocastramAb4G817–24 aa Aβ1:8,000IBR (Dr. R. Kascak)pAb44–34832–40 aa Aβ1:500BiosourcepAb44–34432–42 aa Aβ1:500BiosourcepAbR16435–42 aa Aβ1:500IBR (Drs. D.L. Miller and P.D. Mehta)pAbR16535–42 aa Aβ1:500IBR (Drs. D.L. Miller and P.D. Mehta)pAbR262Fibrillar Aβ 1–42aa1:200IBR (Drs. D.L. Miller and P.D. Mehta)mAbTau1189–207 aa of tau1:100,000IBR (Dr. R. Kascsak) Monoclonal antibody 6E10 reacts with Aβ both on western blots and in formalin fixed material [13, 14]. mAb 6E10 binds APP on western blots, as a native protein, but does not immunoreact with APP after fixation with formalin and dehydration. We have also shown that in cultures of cells overexpressing APP and formalin fixed and dehydrated both antibodies, 6E10 and 4G8 do not detect APP but they detect Aβ [13]. Also, the study of cultured cells with elevated levels of APP and C-terminal fragments of APP revealed that mAb 4G8 does not detect APP and C-terminal fragments of APP by immunocytochemistry [12] (Table 2). Table 2Aβn−40 and Aβn−42 concentration (pmol/g) in temporal cortexControl case #Aβn−40Aβn−42p3p4p3p42472210.325621211,1690.30.30.2a0.2aBrain samples were homogenized in formic acid as described in Materials and methods. Aliquots of the extracts were neutralized and subjected to PAGE and immunobloting. p4 bands migrated above the 3 kDa standards and bound mAb 6E10. p3 bands migrated near the 3 kDa standard and bound R162 or R226 but did not bind mAb 6E10aThere was only one band in this extract. Its level was too low to be revealed by mAb 6E10 and so could not be identified as p3 or p4 The endogenous peroxidase in the sections was blocked with 0.2% hydrogen peroxide in methanol. To enhance immunoreactivity of Aβ, sections were treated with 90% formic acid for 30 min [31]. The sections were then treated with 10% fetal bovine serum in phosphate buffer solution (PBS) for 30 min to block nonspecific binding. The antibodies were diluted in 10% fetal bovine serum in PBS and were incubated with sections overnight at 4°C. The sections were washed and treated for 30 min with either biotinylated sheep anti-mouse IgG antibody or biotinylated donkey anti-rabbit IgG antibody diluted 1:200. The sections were treated with an extravidin peroxidase conjugate (1:200) for 1 h and the product of reaction was visualized with diaminobenzidine (0.5 mg/ml with 1.5% hydrogen peroxide in PBS). After immunostaining, the sections were lightly counterstained with cresyl violet. Phosphorylated tau protein of neurofibrillary tangles was detected with mAb Tau-1 (1:100,000). Tau-1 recognizes an epitope in residues 189–207 of the human tau sequence [16]. To obtain optimum staining with Tau-1, sections were treated with alkaline phosphatase (Sigma, Type VII-L, 400 μg/ml in PBS, pH 7.4, 0.01% H2O2) [20]. A three-point classification was used to estimate semi-quantitatively the difference of immunoreactivity with mAb 4G8 in four brain structures of 25 control adult subjects. The amygdala and cornu Ammonis were selected as structures susceptible to amyloidosis β and neurofibrillary degeneration. The lateral geniculate body and dentate nucleus were selected as structures resistant to AD pathology and almost free of plaques and NFTs even in severe AD. Grade 1 corresponded to weak immunoreactivity present in about 50% of neurons; grade 2 corresponded to moderate immunoreactivity in almost all neurons; and grade 3 corresponded to strong immunoreactivity present in almost all neurons in a given brain structure. Sections from brain of DS subjects were used to determine whether intraneuronal Aβ immunoreactivity differs in people with an extra copy of the gene encoding APP. The difference between intraneuronal Aβ and amyloid in plaques was determined with antibodies detecting the amino- and carboxyl-terminus and the middle portion of Aβ peptide. The relationship of neurofibrillary changes to Aβ immunoreactivity in neurons was evaluated in brains of people with DS/AD and sporadic AD. Partial purification and identification of Aβ peptides Samples of the brain cortex (250 mg) from control males that were 31, 32 and 59 years-old were dispersed in 2.5 ml of 99% formic acid by brief sonification and centrifuged for 30 min at 100,000 g. Supernatants were dried by centrifugal evaporation, the residues were resuspended in 1 ml of 70% (v/v) formic acid and centrifuged as above. The solutions were subjected to size-fractionation on a 1 × 30 cm Pharmacia HR-12 column equilibrated with 70% formic acid as previously described [36]. The Aβ-containing fractions were identified by Western blotting and their contents were quantified by photodensitometry, as previously described [44]. The Aβ-containing fractions were dried under vacuum and were re-dried out of ammoniacal methanol to neutralize traces of formic acid. To solubilize Aβ peptides, each residue was treated with 250 μl of 50 mM ammonia and centrifuged as above, and the supernatant liquid was adjusted to pH 7.4 with KH2PO4. Antibody R287, raised to Aβ27–37 using the previously described methods [40] was purified on a peptide affinity column and was coupled to an Aminolink matrix (Pierce-Endogen) at a concentration of 360 pmol per 200 μl of settled matrix. The Aβ isoforms were immuno-adsorbed to 8 μl of R287-agarose during a 3 h incubation. Following washes with PBS and water the Aβ isoforms were eluted with 2 × 100 μl portions of 2.5% trifluoroacetic acid in 50% acetonitrile. About 50% of total Aβ peptides present in the cortex samples were recovered by this method, as evaluated by Western blotting. The shape and size of the Aβ band in Western blotting of the size-fractionated material and the immuno-purified peptide were the same, which suggested that the process did not selectively enrich any of the Aβ peptides. Mass spectrometry and Western blotting The peptide preparation isolated by immunoadsorption on R287-agarose was dissolved in 30 μl of 40% (v/v) formic acid and approximately 82% was loaded on a Symmetry® C18 nanoAcquity column (180 μm × 20 mm) as six sequential 4.1 μl injections. Following the sixth injection, the peptides were resolved on an Atlantis dC18 nanoAcquity column (100 μm × 100 mm) with a 20 min gradient of 10 to 50% acetonitrile in water containing 0.1% formic acid at flow rate of 0.4 μl/min. The eluate from the column was analyzed directly on a Qtof Micro (Waters Corp.) mass spectrometer equipped with a nanoflow electrospayer and scanned for m/z 50–1,500 at 1.1 s intervals. An external lock mass standard (leucine enkephalin; m/z = 556.2771) was analyzed at 11 s intervals through a separate orthogonal electrosprayer. The data were processed using MassLynx 4.0 (Waters Corp.) software, including Accurate Mass Measure and MaxEnt3 algorithms. Briefly, chromatograms were created from the data set by plotting a 1 Da window around the m/z’s of interest. The peaks were confirmed to have the component of interest by summing the mass spectra across the peak, determining the centroid m/z values, followed by deconvolution and transformation to the accurate mass value of the 1+ ion. The resulting ion envelope encompassing mass isomers containing zero to four 13C atoms was plotted to create the chromatogram of the specific peptide. For Aβ17–40, the 2+ ion was by far the most prevalent and its ion envelope spanned m/z values from 1196.3 through 1198.7. For Aβ1–40, the 4+ and 5+ ions were the most prevalent, spanning m/z values from 1082.8 through 1084.8 and 866.4 through 868.0, respectively. Western blotting was performed by a previously described method, which allows the detection of sub-femtomol quantities of Aβ40 or Aβ42 [44]. Samples and appropriate standards were subjected to PAGE in tris-tricine 16% gels. Rabbit polyclonal antibodies R162 (raised to Aβ31–40) and R226 (raised to Aβ32–42) were used to detect the C-terminal sequences of Aβ isoforms. These antibodies are highly selective for their targets. R162 showed no cross-reactivity with a 100-fold excess of Aβ42 [44], and R226 is 2500-fold more reactive with Aβ42 than with Aβ40 [37]. Monoclonal antibody 6E10 [30] was used to detect the Aβ sequence 4–13 [59]. The blots were developed wth NBT and BCIP and the bands were quantified by photodensitometry [44]. Results Age-associated changes of Aβ immunoreactivity in neurons in control and DS brains Infants and children. In the majority of the brain structures examined, including the entorhinal cortex and neocortex, amygdala, and hippocampus, Aβ immunoreactivity was present in neurons of the 11 and 13 month-old normal infant brains and the 9 month-old infant diagnosed with DS (Fig. 1). At this age, Aβ-immunoreactivity was observed in only about 5–10% of the neurons. However, fine and randomly dispersed Aβ-immunoreactive granules were present in the cytoplasm in almost all neurons in the dentate gyrus. In infants, strong Aβ-immunoreactivity was found in clusters of cytoplasmic granules in almost all neurons in the magnocellular portion of the lateral geniculate body, but reaction in small neurons was rare and weak. In the brains of 4 and 8 year-old children, about half of neurons in the cortex and basal ganglia were Aβ-immunopositive, whereas in the brain of the 14 year-old child, the percentage of immunoreactive neurons was more than 50%. Fig. 1Intraneuronal Aβ immunoreactivity with mAb 4G8 (17–24aa) in the CA1 sector, dentate gyrus (DG), amygdala (AMY), lateral geniculate body (LGB), and cortex (COR; temporal superior gyrus) in 9 month-old infant with DS, 11 month-old control infant, and 33 and 83 year-old control subjects (C) reveals age-associated and structure-specific differences. A few immunopositive neurons are present in CA1 sector, amygdala, and cortex in infants; however, immunopositivity is detectable in the majority of neurons in infant dentate gyrus and lateral geniculate body. In young adult (33 year-old) and aged subject (83 year-old) immunopositivity is present in the majority of neurons and is much stronger than in infants. Dense deposits are present in almost all neurons in the LGB and amygdala, and dispersed fine granular staining is present in the majority of neurons in the dentate gyrus. Loose granular immunopositive material appears in about 80% of neurons in CA1 sector and cortical neurons Adults. In adults, Aβ immunoreactivity was present in the majority of the neurons, but the amount and pattern of distribution of immunopositive material showed a broad range of cell-type- and brain structure-specific differences. Strong Aβ immunoreactivity characterized all nuclei in the amygdala in adults, with the strongest reaction in the lateral and ventral subdivisions (mean grade, 2.3; SD ± 0.4). In the amygdala, numerous Aβ-positive cytoplasmic granules were concentrated at one pole of the cell and formed a perinuclear cap. In the cornu Ammonis (CA), Aβ immunoreactivity showed marked sector-specific differences. Strong and uniform reaction characterized neurons in the CA4 sector, and moderate reaction was observed in CA2 and 3 sectors, but Aβ immunoreactivity in neurons in the CA1 sector was much weaker and less uniform (mean grade in the CA, 1.8; SD ± 0.7). Apical dendrites in pyramidal neurons in the CA and subiculum proper were often marked with rows of Aβ-immunoreactive granules. Granule cells in the dentate gyrus contained numerous randomly dispersed Aβ-positive granules. Strong immunoreactivity appeared in large neurons, and moderate staining appeared in small neurons of the caudate nucleus and putamen. Reactions in neurons in the thalamus, globus pallidus, and n. accumbens were less uniform and weaker. Aβ immunoreactivity in cortical pyramidal neurons was stronger and more common than in cortical granule cells. In the cerebellum, a moderate amount of fine, granular Aβ- immunoreactive material was found in the majority of Purkinje cells, but in granule cells, the reaction was weak and was present only in a minority of cells. Very strong Aβ immunoreactivity was found in the cytoplasm in all neurons in the LGB and dentate nucleus (mean grade 2.8 ± 0.4), as well as in the nucleus olivaris. Aged subjects. In general, in control people older than 65 years of age, Aβ immunoreactivity was reduced in about 20% of neurons of the amygdala, nucleus basalis of Meynert (NBM), cornu Ammonis, large neurons in the caudate-putamen, and cortex, as compared to control adults. However, the amount and pattern of Aβ immunoreactivity in the dentate gyrus, LGB, dentate nucleus, and nucleus olivaris inferior were comparable in aged subjects and normal adults. In addition to the structure- and age-associated differences in neuronal Aβ immunoreactivity, there were interindividual differences between people of the same age and gender. Strong and uniform Aβ immunoreactivity was found in 44% of the control subjects; moderate and uniform in 40%, and weak and nonuniform in 16%. No difference between control and DS infants (Fig. 1) and adult with DS who has incipient neurofibrillary degeneration (28 years-old) was detected. Intraneuronal Aβ immunoreactivity in control brains with neurofibrillary degeneration and in DS/AD or sporadic AD In the control group, mAb Tau-1 revealed neurofibrillary degeneration in the brains of all persons older than 43 years of age. In the entorhinal cortex, subiculum, and amygdala, the number of neurons with NFTs increases with age. The increase in Tau-1-positive material was paralleled by the reduction and loss of cytoplasmic Aβ immunoreactivity in affected neurons in the older subjects. All subjects with DS older than 28 years-old were affected with neurofibrillary degeneration, and all people with DS older than 38 years of age had developed β amyloidosis. The increase in neurofibrillary degeneration, neuronal death, and β amyloidosis was associated with marked reductions of intraneuronal Aβ immunoreactivity in the entorhinal cortex (Fig. 2a, b), hippocampus, amygdala, NBM, and neocortex. Fig. 2Development of neurofibrillary tangles (arrowheads mAb Tau-1, a) in neurons in the second layer of the entorhinal cortex of a 43 year-old subject with DS/AD is associated with loss of cytoplasmic Aβ immunoreactivity (arrowheads mAb 4G8, b) without plaque formation in islands of stellate neurons. Amyloid plaque in the third layer (asterisk) is mAb 4G8-positive. The insula of a subject with mild AD (GDS 4) is affected by early and severe amyloidosis-β. Plaques are marked with arrowheads (c). mAb 4G8-positive material is present in both plaques (asterisk) and neurons (arrowheads, d). The strongest Aβ immunoreactivity is present in almost all neurons (arrowheads) in the lateral geniculate body (low and high magnification; AD, GDS 6; e), dentate nucleus (DS/AD, GDS 7; 72 year-old; f), and nucleus olivaris inferior (DS/AD, GDS 7; 72 years-old; g); however, amyloid plaques do not develop in these structures, even in severe AD A similar pattern of reduction of Aβ immunoreactivity associated with neurofibrillary degeneration, amyloidosis β, and neuronal loss was found in people with MCI and subjects with sporadic AD (GDS stage 4–7). In severe AD (GDS stage 7), almost all neurons in the second layer of the entorhinal cortex were affected by neurofibrillary degeneration detected with mAb Tau-1. These neurons contained only traces of Aβ immunoreactivity or were free of cytoplasmic Aβ. However, in neocortex, the reduction of intracellular Aβ immunoreactivity was less pronounced (Fig. 2c, d). Both in persons with DS/AD and persons with sporadic AD, the percentage of Aβ- immunopositive Purkinje cells was reduced by about 25% compared to age-matched control subjects. The majority of granule cells failed to manifest Aβ immunoreactivity. In the cerebellum, the decrease in neuronal Aβ immunoreactivity was observed in the almost total absence of intraneuronal NFTs. In contrast to these patterns of reduction in Aβ immunoreactivity, a constant level of strong reaction was present in almost all neurons in LGB, dentate nucleus, and nucleus olivaris inferior in people with DS/AD or sporadic AD. These structures were almost free of NFTs and amyloid plaques in all AD subjects, including those with severe AD (Fig. 2e–g). Immunoproperties of intraneuronal Aβ The lack of Aβ immunoreactivity or presence of very few cytoplasmic grains immunoreactive with antibodies detecting an epitope in residues 4–13 (mAb 6E10) and 8–17 (mAb 6F/3D) and the presence of cytoplasmic immunoreactivity with antibodies detecting 17–24 (mAb 4G8), 32–42 (pAb 44–344), and 35–42 (pAbs R164 and 165), indicates that neurons harbor mainly amino-terminally truncated Aβ (Fig. 3a, b, c, d, e, f). Stronger staining of intracellular Aβ with pAb 44–344 (residues 32–42) and pAb R164 (35–42) than with pAb 44–348 (32–40), indicates that the major component of cytoplasmic Aβ in neurons is Aβ17–42. Intraneuronal Aβ was not labeled with pAb R262 that detects fibrillar amyloid in cored plaques and in vessel wall (not shown). The pattern of immunostaining of Aβ in fibrillar plaques confirmed that parenchymal plaques contain mainly Aβ1–42. Fig. 3 a–f shows immunoreactivity of intraneuronal Aβ in the CA4 sector of a 32 year-old control subject. No reaction in neurons stained with mAb 6E10 (4–13 aa) (a) and mAb 6F/3D (8–17 aa) (b) indicates that in the majority of neurons, the amino-terminal portion of Aβ is not detectable. Immunoreactivity of intraneuronal Aβ is shown in sections stained with mAb 4G8 (17–24 aa) (c), pAb R164 (35–42 aa) (d), pAb 44–348 (32–40 aa) (e) and pAb 44–344 (32–42 aa) (f). In the brain of subjects with AD some extracellular ghost tangles contain full length Aβ peptides immunoreactive with antibodies 6E10 (g), 4G8 (h), and pAb 44–344 (i) Aβ immunoreactivity in ghost tangles While neuronal neurofibrillary changes were associated with loss of cytoplasmic Aβ immunoreactivity, some ghost tangles (extracellular NFTs) were Aβ-positive (Fig. 3g, h, i). In the sporadic AD group, among 31 examined subjects, Aβ-positive ghost tangles were found in 24 brains (77%). The proportion of subjects with Aβ-reactive ghost tangles increased from 50% in people with MCI to 70% in people with moderately severe AD and to 100% in people with severe AD. Aβ-positive ghost tangles were found in 100% of these subjects in the entorhinal cortex, in 87% in the CA1, 29% in the amygdala, 21% in the CA2 and CA4, 8% in the subiculum proper, and 4% in the CA3 and temporal cortex. Of AD subjects with Aβ-positive ghost tangles, they were numerous in 41%, moderate in number in 17%, and were rare in 21% of the subjects. In 13 of 24 brains of persons with DS (54%), Aβ-immunoreactive ghost tangles were found in the entorhinal cortex, amygdala, and CA1 and CA2 sectors. The youngest subject with positive staining was 43 years of age at demise. Aβ-positive ghost tangles were found in the brains of all DS subjects from 55 to 72 years old. Of DS subjects in whom Aβ-immunoreactive ghost tangles were present, they were numerous in 46%, moderate in number in 39%, and rare in 15% of subjects. In the control cases, no Aβ-immunoreactive ghost tangles were observed in spite of the presence of neurofibrillary changes in all subjects older than 43 years of age. Aβ-positive ghost tangles were strongly immunoreactive with mAbs 6E10 (residues 4–13), 6F/3D (8–17), and 4G8 (17–24), and pAbs 44–344 (32–42) and R165 (35–42). The reaction with pAb 44–348 (32–40) was weak. Aβ isoforms could be detected on Western blots of crude control brain extracts Antibody R162 revealed 2 bands containing the Aβ32–40 sequence (Fig. 4, lane 3), whereas antibody R226 revealed 2 bands containing the Aβ33–42 C-terminal sequence (Fig. 4, lane 4). Immunoreactivity with mAb 6E10 (Fig. 4, lane 5) unequivocally showed that the upper band contained Aβ peptide sequences possessing amino acid residues 4–13. This antibody does not bind to Aβ17–40 even at a 15-fold higher concentration (Fig. 4, lane 6). We found that the large amounts of other proteins in the extracts affected the electrophoretic migration rates and band shapes of the peptides, so that their migration rates did not unambiguously distinguish Aβ1–40 from Aβ17–40. Using synthetic Aβ standards we showed that substances in a crude brain extract retarded the migration of Aβ17–40 so that it migrated at the rate of Aβ1–40 (in the absence of brain extract). Fig. 4Aβ isoforms in extracts from control brain 247 revealed by immunoblotting. A formic acid extract of cerebral cortex from brain C247 was prepared as described in the methods section. A portion of the dried extract was dissolved in PAGE sample buffer and a 15 μl aliquot containing 30 μg of protein (equivalent to 300 μg of tissue) was applied to the lanes developed with R162 and R226. Because mAb 6E10 is much less sensitive than R162 or R226, the formic acid extract was fractionated by size-exclusion as described in the methods section. The fraction containing peptides in the mass range 2,000–12,000 Da was concentrated, and an aliquot containing peptides from 12 mg of tissue was applied to the lane developed with 6E10. The extracts were subjected to PAGE and blotting along with synthetic Aβ standards and molecular mass markers as described in the text. The blots were developed with antibody R162 (to the C-terminus of Aβ32–40), R226 (to the C-terminus of Abeta33–42) or mAb 6E10 (to Aβ4–13). Lane 1 2 fmol Aβ1–40, lane 2 2 fmol Aβ17–40, lane 6 150 fmol Aβ17–40, lane 7 10 fmol Aβ1–40. Molecular mass markers denoted on the right-hand margin are insulin single chains and aprotinin Peptides with a defined C-terminus may have distinct N-termini, hence, we operationally define the p4 peptides as those that bind mAb 6E10 and migrate between the 3 and 6 kDa standards. Monoclonal antibody 6E10 binds to Aβ residues 4–13; therefore, p4 may contain peptides whose N-terminal residues are Asp1–Arg5. We define the p3 peptides as those that migrate near the 3 kDa standard and do not bind 6E10. The levels of Aβ40 peptides in three normal brain cortex samples were between 0.3 and 2 pmol/g of p3 and between 0.3 and 3 pmol/g, and the levels of Aβ42 peptides were between 0.2 and 2 pmol/g of p3 and between 0.2 and 1 pmol/g. Although these amounts are readily detectable, they are 3–4 orders of magnitude below the Aβ levels in an AD brain sample that we analyzed (not shown). Detection of Aβ17–40 by mass spectrometry We immunoprecipitated 650 fmol of Aβ peptides from brain C247 and characterized both the chromatographic behavior and the masses of the peptides by mass spectrometry using the coupled capillary HPLC-MS system. As shown in Fig. 5, the identity of the Aβ peptide Aβ17–40 from the brain was confirmed by comparison to synthetic Aβ17–40, which eluted at the same time and displayed the same m/z profile. Deconvolution and transformation of the Aβ17–40 2+ ion spectrum to the 1+ monoisotopic mass gave mass values of 2392.2852 Da (brain sample) and 2392.2813 Da (synthetic) compared to the theoretical value of 2392.2950 Da (4.1 ppm and 5.7 ppm errors, respectively). We also observed ion profiles consistent with Aβ1–40 4+ and 5+ ions in the mass spectra. The presence of detergent in the sample and the complexity of the chromatogram and spectra complicated the identification of other Aβ species. Fig. 5Chromatographic elution profile of Aβ17–40. Samples from (a) control brain C247 and (b) synthetic Aβ17–40 were prepared and analyzed by LC-MS as described in “Materials and methods”. The chromatograms show only the MS ion counts for the M+2H+ ions of Aβ17–40 mass isomers containing zero to four 13C atoms (m/z = 1196.3–1198.7). The ion intensities determined at 1.1 s intervals were summed, background subtracted and smoothed. The y-axis scale is 400 ions per second full scale (100%). The identification of Aβ17–40 in the peaks at 24.4 (a) and 24.5 (b) minutes was demonstrated by the characteristic ion envelope obtained from the centroid accurate mass spectra for the Aβ17–40 M+2H+ ions from each peak (24.2 to 24.7 min) shown in the insets Discussion Aβ immunoreactivity in neurons in normal brain Previous studies either did not find intracellular Aβ immunoreactivity in brains of control subjects [64] or found only punctate Aβ immunoreactivity in a minority (15%) of examined adults [22]. Therefore, it was proposed that the accumulation of intraneuronal Aβ is involved in early AD pathology [17]. However, this study of control cases, from several months to 102 years of age demonstrates that intraneuronal Aβ immunoreactivity appears in the first year of life, increases in childhood, stabilizes in the second decade of life, and remains high throughout adulthood. Detection of stable intracellular Aβ immunoreactivity in neurons in control cases throughout the entire life may indicate that Aβ-immunoreactive material in the cell body reflects normal neuronal metabolism and is not neuronal pathology. The observation that intraneuronal Aβ in control, DS/AD, and sporadic AD cases was almost exclusively amino-terminally truncated confirmed and extended the findings of the Mori et al. [39] study of people with DS and the Sergeant et al. [51] study showing that amino-truncated Aβ peptides represent more than 60% of all Aβ species in subjects with preclinical AD. Aβ peptides with N-terminal deletions exhibit enhanced peptide aggregation relative to full-length species [43] and retain the neurotoxicity and β-sheet structure. It was hypothesized that Aβ17–42 peptides may initiate and/or accelerate plaque formation, perhaps by acting as nucleating centers that seed the subsequent deposition of relatively less amyloidogenic but apparently more abundant full-length Aβ [18, 43, 48]. Gouras et al. [17] considered intracellular Aβ42 accumulation to be an early event leading to neuronal dysfunction. However, the appearance of intraneuronal Aβ immunoreactivity in the first year of life, and the stable and strong immunoreactivity throughout adulthood, in the absence of morphological signs of cell injury or degeneration, suggests that Aβ detected in neurons with applied antibodies does not adversely affect cell structure. This lack of fibrillization and toxicity may indicate that these Aβ species remain in inert form. Possibly this inert state is maintained by binding with blockers of fibrillization and toxicity. The transport of Aβ within cytoplasmic vesicles or vacuoles [57] might be another factor preventing the expression of their cytotoxic activity. The detected intraneuronal Aβ appears to be the physiological metabolite with unknown function. Higher levels of secreted APP and nonamyloidogenic secreted APP and lower levels of Aβ 40 in children with severely autistic behavior and aggression compared with controls [54] may indicate that modifications of APP processing and potentially Aβ trafficking are clinically significant in absence of neurodegeneration or neuronal loss. The very strong Aβ immunoreactivity in the LGB, nucleus olivaris inferior, and dentate nucleus during the entire course of human adulthood and the absence of or minimal amyloid load in very severe AD oppose the hypothesis that strong intraneuronal Aβ immunoreactivity is a predictor of fibrillar plaque formation. Convergent findings in PS1 tg mice have been reported by Chui et al. [9]. This study suggests also that the intensity of intraneuronal Aβ immunoreactivity is not a predictor of neurofibrillary degeneration. The strongest Aβ immunoreactivity was observed in neurons in the LGB, nucleus olivaris inferior, and dentate nucleus, which remain free of neurofibrillary degeneration in very severe AD. The moderate Aβ immunoreactivity in the second layer of the entorhinal cortex and pyramidal neurons in the CA1, observed in this study, and the early onset of NFTs observed in these neuronal populations [5–7] also question the link between Aβ immunoreactivity and susceptibility to neurofibrillary degeneration. The lack of pathological changes in brain structures with the strongest intraneuronal Aβ immunoreactivity during the entire human lifespan may indicate that amino-terminally truncated Aβ is a product of cell metabolism that does not (1) limit cell survival, (2) predispose to fibrillar Aβ deposition in plaques, nor (3) cause neurofibrillary degeneration. Reduction of intraneuronal Aβ immunoreactivity in DS/AD and sporadic AD In this study, the strongest intraneuronal Aβ immunoreactivity was found in control cases, compared to those with AD and in DS/AD. Significant weaker Aβ immunoreactivity in neurons in people with DS/AD and sporadic AD appear to be the result of AD pathology, including neurofibrillary degeneration and neuronal loss. In the memory systems of persons with DS/AD or sporadic AD, neurofibrillary degeneration is the major cause of neuronal loss and, therefore, the major factor contributing to the reduction of intraneuronal Aβ. Over the course of 22 years, an AD patient loses 87% of the neurons in the CA1 and 63% in the CA4 sector, and 77% in the subiculum [5]. People with DS/AD older than 50 years of age are affected by severe loss of neurons including 98% loss in the second layer of the entorhinal cortex, 74% in the CA1 sector, 57% in the subiculum proper, and 71% in the amygdala [32, 60]. Large portions of surviving neurons in the brains of people with AD and DS/AD are affected with neurofibrillary changes, yet they show little or no Aβ immunoreactivity. These changes explain, in part, the loss of Aβ immunoreactivity in neurons in the plaque perimeter, as noted in persons with DS/AD [39]. The lower intraneuronal Aβ immunoreactivity observed in people with sporadic AD or DS/AD may also reflect a shift in APP processing. This shift would be from amino-terminally truncated Aβ accumulation in cell cytoplasm in normal brain to APP processing with enhanced generation and secretion of Aβ1–40/42 in people with AD. Accumulation of Aβ1–40/42 in ghost tangles Intracellular and extracellular NFTs have been shown to be Aβ immunoreactive [1, 21, 27, 34, 42, 49, 56]. However, studies of purified PHF revealed the absence of beta-pleated sheet conformation or any other characteristic of an amyloid [50]. It was proposed that the staining of intraneuronal PHF with antibodies to amyloid is due to the proximity of Aβ molecules to the PHF in the cytoplasm of neurons [47]. Deposition of Aβ within extracellular NFTs has been considered a secondary event [67]. Strong immunoreactivity of extracellular NFTs with antibodies 6E10 (residues 4–13), 6F/3D (8–17), 4G8 (17–24), 44–344 (32–42) and R165 (35–42) could be an indicator of the presence of full length of Aβ in extracellular space. The weak reaction of ghost tangles with pAb 44–348 (32–40) may suggest a lower concentration of this form of Aβ in the extracellular space. An increase in the percentage of Aβ-positive ghost tangles in people with progressing AD appears to reflect the increase in the percentage of neurons dying due to neurofibrillary degeneration and an increasing concentration of full length Aβ in the extracellular space. Spatial and temporal separation of plaques and Aβ-positive ghost tangles indicates that ghost tangles may bind Aβ, but do not initiate fibrillar plaque formation. Interindividual differences in intraneuronal Aβ immunoreactivity Intraneuronal Aβ immunoreactivity revealed some interindividual differences. Gouras et al. [17] observed increased apo-E immunoreactivity in Aβ42-immunoreactive neurons and suggested that apo E might be a powerful modifier of intraneuronal Aβ accumulation. Neuron-generated apo-E may bind to neuronal Aβ and affect both its secretory pathway and its storage in the cell cytoplasm. The lack of apo-E in the offspring of βAPP tg mice crossed with apo-E knockouts reduces significantly the amyloid load [3]. Chronic inflammation is another factor that may influence amyloid formation [53]. In the human brain, there appear to be several pools of Aβ in equilibrium [35]. In this study, we demonstrate by the immunocytochemical method abundant intraneuronal amino-terminally truncated Aβ17–40/42 with minor presence of the intraneuronal Aβ1–40/42. The presence of the Aβ17–40 peptide in control cortex samples was confirmed by mass spectrometry. However, in extracts from cerebral cortex we identified by Western blotting the Aβ40 peptides that were amino-terminally truncated (Aβ17–40) and full length Aβ1–40 in similar quantities. We also detected both the Aβ17–42 peptide and the full length Aβ1–42 peptide, although the levels of the latter were 2 or 3 times lower than those of the amino-terminally truncated one. The differences between the Aβ peptides detected in the brain by the immunocytochemical and biochemical methods suggest that the Aβ peptides accumulated intraneuronally are mainly those amino-terminally truncated, while the full length peptides are mainly dispersed in extracellular space, and hence are not detected by the immunocytochemical method as intracellular deposits. Deposition of full length Aβ in ghost tangles and the increase of Aβ-positive ghost tangles could be considered an indicator of the presence of Aβ1–42 in extracellular space and an increase of this extracellular form of Aβ in advanced stages of AD. It appears that the immunocytochemical methods applied in this and other studies [11–14, 17, 22, 39] do not monitor the neuronal secretory pathway of APP processing leading to normal secretion of Aβ1–40 and Aβ1–42. Stern et al. [55] have shown that full length APP is extremely sensitive to fixation methods and its immunogenicity could be easily destroyed. The possibly enhanced secretion of these peptides in AD that may result in fibrillar Aβ deposition in plaques also is not detected. The lack of detection or poor detectability of the product of β- and γ-secretases in autopsy material might be the result of tissue preservation method, the minute amount of Aβ1–40/42 that might be secreted without long-term storage, the prevalence of amino-terminally truncated Aβ masking the detection of Aβ1–40/42, or the masking of the amino-terminal epitopes by interactions with other proteins. One may speculate that deposition of amino-terminally truncated Aβ might be up- or down-regulated by physiological and pathological factors independently of any changes in the secretory pathway. This study suggests that neuron-type- and brain-structure-specific patterns of intraneuronal Aβ immunoreactivity, established in teenagers and maintained at a constant level during adulthood and aging, reflect normal cell metabolism rather than pathological changes. Reduction and loss of immunoreactivity in neurons with neurofibrillary changes appears to be a response to the progression of pathological changes in the neuron. The absence of a link between the age at onset and the progression of accumulation of intraneuronal Aβ-positive material in control subjects with the fibrillar plaque formation or neurofibrillary degeneration in people with AD indicates that the form of intraneuronal Aβ, detected by applied methods, does not predict plaque formation or neurofibrillary degeneration.

Pediatr_Nephrol-2-2-1764601

Focal segmental glomerulosclerosis – epidemiology aspects in children and adults

The histologic features of idiopathic forms of focal segmental glomerulosclerosis (FSGS) were first described by Theodor Fahr in the Handbuch der speziellen pathologischen Anatomie und Histologie in 1925 [1]. Over the subsequent eight decades much has been written about the histologic features and clinical characteristics of patients with FSGS, but it is only in recent years that attention has been directed to the incidence and prevalence of the disorder in various populations. This article will provide an overview of the epidemiology of FSGS by reviewing published surveys of renal biopsies, experiences from clinical registries of children with renal insufficiency, and data from the U.S. Renal Data Systems (USRDS). Diagnostic considerations Primary FSGS can present at any age, and it always causes proteinuria. It is most commonly diagnosed in patients with overt nephrotic-range proteinuria, but any proteinuria that is fixed and persistent over several months (i.e., not orthostatic or transient proteinuria) may signal underlying FSGS [2, 3]. Since the diagnosis of FSGS depends on obtaining renal histologic material, the observed incidence and demographics of FSGS in both children and adults depend on the population examined and on the indications used for renal biopsy. The typical findings on renal biopsy in idiopathic FSGS are described elsewhere in this series of articles. Incidence and prevalence of FSGS in children An estimation of the incidence and prevalence of FSGS in children is hampered by the fact that most children with nephrotic syndrome (NS), unlike the majority of adults, are not routinely subjected to renal biopsy. Therefore, prevalence and incidence figures are extrapolated from clinical reports, registries, and renal biopsy materials. Clinical reports often make the presumptive diagnosis of minimal change nephrotic syndrome (MCNS) based on steroid responsiveness [4, 5]. This usually leads to an under-diagnosis of FSGS because up to 15–20% of FSGS patients initially respond to steroids [6]. The largest pediatric registry is that of the North American Pediatric Renal Transplant Cooperative Study (NAPRTCS), which collects data not only on transplant patients but also on dialysis and chronic kidney disease (CKD) patients [6]. Based on results from nearly 6000 children with CKD, in whom the estimated glomerular filtration rate (GFR) was less than 75 ml/min per 1.73 m2, it has been shown that FSGS carries the highest likelihood of progressing to end-stage renal disease (ESRD), exceeding the risk of hypoplasia, obstructive uropathy, and reflux nephropathy. According to NAPRTCS data, nearly 60% of such children with a diagnosis of FSGS progress to dialysis or transplantation within 24 months of entry into the registry. However, because reporting is entirely voluntary and involves mainly pediatric centers, there is significant underreporting, and selection bias may exist. Clinical surveys from North America and the United Kingdom have reported the incidence of NS to be between two and four new cases per 100,000 children per year, with biopsy-confirmed FSGS comprising 15–20% of the total [5, 7–10]. However, as noted above, many children with NS do not undergo a diagnostic renal biopsy unless they are shown to be steroid-resistant. A study from Canada reported an incidence of FSGS of 0.37 to 0.94 new cases per 100,000 children per year [5], and similar results were obtained in the U.S. Midwest [9]. In the NAPRTCS database, patients with FSGS account for 14.2% of dialysis patients and for 11.5% of transplant patients; the differences may reflect a reluctance by some centers to transplant FSGS patients for fear of disease recurrence. Members of the Southwest Pediatric Nephrology Study Group (SPNSG) reported that FSGS was diagnosed in 75 of 1053 (7.1%) patients with NS who underwent a renal biopsy between 1972 and 1981 [11]. While there is a male preponderance among children with MCNS, most reports have described no clear gender difference in patients with FSGS [4, 11, 12], although in the SPNSG series, all 18 children with FSGS below the age of 3 years were boys [11]. Ethnicity and age appear to play important roles in the prevalence of FSGS. The NAPRTCS and SPNSG reports include approximately equal numbers of Caucasian and African American children with FSGS despite the large difference in the racial distribution of the underlying populations [6, 11]. According to the NAPRTCS report, FSGS is the most common cause of ESRD in African-American children in that cohort, accounting for 23% of all pediatric patients with ESRD. In Caucasian children, FSGS is only the third most common primary disease and is responsible for approximately 10% of all pediatric patients with ESRD [5]. In contrast, African-American children with NS appear to be more likely to have FSGS than MCNS [4, 10, 13]. Whether the frequency of FSGS in Hispanic children differs from that in Caucasians is unclear [14, 15]. The prevalence of FSGS among all children with NS who undergo a renal biopsy increases with age. However, it is not clear if the true population-based incidence of FSGS is age-dependent. Two thirds of children with NS present before the age of 6 years, but only a minority of the younger children exhibit FSGS, whereas FSGS is the most common histology in the older group. The frequency of FSGS in NS patients presenting before 6 years of age is less than 10%, but increases to 20–50% or more in patients presenting in adolescence [4, 9–11, 16]. The combined effect of age and ethnicity is responsible for the reported high frequency of FSGS in kidney biopsies from adolescent African-American patients with NS [4, 10, 13]; both age and ethnicity are probably independent risk factors for FSGS histology. An intriguing recent development has been the apparent increase in the incidence of FSGS in children [4, 5, 10, 17]. Differences in the ages and ethnicity of the subjects do not fully explain these findings. Filler et al., for instance, studied mostly Caucasian children in Canada and found that the calculated incidence of FSGS increased from 0.37 to 0.94 per 100,000 children per year from the first to the second half of a 17-year period between 1985 and 2002 [5]. The increase appears to be present mostly in older children [4]. Similar increases in incidence have not been seen in other types of NS in children [4, 5]. The cause of the increased incidence is not known but is unlikely to be due to genetic factors. The role of environmental pollution has been proposed but remains hypothetical [18]. Morbid obesity-associated FSGS has increased in renal biopsy material [19], but because it remains uncommon, the current adult and pediatric obesity epidemic probably does not explain the increase in FSGS. Incidence and prevalence of FSGS in adults Several publications suggest that the prevalence of idiopathic FSGS in adults may also be increasing [20, 21]. Observations from two metropolitan centers suggest that FSGS was found in 2.5–4% of native renal biopsies in the 1970s, but in 12.2–18.7% in this decade, making FSGS the most common diagnosis based on native kidney biopsies. FSGS becomes less common with advancing age, but it is still a significant problem in the elderly. In a series of 1368 renal biopsies from patients over 60 years of age, FSGS was present in 5.4% of those patients with nephrotic syndrome [20]. These trends in increased numbers of FSGS cases are observed in both African Americans and Caucasians and occurred despite stable or falling rates of other glomerular diseases, such as minimal change disease and membranous nephropathy [21]. Two large series in adults support the fact that primary FSGS is a risk factor for developing ESRD. In one report of outcomes from immunosuppressive therapy in 59 adult patients with FSGS, 30% of patients developed some level of renal insufficiency within 5 years [22, 23]. Wehrmann and colleagues reported results from 250 patients with idiopathic FSGS (average age: 32 years) in which they found an overall 10-year renal survival of 67% [23]. The risk of developing ESRD in idiopathic FSGS can be predicted at the time of diagnosis. Heavy proteinuria, either at the time of biopsy but particularly following treatment, portends a poor long-term outcome. Other features that increase the risk for progressive renal disease include elevated serum creatinine at the time of diagnosis, hypertension, age at time of diagnosis, and male gender [22, 24, 25]. The most compelling predictor for progression of FSGS, however, appears to be the response of proteinuria after treatment is initiated. One series suggests that if a patient with primary FSGS fails to respond to initial therapy (most often steroid-based), the likelihood of renal death is as great as if the patient received no treatment at all [25]. However, a full assessment of relative risk will require larger cohorts of patients with FSGS who received consistent interventions. The risk of FSGS-related ESRD in adults is also supported by data published by the USRDS [26]. Documented FSGS accounts for more than 7000 patients currently receiving ESRD therapy in the United States, and it is likely that primary FSGS accounts for at least some of the 25,000 patients with unspecified forms of glomerulonephritis, the 100,000 patients with ESRD attributed to hypertension, and the 20,000 patients with an unknown cause of ESRD [26]. The USRDS also demonstrates a dramatic increase in the incidence of ESRD due to FSGS, particularly over the last decade in the African-American population. In the Caucasian U.S. population, new-onset ESRD due to FSGS has slowly increased to a rate of approximately five cases/million population in 2003. In contrast, the incidence has increased nearly five-fold in the African-American population since the early 1980s to 30–40 cases per million population [26]. Familial FSGS Occasional cases of familial FSGS have long been recognized. In 1973 Habib suggested that 12% of all pediatric cases of FSGS had an affected sibling [2]. Although most cases of FSGS are sporadic with unknown pathophysiology, it is now clear that mutations in structural podocyte-associated proteins account for a portion of familial cases and for at least some of the sporadic cases. A mutation in the NPHS2 gene, which encodes the protein podocin, has been linked to autosomal-recessive steroid-resistant nephrotic syndrome in children [27]. However, there are large differences between different ethnic populations. For example, approximately one third of Israeli-Arab children but none of Israeli-Jewish or Japanese children with sporadic FSGS were reported to have NPHS2 mutations [27, 28]. The frequency of NPHS2 mutations in North American children is not known. Mutations in the a-actinin 4 gene (ACTN4) have been associated with autosomal-dominant FSGS in adults, a disease characterized by a variable risk of progression to ESRD. Most recently, autosomal-dominant FSGS has been linked to the gene that encodes the transient receptor potential cation channel, subfamily C, member 6 (TRPC6) [29]. Although more data are needed, it is likely that mutations in the known FSGS causative genes account for only a minority of cases of FSGS in children and adults. A more complete review of the genetic aspects of FSGS is provided elsewhere in this series of articles. Conclusion It is clear that there is still much to learn about the epidemiology of idiopathic FSGS. However, there are encouraging signs that the body of knowledge will be enhanced in coming years by the various registries and population-based studies currently underway. Questions (Answers appear following the reference list) Which of the following statements is correct? Patients with idiopathic FSGS often present with microscopic hematuria without proteinuria.All patients with FSGS have nephrotic range proteinuria.Patients with FSGS have proteinuria of varying severity.None of the above.Which of the following carries the highest risk for a patient with FSGS to progress to ESRD? Male gender.Persistent proteinuria after therapy.Hyperlipidemia.Caucasian race.Age.Which of the following is incorrect? Accurate data concerning the outcome of patients with FSGS is lacking since the disease was first described only 30 years ago.The incidence of the disease appears to be decreasing in the last 20 years.Most cases of FSGS can now be linked to specific gene mutations that disturb podocyte function.All of the above.None of the above.Which of the following is correct? FSGS is a more frequent cause of ESRD in Caucasian children than in African-American children.FSGS is more commonly associated with nephrotic syndrome in Hispanic children than in white children.Children with FSGS account for less than 10% of African-American children who progress to ESRD.All of the above.None of the above.

Purinergic_Signal-3-4-2072912

[3H]Adenine is a suitable radioligand for the labeling of G protein-coupled adenine receptors but shows high affinity to bacterial contaminations in buffer solutions

[3H]Adenine has previously been used to label the newly discovered G protein-coupled murine adenine receptors. Recent reports have questioned the suitability of [3H]adenine for adenine receptor binding studies because of curious results, e.g. high specific binding even in the absence of mammalian protein. In this study, we showed that specific [3H]adenine binding to various mammalian membrane preparations increased linearly with protein concentration. Furthermore, we found that Tris-buffer solutions typically used for radioligand binding studies (50 mM, pH 7.4) that have not been freshly prepared but stored at 4°C for some time may contain bacterial contaminations that exhibit high affinity binding for [3H]adenine. Specific binding is abolished by heating the contaminated buffer or filtering it through 0.2-μm filters. Three different, aerobic, gram-negative bacteria were isolated from a contaminated buffer solution and identified as Achromobacter xylosoxidans, A. denitrificans, and Acinetobacter lwoffii. A. xylosoxidans, a common bacterium that can cause nosocomial infections, showed a particularly high affinity for [3H]adenine in the low nanomolar range. Structure–activity relationships revealed that hypoxanthine also bound with high affinity to A. xylosoxidans, whereas other nucleobases (uracil, xanthine) and nucleosides (adenosine, uridine) did not. The nature of the labeled site in bacteria is not known, but preliminary results indicate that it may be a high-affinity purine transporter. We conclude that [3H]adenine is a well-suitable radioligand for adenine receptor binding studies but that bacterial contamination of the employed buffer solutions must be avoided. Introduction Purinergic receptors play an important role in transmembrane signaling [1]. Currently, two distinct families are officially recognized by the International Union of Pharmacology (IUPHAR), i.e. receptors for the purine nucleoside adenosine (P1 or adenosine receptors) and receptors for purine and/or pyrimidine nucleotides (P2 receptors) [2–4]. Whereas the P1 receptor family comprises four subtypes, A1, A2A, A2B, and A3, all of which are G protein-coupled receptors (GPCRs), the P2 family is further subdivided into two subfamilies, P2Y (GPCRs) and P2X (ligand-gated ion channels) [2–4]. In addition to the nucleoside (adenosine) and nucleotide receptors, a receptor for the nucleobase adenine has recently been discovered by a reverse pharmacological approach identifying adenine as the natural ligand for a rat orphan GPCR [5]. A mouse orthologue (mMrgA10) of the rat adenine receptor was subsequently identified by sequence comparison [5]. Very recently, a new adenine receptor has been cloned from mice showing 82% identity in its amino acid sequence to mMrgA10 and 76% to the rat adenine receptor, indicating that the new receptor is a distinct adenine receptor subtype (Genbank nucleotide sequence accession numbers: new mouse adenine receptor, DQ386867; mMrgA10, XM_195647; rat adenine receptor, AJ311952) [6, submitted]. Both adenine receptors that have been pharmacologically characterized are coupled to the inhibition of adenylate cyclase by Gi protein [5, 6]. So far, no human receptor for adenine has been identified, although initial clues for the possible existence of human adenine receptors have been found [7]. Adenine receptors are structurally unrelated to P1 and P2 receptors and therefore constitute a new family of purinergic receptors for which we proposed the designation P0 (“P zero”) receptors [8] based on the structural relationships of the physiological agonists, adenine (P0) representing a partial structure of adenosine (P1) and adenosine again being a partial structure of adenine nucleotides (P2), such as adenosine triphosphate (ATP) or adenosine diphosphate (ADP). Radioligand binding studies are widely used to characterize GPCRs on the protein level [9]. Adenine, the natural ligand of adenine receptors, is commercially available in tritium-labeled form ([3H]adenine) and has been used to characterize recombinant rat adenine receptors expressed in Chinese hamster ovary (CHO) cells [5] as well as natively expressed rat adenine receptors in rat brain [7, 10] using membrane preparations. The radioligand has been shown to be stable under the incubation conditions [10]. Furthermore, mouse adenine receptors natively expressed in NG108-15 (neuroblastoma × glioma hybrid) cell membranes were labeled by [3H]adenine [7]. Recently, we successfully applied [3H]adenine binding to detect the mouse adenine receptor protein recombinantly expressed in Sf21 insect cell membranes, which constitute a null background because they do not endogenously express any high affinity binding site for adenine [6]. Two recent poster presentations reported on problems with [3H]adenine binding to adenine receptors. In one study using whole rat brain membrane preparations, a high-affinity binding site was detected (apparent Ki 57.5 nM) [11]. However, a very high Bmax value was found (281 pmol/mg protein), and the binding was almost completely blocked by 10 μM of hypoxanthine and abolished in the absence of Mg2+, indicating that the detected binding site was not identical with the G protein-coupled rat adenine receptor [11]. In another study, high affinity binding of [3H]adenine was detected even in the absence of added protein, and the authors suggested that [3H]adenine bound in a highly specific manner to the glass fiber filters used in the filtration assays [12]. IC50 values were determined for five adenine derivatives and three compounds [adenine (18 nM [12]; rat: 29.9 nM [7], 18 nM [5]), 7-ethyladenine (30 μM [12]; rat: 47.3 μM [7]), 8-bromoadenine (14 μM [12]; rat: 17.3 μM [7]] showed similar IC50 values as those previously determined for the rat adenine receptor, but two compounds [5′-deoxyadenosine (725 μM [12]; rat: 0.823 μM [10]), 2-fluoroadenosine (19 μM [12]; rat: 0.62 μM [7]] had much lower affinities for the unknown binding site labeled in the absence of added protein [12] than for the rat adenine receptor as previously determined [7, 10], indicating that the unknown binding sites labeled in the absence of added rat tissue were very different. Based on their results, the IJzerman group had suggested to avoid the use of [3H]adenine as a radiolabeled probe for the adenine receptor due to its putative specific, high-affinity binding to glass fiber filters [12]. That report prompted us to carefully reanalyze radioligand binding data obtained with [3H]adenine in our laboratory with the goal to find out the reason for the problems encountered in other laboratories. As a matter of fact, we had occasionally observed unusually high counts in a few experiments, which could, however, be avoided by repeating the experiments under carefully controlled experimental conditions, including the use of freshly prepared buffer solutions. We have now performed a systematic study clearly showing that [3H]adenine is a suitable radioligand for the labeling of adenine receptors in various cells and tissues if bacterial contaminations are excluded. Furthermore, we identified three common gram-negative aerobic bacteria that grow in cold Tris buffer and express high-affinity binding sites for [3H]adenine. Materials and methods Chemicals [8-3H]Adenine (27 Ci/mmol) was obtained from Amersham Biosciences (Munich, Germany). Tris was obtained from Acros Organics (Leverkusen, Germany) and dimethyl sulfoxide (DMSO) was from Fluka (Switzerland). All other chemicals and reagents were obtained from Sigma unless otherwise noted. Cell culture Human embryonic kidney (HEK) 293 and CHO-K1 cells were grown as monolayers at 37°C (5% CO2) in Dulbecco’s modified Eagle medium (DMEM) containing 10% fetal calf serum, 100 U/ml penicillin, 100 μg/ml streptomycin, and 2 mM L-glutamine. Membrane preparations Frozen rat brains were obtained from Pel Freez (Rogers, AR, USA) and thawed at 4°C. Cortex and striata were dissected, and membrane fractions were prepared as previously described [7]. Membrane preparations from CHO-K1 cells, and HEK293 cells were prepared as described [13, 14]. Membrane preparations from bacteria were obtained after growing them on agar plates and subsequent amplification of single strains in Lennox broth (LB) medium over night. Membranes were then prepared in analogy to the procedures used for mammalian-cell membranes [7, 13]. Protein concentrations were determined according to the method of Lowry [15]. [3H]Adenine binding assays Adenine binding assays were carried out as previously described [7]; however, in the absence of Mg2+ and ethyleneglycoltetraacetic acid (EGTA), unless otherwise noted, as Mg2+ and EGTA were found to have no effect on determined Ki values (data not shown). In brief, membrane preparations (50 μg of protein, unless otherwise indicated) were incubated with 10 nM [3H]adenine in 50 mM Tris-HCl, pH 7.4 in a total volume of 200 μl. Inhibition curves were determined using six to nine different concentrations of adenine, spanning three orders of magnitude. Three separate experiments were performed, each in triplicate, unless otherwise noted. Nonspecific binding was determined in the presence of 100 μM unlabeled adenine. Incubations were carried out for 1 h at room temperature and terminated by rapid filtration through GF/B glass fiber filters (Whatman, Dassel, Germany). Filters were washed three times, 2 ml each, with freshly prepared ice-cold 50 mM Tris-HCl buffer, pH 7.4 and immediately transferred to mini vials. Scintillation cocktail (Ultima Gold, Canberra Packard, 2.5 ml) was added and after an incubation of 9 h filter-bound radioactivity was measured by liquid scintillation counting at an efficiency of 54%. In some experiments, the addition of mammalian protein was omitted and replaced by 100 μl of different buffer samples. For competition experiments with bacteria, either 100 μl of 1:100 or 1:1000 dilution of an overnight culture of the bacteria in Tris buffer, 50 mM, pH 7.4, or a membrane preparation of bacteria (containing 0.4–10 μg of protein) was used. After isolation and classification of the bacteria from contaminated buffer solutions, experiments were performed with intact bacteria (approximately 6 × 104 bacteria/sample, which roughly equals 0.5 μg of total protein/sample), unless otherwise indicated, using the standard procedure (see above). The cell number was estimated from the optical density (OD) of the overnight culture suspension, and the assumption that [16]. Data analysis Data were analyzed using Prism 4.03 (Graph Pad, San Diego, CA, USA). IC50 values were determined by fitting data to sigmoidal concentration-inhibition curves. Results are presented as means ± standard error of the mean (SEM) from the number of observations. Isolation of microorganisms from contaminated buffer solutions and classification Microorganisms were isolated from Tris-HCl buffer, pH 7.4, by plating 300 μl of the buffer on LB agar plates. Single colonies were isolated and incubated overnight in LB medium at 37°C with constant shaking at 230 rpm. Further separation was achieved by plating bacteria from overnight cultures on plate count (Merck, Darmstadt) and blood agar (Oxoid, Wesel). Morphology of colonies was visually examined. Conventional physiological and biochemical characterization assays, including gram staining, catalase and oxidase activity, motility, and oxidation/fermentation (O/F) test were carried out and analyzed according to Bergey’s Manual of Systematic Bacteriology using reagents from Merck (Darmstadt) [17]. In addition, the following kits and appliances were used: BBL OXI/FERM Tube II (Schwarz Pharma GmbH, Germany), API 20 NE strips and apiweb software (Biomerieux, Nürtingen, Germany), and VITEK 2 fully automated system (Biomerieux, Nürtingen, Germany). Results Protein dependence of [3H]adenine binding As a first step, we reevaluated (unpublished) data obtained in initial studies that had been performed to investigate the suitability of [3H]adenine as a radioligand for labeling adenine receptors. Figure 1 shows [3H]adenine binding to different membrane preparations from (a) rat brain striatum, (b) rat brain cortex, and (c) HEK293 cells and CHO-K1 cells. For each membrane preparation, different amounts of protein were investigated (25, 50, 100, and 200 μg for striatum; 0, 25, 50, 100, and 200 μg for cortex; 50 and 100 μg for HEK and CHO cells). In all cases, we found a large, approximately linear, increase in specific binding with increasing protein concentration, whereas the increase in nonspecific binding was small. No specific binding was detected in the absence of protein (see Fig. 1b). Fig. 1Protein dependence of [3H]adenine binding to adenine receptors in rat brain striatal membranes (a), in rat brain cortical membranes (b), and in human embryonic kidney (HEK293) cell membranes and in Chinese hamster ovary cell (CHO) membranes (c). Different amounts of protein were incubated for 60 min with 10 nM of [3H]adenine in Tris-HCl buffer, pH 7.4 (n = 3). Nonspecific binding was determined in the presence of 100 μM adenine Microbial contaminations in buffer solutions As a next step, we investigated whether microbial contaminations present in incubation buffers that were not freshly prepared might be responsible for high counts occasionally observed in [3H]adenine binding studies in our laboratory. We performed a systematic analysis of buffer solutions (50 mM Tris-HCl, pH 7.4) stored at different conditions (different periods of time and temperatures), in different containers (plastic, glass, different sizes) used in our laboratory. Samples of buffer solutions were taken, and radioligand binding studies were performed with [3H]adenine (10 nM) using the same procedure as for the labeling of adenine receptors, except that no tissue or cell membrane preparation was added. Most investigated buffer solutions did not show any specific [3H]adenine binding. However, one sample of Tris buffer that had been taken from a 5-l plastic container with a small orifice, stored at 4°C, exhibited high affinity binding of [3H]adenine (data not shown). In contrast, radioligands used for the labeling of adenosine receptors ([3H]2-Chloro-N6-[3H]cyclopentyladenosine (CCPA) (A1) [18], [3H][3H]3-(3-hydroxypropyl)-7-methyl-8-(m-methoxystyryl)-1-propargylxanthine (MSX)-2 (A2A) [19], [3H]8-Ethyl-4-methyl-2-phenyl-(8R)-4,5,7,8-tetrahydro-1H-imidazo[2,1 ]purin-5-one (PSB)-11 (A3) [20]), or P2Y12 receptors ([3H]2-propylthioadenosine-5′-adenylic acid (1,1-dichloro-1-phosphonomethyl-1-phosphonyl) anhydride (PSB)-0413) [21] did not show any specific binding to that buffer solution. Figure 2 shows [3H]adenine binding determined in differently treated Tris-buffer solutions (in the absence of added protein). Freshly prepared buffer did not show any specific [3H]adenine binding. After storing the buffer for 1 day at 4°C, a small degree of [3H]adenine binding could be observed, and after 2 weeks, [3H]adenine binding was significant (562 ± 19 cpm specific binding). There was a large, exponential increase in specific [3H]adenine binding with time, and after 6 weeks at 4°C, approximately 7,000 cpm (specific binding) were measured. After the 6-week-old buffer was filtered through 0.2 μm filters, [3H]adenine binding was completely abolished. Heating of the buffer for 1 min at 80°C dramatically reduced [3H]adenine binding, whereas heating for 3 min at 80°C or heating at 121°C for 20 min in an autoclave led to complete loss of specific [3H]adenine binding. Fig. 2[3H]Adenine binding in the absence of added protein (cells or cell membranes). Incubation buffer (Tris-HCl, pH 7.4) was kept in a 5-l plastic container at 4°C for up to 6 weeks. Differently treated buffer samples (100 μl) were tested for [3H]adenine binding. The results shown represent means of three independent experiments ± standard error of the mean Isolation of microbial buffer contaminants As our results indicated that adenine binding was due to microbial contaminations growing in the incubation buffer, we decided to isolate the contaminants in order to characterize and eventually identify them. The microorganisms were isolated by plating contaminated incubation buffer on agar plates. This led to identification of three bacterial strains differing in the morphology of the formed colonies. The three strains were then separately amplified in medium over night, and membrane preparations were obtained to perform homologous competition binding assays using [3H]adenine (Fig. 3a). Fig. 3Competition curves for adenine versus 10 nM [3H]adenine obtained with membrane preparations from rat brain cortex and isolated microorganisms (contaminants 1–3) (a) and intact bacteria (b). IC50 values (a): ; ; ; contaminant 3 = 2,350 nM, (n = 1). IC50 values (b): ; ; Membrane preparations of contaminant 1 showed a very high affinity for adenine (, n = 3), whereas membrane preparations of the other two contaminants appeared to have considerably lower affinities (contaminant 2: , n = 2; contaminant 3: IC50 = 2351 nM, n = 1). For comparison, the binding curve for adenine at rat brain cortical adenine receptors () is shown (Fig. 3a). Using amounts of membrane preparations of contaminant 1 that contained more than 1 μg protein/assay tube led to depletion of the radioligand (more than 50% of the added radioligand was bound to the protein) (data not shown). Identification of microbial contaminants In order to identify the contaminants, standard procedures and classification kits were used (see Table 1). Standard tests, including catalase, oxidase, fermentation, motility test, and gram staining indicated that contaminant 1 is Pseudomonas spp. or a strain of Achromobacter spp., contaminant 2 is Achromobacter spp., and contaminant 3 Acinetobacter spp. [17]. All three bacteria are strictly aerobic gram-negative rods. For further classification, a standardized system (Api 20 NE) for classification of bacteria was used, which combines conventional and assimilation tests for identification of gram-negative rods not belonging to the Enterobacteriaceae. It was found that contaminant 1 was positive for oxidase, nitrate reduction, glucose degradation, gluconate, caprate, adipate, maltose, citrate, and phenylacetate, whereas contaminant 2 was negative for glucose, and caprate and contaminant 3 only showed positive results for caprate, maltose, and phenylacetate. Those results confirmed contaminant 3, already presumed to belong to the species Acinetobacter, as A. lwoffii, with a probability of 98.1%, and led to the identification of contaminant 1 as A. xylosoxidans (94.5% probability) and of contaminant 2 as A. denitrificans (82.2% probability) (see Table 1). Identification of the two Achromobacter strains with the VITEK fully automated system for rapid bacterial identification and antibiotic susceptibility confirmed both as members of the Achromobacter spp. with a probability of 90% each; however, differences were found in the results for phosphate, citrate, and proline assimilation (for details, see Table 1). Table 1 summarizes selected test results, which were positive for at least one of the strains. Table 1Summary of classification test results using standard procedures and kitsTestContaminant 1Contaminant 2Contaminant 3Morphology of colonies  FormRoundRoundRound  Diameter1–2 mm1 mm1 mm  ColorGreyBeigeWhite  BrimSmoothSmoothSmooth  SurfaceShinyShinyShiny  Cross sectionRaisedRaisedRaised  Gram stainNegativeNegativeNegativeBiochemistry  Catalase+++  Oxidase++–  O/F test–/––/––/–  Motility++–  Lysine/BBL Ox/Ferm Tube II–+n.d.  ResultAchromobacter spp. or Pseudomonas spp.Achromobacter spp.Acinetobacter spp.Api 20 NE  NO3, nitrate reduction++–  GLU, glucose+––  GNT, gluconate++–  CAP, caprate+–+  ADI, adipate++–  MLT, maltose+++  CIT, citrate++–  PAC, phenylacetate+++  OX, oxidase++–  ResultA. xylosoxidans (94.5%)A. denitrificans (82.2%)A. lwoffii (98.1%)VITEK 2  N-LGGH, γ-L-glutamyl (nutrient assimilation, NA)++n.d.  P-BPHO, bis(p-nitrophenyl)phosphate+–n.d.  C-CIT, citrate–+n.d.  C-KGA, α-acetoglutaryc acid++n.d.  N-LPROT, L-proline-NA+–n.d.  ResultAchromobacter spp. (90%)Achromobacter spp. (90%)n.d. n.d. not determined [3H]Adenine binding assays with isolated, intact bacteria Adenine competition binding studies were performed using isolated, intact bacteria (Fig. 3b). The IC50 value obtained with whole bacterial cells of A. xylosoxidans () was in the same concentration range as that obtained with membrane preparations of the same bacteria, previously designated contaminant 1 (). For the other two bacteria, A. denitrificans and A. lwoffii, IC50 values were 5- to 10-fold lower when determined in whole bacterial cells compared with membrane preparations. From the homologous competition experiments, KD and Bmax values were estimated for A. xylosoxidans (membranes and intact cells). For bacterial membrane preparations, a KD value of 5.84 ± 1.12 nM and a Bmax value of 266 ± 65 pmol/mg protein was calculated (n = 3). For the intact bacteria, a KD value of 11.0 ± 1.2 nM and a Bmax value of 780,000 ± 120,000 sites/cell (n = 8) was obtained. Structure–activity relationships Affinities for selected compounds at adenine binding sites of A. xylosoxidans were determined in competition assays using whole bacterial cells and compared with data obtained in binding studies at the rat brain adenine receptor [7]. Figure 4 shows competition curves for selected compounds, which exhibited high affinity, i.e., adenine (), hypoxanthine (), and 2-fluoroadenine (). The results for all compounds tested are summarized in Table 2. Whereas the affinity of adenine for the binding sites of A. xylosoxidans was in the same range as for the rat adenine receptor, the affinities for hypoxanthine and 2-fluoroadenine differed substantially from those determined for the rat adenine receptor. Hypoxanthine showed very low affinity for the rat adenine receptor (Ki 45,000 ± 19,400 nM), but high affinity for the bacterial [3H]adenine binding sites, with an IC50 value in the low nanomolar range (59.1 ± 2.0 nM). For 2-fluoroadenine, the opposite was true: the Ki value for rat brain cortical adenine receptor was 620 ± 140 nM [7], whereas the IC50 value for the A. xylosoxidans binding site was in the micromolar range (32,100 ± 3,000 nM). Fig. 4Competition curves for adenine, hypoxanthine and 2-fluoroadenine versus 10 nM [3H]adenine obtained with Achromobacter xylosoxidans (intact bacteria) (; ; Table 2Comparison of affinities of adenine and selected compounds for the rat adenine receptor and bacterial binding sites determined in radioligand binding studiesCompoundAchromobacter xylosoxidans (intact bacteria)Rat brain cortical membranesIC50 ± SEM [μM]a (% inhibition ± SEM at 10 μM)Ki ± SEM [μM]a (% inhibition ± SEM at 100 μM)Adenine0.0138 ± 0.0027b0.0299 ± 0.0034c2-Fluoroadenine32.1 ± 3.00.62 ± 0.14c2-Hydroxyadenine≫10 (7 ± 15)d29.4 ± 6.42,6-Diaminopurine≫10 (3 ± 8)d4.95 ± 0.751,N6-Ethenoadenineca. 10 (53 ± 6)d34.4 ± 8.1Hypoxanthine0.0591 ± 0.019645.0 ± 19.4Uracil≫10 (6 ± 1)≫100 (−6 ± 2)Xanthine>10 (17 ± 6)>100 (23 ± 7)Uridine≫10 (5 ± 5)d≫100 (−9 ± 3)Adenosine≫10 (12 ± 9)d19.4 ± 6.3caResults are from three independent experiments performed in triplicate, unless otherwise notedbResults are from eight independent experiments performed in triplicatecKi values and % inhibition values are from [7]dResults from two independent experiments performed in triplicate Discussion [3H]Adenine has been successfully used by us [6, 7] and other laboratories [5, 10] to label the recently discovered rat and mouse adenine receptors. However, problems with [3H]adenine binding assays have been reported by two laboratories [11, 12]. These have led to the suggestion that [3H]adenine was not a suitable radioligand for the labeling of G protein-coupled adenine receptors [11, 12]. IJzerman and coworkers [12] reported that [3H]adenine binding was not protein dependent and that it bound with nanomolar affinity to glass fiber filters in the absence of added protein. These results, which were contradictory to our own data, prompted us to reexamine the [3H]adenine binding results that we had obtained during the past years, trying to find an explanation for the discrepancies. In contrast to the results reported by IJzerman and coworkers [12], binding of [3H]adenine to various membrane preparations was strictly protein dependent in our hands, as expected (Fig. 1). Specific binding linearly increased with increased amounts of protein and thus increased numbers of adenine receptors. Nonspecific binding was generally low for [3H]adenine, and there was only a minor increase in nonspecific binding with increasing protein concentrations. In the absence of protein, no specific binding was observed (Fig. 1b). These results indicated that [3H]adenine was a suitable radioligand for labeling adenine receptors. In fact, we could recently perform [3H]adenine binding assays on a null background for the first time, namely, at the mouse adenine receptor heterologously expressed in Sf21 insect cells. Whereas membrane preparations of the nontransfected Sf21 cells did not exhibit any specific [3H]adenine binding, cell membranes prepared from cells infected with recombinant baculoviruses bound [3H]adenine with high affinity [6]. However, when we looked carefully at all of our previous [3H]adenine binding data, we found a few [3H]adenine binding experiments that could not be evaluated due to unusually high radioactivity counts. These occasional problems had been solved by carefully controlling the experimental conditions, e.g., by using freshly prepared buffer solutions. Stimulated by the experiences reported by the IJzerman group [12], we decided to perform a systematic study to find out the reasons for those problems, which might also be causative for erroneous [3H]adenine binding results in other laboratories [11, 12]. High affinity binding of [3H]adenine to filter paper, as suggested by IJzerman and coworkers [12], could be excluded by our experiments, as buffer solutions that were freshly prepared did not show any specific [3H]adenine binding in filtration assays using glass fiber GF/B filters, the same filters that had been used by the IJzerman group [12]. On the contrary, we discovered that bacterial contaminations, which can be present in buffer solutions, express high-affinity binding sites for [3H]adenine and therefore impede adenine receptor binding assays. When we examined different buffer solutions stored in our laboratory, we discovered high [3H]adenine binding in Tris-HCl buffer solution (pH 7.4) stored at 4°C in a 5-l plastic container. That container had only a small orifice and was therefore difficult to purify. Thus, microbial contamination in this container was carried over when new buffer solution was prepared. When the contaminated buffer was filtered through a bacteria-tight filter (0.2 μm) or heated in order to denature proteins, specific [3H]adenine binding was abolished, strongly indicating that microbial proteins were responsible for the high affinity binding of adenine. A further indication that a living organism was involved was the fact that adenine binding increased exponentially with time. Interestingly, the microorganisms grew better at 4°C than at room temperature. Our first presumption, that the contaminants might belong to yeast, could not be proven. Saccharomyces cerevisiae was used as a control organism for binding studies but showed no [3H]adenine binding (data not shown). Three microorganisms were isolated from the contaminated Tris-buffer solution and identified using standard procedures and kits. Two of these contaminants were assigned to the genera Achromobacter and one to Acinetobacter. Both bacteria species are gram-negative rods and are strictly aerobic. They are commonly found in soil and water [17]. For healthy humans or animals, they are not pathogenic, but especially A. xylosoxidans and A. lwoffii have gained increasing importance due to their ability to cause nosocomial infections [22–32]. These bacteria are able to grow under nonoptimal conditions, such as low temperature and restricted nutrient supply [17, 24–26]. They are able to metabolize a wide variety of organic substances, such as chemical pollutants in the environment, and can therefore be used as bioreporters and for the degradation of pollutants [33–36]. This is consistent with the fact that these bacteria are able to grow in simple Tris-HCl buffer at low temperature. Both A. lwoffii and Achromobacter spp., exhibit specific binding sites for adenine, with IC50 values as low as 13 nM for A. xylosoxidans. Thus, the detected adenine binding site in A. xylosoxidans exhibits an even higher affinity than the rat (29.9 nM [7], 18 nM [5]) or mouse (54.9 nM [6]) adenine receptor. The binding affinities for A. lwoffii and A. denitrificans were more that 70-fold lower, with affinities in the low micromolar range (1 μM and 2.4 μM, respectively) when membrane preparations were analyzed and about 20-fold lower when intact cells were investigated for binding (299 nM and 253 nM, respectively). For A. xylosoxidans, only a threefold difference was found when binding affinities for membranes were compared with those with intact cells (Fig. 3). The specific, high-affinity [3H]adenine binding site in A. xylosoxidans appeared to be expressed in extraordinarily high density, as amounts of membrane preparations that contained more than 1 μg of protein/assay tube led to depletion of the radioligand (i.e., more than 50% of the added radioligand was bound to the protein). Rough estimations of receptor densities by homologous competition assays confirmed the high expression levels. So far, the nature of these high-affinity adenine binding sites in bacteria is not known. However, bacteria express a large number of transporters, including nucleobase transporters, in order to secure their nutrient supply (for review see [37]). Nucleobase transport in bacteria has been mainly studied in Escherichia coli and Bacillus spp., as well as in the fungi Aspergillus nidulans and Neurospora crassa [37–39]. Usually, these transporters fulfill two main functions. Firstly, purines can serve as preformed bases for nucleotide biosynthesis, and secondly, they serve as nitrogen sources [40, 41]. Distinct adenine uptake systems have been identified, e.g., in E. coli [42]. The high density of the detected [3H]adenine binding sites in A. xylosoxidans would be consistent with its function as a purine transporter and points to an important role of this protein, which appears to be upregulated when the bacteria are transferred to Tris buffer (unpublished observation), a medium poor in nutrients. Such an effect has been described for protozoa [43]: purine salvage enzymes and transporters can be dramatically up- or down-regulated according to growth stage and availability of purine sources [44]. Examples are the high-affinity hypoxanthine transporter in Trypanosoma brucei brucei, which shows a 450% increased transport rate after 24 h of purine deprivation, and the adenine transporter in Crithidia luciliae, which shows a >100-fold increase of adenine uptake after purine starvation [44, 45]. For E. coli as well as for B. subtilis, two adenine transport systems have been described: a low- and a high-affinity transport system [42, 46, 47]. The latter system is important when the concentration of adenine is low [46, 47]. Differences in adenine-binding affinity observed in our studies when intact cells were compared with membrane preparations (Fig. 3) might be explained by the existence of different transporters in the bacteria (Achromobacter and Acinetobacter). Whereas bacterial membrane preparations were obtained directly from an overnight culture grown in complete medium, binding studies at whole bacteria were performed after growing them in Tris-HCl buffer, a nutrient-poor medium, in which they may have upregulated the high-affinity transporters [42, 47]. In order to investigate the structure–activity relationships of the high-affinity adenine binding site in A. xylosoxidans, a series of compounds, including adenine derivatives, other purines, and pyrimidines (uracil, xanthine, hypoxanthine), and nucleosides (uridine, adenosine) were investigated in binding studies, and the results were compared with those obtained at the rat adenine receptor. As expected, structure–activity relationships at the rat adenine receptor were very different from those at the bacterial adenine binding site. Hypoxanthine, for example, which exhibits a low affinity for the rat adenine receptor (45,000 nM), bound to the bacterial site with 760-fold higher affinity (59.1 nM), and 2-fluoroadenine, which showed a high affinity for murine adenine receptors (620 nM), bound with a 50-fold lower affinity (32,100 nM) to the bacterial site. Several compounds that had shown affinity in the micromolar range at rat adenine receptors (adenosine, 2-hydroxyadenine, 2,6-diaminopurine) were completely inactive at Achromobacter binding sites. Alexander had previously reported that hypoxanthine at a concentration of 10 μM completely blocked [3H]adenine binding to rat brain membranes in his experiments, the results of which were not consistent with the labeling of a G protein-coupled adenine receptor [11]. It might be speculated that he actually labeled a bacterial adenine binding site rather than the rat adenine receptor, which would explain the discrepant results. The limited number of compounds (five) investigated by IJzerman and coworkers [12] do not allow a full comparison of the structure–activity relationships, but large differences were observed for two compounds—5′-deoxyadenosine (725,000 nM [12] vs. 823 nM (rat adenine receptor) [10]) and 2-fluoroadenine (19,000 nM [12], 620 nM (rat adenine receptor) [7])—indicating the labeling of a very different, presumably a bacterial, binding site by the authors [12]. The fact that hypoxanthine exhibits high affinity for the [3H]adenine binding site in A. xylosoxidans in the same concentration range as adenine itself is another indication that the labeled protein may be a purine transporter. Nucleobase transporters have been identified in bacteria, fungi, protozoa, algae, plants, and mammals, but only few have been cloned and analyzed in detail [37, 48]. Five basic families of nucleobase transporters have been described: the nucleobase-ascorbate transporters (NAT), which include members from archaea, eubacteria, fungi, plants, and metazoa; the purine-related transporters (PRP), which are restricted to procaryotes and fungi; the purine permeases (PUP), which are purine transporters exclusively found in plants; and the equilibrative (ENT) and concentrative (CNT) nucleoside transporters, which not only transport nucleosides but may also transport nucleobases [37, 48–52]. Bacteria have developed different transport systems for related compounds, which allow them to independently absorb those compounds. This is an advantage when growing under nutritional deprivation [46, 53]. For C. luciliae, a nucleobase transporter that recognizes adenine and hypoxanthine equally well has been described [45]. In E. coli, adenine and uracil have different, specific transport systems, as do xanthine and guanine, whereas hypoxanthine might utilize the guanine transporter [53]. For B. subtilis, specific transport systems for guanine and hypoxanthine, for guanosine and inosine, as well as three independent uptake systems for adenine, adenosine and uracil, have been identified [46]. From the described observations, we conclude that the high-affinity binding sites found in bacteria isolated from Tris-buffer solutions have not much in common with the adenine receptors found in mammals. The genomes of the bacteria A. baumannii and of several closely related bacteria, such as Pseudomonas spp. are known. Therefore, we performed a search to identify potential sequences with homology to the rat and mouse adenine receptors, which, however, yielded no hits. It appears likely that [3H]adenine labels a high-affinity nucleobase transporter for adenine and hypoxanthine in Achromobacter spp., a bacterium isolated as a contaminant from Tris-HCl buffer. The high affinity in the low nanomolar range is in fact unusual; therefore, it may be speculated that the labeled protein might belong to a new type of high-affinity bacterial nucleobase transporter. Conclusions In conclusion, we have demonstrated that [3H]adenine is a well-suited radioligand for the labeling of G protein-coupled adenine receptors, but precaution is advised for preparing and storing buffers used for the assays to avoid bacterial contaminations. After systematically analyzing occasionally encountered irregularities in [3H]adenine binding assays, we were able to isolate three bacteria, commonly found in soil and water, from Tris-HCl buffer. They were identified as A. lwoffii, A. xylosoxidans, and A. denitrificans and revealed high-affinity binding sites for [3H]adenine.

Surg_Endosc-3-1-2169271

Expert and construct validity of the Simbionix GI Mentor II endoscopy simulator for colonoscopy

Objectives The main objectives of this study were to establish expert validity (a convincing realistic representation of colonoscopy according to experts) and construct validity (the ability to discriminate between different levels of expertise) of the Simbionix GI Mentor II virtual reality (VR) simulator for colonoscopy tasks, and to assess the didactic value of the simulator, as judged by experts. Training skills in endoscopy for diagnostic and therapeutic procedures is essential and requires a great deal of hands-on training [1]. Virtual reality (VR) simulators offer a promising option to train these skills extensively prior to training in real-life colonoscopy, without jeopardizing patients or causing them unnecessary discomfort [2]. The use of VR training prior to performing real flexible endoscopy on patients enables novice endoscopists to go through part of their proficiency curve before submitting patients to their relatively insufficient endoscopy skills. This might not only be advantageous for the patients undergoing endoscopy, but might also prevent complications and potential consequences resulting in medicolegal litigation. One of the simulators in the field of flexible endoscopy is the GI Mentor II (see Figure 1). VR simulators have been used extensively in different fields of expertise before applying these procedures to patients. In the United States of America simulator training is mandated by the Accreditation Council for Graduate Medical Education (ACGME) in laparoscopic procedures for surgical residents [3]. The first step is to validate the simulator construct properly and verify its didactic value, before implementing simulators in teaching programmes or developing a new curriculum for flexible endoscopy around them. Figure 1.The GI Mentor II virtual reality simulator, the setup for training in lower endoscopy. Some studies have already been published on this subject [4–6], but the presented outcomes lacked power due to their relatively small sample sizes. In addition, some cases did not study the validity of endoscopy, but for example only the EndoBubble module, a computer simulation skills test measuring how long it takes a person to pop 20 balloons in a virtual tunnel. Objective The main objectives of this study were: (1) to establish the degree of representation of real-life colonoscopy on the Simbionix GI Mentor II VR colonoscopy simulation, as judged by experts (expert validity), (2) to determine whether the GI Mentor II simulator can distinguish between various degrees of expertise in endoscopy, judged by novice, intermediate experienced, experienced and expert endoscopists performing VR colonoscopy (construct validity), and (3) to assess the didactic value of the simulator, as judged by experts. Material and Methods Simulator The simulator used in this study was the Simbionix GI Mentor II (Simbionix Ltd. Israel, software version 2.7.3.0) (Figure 1). The GI Mentor II can simulate upper GI tract endoscopies such as esophagogastroduodenoscopy, endoscopic retrograde cholangiopancreatographies, and endoscopic ultrasound. The lower GI tract endoscopies it simulates are sigmoidoscopy and colonoscopy. The simulator records a range of parameters upon each exercise, which can be used to assess performance objectively. The endoscope used is a customized Pentax ECS-3840F endoscope. Participants Participants were allocated to four groups to assess the validity and didactic value of the GI Mentor II simulator. The first group, the novices, was defined as participants without any flexible endoscopy experience; they were all medical interns or residents. The second group was intermediate experienced, with fewer than 200 colonoscopies performed before. In the third group experienced participants all performed more than 200 colonoscopies but fewer than 1,000. The fourth group consisted of experts, all of whom had performed more than 1,000 colonoscopies. These categories were chosen based upon several other studies, the demands for Dutch accreditation for colonoscopy, and the accreditation demands of the British Society of Gastroenterology, which advocates 200 colonoscopies under supervision during training [4, 6–8]. All persons were either invited to participate within our hospital, or participated during a national congress of the Dutch Society of Gastroenterology in spring 2006. The groups consisted of at least 28 persons to ensure sufficient statistical power [9]. A post hoc sample size calculation based on the results for time to finish the EndoBubble task showed a minimal sample of 26 participants in the novices group to achieve a power of 0.95. Originally, the intermediate experienced and experienced participants formed one group, but as the expertise level and performance within this group varied considerably, this groups was split. A schematic setup of the study design is presented in Figure 2. Figure 2.The study design. Questionnaire All participants were asked to fill out a questionnaire on demographics and their general medical and endoscopy experience. It also included the number of endoscopies performed annually and number of years registered as a skilled professional endoscopist. After the simulator run the participants were asked to answer questions about their appreciation of the realism of the colonoscopy exercises performed. Appreciation was expressed on a four-point Likert scale [10] varying from very unrealistic (1) to very realistic (4). Questions were asked about the realism of imaging, simulator setup, endoscope control and both haptic and visual feedback. Experts were asked whether the GI Mentor II could be used as a teaching device for novice endoscopists and whether experience on the simulator could be useful in practice. Simulation modules All participants first performed the hand–eye coordination task (EndoBubble level 1) of popping all 20 balloons in the test as quickly as possible, without touching the walls. Next, the participants performed VR case numbers 1 and 3, both from colonoscopy module 1. These cases were carefully selected for their discriminative value; both cases are straightforward colonoscopies, without any abnormalities such as polyps, tumours, or inflammation. Case number 1 is a relatively easy colonoscopy to perform, whereas case number 3 is more difficult, requiring the endoscopist to apply techniques such as straightening the endoscope during loop formation and applying torque to the endoscope shaft. The assignment given for the VR colonoscopies was to reach the cecum as quickly as possible with as little patient discomfort as possible. Patient discomfort was defined as the estimated percentage of time the virtual patient was in excessive pain and the number of times excessive local pressure was caused. Other relevant test parameters were the percentage of time spent with clear view and the number of times view of the lumen was lost. The task was considered accomplished when the cecum was reached. Data analysis SPSS 13.0 software was used to perform descriptive statistics and Kruskal–Wallis tests for statistic analysis of the data. A separate analysis between groups was performed using a two-tailed Mann–Whitney exact U test. A p-value of less than 0.05 was considered significant. The data showed a nonparametric distribution, therefore the median and range of performance parameters are presented as primary values. Results Participants Thirty-five novices, 15 intermediates, 20 experienced, and 35 expert endoscopists participated in the study. The average number of colonoscopies performed annually by experts was 445, and their mean number of years registered as a gastroenterologist was 7.7 (range 0–35 years). Construct validity Data output by the simulator are presented in Tables 1 and 2. The EndoBubble task was completed faster by the experts and experienced endoscopists than by novices, with fewer wall collisions. These differences were statistically significant (Kruskal–Wallis test) (Table 1). Also the colonoscopy tasks were completed faster (p < 0,001, Kruskal–Wallis test), with less patient comfort and better visibility by experts and experienced endoscopists (Table 3). Novice endoscopists (N = 35) reached the cecum in a mean time of 29:57 (min:sec) in colonoscopy case 3, intermediate experienced (N = 15) in 5:45, experienced (N = 20) in 4:19, and experts (N = 35) in 4:56. Novices lost view of the lumen significantly more often than the other groups. Table 1.EndoBubble hand–eye coordination taskExperienceTime to finish (min:sec)Number of times wall touchedNovice N = 35Mean6:561.9Median5:581.0Range1:24–20:250–20Intermediate N = 15Mean1:561.1Median1:410.0Range0:54–4:020–5Experienced N = 20Mean1:370.9Median1:210.0Range0:43–5:330–9Expert N = 35Mean1:240.3Median1:130.0Range0:49–3:250–2Kruskal-WallisChi- square63.1519.374Asymp. sign.0000.025Table 2.Colonoscopy module 1, cases 1 and 3ExperienceTime to reach cecum (hour:min:sec)% of time spent with clear viewLost view of lumenExcessive local pressure% of time patient was in painExcessive loop formedCase 1Novice N = 35Mean6:47960.40.513.30.83Median6:169700110Range 1:53–15:08 82–99 0–30–30–44 0–6Intermediate N = 15Mean1:36970080.6Median1:40980050Range0:55–2:5291–100000–300–3Experienced N = 20Mean1:239800.29.20.7Median1:21980081Range0:48–2:4389–10000–10–270–3Expert N = 35Mean1:23980014.51.49Median1:179800121Range0:42–3:1694–1000–100–570–10Case 3Novice N = 35Mean29:57863.23.892.24.77Median23:42853301Range4:48–1:28:1972–960–121–140–240–34Intermediate N = 15Mean5:45891.12.10.91.13Median4:21921200Range2:28–13:4178–970–40–60–40–8Experienced N = 20Mean4:19910.61.91.01.6Median3:50910101Range2:27–7:0273–990–30–80–40–9Expert N = 35Mean4:56890.91.622.51Median4:0390111Range1:38–15:3968–990–40–60–100–12Table 3.Statistics colonoscopy module 1, cases 1 and 3Time to reach cecum% of time spent with clear viewLost view of lumenExcessive local pressure% of time patient was in painExcessive loop formedCase 1Chi square69.04313.88918.41519.7837.10110.691Asymp. sig.0.0000.0030.0000.0000.0690.014Case 3Chi Square65.5596.97841.93628.7944.2844.856Asymp. sig.0.0000.0730.0000.0000.2320.183Kruskal-Wallis test A separate analysis between groups using a Mann–Whitney exact U test demonstrated no significant difference between the intermediate, experienced and expert groups on all parameters. They all completed the task faster than the novices (see Table 4). Table 4.Differences between groups module 1, cases 1 and 3Time to reach cecum% of time spent with clear viewEost view of lumenExcessive local pressure% of time patient was in painExcessive loop formedCase 1Novice vs. intermediate0.0000.1770.0390.0130.0700.743Intermediate vs. experienced0.1660.6171.0000.2440.3850.547Experienced vs. expert0.9620.6211.0000.0430.0770.020Intermediate vs. expert0.1410.2591.0001.0000.0180.009Case 3Novice vs. intermediate0.0000.1040.0000.0040.5840.040Intermediate vs. experienced0.2570.3940.2850.5030.7710.184Experienced vs. expert0.9690.2970.1530.9420.1540.726Intermediate vs. expert0.3260.7570.8700.4160.1110.090Mann–Whitney two-tailed test, exact significance Expert validity The group of expert endoscopists rated the colonoscopy simulation 2.95 on a four-point Likert scale for overall realism. Anatomical representation was rated 2.58, and the simulator setup 3.14. Endoscope control scored 3.21. Haptic feedback was rated 2.57. Didactic value Expert opinion was that the GI Mentor II simulator should be included in the training of novice endoscopists (3.51 on a four-point Likert scale) and that expertise gained on the simulator was considered applicable in a clinical curriculum (rated 3.29 out of 4). The simulator was not considered suitable for certification of trained endoscopists (rated 2.29 out of 4). Discussion This study represents the largest and most detailed study on the validity of this type of colonoscopy simulator so far. The data show that the simulator can discriminate clearly between endoscopists of different expertise levels performing different colonoscopy tasks. Differences were statistically significant using relatively large sample sizes in all three exercises, the EndoBubble task as well as cases number 1 and 3. The difference between our study and previous studies by others is that we focused on the basic aspects of navigation for colonoscopy itself, rather than on the hand–eye coordination task alone, used for example in the study by Ritter et al. [4], and that we included more participants in four separate groups with different levels of expertise [4–6, 11, 12]. in this way we were able to demonstrate that the GI Mentor II can distinguish between expertise levels up to the level of an intermediate experienced endoscopist, who has performed around 200 colonoscopies. In a similar study Sedlack et al. [5] describe a limited construct for a different simulator (AccuTouch, Immersion Medical). Felsher et al. [11] demonstrated differences between novices and experts in large sample sizes but did not compare novices to intermediate levels of expertise. In this study we have demonstrated convincing expert validity for colonoscopy on the GI Mentor II virtual simulator. This in contrast to other studies focusing on the EndoBubble task as a validation study [4] and not dealing with the subject of expert validity [4, 6, 7, 11, 12]. The colonoscopy tasks were considered as accomplished once the participants reached the cecum. Asking the participants to inspect the mucosa on the way back through the colon does not, in our opinion, provide a proper representation of the endoscopist’s skills in manoeuvring through the colon, as other aspects besides the basic navigation skills of the endoscopist could influence the performance parameters provided by the simulator considerably in this case. This might lead to very different end times depending, for example, on the carefulness of the endoscopist. This study demonstrates that the GI Mentor II simulator offers a convincing, realistic representation of colonoscopy according to experts. The overall assessment was good. Expert opinion was that the simulator can be used as a teaching tool for novice endoscopists. The simulator’s haptic feedback is doubtful. Inexperienced residents can be trained in the skills necessary in flexible endoscopy such as steering control, straightening the endoscope during loop formation and applying torque up to a certain level. Conclusion The current study demonstrates that the GI Mentor II simulator offers a convincing, realistic representation of colonoscopy according to experts (expert validity) and that the simulator can discriminate up to the level of intermediate experienced endoscopists (construct validity) in colonoscopy. In the cases used the simulator could not discriminate between intermediate, experienced and expert endoscopists. The next step will be a study to determine whether novice endoscopists can develop a learning curve that will actually improve their endoscopic skills applied to real patients.

Virchows_Arch-3-1-1888715

Discrepancies in the diagnosis of intraductal proliferative lesions of the breast and its management implications: results of a multinational survey

To measure discrepancies in diagnoses and recommendations impacting management of proliferative lesions of the breast, a questionnaire of five problem scenarios was distributed among over 300 practicing pathologists. Of the 230 respondents, 56.5% considered a partial cribriform proliferation within a duct adjacent to unequivocal ductal carcinoma in situ (DCIS) as atypical ductal hyperplasia (ADH), 37.7% of whom recommended reexcision if it were at a resection margin. Of the 43.5% who diagnosed the partially involved duct as DCIS, 28.0% would not recommend reexcision if the lesion were at a margin. When only five ducts had a partial cribriform proliferation, 35.7% considered it as DCIS, while if ≥20 ducts were so involved, this figure rose to 60.4%. When one duct with a complete cribriform pattern measured 0.5, 1.5, or 4 mm, a diagnosis of DCIS was made by 22.6, 31.3, and 94.8%, respectively. When multiple ducts with flat epithelial atypia were at a margin, 20.9% recommended reexcision. Much of these discrepancies arise from the artificial separation of ADH and low-grade DCIS and emphasize the need for combining these two under the umbrella designation of ductal intraepithelial neoplasia grade 1 (DIN 1) to diminish the impact of different terminologies applied to biologically similar lesions. Introduction The concept of atypical hyperplasia was introduced decades ago within the continuum of intraductal breast proliferations, a continuum which encompasses benign proliferations of usual ductal hyperplasia to high-grade ductal carcinoma in situ (DCIS) [3–5, 7, 15, 32]. The term atypical ductal hyperplasia (ADH) was used initially for a vaguely defined group lesions that had “some but not all of the requisite features of ductal carcinoma in situ” [21]. Subsequently, the qualitative similarity to low-grade DCIS (LG-DCIS) was stressed, and quantitative measures were introduced to separate ADH from DCIS [21, 31]. Totally arbitrary boundaries were drawn to separate ADH from LG-DCIS, a separation that resulted in drastically different management approaches in the 1980s —i.e., mastectomy for those diagnosed as DCIS and follow-up for those with a diagnosis of ADH. Now, even with widespread use of conservative surgery, patient management differs significantly depending on whether a lesion is diagnosed as ADH or DCIS. The two most notable approaches for this separation have been based on: (a) the definition by Page et al. [21] who defined the minimum requirement for LG-DCIS as complete involvement of at least two spaces by a proliferation that cytologically and architecturally resembles DCIS and (b) the definition of Tavassoli and Norris [31] who required complete involvement of one or more ducts by a cribriform or micropapillary proliferation of uniform cells with low-grade nuclei exceeding 2 mm in aggregate diameter for a diagnosis of LG-DCIS. According to either of these two definitions, any intraductal proliferative lesion that demonstrated the qualitative cytologic and architectural features of LG-DCIS but failed to pass the defined quantitative threshold was to be designated as ADH. Since then, no objective molecular, ultrastructural, immunohistochemical, or morphometric feature has been identified to reliably distinguish ADH from LG-DCIS [1, 11, 15, 16, 18, 25]. Therefore, the distinction of the two continues to be based solely on application of the arbitrary criteria mentioned above. Frequently cited studies in the literature, which have attempted to calculate the relative risk of various proliferative lesions, may have claimed to adhere to the criteria of Page et al. have actually deviated from the two-space rule and have accepted single-space involvement as DCIS, thereby blurring the distinction between levels of hyperplasia and atypia and diminishing the practical utility of such a distinction [8]. We conducted a survey 15 and 20 years postintroduction of these two sets of criteria to determine how much uniformity or discrepancy exists in the interpretation of these lesions and how it would impact current patient management. Materials and methods To investigate pathologists’ approach to selected problems in the diagnosis and management of intraductal proliferative lesions of the breast, a questionnaire was prepared with diagrammatic representations of five potentially problematic scenarios in breast pathology dealing with intraductal proliferative lesions (Figs. 1, 2, 3, 4, 5). It was decided to use diagrammatic representations rather than glass slides to facilitate distribution of the survey among a greater number of pathologists and to ensure that the participants focused on the same diagnostic issue, given the presence of certain lesions in the background. The questionnaire also inquired whether the respondents considered breast pathology their area of expertise, the average number of breast biopsies and lumpectomies or mastectomies they signed out each week, the practice setting in which they worked in (academic, community, or private), and how long they had been in practice. Fig. 1The first scenario assessed how pathologists would diagnose a partially involved duct adjacent to unequivocal cribriform DCIS, and whether they would recommend reexcision if it were less than 0.1 mm from a resection marginFig. 2In question 2, participants were asked whether the number of partially involved ducts affects their decision to make a diagnosis of ADHFig. 3Responses to question 3 demonstrated what pathologists thought was the lowest size threshold required to make a diagnosis of DCISFig. 4Question 4 evaluated how participants would manage flat epithelial atypia close to a resection marginFig. 5Question 5 surveyed how pathologists measured invasive carcinoma in the presence of multifocal microinvasion The questionnaire was distributed among over 300 pathologists who were either known for their expertise and specialization in breast pathology in various countries or who were interested in the field as evidenced by their participation in educational courses dedicated to breast pathology. If a participant left a response in the questionnaire blank, that response was considered null and excluded from calculations. Also, for all questions regarding diagnosis and management of lesions illustrated in the questionnaire, the respondents were given the option of choosing one of the two answers (i.e., ADH vs DCIS, Yes vs No). The responses of the small minority who chose both were similarly excluded from analyses. Also, if a participant responded to the questions regarding average number of biopsies or lumpectomies/mastectomies a week by providing a numeric range, the average value of the range was considered for calculations. The collected responses were anonymized and tabulated in a custom Microsoft Access database (Microsoft Corporation, Redmond, WA, USA), and custom queries were written for data analysis. Statistical significance of the results was evaluated using chi-square and t tests. Results A total of 230 completed questionnaires were received and included in the study. Of the 230 respondents, 93 (40.4%) considered breast pathology their area of expertise, while 130 (56.5%) did not consider themselves experts but had special interest or responsibility for signing out breast pathology. The respondents signed out an average and standard deviation of 10.1±9.5 biopsies and 7.5±7.8 lumpectomies or mastectomies a week. These figures were 14.4±11.4 biopsies and 10.3±9.0 lumpectomies or mastectomies a week for the responding pathologists who considered breast pathology their area of expertise and 6.9±6.0 biopsies and 5.3±6.0 lumpectomies or mastectomies a week for those who did not. T test showed the differences between the means in these two groups to be statistically significant (p<0.0001 for both biopsies and lumpectomies/mastectomies). Academic pathologists constituted 85 (37.0%) of the respondents, while 113 (49.1%) of the respondents worked in community hospitals and 20 (8.7%) in private laboratories. Nine pathologists (3.9%) worked in more than one practice setting. Sixty-five pathologists (28.3%) had been practicing pathology for less than 5 years, 37 (16.1%) for 6 to 10 years, and 125 (54.3%) had over 10 years of practice experience. As shown in Fig. 1, the first question addressed whether respondents considered a partially involved duct adjacent to unequivocal cribriform DCIS as ADH or DCIS and, depending on their response, whether they would recommend reexcision if this partially involved duct was less than 0.1 mm from the excision margin. Of the 230 respondents, 130 (56.5%) considered this partial cribriform proliferation as ADH and, among these respondents, 49 (37.7%) recommended reexcision of the ADH if it were close to the excision margin. On the other hand, of the 100 (43.5%) who diagnosed this partially involved duct as DCIS, 28 (28.0%) would not recommend reexcision if the lesion were close to the excision margin. Although chi-square analysis showed that pathologists who made a diagnosis of DCIS were significantly more likely to recommend reexcision compared to those who made a diagnosis of ADH (p<0.001), the final impact on patient management was that, regardless of the diagnosis, 116 (50.4%) would recommend to have a reexcision while 108 (47.0%) would not. Question 2 documented how respondents categorized a partial cribriform proliferation involving a few (five) or numerous (20 or more) ducts (Fig. 2). When only five ducts had a partial cribriform proliferation, 82 (35.7%) of the respondents considered it as DCIS. With ≥20 ducts so involved, 139 (60.4%) of the respondents designated the changes as DCIS. Chi-square testing showed this difference to be statistically significant (p<0.001). In Question 3, as shown in Fig. 3, the minimum size requirement for diagnosis of LG-DCIS in a single duct was addressed. When a single duct with a complete cribriform pattern measured 0.5, 1.5, or 4 mm, a diagnosis of DCIS was made by 52 (22.6%), 72 (31.3%), and 218 (94.8%) of the respondents, respectively. Chi-square testing confirmed that pathologists were significantly more likely to diagnose DCIS if the single duct measured >2 mm (p<0.001). Question 4 addressed the approach to management of flat epithelial atypia when it was present in multiple ducts located within less than 0.1 mm from the excision margin (Fig. 4). In this scenario, when multiple ducts with flat epithelial atypia were close to an excision margin, 48 respondents (20.9%) recommended reexcision. Finally, question 5 evaluated how respondents measured invasive carcinoma when two foci of stromal microinvasion (measuring less than 1 mm each) emanated from two opposite poles of a 1-cm duct with high-grade comedo-type DCIS (Fig. 5). Of the 230 respondents, 185 (80.4%) measured these as two separate foci of microinvasion (less than 1 mm each), while 37 (16.1%) considered the total size of invasive carcinoma as the aggregate diameter of the high-grade DCIS plus the two microinvasive foci, namely, 1.19 cm. Chi-square testing was performed to compare the responses of expert breast pathologists and nonexperts to all five questions. No statistically significant difference was found between the two groups in the proportion of responses to any of the questions. Discussions These results indicate that 15 and 20 years postintroduction of criteria for separation of ADH from DCIS [21, 31], interobserver variability in the diagnosis of intraductal proliferative lesions of the breast has not diminished. Given the significance of margin involvement or proximity in current management of DCIS, this survey demonstrates an even wider variability that exists in the management of such lesions and ultimately patient care. In a survey done by Rosai in 1991, 17 ductal and lobular proliferative breast lesions were distributed among five experts in breast pathology [24]. The participants were asked to categorize such lesions, which had already been circled on glass slides, as either hyperplasia, atypical hyperplasia, carcinoma in situ, or “other” (to be specified) based on the diagnostic criteria they used in their daily practice. In that study, there was no a single case in which all five pathologists agreed on the diagnosis, and there were only three cases (18%) in which four of the five pathologists agreed. Also, some pathologists tended to make more malignant diagnoses than others. In his report, Rosai considered this interobserver variability to be unacceptably high and suggested the adoption of a terminology such as mammary intraepithelial neoplasia with a grading system similar to that which was being used for the uterine cervix. A possible explanation for such a high degree of interobserver variability set forth in that report was that the pathologists were not using a standard set of criteria. Therefore, a year later, Schnitt et al. [27] tried a different approach to assess interobserver variability in the diagnosis of intraductal proliferative lesions of the breast. In their 1992 survey, they asked six experts in breast pathology to evaluate 24 proliferative ductal lesions. In this survey, the participating pathologists agreed to use the criteria of Page et al., and 15 teaching slides representing classic examples of usual ductal hyperplasia, atypical ductal hyperplasia, and noncomedo DCIS were circulated among the six pathologists to foster concordance before initiation of the study. The specific area of interest on each study slide was indicated by masking all the surrounding tissue so that all the participants focused on the same lesion and to prevent any bias that may result from assessment of changes in the surrounding breast tissue. The participants in this survey were asked to adhere to the provided criteria rather than the ones they used in their daily practice. Despite all these efforts, there was complete agreement among all six pathologists in only 14 (58%) of the 24 cases and among five or more pathologists in 17 cases (71%). The most common diagnostic problem was the distinction of atypical hyperplasia from DCIS in six cases. Although this study showed a significant improvement in interobserver agreement compared to Rosai’s survey, the persistence of significant differences among expert breast pathologists even under optimal and highly artificial conditions conveyed a more widespread problem in the pathology community with potential impact on patient management. Responses to the current survey also demonstrate that although criteria to distinguish ADH from DCIS have been introduced over two decades ago, there is still no agreement on how to categorize these two types of mammary intraductal proliferation. While the sole partially involved duct described in question 1 would not qualify as DCIS according to the criteria set forth by either Page et al. or Tavassoli and Norris, 43.5% of practicing pathologists, nevertheless, diagnosed it as DCIS. This might seem logical because the partially involved duct probably reflects an extension of the same process present in the nearby unequivocal cribriform DCIS. However, the fact remains that there is no consensus on what to designate a partially involved duct adjacent to unequivocal LG-DCIS and how to manage such a lesion when it is near or on the surgical excision margin [10]. Even among the respondents who considered the partially involved duct in question 1 as ADH, more than a third recommended reexcision if it were close to the margin of resection, and of those who considered this partially involved duct as DCIS, close to 30% would not recommend reexcision even if the in situ carcinoma were within 0.1 mm of the resection margin. This interobserver variability in diagnosis and management was so high that, in the end, patients would have an almost 50:50 chance of having a reexcision or not, regardless of the diagnosis of the partially involved duct as ADH or DCIS. This was due to a combination of the high proportion diagnosing it as DCIS and the high proportion of those who recommended reexcision even though they interpreted it as ADH. Obviously, the impact of this issue goes beyond simply margin involvement; it is also crucial in the assessment of lesion size, resulting in widely varied assessment of size/extent of LG-DCIS. Question 2 addressed the concept of extensive ADH. Cognizant of the fact that available criteria require complete involvement of duct cross sections for a diagnosis of low-grade (cribriform or micropapillary) variants of DCIS, it was surprising to find that even when only five ducts were involved by a partial cribriform proliferation, over 35% of the respondents considered it as DCIS. Furthermore, when greater than 20 ducts were partially involved by such a cribriform proliferation, the proportion of respondents who treated it as DCIS rose significantly to over 60%. Chi-square testing showed that as the number of partially involved ducts increased, the number of patients diagnosed with cancer significantly increased. Nevertheless, even with drastic differences in the number of involved ducts (5 vs ≥20), there was no unanimity in the diagnosis of carcinoma in situ vs atypical hyperplasia. This scenario has become increasingly more frequent since the introduction of screening mammography. It was not addressed in the criteria introduced by either Page et al. or Tavassoli and Norris because both those retrospective studies were based on predominantly premammographic era biopsies. With the increasing number of such “extensive ADH” cases (≥ 20 partially involved ducts) seen in her consultation practice, Tavassoli suggested that even when these are designated as extensive ADH due to strict adherence to previously defined criteria, they should be managed as DCIS and reexcised if close to or at the margin of resection [29]. Question 3 demonstrated that the majority of respondents used the 2-mm size criterion rather than the two spaces in diagnosing DCIS. When the single duct described in the question measured greater than 2 mm (4 mm), close to 95% of the respondents considered it DCIS. However, if this single duct measured less than 2 mm, i.e., 0.5 or 1.5 mm, 22.6 and 31.3% of the respondents considered it DCIS, respectively. Although this difference in the probability of diagnosing DCIS when the lesion measured less than or greater than 2 mm was statistically significant, it is interesting to note that over 30% of the respondents considered a single duct measuring 1.5 mm as DCIS and over 20% designated a single duct measuring only 0.5 mm as carcinoma in situ despite the fact that these two scenarios meet neither the criteria of Page et al., which require two completely involved spaces not otherwise specified as to size [21], nor the criteria of Tavassoli and Norris, which require one or more spaces greater than 2 mm in maximal cross-sectional diameter [31]. Even with a substantial 4 mm duct, 5.2% did not designate the lesion as DCIS presumably because of the two-space requirement of Page et al. These responses demonstrate that in the current day to day practice of pathology, some women with a 0.5-mm lesion would be diagnosed as having DCIS, with a reexcision if it were close to or at the resection margin, and most probably radiation therapy, whereas some women with 4 mm lesions interpreted as ADH would get no reexcision, would be simply followed up, or at most would enter some form of prevention trial with hormone therapy [2, 9]. Question 4 in our survey documented the confusion that exists regarding the management of flat epithelial atypia, with over 20% of respondents recommending reexcision if the lesion were close to an excision margin. As a result, over a fifth of the patients would have reexcision, while the remaining 80% would not. The last question showed the various approaches pathologists take in measuring invasive carcinoma when multifocal early invasion emanating from a single duct is present. Although over 80% of respondents would consider foci of invasion less than 1 mm emanating from opposite poles of a duct with DCIS as microinvasion, slightly over 16% of respondents would measure DCIS with its associated microinvasion from opposite poles as one continuous invasive carcinoma, which would entail a drastically different treatment approach. The results of this survey raise numerous questions about studies performed in different countries and even different institutions within the same country regarding risk factors, treatment, prognosis, and outcome of intraductal proliferative lesions of the breast, which include LG-DCIS and/or ADH—a significant proportion of mammographically detected noninvasive lesions. Even when the criteria used are explicitly stated, application of criteria varies remarkably among pathologists and from one study to the next. Certainly, the issues raised in this study are not uncommon but have not been specifically addressed in any of the major single-, multiinstitutional, or multinational studies on DCIS cases that include LG-DCIS. Most if not all such studies lack a central review of the diagnosis and even many rely on multiple pathologists at sometimes multiple institutions. This study also illustrates that 15 to 20 years of education of pathologists at local, national, and international courses has not helped much in increasing the level of agreement and uniformity in the diagnosis and interpretation of these common lesions using the criteria available for separating ADH from LG-DCIS. Chi-square testing failed to reveal any statistically significant difference in the response behavior of expert breast pathologists and pathologists who did not consider themselves experts. Since the institution of widespread screening mammography, the number of early, LG-DCIS lesions that are diagnosed has dramatically increased [13, 14, 19]. How partially involved ducts, often spread around completely involved ducts, are interpreted could change the size/extent of the lesion by several centimeters. This potentially undermines the internal consistency and comparability of epidemiological studies and clinical trials regarding such lesions. In his 1991 survey, Rosai lamented the fact that there were no known morphometric, ultrastructural, immunohistochemical, or molecular features to distinguish ADH from LG-DCIS [24]. Unfortunately, this issue continues to be the case [1, 11, 15, 16, 18, 25]. Pathologists must assign lesions within the ADH–DCIS continuum to one end of the spectrum or the other based on the morphologic features present on an H&E-stained slide and an arbitrary set of quantitative criteria that cannot be applied to every lesion encountered. Not only the subjective nature of interpreting the morphologic findings but also the existence of different and in some cases conflicting diagnostic criteria create considerable interobserver variability in distinguishing ADH from LG-DCIS. This diagnostic variability in turn leads to confusion regarding the optimal management approach to such intraductal proliferative lesions because a diagnosis of LG-DCIS is automatically associated with a significantly worse prognosis and usually requires a more drastic surgical approach compared to ADH. Furthermore, the negative impact of the diagnosis of “carcinoma”, albeit an in situ one, on patients’ psychological well-being (depression and anxiety) has been well documented [12, 23] and should be a consideration in the choice of optimal terminology. Molecular, immunohistochemical, and morphologic similarities indicate that the only difference between ADH and LG-DCIS is a quantitative one [1, 11, 15, 16, 18, 25]. The claim that separation of ADH from LG-DCIS is justified based on the development of carcinomas post-ADH in either breast while those that develop after LG-DCIS occur in the same breast is questionable [20, 22, 26]. Over a period of 23 years, three papers have appeared on this topic based on a group of 28 women with a median follow-up of 31 years, noting that the invasive carcinomas that occur after LG-DCIS treated by biopsy alone develop in the same breast and in the same quadrant from which the original biopsy with DCIS was taken. Seven of the 11 women who developed subsequent invasive carcinoma were diagnosed within 10 years of the DCIS biopsy, 1 was diagnosed within 12 years, and the remaining 3 developed infiltrating carcinomas over 23–42 years. The authors conclude that these results “indicate a striking dividing point biologically and histopathologically between LG-DCIS lesions and the cytologically similar but lesser lesions of ADH” [26]. When 7 of the 28 women developed invasive carcinoma within 10 years, one wonders why the remaining patients were not contacted to receive appropriate therapy for their disease rather than continuing the study for another 20 years. Judging from the microscopic description and the sampling documented for these 28 cases, a substantial number of these lesions were probably not low grade because they had some atypia and/or necrosis, albeit not diffusely, suggesting that those who died within 10 years most likely had substantial amount of residual disease and/or higher grade DCIS lesions [6]. The gross and pathologic features of these 28 cases are described in detail in the initial paper published in 1982, which did not refer to them as low grade [20]. Therefore, it is more likely that this study reflects the natural history of a wide variety of intraductal carcinomas, only some of which were LG-DCIS. A more simple explanation for any differences that might exist between ADH and LG-DCIS in the frequency of subsequent progression would be that a minuscule lesion, once totally removed, results in near equalization of the risk for subsequent development of carcinoma in the two breasts of that individual patient. A more extensive process, however, is less likely to be completely eliminated by conservative surgical excision alone, leading to the development of recurrences at the same site due to residual disease [6]. A recent epidemiological study has shown a nearly equal incidence rate (5.4 vs 4.5/1,000 person-years) for development of subsequent invasive carcinoma in either breast after a diagnosis of DCIS among close to 37,700 patients diagnosed between January 1988 and December 2002, many of whom were treated by surgery and radiation therapy [17]. This confirms the fact that once a DCIS is adequately treated, it results in equalization of risk for either breast. Furthermore, it indicates that complete excision is an adequate therapy for small lesions. Therefore, complete excision with adequate margins should be explored further as the only therapy for small DIN 1 (DCIS grade 1) lesions that do not exceed 3–4 mm in maximum extent, particularly when there is no evidence of even any flat epithelial atypia around the DCIS. It seems only logical to adopt a terminology in which two morphologically identical lesions are not given diagnostic designations as different as “hyperplasia” and “carcinoma”. The use of the DIN terminology may help solve this problem [28, 30]. Although the DIN terminology does not claim to reduce interobserver variability, it diminishes the effect of drastically different designations for similar lesions by including the spectrum of atypical ductal hyperplasia and LG-DCIS under the umbrella designation of grade 1 DIN.

Environ_Health_Perspect-114-9-1570082

GIS Approaches for the Estimation of Residential-Level Ambient PM Concentrations

Spatial estimations are increasingly used to estimate geocoded ambient particulate matter (PM) concentrations in epidemiologic studies because measures of daily PM concentrations are unavailable in most U.S. locations. This study was conducted to a) assess the feasibility of large-scale kriging estimations of daily residential-level ambient PM concentrations, b) perform and compare cross-validations of different kriging models, c) contrast three popular kriging approaches, and d ) calculate SE of the kriging estimations. We used PM data for PM with aerodynamic diameter ≤10 μm (PM10) and aerodynamic diameter ≤ 2.5 μm (PM2.5) from the U.S. Environmental Protection Agency for the year 2000. Kriging estimations were performed at 94,135 geocoded addresses of Women’s Health Initiative study participants using the ArcView geographic information system. We developed a semiautomated program to enable large-scale daily kriging estimation and assessed validity of semivariogram models using prediction error (PE), standardized prediction error (SPE), root mean square standardized (RMSS), and SE of the estimated PM. National- and regional-scale kriging performed satisfactorily, with the former slightly better. The average PE, SPE, and RMSS of daily PM10 semivariograms using regular ordinary kriging with a spherical model were 0.0629, −0.0011, and 1.255 μg/m3, respectively; the average SE of the estimated residential-level PM10 was 27.36 μg/m3. The values for PM2.5 were 0.049, 0.0085, 1.389, and 4.13 μg/m3, respectively. Lognormal ordinary kriging yielded a smaller average SE and effectively eliminated out-of-range predicted values compared to regular ordinary kriging. Semiautomated daily kriging estimations and semivariogram cross-validations are feasible on a national scale. Lognormal ordinary kriging with a spherical model is valid for estimating daily ambient PM at geocoded residential addresses. Large-scale, population-based epidemiologic investigations of the health effects of ambient air pollution often rely on measurements from a network of air quality monitors maintained by the U.S. Environmental Protection Agency (U.S. EPA 1995a, 1995b, 2005). The Air Quality System (AQS) is the only national ambient air pollution database currently available for public use in the United States. The availability of individual-level health outcome and covariable data from national-scale studies that often characterize participants over the course of several years enables researchers to study the acute effects of ambient air pollution using individual-level data (Liao et al. 2004, 2005a; Sullivan et al. 2005; Wellenius et al. 2005; Whitsel et al. 2004). This approach requires measures of daily particulate matter (PM) exposures, ideally assessed as close to the individual level as possible, such as at participant residences or in immediate proximity to participants themselves. Because daily measures of ambient PM concentrations from the AQS are unavailable in the large majority of locations, spatial estimation methods using geographic information systems (GIS) are increasingly being considered to estimate geocoded location-specific ambient PM concentrations, such as kriging methods. Important methodologic and practical issues still need to be resolved, however. This study was designed to a) assess the feasibility of large-scale kriging estimation of daily residential-level ambient PM concentrations, b) perform and compare cross-validations of different kriging models, c) determine and contrast the most appropriate kriging approaches, and d) calculate the SEs of the kriging estimations. Materials and Methods We obtained from AQS the PM10 and PM2.5 (PM with aerodynamic diameter ≤ 10 and 2.5 μm, respectively) data from 1993–2004 (U.S. EPA 2005). The data from 2000 were used for this study after eliminating duplicate records and converting all measures to the same units and denominator. We calculated “monitor-specific” daily averages based on ≥ 18 hourly measures. Monitor-specific daily averages were set to missing for monitors reporting < 18 hourly measures on any given day. If more than one monitor was operating at the same location on a given day, we then computed “site-specific” daily PM10 and PM2.5 averages by taking the mean of the monitors’ measures. We also obtained the longitude and latitude for each site from the AQS database. These data served as pollutant- and site-specific daily source data for our study (Liao et al. 2005b). We geocoded 94,135 addresses of Women’s Health Initiative (WHI) Clinical Trial (CT) participant residences and examination sites in the contiguous 48 United States and District of Columbia, after assessing geocoding vendor error (Whitsel et al. 2004, 2005). Daily PM10 and PM2.5 concentrations and the associated estimation errors (SEs) are estimated at these geographic locations by the Environmental Epidemiology of Arrhythmogenesis in WHI study (Whitsel 2006). We used ArcView GIS (version 8.3) and its Geostatistical Analyst Extension (ESRI Inc., Redlands, CA) for semivariogram determination and cross-validation and for subsequent spatial estimation of daily location-specific PM concentrations. Three frequently referenced spatial models (spherical, exponential, Gaussian) (Cressie 1993a; Davis 2002) were considered using the weighted least-squares method (Gribov et al. 2004; Jian et al. 1996) to obtain the “optimal” daily semivariogram parameters (range, partial sill, and nugget). Based on the daily semivariograms, we performed ordinary kriging to estimate the daily mean PM concentration and its SE at each of the 94,135 geocoded addresses. Next, we performed the standard cross-validation—an iterative procedure that omits site-specific PM data points one at a time and refits the model using the remaining data to estimate the PM concentration at the site of the omitted observation. We assessed the validity (also termed “goodness of fit”) of each semivariogram using three cross-validation parameters readily available from the ArcView software package: a) the average of prediction error (PE), where PE is the average of the difference between the predicted and measured daily PM values at each monitoring site; b) the average of standardized prediction error (SPE), where SPE is the PE divided by the SE of estimation across all sites; and c) root mean square standardized (RMSS), the standard deviation (SD) of all SPEs across all sites. Additionally, we assessed the goodness of fit of each semivariogram by the average of the SEs of the estimations, generated by the kriging procedure, across all 94,135 geocoded addresses. The expectations for a good-fitting semivariogram and kriging model are an average PE and SPE near 0, an RMSS near 1, and a small SE. If RMSS < 1, there is a tendency toward overestimation of the variance; if > 1, there is a tendency toward underestimation (ESRI Inc. 2001). These criteria were consistently used to guide our model selection processes throughout this study (Liao et al. 2005c). As an alternative to using the automatically calculated semivariogram (calculated using the weighted least-squares method (Gribov et al. 2004; Jian et al. 1996), one can also manually specify the semivariogram parameters to improve the cross-validation parameters in ArcView. We selected six least satisfactory daily semivariograms throughout year 2000 and manually adjusted the semivariogram parameters to obtain the best achievable average RMSS and SPE (RMSS as close to 1 and average SPE as close to 0 as possible). The cross-validation parameters from the weighted least-squares method–calculated semivariograms were then compared to those of the manually adjusted semivariograms. We performed daily ordinary krigings on both the original scale (regular ordinary krig-ing) and the lognormal scale (lognormal ordinary kriging) (Cressie 1993b; Johnston 2001) for all WHI CT addresses for the year 2000 and compared the cross-validation parameters between the two kriging procedures. Log-normal ordinary kriging was used because it has the ability to eliminate the negative predicted values, which is a problem in ordinary kriging, especially when the source data contain extreme values. Results Characteristics of the site-specific daily average PM10 and PM2.5 concentrations During 1994–2003, the number of monitoring sites that provided GIS-usable daily PM10 data varied widely (range, 120–1,340). On 17% of days, GIS-usable data were provided by ≥ 400 monitoring sites; on 39% of days, by 200–400 sites; and on 44% of days, by 120–200 sites. The corresponding values for PM2.5 during 1999–2003 were 33% of days by ≥ 400 sites and 67% of days by 148–400 sites. Specific to the year 2000, there were averages of 325 PM10 and 456 PM2.5 monitoring sites operating per day across the contiguous United States, with minima and maxima of 148 and 1,061 sites for PM10 and 178 and 1,019 sites for PM2.5. As a result, there were 118,791 site-days during 2000 for which we can retrieve measured PM10 data and 166,796 site-days for PM2.5 data. The mean (± SD) of PM10 and PM2.5 from these retrievable site-days were 26.29 ± 58.13 and 13.14 ± 8.59 μg/m3, respectively, with medians of 21.33 and 11.20 μg/m3, respectively. A right-skewed distribution of both PM10 and PM2.5 are evident, especially for PM10. Figure 1 illustrates the spatial relationships between the geocoded addresses and the PM monitoring sites on an optimal day and a typical day. The mean distance between each address and its nearest PM monitor was 12.35 km, with an SD of 13.98 km, a median of 7.81 km, an interquartile range of 10.53 km, and 99th percentile of 68.36 km. Comparisons of three widely used spatial models Tables 1 and 2 present summary statistics of the cross-validation parameters (PE, SPE, and RMSS) comparing three widely used spatial models (spherical, exponential, Gaussian) for PM10 and PM2.5, respectively. In general, both average PE and average SPE are very close to 0, with a very narrow range of variation from the 366 daily cross-validations. More specifically, > 95% of average PEs were within ± 2 μg/m3 of measured PM10, and ± 0.5 μg/m3 of measured PM2.5, an average measurement error that we considered acceptable. In terms of RMSS, we considered > 95% of cross-validations as acceptable, but there were days when RMSS indicated a slight over-or underestimation of the prediction variability. These data support the overall validity of using kriging-based estimation approaches to estimate location-specific PM concentrations across the contiguous United States. Comparisons of default and manually adjusted semivariograms Table 3 presents the cross-validations and actual kriging estimations from the weighted least-squares mean method calculated semivariogram and manually adjusted semivariogram. For the 6 days when the PE, SPE, or RMSS indicated a less satisfactory default-calculated semivariogram, these three cross-validation parameters could be improved satisfactorily through adjustment of the semivariogram parameters by an operator. However, the application of such “improved” semivariograms to the estimation of PM10 concentrations at geocoded locations across the United States did not necessarily provide better estimation of location-specific PM (i.e., smaller SEs). To the contrary, the average SEs from the default semivariograms were smaller than those from manually adjusted semivariograms. Because each average SPE of the default-calculated daily semivariograms was close to 0, and each default-calculated daily semivariogram produced a smaller estimation error, we recommend using the default-calculated semivariogram, even though the RMSS from the default-calculated semivariogram was not fully satisfactory. Comparisons of regular versus lognormal ordinary krigings We applied regular ordinary kriging (spherical model, default-calculated daily semivariograms) to estimate daily PM10 concentrations at geocoded addresses (n = 94,135) of WHI CT participants and examination sites in the contiguous United States. We examined the estimated PM10 concentrations and identified 22 days during 2000 when estimated values exceeded the range of observed values. In some cases, the estimated values were negative. The number of addresses affected by this problem ranged from a few on most days to 3.5% of all addresses. This problem was related to skewed PM10 distributions and to small numbers of extreme outlying values or operating sites on some days. We therefore compared regular ordinary kriging and log-normal ordinary kriging anticipating that log-normal kriging would attenuate this problem. Table 4 lists the 22 days on which regular ordinary kriging yielded estimated PM10 values that were outside the range of measured values. For comparison, the minima and maxima of the measured and estimated PM10 concentrations from both regular and log-normal ordinary krigings are also listed in Table 4. In summary, during 2000, lognor-mal ordinary kriging effectively reduced the number of problematic days from 22 to 1. Even on this one day, lognormal ordinary kriging yielded a minimum value that was closer to the range of measured data than that from regular ordinary kriging. Table 5 shows the mean values of cross-validation parameters of daily PM10 semivariograms for both regular ordinary kriging and lognormal ordinary kriging. Cross-validation parameters were within the acceptable range from both regular and lognormal ordinary krigings, except for the 22 “out-of-range” days as defined above. On these out-of-range days, the SPE was well within the acceptable range for both regular and lognormal krigings, but the RMSS was > 1 from both approaches. Even so, for these out-of-range days RMSS from lognormal ordinary kriging was closer to 1 than that from regular ordinary kriging. We then performed regular and lognormal ordinary kriging to estimate PM10 concentrations at geocoded addresses of WHI CT participants and examination sites, based on year 2000 PM10 data (94,135 locations and 366 days). The mean, SD, median, and maximum of the daily mean SE of the estimated PM10 from the regular ordinary kriging were 27.36, 83.35, 13.93, and 1160.20 μg/m3, respectively. In contrast, those from the lognormal ordinary kriging were 16.29, 6.65, 15.05, and 67.46 μg/m3. Clearly, the distribution of the estimation errors from lognormal ordinary kriging was considerably less skewed and had fewer outlying values than that from regular ordinary kriging. Alternative methods (winsorizing extreme PM10 values; using ArcView’s “no-sector” option to search for measured data points from a circle centered around a location that needs of an estimation—i.e., disabling the default “sector” search for measured data points in the four sectors of a circle, reducing the range or nugget) were less effective in estimating predicted values within the range of measured values (data not shown). Similar to the situation observed in PM10 estimations, lognormal ordinary kriging also effectively eliminated the negative or out-of-range problem that occurred in about 5% of PM2.5 data when using regular ordinary kriging. Other cross-validation parameters were comparable between the lognormal and regular ordinary krigings (data not shown). Comparisons between national and regional krigings From the 61 days when 900 or more monitoring sites were operating in the year 2000 in the 48 contiguous states, the first of such days from each month was selected for comparisons between ordinary kriging models on a national versus regional scale. National krigings and cross-validations were performed on these 12 selected days using daily site-specific PM10 data. Regional krigings and cross-validations were performed on the same data using the regional map (Figure 1) that divides the U.S. continent into five regions (northwest, southwest, middle north, southeast, and northeast). These five regions were created based on the assumption that different semivariogram parameters would be needed for different geographic areas. In general, for both regional and national krigings, the average SPE and RMSS from cross-validations of semivariograms calculated for the 12 selected days were very close to 0 and 1, respectively (Table 6) Discussion Classical methods often assume that measures are uniformly or randomly distributed. The assumptions are often inappropriate for analysis of environmental measures because values at neighboring locations are rarely independent, particularly over short distances. This form of dependence (spatial autocorrelation) nonetheless makes it possible to interpolate values at unmonitored locations from known values at monitored locations. Kriging is one such interpolation method originally developed by mining engineers (Krige 1966). It is especially attractive in this setting because it takes the spatial autocorrelation structure function (variogram) into account by considering known values from monitored locations, weighting them with values read from the variogram at corresponding distances, and splitting weights among adjacent locations. The method thereby ensures that interpolations do not depend on monitor density (Legendre and Fortin 1989). By doing so, kriging yields best linear unbiased estimates, in this setting, of location-specific daily mean ambient PM concentrations and their SEs. Large-scale population-based epidemiologic investigations of the health effects of ambient air pollution often rely on data collected from a network of air quality monitors maintained by the U.S. EPA—the AQS data (U.S. EPA 1995a, 1995b, 2005). It is revealing to compare kriging with interpolation methods used in the well-known time-series and cohort studies of PM effects on mortality and cardiovascular disease (Abbey et al. 1991, 1999; Dockery et al. 1993; Katsouyanni et al. 1996, 2001; Miller et al. 2004, 2005; Pope et al. 2004; Samet et al. 2000a, 2000b). These studies uniformly estimated PM exposures using area-based arithmetic averaging or nearest-neighbor imputation—alternative methods that have important limitations (Moore and Carpenter 1999). Such limitations include the assumption of homogeneous exposures within study areas and the inability (or failure) to estimate exposures or associated PEs. For example, when daily exposure was of interest and there were no operating PM monitors with a study area, data pairs (daily PM concentrations, death counts) were unavailable in these studies. In addition, when longer-term (monthly to yearly) exposure was of interest, area aggregated exposures were based on available measurements within a given time frame. If there were five 24-hr measures in a month, for example, the monthly average exposure was calculated as the mean of the five readings. In contrast, our kriging-based approach estimated daily mean exposures and SEs at geocoded addresses of participants and their examination sites across the contiguous United States that can be readily integrated over time with little influence of missing data. Studies in the geosciences have also found that kriging provides consistently improved interpolation accuracy over traditional inverse-distance weighting and other, simpler spatial interpolation methods (Zimmerman 1999). Another important advantage of GIS-based estimation over the traditional area-average approach is the availability of both the location-specific estimated pollutant concentrations and their SEs. Our goal in this study was to contribute methodologic and practical insights toward standardized, semiautomated GIS approaches to estimation of daily air pollution concentrations and their associated estimation errors. The air pollution data estimated using these approaches will support the Environmental Epidemiology of Arrhythmogenesis in WHI study (Whitsel 2006) examining the cardiac effects of air pollution in 68,133 post-menopausal women 50–79 years of age at baseline in the WHI CT (WHI Study Group 1998). Here we describe our experience resolving several important methodologic and practical issues in adopting a systematic, standardized, and semiautomated kriging approach to estimate daily air pollution concentrations and the associated estimation errors at geocoded addresses across the contiguous United States over 10 years. We successfully downloaded from AQS the PM10 and PM2.5 raw data from 1993–2004. We then cleaned, calculated, and reconstructed site-specific daily PM concentration data ready for GIS applications. It is well known that the monitoring sites in AQS are not randomly distributed, which is one of the assumptions in kriging estimation, and the density of the monitoring sites is relatively low given the size of the contiguous United States. However, the AQS is the only currently available nationwide database. Our cross-validation studies suggest that the AQS data can be used as source data for kriging estimation of ambient pollution concentrations at various locations across the 48 contiguous states. In this study, we performed cross-validation to assess the goodness of fit of various semivariogram and spatial models using four major parameters: the average PE, SPE, RMSS, and SE of estimation. Details can be found elsewhere (Webster and Oliver 2001), but it is worth noting that in addition to using the SE as a measure of the goodness of fit of a kriging model, one could improve the health effects models by incorporating SE in the models to account for the error in the estimation of location-specific PM concentrations. We consider this an important advantage of GIS-based estimation over the traditional area-average approach and are performing studies of using SE in health effects models. We compared the performance of three widely spatial models (spherical, exponential, Gaussian) for PM10 and PM2.5 estimations using regular ordinary kriging on a national scale (Tables 1 and 2). In general, the cross-validation parameters suggest that all three models performed fairly well. Overall, the spherical model seemed to perform slightly better, consistent with the observation that the spatial distribution pattern of ambient air pollutants is closest to the assumption of the spherical model. The spherical model has been used most often in modeling spatially distributed data, providing a further rationale for its use in our large-scale population-based study of the health effects of PM. Furthermore, from the perspective of the cross-validation results, both average PE and average SPE are very close to 0, with a very narrow range of variation from the 366 daily cross-validations. These data support the overall validity of using kriging-based estimation approaches to estimate location-specific PM concentrations across the contiguous United States. We completed an empirical analysis to investigate whether manually adjusting semivariogram parameters improves a) cross-validation parameters and b) estimated PM10 concentrations and their SEs (Table 3). From these data, we conclude that manually adjusting semivariogram parameters improves cross-validation parameters. However, the application of such “improved” semivariograms to the estimation of PM10 concentrations at geocoded locations across the United States did not necessarily provide better estimation of location-specific PM. Therefore, we recommend using the default-calculated semivariogram. Semivariograms are sensitive to strong positive skewness. As a result, regular ordinary kriging can yield negative predicted values or values exceeding the range of the source data. Kriging works best if the input data have a normal distribution. One solution is to log-transform the input data—using “lognormal kriging.” In the ArcView software package, performing lognormal kriging is a standard option. This option log-transforms the input data to normalize its distribution and attenuate the impact of very large values. It also back-transforms the estimated values and the “unbiased” SE of the estimation to the original scale (Cressie 1993b; Johnston 2001). Our results comparing lognormal ordinary kriging versus regular-scale ordinary kriging suggest that lognormal ordinary kriging not only effectively estimated location-specific PM concentrations within the range of the measured data for the days regular ordinary kriging yielded negative or “out of range” PM estimations, but also yielded a smaller average SE than did regular ordinary kriging and estimations. Therefore, our results support the use of lognormal ordinary kriging as an acceptable solution to the problem commonly posed by positively skewed distributions of environmental data. Our comparisons of national- versus regional-scale kriging indicate that, in terms of cross-validation results, both performed similarly. However, such comparisons are based on krigings using the source data from optimal days (when > 900 sites across the country were reporting data), which account for only 17% of all days in a year. Therefore, there is additional justification for using national-scale kriging: Usually, there were very few operating sites within a region. On typical days—when only about 200 monitoring sites were operating—ability to derive stable and meaningful semivariograms was greatly impaired. Regional kriging also poses problems for estimation at locations near regional borders. For example, at locations within Washington State but near the Washington–Idaho border, regional kriging is based solely on PM10 concentrations in the “Washington/Oregon, Northern California” region. It is not based on PM10 concentrations measured immediately across the border in Idaho, despite the real possibility that they would have the largest weights in national-scale kriging estimation. For all these reasons, we recommend national-scale kriging. Considering the number of study participants and the length of study period (1994–2003) for the Environmental Epidemiology of Arrhythmogenesis in WHI study, development of an automated procedure enabling large-scale daily krigings and semivariogram cross-validations was critical. In this study, we decided to use ArcView for predicting individuals’ PM exposure concentrations because of the flexibility it offers for automation. Because ArcView GIS relies on either the weighted least-squares method or visual adjustment to create semivariograms, we did not compare the relative performance of semivariograms generated using alternative methods such as maximum likelihood and restricted maximum likelihood. For generating semivariograms, we compared only three popular spatial models (spherical, exponential, and Gaussian). Our results, however, do not invalidate alternative spatial models (e.g., power). In the end, we selected the spherical model for our study because it is the most studied model, and its assumption pertaining to the spatial correlation of data is probably closest to our pollutant data. Furthermore, the spherical model seemed to perform as well as or slightly better than the remaining models in terms of cross-validation parameters. We chose ordinary kriging instead of universal or simple kriging for several reasons. First, the assumption for simple kriging of a known mean concentration on any given day across space is not practical for our data. Although it may seem more appropriate because of the “varying mean” concentration across the contiguous U.S. assumption, universal kriging requires a predetermined set of “exploratory variables” to explain the varying means. The candidates, many of which are spatial variables, include emissions, land use, population, road network distribution, altitude, rainfall, latitude, climatology, and other quality data. Denby et al. (2005) recently recommended a method that uses measured concentration data in combination with some “exploratory variables” as suggested above. However, their approach may not be feasible for a national-scale study such as ours, because little guiding information is available as to how to identify a set of widely acceptable variables that can be applied to the entire nation. Moreover, even if we could identify a set of exploratory variables, we do not know the forms or shapes of their independent and joint relations to the air pollution measures. Further studies that involve large-scale national data using universal kriging are still needed. In this study, we empirically tested whether the non-constant mean assumption for universal kriging was needed; we performed five regional ordinary krigings so that different parts of the country would assume a different mean PM concentration. Our data suggested that regional and national ordinary kriging performed similarly. Therefore, our data indirectly validated and supported the use of national ordinary kriging. Although the primary objective of our study is to assess the short-term relationship between PM and cardiac responses, the proposed kriging method also enables us to calculate the long-term cumulative exposure of an individual by taking into account the change of his or her residences over time, because the WHI study recorded the residential location history over 10 years. Nevertheless, from the environmental perspective, an inherited limitation of the kriging-based approach is that the estimations of the PM concentrations will provide only surrogates, or the best guesses, of the true exposure levels at the locations of interest. Thus, the accuracy of the estimations depends highly on the quality of the measured data and their spatial correlation. Even if the estimations were made with a high level of confidence, they cannot be directly interpreted as the true individual-level exposures. However, to correlate individual level cardiac responses with a surrogate of location-specific exposure, our approach represents one of the best available methods for a large-scale population-based study. In summary, our investigation of GIS approaches for estimating daily mean geocoded location-specific air pollutant concentrations and their SEs supports the use of a spherical model to perform lognormal ordinary kriging on a national scale. Our findings also support the use of default-generated semivariograms (estimated using the weighted least-squares method) without visual adjustment. We developed a semiautomated program to access and execute ArcView to implement these approaches for large-scale daily kriging estimations and semivariogram cross-validations. Detailed information about this program can be obtained on request.

Rev_Endocr_Metab_Disord-3-1-1894829

Monogenic diabetes in children and young adults: Challenges for researcher, clinician and patient

Monogenic diabetes results from one or more mutations in a single gene which might hence be rare but has great impact leading to diabetes at a very young age. It has resulted in great challenges for researchers elucidating the aetiology of diabetes and related features in other organ systems, for clinicians specifying a diagnosis that leads to improved genetic counselling, predicting of clinical course and changes in treatment, and for patients to altered treatment that has lead to coming off insulin and injections with no alternative (Glucokinase mutations), insulin injections being replaced by tablets (e.g. low dose in HNFα or high dose in potassium channel defects -Kir6.2 and SUR1) or with tablets in addition to insulin (e.g. metformin in insulin resistant syndromes). Genetic testing requires guidance to test for what gene especially given limited resources. Monogenic diabetes should be considered in any diabetic patient who has features inconsistent with their current diagnosis (unspecified neonatal diabetes, type 1 or type 2 diabetes) and clinical features of a specific subtype of monogenic diabetes (neonatal diabetes, familial diabetes, mild hyperglycaemia, syndromes). Guidance is given by clinical and physiological features in patient and family and the likelihood of the proposed mutation altering clinical care. In this article, I aimed to provide insight in the genes and mutations involved in insulin synthesis, secretion, and resistance, and to provide guidance for genetic testing by showing the clinical and physiological features and tests for each specified diagnosis as well as the opportunities for treatment. Introduction Monogenic diabetes is diabetes that results from one or more mutations in a single gene. The mutation might have arisen de novo and hence be a spontaneous case or might be dominantly or recessively inherited. Molecular genetic testing specifies a diagnosis in 1–2% [1]. The mechanism can be easily understood from the underlying pathophysiology. In children, almost all monogenic diabetes results from mutations in genes that regulate β-cell function and infrequently from mutations resulting in very severe insulin resistance [2]. This article discusses the pathophysiology and clinical manifestations used to select patients eligible to genetic testing, and demonstrates the importance in the treatment of monogenic diabetes. Why diagnose monogenic diabetes? Many patients with genetically proven monogenic diabetes were initially incorrectly diagnosed. They received a diagnosis of neonatal diabetes or diabetes diagnosed before the age of 6 months that was not further specified [3] or diagnosed as type 1 or type 2 diabetes [3–5]. It is important to correctly diagnose monogenic diabetes because not only does it help to elucidate the aetiology of the patient’s diabetes and explain other associated clinical features, it can also predict the clinical course of the patient and guide the most appropriate treatment. For instance, patients might not need any treatment or might be able to switch from insulin injections to tablets such as sulfonylurea [6, 7]. Finally a diagnosis has implications for other family members often correcting their diagnosis, prognosis and treatment as well as allowing appropriate genetic counselling. Normal insulin release and normal insulin sensitivity The β-cells form the core of the islets of Langerhans and are responsible for the synthesis and secretion of insulin and C-peptide (Fig. 1, upper part). Insulin’s main function, stimulating glucose uptake in peripheral tissue, is regulated by insulin binding to insulin receptors on the cell membrane of peripheral cells such as muscle cells. This binding initiates an intracellular signalling cascade that leads to an increase in the glucose influx (Fig. 2). Fig. 1Pancreatic β-cell and the genes involved in monogenic diabetes. Mutations in different genes result in different phenotypes (Tables 1, 2, 3, and 4). Also, different mutations in the same gene might lead to different phenotypes as shown in the spectrum of phenotypes in for instance Kir6.2.The upper part shows the physiological situation from insulin synthesis to packaging and from glucose sensing to insulin secretion. From centre to right: Insulin synthesis and packaging: Insulin (Ins) is synthesized in the nucleus regulated by transcription factors and after translation in the endoplasmatic reticulum (ER) and Golgi apparatus (Golgi) stored in granules. From left to right down: Glucose sensing and insulin secretion: Glucose enters the β-cell by passive diffusion facilitated by the glucose-transporter-2 (GLUT2). It is phosphorylated by the enzyme glucokinase (GCK) to glucose-6-phosphate (G6P) and metabolised to ATP via glycolysis or even further via the Krebs cycle in the mitochondria (Mito). ATP closes the KATP channel, preventing K+ efflux, depolarising the cell membrane. Depolarisation opens voltage dependent calcium channels (VDCC) allowing calcium influx. The rise in intracellular calcium (Ca2+) helps the insulin granules to fuse with the cell membrane resulting in insulin secretion.The lower part shows the pathological situation due to mutations in the genes involved in monogenic diabetes. The proteins encoded by the genes involved are given in bold type followed by the clinical presentation if mutated (in brackets and italics). From centre to right: Insulin synthesis is mainly influenced by nuclear transcription factors that may also be involved in pancreatic development and hence mutations may result in pancreatic atrophy (PTF1α, HNF1β) or agenesis (IPF1) rather than reduced insulin synthesis per se as in TNDM (ZAC) or MODY (HNF1α, HNF4α, NEUROD1). Mutations in genes that are involved in packaging of the insulin into granules in the ER and Golgi apparatus result in TNDM (HYMAI), β-cell destruction and PNDM (EIF2AK3 in Wolcott Rallison Syndrome—WCRS), diabetes at a mean age of six as part of the Diabetes Insipidus, Diabetes Mellitus, Optic Atrophy and Deafness syndrome (DIDMOAD) alternatively called Wolfram syndrome (WRS) (WSF1), or in diabetes as part of the Thiamine Responsive Megaloblastic Anaemia Syndrome (TRMAS) alternatively called Roger’s syndrome. Mutations in T-lymphocytes may also lead to β-cell destruction and PNDM as seen with mutations in FOXP3 as part of the Immunodysregulation, Polyendocrinopathy, Enteropathy, X-linked (IPEX) syndrome. From left to right down: Glucose sensing is reduced by mutations in GLUT2- leading to TNDM as part of the Fanconi-Bickel Syndrome, GCK- resulting in MODY2 and mitochondrial DNA-interfering with oxidative phosphorylation. These latter three mutations reduce ATP and increase ADP leading to a decreased ATP/Mg-adenosine diphosphate (ADP) ratio that activates the KATP channel to remain open. Activating mutations in the KATP channel itself (Kir6.2/SUR1) reduce sensitivity to ATP and hence also favour the open state of the channel. The subsequent efflux of potassium prevents depolarisation of the cell-membrane and hence prevents insulin secretionFig. 2Peripheral cell (e.g. muscle cell) showing the impact of diverse mutations that result in insulin resistance. In the physiological situation, insulin binds to the insulin receptor resulting in phosphorylation (P) of a tyrosine (Y) residue of the insulin receptor substrate-1 (IRS1). This phosphorylation activates phosphatidylinositol 3-kinase (PI3K) resulting in glucose transporter 4 (GLUT4) being translocated to the cell membrane which in turn leads to glucose influx. In the pathological situation, mutations in the insulin receptor interfere with insulin receptor synthesis, posttranslational processing and intracellular-transport of the receptor to the cell membrane or lead to reduced binding of insulin, reduced activation or increased degradation of the receptor. The result is no phosphorylation of Y and hence eventually no glucose uptake. This situation occurs in Type A severe insulin resistance, Rabson-Mendenhall and Leprechaunism. In the presence of high levels of triglycerides (TG) and hence free fatty acids (FFA) a serine (S) residue is phosphorylated preventing phosphorylation of Y and hence eventual glucose uptake. This situation occurs in congenital generalised lipoatrophy and familial partial lipodystrophy. In all these pathological situations, higher levels of insulin are needed for glucose uptake and hence lead to reduced insulin sensitivity and insulin resistance Abnormal insulin release and abnormal insulin sensitivity Most monogenic diabetes in children results from mutations in genes causing β-cell loss or β-cell dysfunction and hence affect insulin synthesis, packaging, glucose sensing or insulin secretion (Fig. 1, lower part) [3, 8]. Mutations of pancreatic transcription factors may result in reduced pancreatic growth, (sub)total pancreatic agenesis or reduced insulin synthesis per se and may present as neonatal diabetes (ND), maturity onset diabetes of the young (MODY), or diabetes as part of a syndrome (Fig. 1, lower part centre). Other mutations that affect insulin synthesis are not expressed in the pancreatic β cell but in the CD4+ CD25+ regulatory T lymphocytes, where overactivation results in auto-immunity against β-cells typically causing diabetes in the first 3 months of life. Mutations affecting translation, cleavage and packaging of insulin may result in reduced β-cell number or reduced number of granules and lower concentrations of insulin in these granules (Fig. 1 lower part right). Mutations that affect insulin secretion involve genes that regulate glucose sensing and affect beta-cell function rather than its development or destruction. They include functional mutations in mitochondrial DNA (Fig. 1, lower part left). All these mutations reduce sensitivity to glucose and its metabolism favouring the open state of the potassium channels prevent depolarisation and hence insulin secretion. In children, monogenic diabetes infrequently occurs from mutations resulting in very severe insulin resistance (Fig. 2) [2]. They are mainly caused by mutations in the insulin receptor gene, that affect its synthesis and posttranslational processing, increase receptor degradation, reduce binding of insulin or receptor activation [9]. These give rise to type A severe insulin resistance, Rabson-Mendenhall syndrome or Leprechaunism [2, 9–11]. Alternatively, insulin resistance may be the result of hypertriglyceridaemia associated with congenital generalised lipoatrophy or familial partial lipodystrophia [12, 13]. When to consider a diagnosis of monogenic diabetes? Given the limited resources available it is vital that genetic tests are used in situations where they are likely to be positive and will alter clinical care. This will involve careful clinical selection and physiological tests like C peptide and autoantibody measurement as well as examination of other family members before doing molecular genetic tests. Monogenic diabetes should be considered in any diabetic patient who has features inconsistent with their current diagnosis and clinical features of specific subtypes of monogenic diabetes. No specified diagnosis or features inconsistent with current diagnosis A diagnosis of diabetes might have been made without further specifying the cause as in neonatal diabetes or diabetes diagnosed below the age of 6 months. Also, the majority of patients with genetically proven monogenic diabetes are initially incorrectly diagnosed as Type 1 or Type 2 diabetes [5]. No specified diagnosis of neonatal diabetes or of diabetes diagnosed below the age of 6 months Neonatal diabetes was defined as insulin requiring diabetes diagnosed within the first 3 months of life. It is an area that has rapidly transitioned from a clinical to a molecular genetic classification [3, 14]. There is good evidence that diabetes diagnosed in the first 6 months of life is not type 1 diabetes because neither autoantibodies typical for diabetes nor an excess of high type 1 HLA susceptibility are found in these patients [15]. Diabetes diagnosed before the age of 6 months irrespective of current age should always lead to further molecular genetic testing. Also because many years after diagnosis, patient and doctor will find difficulty in recalling the exact age of diagnosis and only be aware that it was ‘at very young age’. Features that help differentiate between the different subtypes and hence guide genetic testing are further described in the section below and in Table 1. Table 1Features of diabetes diagnosed before 6 months of age in addition to undetectable to low C-peptideProtein (Chromosome/gene; Syndrome)Clinical pictureNumber of cases describedMedian birth weight ingrams SDS (standard deviation score)Median age at diagnosis in weeks (range)Family history reflected by inheritanceOther clinical featuresOther testsTreatment(% in consanguineous or isolated populations)Pancreatic appearance (present/size)• ZAC/HYMAI (6q24 imprinting defect)TNDM±150 (rare)2,100 (−2.94)0.5 (0–4)- Macroglossia (23%)NormalInsulin/pump > relapse: diet > insulin• Kir6.2 (KCNJ11)TNDM10% PNDM90%±100 (rare)2,580 (−1.73)6 (0–260)- Spontaneous- DKA (30%)NormalHigh dose sulfonylurea- Dominant(10%)- Developmentaldelay 20%)- Epilepsy (6%)• PTF1A(10p13-12)PNDM3 (100%)1,390 (−3.8)Recessive- Severeneurological dysfunctionAtrophyInsulin/pump- Cerebellar hypoplasia• IPF1 (13q12.1)PNDM2 (50%)2,140 (−2.97)- RecessiveNo pancreasInsulin/pump- Parents may have early onset diabetes as heterozygotes• HNF1β (179)TNDMRare1,900 (−3.21)- Dominant (60%)- Renal development disordersAtrophyInsulin/pump- Spotaneous• EIF2AK3(2p;Wolcott-Rallison Syndrome)PNDM30 (90%)13 (6–65)- Recessive- Epiphyseal dysplasia (90%)Exocrine dysfunctionInsulin/pump- Developmental delay (80%)- Acute liver failure (75%)- Osteopenia (50%)- Hypothyroidism (25%)• FOXP3 (Xp11.23; IPEX Syndrome)PNDM14 (rare)2,860 (−1.2)6 (0–30)X-linked Hence only boys affected- Chronic diarrhoea with villous atrophy (95%)Insulin/pump- Pancreatic and thyroid autoantibodies (75%)- Eczena (50%)- Anaemia (30%)- Thyroiditis (20%)- Often die in first year• GLUT2 (3q; Fanconi Bickel Syndrome)TNDMRecessive- Impaired utilisation of glucose and galactoseInsulin/pump- Hepatorenal glycogen accumulation- Proximal renal tubular dysfunction > glucosuria• Glucokinase (GCK11 homozygote)PNDM6 (85%)1,720 (−2.75)- RecessiveNormalInsulin/pump- Parents have fasting hyperglycaemia as heterozygotes Clinical features that are unusual for type 1 diabetes Type 1 diabetes shares clinical features with certain types of monogenic diabetes, such as young age of diagnosis, normal body weight and insulinopaenia [5] so that a monogenic diagnosis can be overlooked. Features in children initially thought to have type 1 diabetes but that should suggest a possible diagnosis of monogenic diabetes are shown below. None of these are absolute and should be considered collectively rather than in isolation [4]. The approximate percentage of patients with type 1 diabetes is given in brackets. Age of diagnosis below 6 months after birth (<1%) [15]Family history with a parent affected (2–4%) [16]Endogenous insulin production after 3 years of diabetes (the honeymoon phase), indicated by detectable C-peptide (>200 nmol/l) in response to raised glucose (>8 mmol/l) (1–5%).Absence of islet autoantibodies, especially when measured at diagnosis (3–30%) [17, 18]. The great variation in antibody prevalence in series probably represents differences in assays and means it is hard to apply published series directly into clinical practice. Absent antibodies should lead to other investigation/consideration rather than leading directly to genetic tests. Clinical features that are unusual for type 2 diabetes Type 2 diabetes in children and young adolescents might share features which meet former classification criteria for MODY such as diagnosed <25 years, autosomal dominant inheritance and non-insulin dependency [19–22]. Patients might have a rather low body mass index and have other features that are unusual for type 2 diabetes. These should raise suspicion for the current diagnosis to be correct as chances for these features occurring in type 2 diabetic children or adolescents are low as shown here (percentage shown in brackets). No evidence of insulin resistance with fasting C-peptide within the normal range (0–20%) [19–22].Ethnic background from a low prevalence type 2 diabetes race e.g. European Caucasian (0–45%) [19–22]Acanthosis nigricans not detected (10%) [19]Not markedly obese or diabetic family members who are normal weight (20%) [22]. Clinical features of specific subtypes of monogenic diabetes and their treatment Typical clinical presentations in children when a diagnosis of monogenic diabetes should be considered can mainly be classified into four categories that are further discussed below: Neonatal diabetes and diabetes diagnosed within the first 6 months of life (Table 1)Familial diabetes with an affected parent (autosomal dominant) (Table 2 upperpart)Mild (5.5–8.5 mmol/l) fasting hyperglycaemia especially if young or familial (Table 2 lower part).Features in other tissues as part of genetic syndromes associated with diabetes (Tables 3 and 4) Neonatal diabetes and diabetes diagnosed before the age of 6 months irrespective of current age Diabetes diagnosed before the age of 6 months is most likely to have a genetic cause, and further clinical aspects should be investigated to guide genetic testing. Clinically two subgroups were recognised: transient neonatal diabetes mellitus (TNDM) that resolved at a median of 12 weeks and then did not require any treatment although as many as 50% of cases relapsed during the paediatric age range [23, 24]. In contrast permanent neonatal diabetes mellitus (PNDM) required lifelong insulin injections from diagnosis onwards. The majority of patients with TNDM have an imprinting abnormality of the transcription factor ZAC gene and HYMAI gene encoding an untranslated RNA chromosome 6q (Table 1) [14, 23]. Apart from macroglossia seen in 23% there are no non-pancreatic features [23]. The second commonest cause of mutations in patients with diabetes diagnosed before the age of 6 months of life are mutations in the KCNJ11 gene encoding the Kir6.2 subunit of the KATP channel and that can result in either TNDM (10%) or PNDM (90%) [3, 25–27]. Despite being a heterozygous mutation most have no family history because 90% of cases are spontaneous mutations. Most patients have isolated diabetes although neurological features are seen in 20% of patients. The most severe is the Developmental delay early onset generalised Epilepsy and Neonatal Diabetes (DEND) syndrome [25]. More common is the intermediate DEND syndrome where patients have less severe developmental delay and do not have early onset generalised epilepsy and that is often associated with the V59M mutation [3, 26]. Recently, activating mutations in the ABCC8 gene encoding the SUR1 subunit of the KATP channel were found to be a similar cause of neonatal diabetes [28]. While the majority of cases in PNDM are the result of mutations in the KCNJ11 gene, in a small minority of patients, many different genetic mutations have been described [3]. If both parents are glucose intolerant, homozygous or compound heterozygous mutations in glucokinase are most frequent [29, 30]. Features that help differentiate between TNDM, PNDM and the different subtypes, guide which gene to test for and are described in Table 1. Concerning treatment, imprinting abnormalities of the 6q24 locus show initially very high glucose values (range 12–57 mmol/l) and so insulin is used initially although the dose can rapidly be reduced. Once patients have relapsed patients should remain under annual follow up due to the risk of diabetes relapsing. Relapse patients are not insulin dependent and can be treated with diet initially although subsequently often need insulin [14]. The long-term response to oral treatment such as sulfonylurea or metformin is uncertain. Patients with Kir6.2 mutations have all the clinical features of insulin dependency as 30% present with ketoacidosis and they usually do not have detectable C peptide and so were treated with insulin [26]. It has recently been shown that these patients cannot only be successfully treated with oral sulfonylureas but can also get better glycaemic control without an increase in hypoglycaemia and with neurological features improving as well [70]. The doses needed are high when calculated on a per kg body weight basis compared to adults, with patients typically needing 0.5 mg glibenclamide/kg/day although some may need as much as 1 mg/kg/day [31–36, 70]. With time many patients have been able to reduce their doses of sulfonylurea while maintaining excellent glycaemic control [7, 70]. Similar results have recently been found for patients with activating SUR1 mutations [37]. All other causes need to be treated with insulin. Some paediatricians find these patients are best managed on subcutaneous insulin pumps due to the fluctuations in glucose levels. In patients with pancreatic aplasia exocrine pancreatic supplements will additionally be required. Familial diabetes with an affected parent Children and young adults with a strong family history of diabetes Genetic causes of a family history of diabetes are autosomal dominant and non-insulin dependent. A diagnosis of Maturity Onset Diabetes of the Young (MODY) should be considered whenever a parent has diabetes even if they are thought to have type 1 or type 2 diabetes (Table 2). MODY comes in two major subgroups: (1) the result of transcription factor mutations (IPF1-MODY4, HNF1β-MODY5, HNF1α-MODY3, HNF4α-MODY1, NeuroD1-MODY6 or CEL-MODY7) described in this section and (2) mutations reducing glucose sensing (glucokinase mutations-MODY2) in which a family history might be less evident because it only presents as mild hyperglycaemia as described in the following section. MODY due to transcription factor mutations often present in adolescence/young adulthood, show progressive hyperglycaemia and frequently lead to complications. Of the transcription factor mutations, MODY3—due to HNF1α mutations—is the commonest form [38]. The clinical characteristics of patients with HNF1α mutations are: Young onset diabetes that shows characteristics of not-being insulin dependent e.g., do not develop keto-acidosis in the absence of insulin, achieve good glycaemic control on a small dose of insulin. Detectable C-peptide is measured when on insulin with glucose >8 mmol/l after 3 years of diabetes (the honeymoon period).Family history of diabetes. This might be insulin treated and considered to be type 1 diabetes. This would typically be diagnosed at their 20s, 30s or 40s. There may also be an affected grandparent although often these are diagnosed after 45 years.Glucosuria at relatively normal blood glucose levels are often seen as these patients have a low renal threshold [39].Oral glucose tolerance tests in early stages tend to show a very large glucose increment usually >5 mmol/l [39]. Some patients might have a normal fasting value while still rise into the diabetic range at 2 h.Marked sensitivity to sulfonylurea resulting in hypoglycaemia despite poor glycaemic control before starting sulfonylurea [40,41].Table 2Familial diabetes diagnosed, or undiagnosed due to mild hyperglycaemiaGene/proteinClinical pictureNumber of cases describedMedian age at diagnosis in weeks (range)Family history reflected by inheritanceOther clinical featuresOther testsTreatmentGlucose at presentation in mmol/l Median (range)OGTTFamilial diabetes diagnosedHNF-1αMODY319714 (4–18)DominantHyperglycaemia is rapidly progressive with age17 (11–26)Large increment (0 h–2 h usually >5 mmol/l)Diet > low dose of sulfonylureaLow renal threshold > glucosuriaSensitive to sulfonylureaHNF-4αMODY12217 (5–18)DominantHyperglycaemia is rapidly progressive with age15 (9–20)Large increment (0 h–2 h usually >5 mmol/l)Low dose of sulfonylureaNormal renal thresholdSensitive to sulfonylureaReduced levels of apoAIII, apoCIII, and triglyceridesOther unusual causes: IPF1 (MODY4), NeuroD1 (MODY6), CEL (MODY7)Familial diabetes undiagnosed due to mild fasting hyperglycaemiaGlucokinase (GCK, heterozygous)MODY215210 (0–18)DominantHyperglycaemia is mild (fasting 5.5–8 mmol/l)11 (5.5–16)Small increment (0 h–2 h usually <3.5 mmol/l)No treatment(The mild hyperglycaemia might not have been diagnosed in relatives/parents)Hyperglycaemia is only slowly progressive with age> usually diagnosis is by incidental findingNormal renal threshold Patients with HNF1α gene mutations can initially be treated through diet although they will have marked postprandial hyperglycaemia after high carbohydrate food as the β-cell defect results in insufficient increase in insulin secretion with hyperglycaemia [42]. Most patients will need pharmacological treatment as they show progressive deterioration in glycaemic control throughout life and are at risk of considerable micro-vascular and macro-vascular complications [43]. The first treatment to be used in children who are not controlled on insulin should be low dose sulfonylureas which results in a 4-fold greater lowering of glucose than metformin [41]. These patients are extremely sensitive to sulfonylurea and as long as they do not have problems with hypoglycaemia can be maintained on these for many decades [40]. Glycaemic control in sulfonylureas is often better than that achieved on insulin especially in children and young adults [44]. The dose of sulfonylureas should initially be low (1/4 of the normal starting dose in adults) to avoid hypoglycaemia. If there is hypoglycaemia despite dose titration of a once or twice daily sulfonylurea preparation such as Gliclazide, a slow release preparation or mealtime doses with short-acting agents like nateglinidine may be considered [45]. Mutations in the hepatocyte nuclear factor 4α (HNF4α)-gene resulting in MODY1 are considerably less common, have similar characteristics but might be diagnosed later and patients have no low renal threshold [46]. They also lead to reduced levels of triglycerides and the apolipoproteins apoAIII and apoCIII [47]. Diagnosis should be considered when HNF1α tests negative while the clinical features are strongly suggestive for HNF1α [46]. Patients with HNF4 α mutations are usually sensitive to sulfonylurea [48]. The handful of families with autosomal dominant non-insulin dependent diabetes that have further been described include mutations in IPF1 (MODY4) [49], NeuroD1 (MODY6) [50,51], and recently the carboxyl ester lipase (CEL) gene (MODY7) [52], but these are so unusual they do not need to be tested for in children with diabetes except in a research setting or when there are additional phenotypes such as pancreatic exocrine dysfunction [52]. IPF has been described under neonatal diabetes, NeuroD1 encodes a transcription factor that binds to a critical Ebox motif on the insulin promoter and plays a role in both pancreatic and neuron development. Mild (5.5–8.5 mmol/l) fasting hyperglycaemia especially if young or familial Raised fasting blood glucose in the range of 5.5 to 8.5 mmol/l is unusual in children and young adults. This always raises concern that they may be about to develop type 1 diabetes or have type 2 diabetes. However a considerable proportion of these patients with persistent mild fasting hyperglycaemia will have a heterozygous mutation in the glucokinase gene. The phenotype associated with glucokinase mutations is remarkably similar for all mutations with the following features suggesting a diagnosis. The fasting hyperglycaemia is persistent and stable over a period of months or years [39]HbA1C is typically just below or just above the upper limit of normal (5.5 to 5.7%)In an oral glucose tolerance test the increment (2 h glucose-fasting glucose) is small (typically <3.5 mmol/l) although because of the variability of the oral glucose tolerance test this should not be considered an absolute criteria [39].Parents may have ‘type 2 diabetes’ or may not be diabetic. On testing one parent will have a mildly raised fasting blood glucose, in the range of 5.5–8.5 mmol/l, as this is an autosomal dominant condition [39]. Testing of fasting glucose in apparently unaffected parents is important when considering a diagnosis of a glucokinase mutation. The fasting hyperglycaemia does not deteriorate significantly and the glucose is regulated at the higher set point [39]. As this is rarely associated with any microvascular or macrovascular complications even when no treatment is given throughout life and as there is very little if any response to either oral hypoglycaemic agents or insulin because the set point remains the same, these patients do not need treating in the paediatric age range [53]. Genetic syndromes associated with diabetes When diabetes in a child is associated with other multi-system disease the possibility of a monogenic syndrome that explains all features should be considered. The online Mendelian inheritance in Man (OMIM) website (access through the NCBI website http://www.ncbi.nlm.nih.gov/entrez/query.fcgi) can help with clinical features and to know if the gene has been defined and hence molecular genetic testing is available. For described and previously undescribed syndromes see the ISPAD rare diabetes collection (contact through link on the ISPAD website or through http://www.diabetesgenes.org). The most common genetic syndromes that include diabetes are described below and in Tables 3 (insulin synthesis/secretion) and 4 (insulin resistance). Table 3Syndromic features in addition to the diabetes: insulin synthesis/secretionGene/proteinClinical pictureNumber of cases describedMedian age at diagnosis in weeks (range)Family history reflected by inheritanceOther clinical featuresTreatmentHNF1βRarely isolated PNDM or MODY 5- Renal developmental disorders, especially renal cysts and dysplasiaInsulin (+possibly treat exocrine deficiency?)HNF1βRenal cysts and diabetes syndrome (RCAD)- Uterine and genitalia developmental anomalies- Hyperuricaemia, gout- Abnormal liver function testsWSF1Diabetes insipidus, diabetes mellitus, optic atrophy, deafness (DIDMOAD) syndrome/Wolfram syndrome (90% have mutations)Especially where consanguineous marriages are frequent6 years(Most <16 years)Dominant- Diabetes insipidus- Optic atrophy - Bilateral sensorineural deafness- Dilated renal tracts- Truncal ataxia- Protean neurological signs 75% has the complete phenotype, increasing with increasing ageInsulinSLC19A2(Thiaminetransporter protein)Thiamine responsiveMegaloblastic anaemia (TRMA) syndrome Roger’s syndromeRareRecessive- Thiamine responsive megaloblastic anaemia- Sensorineural deafnessThiamine > insulin   tRNA(leu(UUR)) gene (3243 A to G; tRNA)- Maternally inherited diabetes (MID)- Mitochondrial myopathy, encephalopathy, lactic acidosis, stroke-like syndrome (MELAS)- Sensorineural deafness- Short stature- Subclinical exocrine deficiency- HeteroplasmyInsulinTable 4Syndromic features in addition to the diabetes: insulin resistanceProteinClinical pictureMedian age at diagnosis in weeks (range)Family history reflected by inheritanceOther clinical featuresOther features/testsTreatmentAcanthosis nigricansInsulin levelsAndrogen excess and hypertrichosisInsulin receptorType AAdolescenceRecessive (usually)Insulin resistance in absence of obesityYes—marked↑↑↑↑↑↑/PCO(Metformin/glitazones) > insulin/pumpInsulin receptorRabson-MendenhallCongenitalRecessive (usually)- Abnormal dentitionYes—marked↑↑↑↑↑/PCO(Metformin/glitazones) > insulin/pump- Extreme growth retardationInsulin receptorLeprechaunism (Donahue syndrome)CongenitalRecessive (usually)- Abnormal faciesYes—marked↑↑↑↑↑↑/PCO(Metformin/glitazones) > insulin/pump- SGA and growth retardation- Large genitalia- Rarely survive infancySeipin&AGPAT2Total lipodystrophyAdolescence or congenitalRecessive- Total loss of subcutaneous fatYes—may be marked↑↑↑↑↑/PCO+/−Recombinant /insulinLamin AC&PPARγPartial lipodystrophyDominant- Partial loss of subcutaneous fatMetformin > insulin MODY-5 due to an HNF-1β mutation (renal cysts and diabetes syndrome The transcription factor hepatocyte nuclear factor 1 β is expressed in primitive pancreatic duct cells involved in early endocrine cell differentiation incidentally leading to PNDM, but mainly leading to MODY5 with pancreatic atrophy as a consistent finding when assessed by abdominal computed tomography [54, 55]. Patients with HNF-1β mutations rarely present with isolated diabetes [56]. Renal developmental disorders especially renal cysts and renal dysplasia are present in almost all patients with mutations or gene deletions [57]. These may be diagnosed in utero and precede the diagnosis of diabetes. A diagnosis of HNF-1β should be considered in any child with diabetes who also has non-diabetic renal disease. Other features are presented in Table 3. Patients with HNF1-β mutations, unlike patients with HNF-1α mutations, are not sensitive to sulfonylurea and hence usually require insulin treatment [58]. Pancreatic size is reduced reflecting a reduction in both the endocrine and exocrine pancreas and sub-clinical exocrine deficiency is present in most patients [57] but it is uncertain if this should be treated if it is asymptomatic. Wolfram or DIDMOAD syndrome (diabetes insipidus, diabetes mellitus, optic atrophy and deafness) This autosomal recessive syndrome shows mutations in the gene for Wolfram syndrome (WFS1) in at least 90% of patients [59–61], and the association of diabetes with progressive optic atrophy below 16 years of age is diagnostic [62]. Other features are described in Table 3 with the complete phenotype seen in 75% of patients with increasing prevalence with age. The order of appearance of the neurological features may vary, even within families. Patients with Wolfram syndrome die at a median age of 30 years [62]. The diabetes is non-autoimmune and insulin deficient and presents at a mean age of 6 years [62]. Patients require insulin treatment from the time of diagnosis but autoantibodies are not present [62]. Roger’s or TRMA syndrome (thiamine responsive megaloblastic anaemia) The diabetes in this rare recessive genetic syndrome, is insulin deficient in nature and both the diabetes and the anaemia are responsive to thiamine. However, all seem to develop an insulin requirement in the long term [63]. Further details are described in Table 3. Mitochondrial diabetes Maternal transmission of mutated or deleted mitochondrial DNA (mtDNA) can result in maternally inherited diabetes [64] although they do not usually occur in the paediatric age range. Patients suffer from sensorineural deafness and short stature. In family members there might be some overlap between MIDD and MELAS syndrome [65, 66]. The heteroplasmy in proportions of mitochondrial DNA affected lead to a variety of phenotypes. The diabetes is characterised by progressive non-autoimmune β-cell failure, usually manifests outside the paediatric age range, and may progress to needing insulin treatment rapidly. Other features are described in Table 3. Insulin resistance syndromes The key features of all insulin resistance syndromes are acanthosis nigricans, massively raised insulin concentrations, and androgen excess while obesity is absent (Table 4) [2]. The more severe the insulin resistance and the earlier the onset, the more likely is diabetes. The mechanism of the main insulin resistance syndromes, Type A, Rabson-Mendenhall, Leprechaunism and Lipodystrophy, is shown in Fig. 2 while the key clinical features of the main insulin syndromes, are presented in Table 4 [2]. Treatment of severe insulin resistance is very difficult and most patients with diabetes have poor glycaemic control and frequently develop long-term complications [2]. Treatment approaches include the use of the insulin sensitisers metformin and glitazones, but effectiveness is limited when the insulin resistance is very severe. Insulin is the main stay of treatment and U500 insulin and insulin pumps are usually required [2]. In partial lipodystrophy metformin may have benefit and insulin is not required in the early stages [67]. In total lipodystrophy the response of diabetes to recombinant lipodystrophy [68] can be dramatic but is only available on a research basis. Testing for a molecular monogenic diagnosis How to test for monogenic diabetes? While in type 1 and type 2 diabetes there is no single diagnostic test, this is not the case in monogenic diabetes where in >80% of cases a molecular genetic diagnosis can be made by DNA testing. These are offered by many laboratories but should preferably be performed in a laboratory that is experienced and specialised in such diagnosis irrespective whether this requires sending samples abroad (http://www.diabetesgenes.org). Some recently described monogenic genes, such as Kir6.2 in patients in whom diabetes was diagnosed before the age of 6 months, can be screened for with no charge (http://www.diabetesgenes.org). Other tests are more expensive (up to 500 Euro or 635 USD) but make future patient care cheaper and hence testing would be cost-effective. For example, in a Swiss centre yearly costs for type 1 diabetes was approximately 2050 Euro (2603 USD) per type 1 patient compared to 410 Euro (521 USD) per MODY2 patient [69]. Approval from the patient’s insurance company should be sought prior to sending DNA when applicable. Future analyses will show cost-effectiveness for the different subtypes and prove earlier diagnosis to be more beneficial. What if a monogenic diagnosis cannot be made? Occasionally molecular genetic test results are negative, despite unusual clinical features or typical features for a certain monogenic subtype. The certainty of such a negative result increases if a specialised centre performed testing. Even then, some cases remain unsolved. These can be referred to the ISPAD rare cases registry (see website or contact a.t.hattersley@exeter.ac.uk) to allow pattern finding and closer investigation by experts if a novel idea evolves, hence increasing chances for new and future insights, novel diagnoses and improved patient care. Summary Molecular genetic testing can define a diagnosis in 1–2% of all diabetic patients with monogenic diabetes. Advances in this field have led to the identification of the genes associated with many clinically identified subgroups of diabetes and explained clinical heterogeneity in conditions defined by age of diagnosis e.g. neonatal diabetes and MODY. Molecular genetic tests are now available to help define the diagnosis, and importantly alter prognosis and optimise treatment of children, young adults and their families with diabetes. As these tests are expensive genetic testing should be limited to those who on clinical grounds are likely to be positive. Considering testing for monogenic diabetes is hence a challenge and should be guided by unusual features of the current diagnosis, specific features concordant with monogenic subtypes and by the possibility of a change in treatment. This article discussed the pathophysiology and clinical manifestations used to select eligible patients and guide genetic testing, and demonstrates its importance in the treatment of monogenic diabetes.

Arch_Sex_Behav-3-1-2042031

Characteristics of Sexual Abuse in Childhood and Adolescence Influence Sexual Risk Behavior in Adulthood

Childhood and adolescent sexual abuse has been associated with subsequent (adult) sexual risk behavior, but the effects of force and type of sexual abuse on sexual behavior outcomes have been less well-studied. The present study investigated the associations between sexual abuse characteristics and later sexual risk behavior, and explored whether gender of the child/adolescent moderated these relations. Patients attending an STD clinic completed a computerized survey that assessed history of sexual abuse as well as lifetime and current sexual behavior. Participants were considered sexually abused if they reported a sexual experience (1) before age 13 with someone 5 or more years older, (2) between the ages of 13 and 16 with someone 10 or more years older, or (3) before the age of 17 involving force or coercion. Participants who were sexually abused were further categorized based on two abuse characteristics, namely, use of penetration and force. Analyses included 1177 participants (n=534 women; n=643 men). Those who reported sexual abuse involving penetration and/or force reported more adult sexual risk behavior, including the number of lifetime partners and number of previous STD diagnoses, than those who were not sexually abused and those who were abused without force or penetration. There were no significant differences in sexual risk behavior between nonabused participants and those who reported sexual abuse without force and without penetration. Gender of the child/adolescent moderated the association between sexual abuse characteristics and adult sexual risk behavior; for men, sexual abuse with force and penetration was associated with the greatest number of episodes of sex trading, whereas for women, those who were abused with penetration, regardless of whether the abuse involved force, reported the most episodes of sex trading. These findings indicate that more severe sexual abuse is associated with riskier adult sexual behavior. Introduction Childhood and adolescent sexual abuse has been associated with a wide variety of adverse mental and physical health outcomes. Research also suggests that the greater the severity of the sexual abuse, the worse the health outcomes. Thus, more severe sexual abuse (e.g., sexual abuse involving force, more intimate sexual acts, a close relative, or repeated sexual abuse) has been associated with poorer social adjustment, less life satisfaction, and more severe psychological symptoms (Callahan, Price, & Hilsenroth, 2003; Carlson, McNutt, & Choi, 2003; Fassler, Amodeo, Griffin, Clay, & Ellis, 2005; Feinauer, Mitchell, Harper, & Dane, 1996). In a meta-analysis on the effects of child sexual abuse, Rind, Tromovitch, and Bauserman (1998) found that force was associated with more negative reactions but not with later psychological symptoms, whereas penetration was unrelated to these outcomes. Sexual abuse severity also has been associated with subsequent sexual risk behavior, including more sexual partners (Merrill, Guimond, Thomsen, & Milner, 2003) and greater likelihood of sex with someone just met, earlier age at first intercourse, and a higher frequency of STD diagnoses (Walser & Kern, 1996). Although these studies suggest that more severe sexual abuse is associated with more sexual risk behavior, they provide only limited information regarding whether specific aspects of the abuse predict such outcomes. Needed is more fine-grained research to determine whether characteristics of the abuse experience (e.g., whether physical force was used during the abuse and the type of sexual act that occurred during the abuse) are associated with sexual health outcomes. Two studies have examined the association between sexual abuse with force and later sexual risk behavior. Cinq-Mars, Wright, Cyr, and McDuff (2003) found that adolescent girls who experienced childhood or adolescent sexual abuse with force were more likely than girls who were sexually abused without force to have engaged in subsequent consensual sex during adolescence, to have more than one partner per year, and to have been pregnant. In a sample of men who have sex with men (MSM), Jinich et al. (1998) reported that sexual abuse perceived as moderately or strongly coerced was associated with higher frequency of unprotected anal sex and higher HIV seroprevalence rates, relative to MSM who perceived the sexual abuse as voluntary or mildly coerced. Thus, in these two studies, sexual abuse involving force has been associated with greater sexual risk behavior. Investigations of the effects of penetrative (vs. non-penetrative) sexual abuse have yielded mixed results. Cinq-Mars et al. (2003) found that adolescent girls who experienced sexual abuse involving penetration were more likely to have engaged in consensual sex and experienced an unplanned pregnancy, compared to girls who experienced non-penetrative sexual abuse. Fergusson, Horwood, and Lynskey (1997) found that, compared to participants reporting no sexual abuse, those who experienced non-penetrative sexual abuse reported higher rates of unprotected sex. However, those who experienced penetrative sexual abuse reported the worst outcomes; compared to participants who were not sexually abused, those who reported penetrative sexual abuse were more likely to have been pregnant, to report more than five sexual partners by age 18, to have had unprotected sex, and to have had an STD. In contrast, in a meta-analysis, Arriola, Louden, Doldren, and Fortenberry (2005) found that the effect size for the relation between sexual abuse and later sexual behavior (i.e., unprotected sex, sex with multiple partners, and sex work) did not differ for studies including non-contact abuse, studies including contact abuse only, and studies including penetration abuse only; their findings suggest that type of sexual act during sexual abuse was not associated with later sexual behavior. However, some of these categories included very few studies (e.g., there were only three studies that included penetration abuse only). Furthermore, effects may have been obscured if studies with less restrictive definitions of sexual abuse included a large number of participants who had exper- ienced more severe (i.e., contact or penetrative) sexual abuse. In sum, evidence from a small number of studies suggests that force and penetration may be associated with adult sexual risk behavior. An important limitation of previous research is that few studies have investigated the differential effects of sexual abuse for men and women. Because women in heterosexual relationships often have less control or power over sexual encounters compared to men (see the Theory of Gender and Power, Connell, 1987, for an explanation of the power imbalance between men and women), it is important to study gender in relation to sexual health behaviors. Indeed, the limited research on this topic suggests that the association between sexual abuse and adult sexual behavior differs by gender (e.g., Futterman, Hein, Reuben, Dell, & Shaffer, 1993; Mason, Zimmerman, & Evans, 1998; Zierler et al., 1991). It is possible that gender interacts with abuse characteristics to lead to different outcomes for males and females, an idea which is supported by research on the psychological sequelae of sexual abuse. For example, in a meta-analysis, Rind et al. (1998) found that whether or not the sexual abuse experience was consensual was associated with later psychological adjustment for men, but not for women. Few studies have investigated the effects of the interaction of gender and abuse characteristic on later sexual behavior. Overall, previous studies investigating the association between abuse characteristics and later sexual behavior have tended to use small samples, or included only males or only females, thus precluding gender comparisons. The primary purpose of this study was to determine whether use of force and type of sexual act was associated with sexual risk behavior in a group of patients receiving outpatient care from a sexually transmitted disease (STD) clinic. Based on previous research, we hypothesized that: (1) the use of force; and (2) sexual abuse involving penetration would be associated with greater sexual risk behavior. The secondary purpose of this study was to determine whether the effects of abuse characteristics on adult sexual behavior differed by gender. Method Participants Participants were men and women attending a public STD clinic in upstate New York. All had been screened for possible inclusion in a randomized controlled trial (RCT) evaluating several different sexual risk reduction programs. Screening criteria for the RCT included: age 18 or older; not HIV positive; and sexual behavior (e.g., unprotected sex, multiple partners) that put them at risk for contracting an STD in the past 3 months. This study used baseline data from the RCT, prior to the receipt of the interventions. Baseline data were available from 1265 eligible participants. Data from participants who refused to answer sexual abuse (n=12) or demographic (n=1) questions, were inconsistent in reporting their sexual behavior (n=5), were outliers on sexual behavior data (n=30; defined as having a studentized deleted residual >4), or were recruited in error (n=1) were eliminated. Outliers on sexual behavior data were excluded because these individuals likely were members of an extremely high-risk population that merits separate investigation (Wegener & Fabrigar, 2000). Overall, the sample was 46% female (n=557), 65% African American (n=785), and 24% Caucasian (n=294). The majority of participants were unemployed (n=620; 51%), had a high school education or less (n=762; 63%), and had a household income of less than $15,000 per year (n=686; 57%). Most participants were single (never married; n=958; 79%); 75 participants (6%) were married, and 183 (15%) were divorced, separated, or widowed. Participants were, on average, 29.2 years of age (SD=9.7). Among women, 504 (90%) reported having sex with only men in the past 3 months; 53 (10%) reported having sex with both men and women. Among men, 612 (93%) reported having sex with only women in the past 3 months; 31 (5%) reported having sex with only men; and 15 (2%) reported having sex with both men and women in the past 3 months. Procedure Patients who registered for a clinic visit were invited to a private exam room by a trained research assistant (RA), and were asked to answer a series of brief screening questions. The RA explained the study to patients who met eligibility criteria and obtained informed consent. Participants then completed a 45-minute, Audio Computer-Assisted Self-Interview (ACASI) that included measures of demographic characteristics, health behaviors and beliefs, and psychosocial functioning, as well as questions about childhood sexual experiences and current sexual behavior. ACASI was used because it optimizes participant privacy (improving data quality) while allowing low-literacy persons to participate (Schroder, Carey, & Vanable, 2003). For the present study, we used data from measures of childhood/adolescent sexual abuse and sexual behavior. After their clinic exam and counseling, participants were paid $20 to compensate them for their time. All procedures were approved by the IRBs of the participating institutions. Measures Childhood/adolescent sexual abuse Three items, adapted from Finkelhor’s (1979) longer survey of childhood sexual experiences, were used to assess sexual abuse (see Appendix A).1 Participants who reported any contact sexual experiences (including kissing, fondling, giving oral sex, receiving oral sex, vaginal sex, or anal sex) (1) before age 13 with someone 5 or more years older or (2) between ages 13 and 16 with someone 10 or more years older, and those who reported (3) any contact sexual experience before age 17 involving force or coercion, were classified as sexually abused; all other participants were classified as not sexually abused. Those who were sexually abused were further categorized according to whether the abuse involved force and/or penetration. Participants who reported a sexual experience before age 17 involving force or coercion were considered to have experienced sexual abuse with force. Sexually abused participants who reported any oral, vaginal, or anal sex were considered to have experienced sexual abuse with penetration. A single, four level categorical variable was created to examine the impact of different abuse characteristics on sexual risk behavior: (1) no sexual abuse; (2) sexual abuse without force or penetration; (3) sexual abuse with penetration but without force; and (4) sexual abuse with both force and penetration. Too few participants reported sexual abuse with force and without penetration (n=39, 5%) to include a sexual abuse with force only category. Current sexual behavior The sexual risk behavior items were developed and tested in previous studies (Carey et al., 1997, 2000, 2004). Participants were asked to report: the number of male and female sexual partners they had in their lifetime and in the past 3 months; the number of times they exchanged sex for money or drugs (lifetime); and the number of times they had been treated for an STD (lifetime). The frequency of unprotected sex was also investigated. Participants were asked to report the number of times in the past 3 months that they had vaginal and anal sex with and without a condom with their: (1) steady partner; (2) other male partners; and (3) other female partners. Responses to these items were used to calculate the absolute number and the proportion (number of unprotected sex episodes/number of protected and unprotected sex episodes) of unprotected sex episodes in the past 3 months. Table 1Sexual abuse characteristics by genderMen (n=643)Women (n=534)Total (n=1177)n% of menn% of womenn% of totalNo sexual abuse227351823440935Sexual abuse without force and without penetration10016591115914Sexual abuse with penetration only (no force)208321052031327Sexual abuse with both force and penetration108171883529625Table 2Demographic characteristics of sexual abuse groups formed by force and penetrationNo Sexual Abusea (n=409)Sexual Abuse (no force or penetration)b (n=159)Sexual Abuse (penetration)c (n=313)Sexual Abuse (force and penetration)d (n=296)n%n%n%n%Sex (male)227c,d56100d63208a,d66108a,b,c36Race (minority)269b,c,d66120a,c75279a,b,d89228a,c77Education (high school or less)212b,c,d52100a,c63244a,b,d78191a,c65MSDMSDMSDMSDAge (in years)28.4d9.628.79.529.29.830.5a9.7ap < .05, compared to No Sexual Abuse.bp < .05, compared to Sexual Abuse, No Force, No Penetration.cp < .05, compared to Sexual Abuse, No Force, Penetration.dp < .05, compared to Sexual Abuse, Force, Penetration. Statistical analyses Analyses of variance (ANOVAs) were used to determine whether the four categories of sexual abuse (no sexual abuse; sexual abuse without force or penetration; sexual abuse with penetration; and sexual abuse with both force and penetration) were associated with later risky sexual behavior. If there was a significant overall effect of sexual abuse, Tukey tests were conducted to determine specifically which groups differed. Demographic variables that differed between groups were controlled for in these analyses. Thus, the ANOVAs included: (1) demographic covariates and (2) a main effect of sexual abuse. Continuous outcome variables that were not normally distributed (i.e., the number of lifetime partners, the number of partners in the past 3 months, the number of episodes of unprotected sex in the past 3 months, the number of times participants exchanged sex for money or drugs, and the number of previous STD diagnoses) were transformed using a log10 of (x+1) transformation (Tabachnick & Fidell, 2001). Unless otherwise stated, analyses associated with these variables used the log transformation. Exploratory analyses were conducted to investigate whether gender moderated the relations between the sexual abuse characteristics and later sexual behavior. ANOVAs were conducted including demographic covariates, a main effect of abuse, and the interaction of abuse and gender. Results Of the 1216 patients who completed the survey, 66% reported childhood/adolescent sexual abuse (n=807). Of these 807 participants who met criteria for sexual abuse, 159 (20%) reported sexual abuse without force and without penetration, 313 (39%) reported sexual abuse with penetration, 39 (5%) reported sexual abuse with force and without penetration, and 296 (37%) reported sexual abuse with both force and penetration. Because few participants reported sexual abuse with force but without penetration, those participants were excluded from the analyses, leaving a final sample size of N=1177. Sexual abuse characteristics by gender are reported in Table 1. Table 3Sexual risk behaviors of participants who reported sexual abuse with force, sexual abuse without force, and no sexual abuse (raw data)No Sexual Abusea (n=409)Sexual Abuse (no force or penetration)b (n=159)Sexual Abuse (penetration)c (n=313)Sexual Abuse (force and penetration)d (n=296)MSDMSDMSDMSDSexual partners (number, lifetime)31.7c,d80.927.6c,d28.260.1a,b211.564.2a,b172.4Sexual partners (number, past 3 months)2.5c,d2.12.72.23.2a2.73.5a4.0Unprotected sex (number of events, past 3 months)15.3c,d24.317.026.820.7a30.122.8a38.2Unprotected sex (proportion, past 3 months)0.680.320.640.330.700.300.670.33Exchanged sex for money or drugs (number, lifetime)4.9c,d55.95.6d42.06.5a,d42.417.6a,b,c89.6STD diagnoses (number, lifetime)2.4c,d3.02.6c,d3.23.4a,b3.64.0a,b4.1ap < .05, compared to No Sexual Abuse.bp < .05, compared to Sexual Abuse, No Force, No Penetration.cp < .05, compared to Sexual Abuse, No Force, Penetration.dp < .05, compared to Sexual Abuse, Force, Penetration. Demographic differences Preliminary analyses examined whether any demographic variables were associated with the sexual abuse characteristics (see Table 2). Sexual abuse was significantly associated with sex, race, education, and current age. Thus, for example, participants reporting a history of sexual abuse were more likely to be less well-educated and more likely to report a minority racial/ethnic identity than nonabused participants. Importantly, sexual abuse involving force and penetration was more likely to be reported by women than by men. All pairwise comparisons for the demographic characteristics are presented in Table 2. Because of these associations, sex, race, education, and current age were used as covariates in subsequent analyses. Relation between sexual abuse characteristics and sexual behavior After controlling for relevant demographic covariates, sexual abuse was significantly associated with the number of lifetime partners, F(3, 1160)=21.08, p < .0001, the number of episodes of unprotected sex in the past 3 months, F(3, 1169)=3.97, p < .01, the number of partners in the past 3 months, F(3, 1169)=7.28, p < .0001, the number of times sex was traded, F(3, 1153)=14.23, p < .0001, and the number of previous STD diagnoses, F(3, 1169)=8.01, p < .0001 (see Table 3). Sexual abuse was not significantly associated with the proportion of unprotected sex episodes in the past 3 months. Follow-up Tukey tests showed that, compared to those who were sexually abused with penetration, those who were not sexually abused had significantly fewer: (1) lifetime sexual partners (Cohen’s d=.40); (2) partners in the past 3 months (d=.23); (3) episodes of unprotected sex in the past 3 months (d=.19); (4) episodes of sex trading (d=.31); and (5) previous STD diagnoses (d=.22; all ps < .05). Similarly, compared to those who were sexually abused with both force and penetration, those who were not sexually abused had significantly fewer: (1) lifetime sexual partners (d=.49); (2) partners in the past 3 months (d=.29); (3) episodes of unprotected sex in the past 3 months (d=.21); (4) episodes of sex trading (d=.46); and (5) previous STD diagnoses (d=.34; all ps < .05). In addition, compared to those who were sexually abused with penetration, those who experienced sexual abuse without force and without penetration had significantly fewer lifetime sexual partners (d=.32) and fewer previous STD diagnoses (d=.17; both ps < .05). Similarly, compared to those who were sexually abused with both force and penetration, those who experienced sexual abuse without force and without penetration had significantly fewer: (1) lifetime sexual partners (d=.34); (2) episodes of sex trading (d=.29); and (3) previous STD diagnoses (d=.30; all ps < .05). Finally, those who were sexually abused with both force and penetration reported significantly more episodes of sex trading than those who were sexually abused with penetration (d=.29; p < .05). Because sex trading likely leads to a greater number of sexual partners, episodes of unprotected sex, and STD diagnoses, follow-up analyses were conducted to determine whether penetration was still associated with the sexual behavior outcomes after controlling for sex trading. All effects remained significant after controlling for sex trading (all ps < .05). Gender as a moderator of the relation between sexual abuse characteristics and sexual behavior To determine whether gender moderated the relation between sexual abuse characteristics and sexual behavior, gender-by-sexual abuse interactions were included in the ANOVAs. Relevant demographic covariates were included. The gender-by-sexual abuse interaction was significantly associated with the number of episodes of sex trading, F(3, 1150)=3.56, p < .05. Analyses of simple main effects revealed that, for both women and men, those who were sexually abused with force and penetration reported significantly more episodes of sex trading than those who were not abused, or than those who were abused without force and without penetration. However, for women only, those who were sexually abused with penetration reported significantly more episodes of sex trading than those who were not abused (all ps < .05; see Fig. 1).Fig. 1The effect of the interaction of gender and sexual abuse status on the number of episodes of sex trading Discussion This study investigated whether characteristics of childhood and adolescent sexual abuse (i.e., force and type of sexual activity) were related to adult sexual risk behavior, and whether these associations differed by gender. This research benefited from several methodologic strengths. For example, we sampled a large group of both men and women who reported sexual abuse; this large and diverse sample allowed exploration of two sexual abuse characteristics and gender differences. We also used psychometrically sound measures and a computer-administered survey, known to result in higher, and presumably more candid, rates of socially stigmatized and sensitive behaviors (Schroder et al., 2003). These strengths increase confidence in the validity and generalizability of the results. A key set of findings was that (1) sexual abuse with penetration as well as (2) sexual abuse with force and penetration were both related to higher rates of adult sexual behavior compared to (3) sexual abuse without force and without penetration and (4) no sexual abuse. This pattern of findings corroborates results from research investigating the mental health sequelae of sexual abuse, which indicate that force (e.g., Bulik, Prescott, & Kendler, 2001; Rind et al., 1998; Rodriguez, Ryan, Kemp, & Foy, 1997) and penetration (e.g., Briere & Elliott, 2003; Bulik et al., 2001) are associated with worse psychological outcomes; the current research also adds to the limited body of research suggesting a relation between force and penetration, and later sexual behavior (e.g., Cinq-Mars et al., 2003; Fergusson et al., 1997). The effect sizes for the association between sexual abuse and later sexual behavior were small to medium, indicating that other variables besides sexual abuse account for a large portion of the variance in adult sexual behavior. The latter finding is consistent with the idea that adult sexual behavior is influenced by multiple environmental as well as individual factors (Smith & Subramanian, 2006). Penetration by itself (i.e., without force) and penetration in combination with force were associated with increased sexual risk behavior relative to those who were abused without force and without penetration, and those who were not abused. The sole difference between the penetration only and the penetration plus force groups involved sex trading, where those who experienced sexual abuse with force and penetration reported engaging in a greater frequency of sex trading, relative to those who experienced sexual abuse with penetration and no force. However, this finding was qualified by a significant gender-by-abuse interaction. Because only a very small number of participants reported sexual abuse with force but without penetration (i.e., forced kissing or fondling), we were unable to investigate the impact of force only. A somewhat unexpected finding was that the group that reported sexual abuse without force and without penetration did not differ significantly from the nonabused group on any of the sexual behavior outcomes. Future investigation of the relation between sexual abuse and adult sexual behavior might find it fruitful to conduct more fine-grained assessments of the sexual experiences that involve only large age differentials to determine how these experiences are perceived by both men and women, and whether such experiences influence subsequent sexual behavior. It may seem counter-intuitive that individuals who experienced more severe sexual abuse (i.e., sexual abuse with force or penetration) would engage in more sexual experiences than those who experienced less severe sexual abuse; that is, one might expect individuals who experienced severe sexual abuse to avoid sex because of the negative consequences. However, relative to individuals who experienced less severe sexual abuse, individuals who experienced more severe sexual abuse may use different strategies to cope with their sexual abuse experience(s). Thus, for both men and women, those who experienced more severe forms of sexual abuse may use alcohol or drugs to cope with the sexual abuse, which, in turn, may lead to the exchange of sex for money or drugs, and/or to a greater number of sexual partners and episodes of unprotected sex. In addition, alcohol and other drug use may lead to a greater number of sexual partners and episodes of unprotected sex due to decreased ability to attend to distal concerns, such as acquiring an STD when intoxicated or high (cf. alcohol myopia; Steele & Josephs, 1990). Indeed, we have reported previously that substance use is an important mediator of the relation between sexual abuse and risky sexual behavior (Senn, Carey, Vanable, Coury-Doniger, & Urban, 2006); future research should explore whether substance use and other potential mediators operate differently for those who experienced different severity levels of sexual abuse. An alternative explanation for the association between more severe sexual abuse and greater adult sexual risk behavior is Finkelhor and Browne’s (1985) traumagenic dynamics model. This model proposes that one consequence of sexual abuse is traumatic sexualization, in which a child develops maladaptive scripts for sexual behavior, when rewarded for sexual behavior by affection. More severe sexual abuse, such as sexual abuse involving force or penetration, may lead to greater traumatic sexualization. As adults, those who experienced traumatic sexualization may believe sex is necessary to obtain affection from others. Thus, traumatic sexualization may lead to, for example, earlier consensual sex or a greater number of sexual partners (e.g., Cinq-Mars et al., 2003; Fergusson et al., 1997). Another consequence of sexual abuse, according to Finkelhor and Browne (1985), is powerlessness, in which a child learns that his or her needs or requests are ignored by others; the child thus fails to develop self-efficacy to stop unwanted sexual advances. More severe sexual abuse, particularly sexual abuse involving force or penetration, may lead to greater feelings of powerlessness. Perhaps because they lack the interpersonal skills or the self-efficacy to stop unwanted sexual advances, these individuals may be less likely to refuse intercourse with aggressive partners, resulting in more sexual partners. Powerlessness could help explain findings linking more severe sexual abuse to more adult sexual risk behavior (e.g., Cinq-Mars et al., 2003; Fergussion et al., 1997). In this regard, Kallstrom-Fuqua, Weston, and Marshall (2004) found that sexual abuse severity had an indirect effect on maladaptive relationships, mediated through powerlessness; thus, having many sexual partners could be a consequence of difficulty forming close relationships. Further research is needed to examine whether the sexual abuse characteristics investigated in this study are associated with Finkelhor and Browne’s (1985) traumagenic dynamics. Another finding yielded by this study is that abuse characteristics were associated with different outcomes for men and women. For men, only abuse with both force and penetration was associated with a greater frequency of sex trading, whereas for women, abuse with penetration, regardless of whether or not force was involved, was associated with more sex trading. In the current cultural context, young males may view sex with an older woman as masculine and mature, rather than abusive. Males, therefore, may tend to view only experiences involving force or coercion as abusive. Women, on the other hand, may be more likely to view intercourse with an older individual as abusive, regardless of whether or not force was involved. This idea is supported by meta-analytic findings that boys’ reactions to sexual abuse were less negative than were girls’ reactions (Rind et al., 1998). Different perceptions of whether or not the experience was abusive may lead to the use of different coping strategies. These results should be interpreted mindful of the limitations of the study. One limitation involved the brevity of the sexual abuse assessment. Use of a brief survey allowed us to obtain a large and diverse sample, but limited the richness of the data collected. The survey did not assess other aspects of sexual abuse, such as duration, frequency, and relationship to the perpetrator, which may be important correlates of later outcomes (e.g., Banyard & Williams, 1996; Briere & Elliott, 2003). In addition, these brief questions did not allow for assessment of reactions to the sexual experience; many participants, especially those who did not report force or coercion, may not have considered themselves sexually abused, but may have viewed these sexual experiences as inconsequential or even consensual. Future research, involving mixed qualitative and quantitative methods, might help to elucidate the empirical relations observed in the current sample. A second limitation involves the correlational nature of the data. Clearly, such data limit causal inferences, although given the temporal sequence of childhood/adolescent sexual abuse and adult sexual behavior, the limits may be less concerning in this context. Nonetheless, we acknowledge that unexplored variables that are related to both sexual abuse and greater sexual risk behavior (e.g., more adverse childhood experiences; Dong, Anda, Dube, Giles, & Felitti, 2003) should be included in future investigations of the sexual abuse–risky sex relation. It is important to recognize that participants in this study were recruited from a sexually transmitted disease clinic, and were included because they were currently engaging in sexual behavior that conferred risk for contracting an STD. The rates of sexual abuse reported in this sample were considerably higher than rates (i.e., 15% for men and 30% for women) reported in national samples (Briere & Elliott, 2003; Finkelhor, Hotaling, Lewis, & Smith, 1990; Vogeltanz et al., 1999). In addition to engaging in sexual risk behavior, patients attending STD clinics may differ from the general population in other important ways as well; for example, patients attending STD clinics often report extremely high rates of alcohol and drug use (Cook et al., 2006). Due to the nature of the sample, these results of the present study may not generalize to other populations. These results have implications for both practice and research. Regarding public health and clinical practice, they suggest that a thorough sexual health assessment should include inquiry about the nature of the sexual abuse, particularly whether force was involved and what type of sexual act occurred. Given the likely impact of sexual abuse on sexual risk behavior (as well as other health outcomes), we recommend a more comprehensive approach to sexual health assessment, education, counseling, and/or therapy. Indeed, these findings highlight the need to develop interventions tailored to the unique needs of persons with a history of sexual abuse to promote (and restore) sexual health and reduce sexual risk. With respect to research, these findings raise many questions about the conditions under which sexual abuse impairs healthy sexual development and expression, and about the mechanisms by which sexual abuse influences sexual development, behavior, and adjustment. This work will require sophisticated methods and analyses to overcome the limitations of what is inherently retrospective and correlational research.

Eur_J_Nutr-2-2-1705489

Digestibility of resistant starch containing preparations using two in vitro models

Background Resistant starch (RS) is known for potential health benefits in the human colon. To investigate these positive effects it is important to be able to predict the amount, and the structure of starch reaching the large intestine. Introduction Resistant Starch (RS) is defined as starch that is not absorbed in the small intestine of humans. Three different types of RS were defined by Englyst et al. [1]. Type 1 is defined as physically inaccessible starch, type 2 (RS2) as native starch granules and type 3 (RS3) as retrograded starch. More recently, a fourth type of RS has been classified, comprising chemically modified starches [2, 3]. RS is claimed to be a good substrate for colonic fermentation and to be beneficial because of its high ratio of butyrate production, which may play a major role in the prevention of colon cancer, as shown in several studies on animal models [4, 5]. The fermentation products from RS are known to lower the pH in the colon which leads to less production, and/or accumulation of potentially harmful by-products of protein fermentation, for example, ammonia or phenols which may promote tumorigenesis [6]. To estimate potential health benefits of RS it is important to be able to predict its behavior in the human gastrointestinal tract, in particular the amount and the structure of starch reaching the large intestine. The digestion of food and absorption of nutrients are spatiotemporal and dynamic processes involving complex enzymatic systems and transport reactions. Thus, the simulation of all these biochemical and physiological events in a single model is illusive. Nevertheless, several in vitro digestion methods, from basic batch systems to sophisticated dynamic models have been developed in order to quantify digestibility of food [7–9]. A realistic approach implies a well-defined system that takes into account the specific contributions of oral, gastric and intestinal digestion. Advantages of using in vitro instead of in vivo models are low costs, relatively easy performance, no limitations by ethical constraints, and the possibility to compare different substrates when applying standardized conditions. Therefore, the aim of this study was to compare a rather simple and a more sophisticated in vitro model as well as the Megazyme RS method to determine the indigestible fractions of two RS preparations containing RS2 and RS3. The digestion residues obtained by the two models were compared and characterized. Furthermore, the RS3 fractions were compared with ileostomy effluents obtained in an in vivo study [10]. Materials and methods Substrates The RS3 containing carbohydrate source, C⋆Actistar®, is a retrograded long chain maltodextrin product obtained after partial enzymatic hydrolysis, and subsequent retrogradation of tapioca starch (RTmd), produced according to United States Patent 6 043 229 [11] and obtained from Cerestar-Cargill (Vilvoorde, Belgium). Physico-chemical characteristics of RTmd were described recently by Pohu [12]. The RS2 containing carbohydrate source is native high amylose maize (HAM) starch and was obtained from Cerestar International (Neuilly-sur-Seine, France). Batch in vitro digestion model The batch model used is a three step incubation at 37°C simulating the digestion in mouth, stomach and small intestine. The method was carried out as described by Lebet et al. [7] and modified by Jörger [13]. A sample of 30 g substrate was suspended in 500 ml phosphate buffer (20 mM, pH 6.9, Na2HPO4 1.42 g/l, KH2PO4 1.36 g/l, NaCl 0.58 g/l), and incubated stepwise with (a) 0.5 ml human salivary α-amylase solution (Sigma A1031, Buchs, CH, 10 mg/ml in CaCl2 1 mM) at pH 6.9 for 15 min, (b) 1.25 ml porcine pepsin suspension (Sigma P7012, 1 mg/ml in NaCl 9 g/l) at pH 2.0 for 30 min and (c) 10 ml porcine pancreatin (Sigma P7545, 0.5 mg/ml in CaCl2 25 mM) at pH 6.9 for 3 h in the presence of 12 g bovine bile (Sigma B8381). Degradation products were removed by dialysis (Servapor 44146, Serva Feinbiochemica GmbH & Co., Heidelberg, Germany; cut-off 12–14 kDa) overnight under continuous movements against running deionized water <20°C. The retentate was freeze-dried. Based on total starch (TS) determinations in the starting materials and digestion residues, the amount of indigestible starch was calculated. Dynamic in vitro digestion model (TIM-1) The dynamic model used in this study has been described by Minekus et al. [8]. The model comprises four serial compartments simulating stomach, duodenum, jejunum and ileum. The in vitro digestion was performed for 6 h at 37°C. A mixture of 60 g substrate, 180 g electrolyte solution (NaCl 5 g/l, KCl 0.6 g/l, CaCl2·2H2O 0.3 g/l, NaHCO3 0.6 g/l), 60 g water, 5 g pepsin solution (Sigma P7012, 0.28 g/l in solution A: NaCl 3.1 g/l, KCl 1.1 g/l, CaCl2·2H2O 0.15 g/l, NaHCO3 7.1 g/l) and 5 g lipase solution (Rhizopus lipase, Amano Pharmaceutical Co. F-AP 15, Ltd. Japan, 0.25 g/l in solution A) were introduced into the gastric compartment. Computer controlled peristaltic valve pumps controlled meal transit through the individual compartments. The pH value was computer monitored by adding HCl (1 M) or NaHCO3 (1 M), respectively. In the stomach the values were preset to pH 4.5, 2.8, 1.8, 1.7 and 1.5 at 0, 20, 40, 60 and 90 min, respectively. In the small intestine, the pH was maintained at 6.5, 6.8 and 7.2 in the duodenum, jejunum and ileum, respectively. About 1 ml trypsin solution (Sigma T4665, 2 g/l in solution A) was added to the duodenum at the beginning of the experiment. Secretions of porcine bile 4 g/100 g in water (Sigma B8631) and pancreatic solution 7 g/100 g in water (Pancrex-V powder, Paines & Byrne, Greenford, UK) entered the duodenal compartment at 0.5 and 0.25 ml/min, respectively. The absorption of water and digestive products from the jejunal and ileal compartments was simulated using hollow-fiber devices (cut-off 5–10 kDa). Ileal effluents (indigestible fraction) were collected after 2, 4 and 6 h, pooled and freeze-dried. Substrate remaining in the jejunum and ileum at the end of the experiment was considered to be indigestible as well and was therefore mixed with the ileal effluent pool. TS determinations of the starting materials, the digestion residues and the remaining substrate in the model after the experiment were carried out. Starch degradation products (up to DP 7) were quantified in dialysates by high-performance anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD) in a BioLC System (Dionex Corp., Sunnyvale CA, USA) using a Carbo-Pac PA1 column and applying a sodium acetate gradient (unpublished, internal method). TS and oligosaccharide quantifications were recalculated as glucose content. The recovered glucose after the experiment (in the digestible and indigestible residues) was assumed as 100%, of which the indigestible fraction is expressed as a fraction. Characterization Both RS preparations were characterized before and after digestion. Additionally, freeze dried RTmd fractions from an ileostomy study feeding RTmd were pooled (7 subjects) and used for characterization analyses [10]. TS was determined using the Megazyme TS assay kit (Megazyme, Co. Wicklow, Ireland) [14]. In brief, starch was hydrolyzed in two phases. In phase 1, starch was pre-treated with dimethyl-sulfoxide (DMSO), totally solubilized and partially hydrolyzed with α-amylase. In phase 2, dextrins were quantitatively hydrolyzed to glucose with amyloglucosidase (AMG). Glucose was subsequently determined enzymatically using the Hexokinase/Glucose-6-phosphate-dehydrogenase assay according to Boehringer [15]. The RS fraction was determined using the Megazyme RS assay kit (Megazyme, Co. Wicklow, Ireland) [16]. The samples were incubated in a shaking water bath with pancreatic α-amylase and AMG at 37°C for 16 h, during which time non-resistant starch was solubilized and hydrolyzed to glucose. RS was recovered as a pellet after centrifugation, which was dissolved by stirring in KOH 2 M and quantitatively hydrolyzed to glucose with AMG [16]. Glucose was determined as described above. Protein was determined as sum of amino acids after acid hydrolysis and ion-exchange chromatography with post-column ninhydrin detection [17]. Wide-angle X-Ray powder diffraction (X-Ray) measurements were performed using a Scintag PADX Diffractometer (PANAlytical, Almelo, NL) operating at 45 mA and 40 kV at an angular range of 2θ from 1 to 40° with a step size of 0.03°. Counting time was 5 s on each step. Differential scanning calorimetry (DSC) measurements were performed using the Thermal Analyst system 2000 (DSC 2910, TA Instument Ltd, Newcastle, UK). Samples at a starch:water ratio of 1:4 were prepared and heated from 4°C to 200°C at 10°C/min, thereafter the samples were cooled at 20°C/min to 4°C and heated again at the same heating rate. All the DSC results were evaluated from the mean of three separate determinations for each sample, unless stated otherwise. Statistical analysis Statistical analyses were carried out using SPSS, version 12.0.1 for Windows (one-way ANOVA and Post-Hoc Tukey). Differences were considered to be significant at P < 0.05. Results Composition and digestibility of RS containing preparations In Table 1 the average TS contents and indigestible starch fractions of RTmd and HAM are shown. All data are mean values of at least five repetitions. Data resulting from the dynamic and the batch model are based on the TS content before and after digestion. Both preparations contain identical amounts of TS. The two in vitro digestion models as well as the Megazyme RS method which is accepted as the official AOAC method for RS determination, led to similar amounts of indigestible RS3 fractions. Digestibility of RS2 was similar using the dynamic model, and the Megazyme RS method, whereas the batch model led to a lower digestibility. Expressing the RS results obtained by the Megayzme RS method based on substrate fresh weight (instead of g/100 g TS as shown in Table 1) yielded to 50.2 ± 2.6, and 45.3 ± 4.1 g/100 g for RTmd and HAM, respectively. This is very well in accordance with respectively, 48.3 ± 2.8 and 46.3 ± 3.9 g/100 g found in a collaborative study published earlier by McCleary et al. [16]. Standard deviations after HAM digestion were found to be higher compared to RTmd digestion. The batch system showed a better reproducibility than the dynamic model and Megazyme RS method for both substrates. Table 1Average total starch (TS) content of retrograded tapioca maltodextrins (RTmd) and high amylose maize starch (HAM) and their indigestible starch fractions obtained by dynamic digestion, batch digestion and the Megazyme resistant starch (RS) method (mean values ± SD)SampleTS (g/100 g dm)Digestion methodIndigestible starch fraction (g/100 g TS dm)RTmd94.3 ± 1.8 (n = 17)Dynamic59.6 ± 2.1 (n = 6)a,bBatch60.6 ± 0.9 (n = 5)aMegazyme57.7 ± 2.9 (n = 12)bHAM94.3 ± 1.1 (n = 7)Dynamic57.5 ± 4.7 (n = 5)dBatch65.8 ± 3.5 (n = 10)cMegazyme54.8 ± 5.0 (n = 17)dMeans with the same letters are not significantly different (Tukey P < 0.05) Composition of digestion residues In Table 2 the average TS and RS contents of RTmd and HAM digestion residues are shown. TS quantifications of in vitro digestion residues were carried out after each single experiment in duplicate. In contrast, RS contents as well as TS in ileostomy effluents were measured in the pooled fractions only. TS contents in digestion residues turned out to be lower than in corresponding starting materials (Table 1), which are due to additional enzymes and bile added during digestion. Moreover, residues obtained from the batch digestion contained more starch than those from the dynamic model. The digestion residues of RTmd and HAM obtained from the dynamic model contained 1.5 g protein/100 g dm and 1.8 g protein/100 g dm, respectively, whereas in the batch digestion residues only 0.6 g protein/100 g dm and 1.2 g protein/100 g dm were determined. Bile contains large amounts of glycine that was correspondingly found in higher amounts in the dynamically digested residues (0.2 and 0.3 g/100 g dm for RTmd and HAM, respectively) compared to the batch digested residues (0.3 and 0.4 g/100 g dm for RTmd and HAM, respectively). Also other amino acids like asparagine, glutamine, alanine, proline, and leucine were found in higher amounts in dynamically digested preparations. The lower amount of TS found in the ileal effluent pool compared to the in vitro digested samples is due to the dilution by additional endogenous and exogenous material such as diet derived protein, dietary fiber and fat. Table 2Total starch (TS) and resistant starch (RS) contents of in vitro and in vivo digestion residues of retrograded tapioca maltodextrins (RTmd) and high amylose maize starch (HAM) (mean values ± SD)SampleTS (g/100 g dm)RS (g/100 g dm)RTmd dynamic72.4 ± 3.7 (n = 12)36.4 ± 1.0 (n = 2)RTmd batch88.6 ± 0.6 (n = 10)61.4 ± 1.2 (n = 3)RTmd in vivo48.4 ± 0.0 (n = 2)31.3 ± 1.1 (n = 2)HAM dynamic74.9 ± 1.7 (n = 10)27.6 ± 2.0 (n = 2)HAM batch80.9 ± 2.6 (n = 12)35.7 ± 0.7 (n = 3) It is expected that RS values close to 100% are found in the in vivo as well as the in vitro digestion residues. As shown in Table 2 lower amounts of RS (28–62%) were found in the digestion residues. This can be explained by the fact that the digestion method carried out prior to the analytical RS determination has an influence on the starch structures leading to less resistant starch. Therefore, analysing RS in digestion residues may be interpreted as characterization method, indicating the highly resistant fraction. Different amounts of RS were found because of additional non-starch material present in the samples as described above. Calculating RS recoveries based on TS led to 50 g/100 g TS for the dynamically digested RTmd fractions, whereas after batch and in vivo digestion similar amounts (69 and 65 g/100 g TS, respectively) were found. In contrast, clearly lower RS amounts were found in both HAM digestion residues (37 and 44 g/100 g) after dynamic and batch digestion, respectively. In Fig. 1 the absorption of digestion products from the jejunum and ileum compartments during the dynamic experiment can be followed. The main digestion products found were maltose and maltotriose despite the fact that larger molecules are able to pass the hollow fiber devices as well. Starch digestion products up to DP3 were found when RTmd was being digested, whereas after HAM digestion minor amounts of maltotetraose (less than 0.1% of total digestible glucose) were detected as well. The kinetics of RTmd and HAM digestion differed considerably. The digestible fraction of RTmd was degraded completely after 4 h, whereas degradation of HAM still continued after 4 h. The lower amounts of degradation products collected during the first 4 h from HAM digestion, and the higher total amounts separated from the ileum compartment show that HAM under these experimental conditions is being digested more slowly compared to RTmd. The enzyme used for starch digestion in the dynamic model is a α-amylase, therefore maltose is theoretically expected to be the smallest degradation product. As can be seen in Fig. 1, glucose was found as well. Glucose could originate from added compounds during the dynamic digestion experiment, such as bile and enzymes. Moreover, a small contamination of α-amylase by AMG or α-amylase containing small amounts of side-activity cannot be excluded. The total amount of starch degradation products absorbed during the dynamic in vitro digestion experiment is slightly higher for HAM compared to RTmd. This is in accordance with a somewhat lower non-significant indigestible residue of HAM compared to RTmd as shown in Table 1. Fig. 1Digestible starch degradation products (up to DP3) of retrograded tapioca maltodextrins (RTmd), and high amylose maize starch (HAM) recovered from the jejunum (a) and the ileum (b) compartments Characterization of digestion residues X-ray diffraction X-ray diffraction patterns of RTmd and its dynamic, batch and in vivo digestion residues are presented in Fig. 2. RTmd and indigestible RTmd fractions all led to a typical A-type pattern. However, the diffraction patterns, with exception of RTmd dynamically digested, exhibited an additional peak at θ2 = 20° which can be interpreted as V-form due to the presence of amylose–lipid complexes. RTmd and RTmd batch digested residues were very similar concerning their crystallinity. The only differences were the peaks at 2θ = 20°, and 2θ = 23°, which were slightly lower and higher, respectively in the batch digested sample. In contrast to RTmd and its batch digested fraction, dynamically digested RTmd showed no peak at 2θ = 27°C and no pronounced double peak at 2θ = 17° and 2θ = 18° (both typical for A-type structure), which indicate a loss in crystallinity due to digestion. The in vivo digestion residue showed an additional peak at 2θ = 29°; this peak could be due to additional structures of proteins or lipids present in this sample. Fig. 2X-ray diffraction patterns of retrograded tapioca maltodextrins (RTmd) and their digestion residues HAM containing samples led to lower crystallinity compared to RTmd (Fig. 3), which correspond to Gerard et al. [18], showing that high amylose maize starches have low crystallinity. HAM and its digestion residues led to B-type pattern, although the peak at 2θ = 20° observed in HAM and batch digested HAM indicated a mixture between B- and V-type which was described by Shi et al. [19] as well. Due to in vitro digestion of HAM and dynamically digestion of RTmd the peak θ2 = 26° disappeared which points to a loss in crystallinity. Similar as for RTmd (Fig. 2) the dynamic digestion led to higher changes in crystallinity compared to batch digestion. Fig. 3X-ray diffraction patterns of high amylose maize starch (HAM) and their digestion residues DSC The thermal behavior of the substrate melting was studied using DSC measurements. The results from the first and second heating are shown in Table 3. During the first heating, endothermic peak temperatures (Tp) turned out to be slightly lower for RS2 compared to RS3. The peak ranges within the RTmd and HAM containing preparations were found to be similar with exception of a slightly narrower peak of in vivo digested RTmd. Enthalpies were found to be lower for HAM containing preparations compared to RTmd containing preparations. The melting transition of RTmd was separated by an exothermic effect (not shown). Such a melting characteristic can be interpreted as partial melting (endothermic), followed by recrystallization (exothermic) and final melting (endothermic) [20, 21]. This exothermic peak was not visible in the digestion residues, which indicates that due to digestion the fractions responsible for these rearrangements were removed. Within the RS3 fractions, in vivo digestion led to significantly lower enthalpies. Dynamic digestion of both preparations led to an exothermic peak when the sample was reheated (Table 3). This indicates that the reorganization of the crystalline structures takes place during the second heating. RTmd and its batch and in vivo digested fractions did not show any heat flow during the second heating (therefore not shown in Table 3), it can be interpreted that recrystallization was completed during cooling in these samples already. HAM containing samples, excluding the dynamically digested fraction, all led to an endothermic peak during reheating which indicates a reversible reconstruction of the amylose–lipid complexes [20]. Table 3Average results (n = 3) from differential scanning calorimetry (DSC) for retrograded tapioca maltodextrins (RTmd) and high amylose maize starch (HAM) and their digestion residues. T0: onset temperatures (°C), Tp: peak temperatures (°C), Tc: completion temperatures (°C), ΔH: transition enthalpy ± SD (J/g) (endothermic if nothing else stated)SampleT0(°C)Tp (°C)Tc (°C)ΔH (J/g)First heatingRTmd (n = 2)60.7112.8140.49.0 ± 1.1aRTmd dynamic58.4108.6132.49.1 ± 0.2aRTmd batch61.8109.1136.410.8 ± 2.0aRTmd in vivo58.4105.0123.55.2 ± 1.1bHAM61.896.9110.02.6 ± 0.3cHAM dynamic55.598.7127.04.2 ± 2.2cHAM batch (n = 2)59.991.5110.53.6 ± 0.7cSecond heatingRTmd dynamic79.787.9144.0*2.4 ± 0.4HAM60.088.2109.62.7 ± 0.9dHAM dynamic84.994.2146.1*2.8 ± 0.2dHAM batch59.883.2101.90.9 ± 0.0e*Exothermic, means with the same letters are not significantly different (Tukey P < 0.05) Discussion Digestibility of RS preparations In vitro digestion models use a strictly standardized procedure based on an average human digestibility. Therefore standard deviations of in vitro experiments are expected to be low. In vitro obtained HAM fractions showed higher standard deviations compared to in vitro digested RTmd fractions. Possibly the molecular structures of HAM have been more sensitive to small differences in experimental conditions such as mechanical stress, temperature or pH between experiments. RTmd in vitro digestion residues result in quantitatively similar RS fractions (Table 1) which are very well in accordance with 59 g/100 g found in vivo after feeding RTmd to ileostomy patients [10]. In contrast, in vitro digestion using HAM as a substrate led to different amounts of RS. These differences may be explained by methodological variations in the in vitro models used, suggesting that HAM was more affected by the experimental conditions. During the dynamic digestion experiment the substrate is exposed to the movements of the flexible walls [8], in the batch model the slurry is stirred [7], and in the Megazyme RS method the sample is shaken [16]. An effect of mechanical treatment on starch degradation of HAM but not of RTmd was shown by McCleary and Monaghan [22] before. They performed the Megazyme RS method comparing shaking and stirring. Stirring led to a higher digestibility of HAM, whereas RTmd did not show any differences. In the present study however, stirring applied in the batch model led to a lower digestibility compared to shaking as used in the Megazyme RS method. This shows that apart from mechanical forces, also other factors may have a significant impact on starch degradation. For example, different amounts and types of enzymes, their accessibilities to the substrate as well as the formation of the enzyme–substrate complexes could have a larger influence on starch degradation of HAM compared to RTmd using in vitro digestion models. Furthermore, the incubation time during the dynamic digestion is nearly twice as long compared to the batch digestion and over four times as long in the Megazyme RS method. As pointed out in Fig. 1, 99% of in vitro digestible RTmd fractions were degraded after 4 h, whereas only 86% of those of HAM were digested at this stage. If the dynamic digestion experiment would have been stopped after 4 h (the duration of the batch digestion experiment is 3.75 h), the remaining indigestible fraction of HAM would be identical to that obtained by the batch in vitro digestion. Whether a prolongation of the batch incubation would lead to HAM fractions similar to those obtained with the dynamic model, remains to be elucidated. The analytical RS contents found in the dynamically digested residues was lower in both preparations when compared to the batch digested fractions (Table 2). This leads to the assumption that in the dynamic approach starch structures are attacked to a higher extent compared to the batch model. Nevertheless, the amount of TS recovered after digestion of RTmd was not affected, whereas the TS recovery of HAM possibly was influenced by the different amount of analytical RS present in the sample; the results strongly indicate that structural features play an important role in digestibility. Structural aspects of the RS preparations X-Ray diffraction A mixture between A- and V-type structures for RTmd, and its in vivo digestion residues (Fig. 2) was found by Pohu [12] as well. The tightly packed A-type structure is known to be very heat stable. Shamai et al. [23] investigated RS3 fractions from high amylose maize, maize flour and wheat starch, respectively. They found that RS3 produced at low retrogradation temperatures of 40°C lead to B-type patterns, whereas incubation at 95°C produced a mixture of A- and V-type polymorphs. This is in contrast to the present study, where RS3 produced from long chain tapioca maltodextrins was found to consist of a mixture between A- and V-type crystal structures, even though rather low retrogradation temperatures of 52–54°C were applied [11]. This shows that the structural behavior of RS3 cannot be predicted without knowing the exact process parameters and these parameters are rarely disclosed, particularly in patents. HAM led to a mixture of B- and V-type polymorphs, which is characteristic for native high amylose maize starch, as shown by several research groups [23–25]. The dynamically digested fraction did not show the V-type characteristic peak at 2θ = 20°. This peak is known to represent amylose–lipid complexes. It is possible that the extra-addition of lipase in the dynamic digestion model was able to degrade lipids present in HAM and RTmd to an extent where no complexation was possible anymore. During batch digestion no additional lipase was used since this enzyme is present in the pancreatin preparation. In earlier experiments it has been shown that the pH-optimum of the pancreatic lipase is approximatively 9 [26]. Therefore, the incubation at pH 6.9 in the batch model is expected to lead to a small lipid degradation only. This could explain that the peak at 2θ = 20° was only reduced during batch procedure, but disappeared using the dynamic digestion model. In vivo digested RTmd showed amylose–lipid complexes (Fig. 2) as well. However, it can not be stated whether the lipid degradation in the dynamic model was more efficient compared to in vivo digestion because diet-derived lipids led to a much higher lipid content before in vivo digestion (results not shown). The X-ray patterns of RTmd and HAM samples indicate that the molecules within the crystals are not packed in the same way which leads to different functional properties of HAM, and RTmd. Batch digestion of both preparations did not lead to a distinct change in crystallinity. However, in vivo digestion caused slightly broader, and dynamic digestion distinctly broader diffraction peaks, which indicate the presence of either imperfect or relatively small crystallites [27]. The latter could have occurred due to the mechanical stress which underlines that the dynamic model attacks the starch structures more strongly. DSC Endothermic transitions of the investigated samples took place at temperatures which are known to destroy the amylose–lipid complexes. At lower temperatures (45–60°C) no peak was detected indicating that amylopectin is neither present in RTmd nor in HAM [20]. RTmd in vivo digested showed the lowest peak maximum and enthalpy of all the digestion residues. This may be due to the higher amount of non-starch material in the sample (Table 1) which may compete for water binding and thus reduce the starch swelling. Dynamically digested HAM showed a slightly broader endothermic peak compared to the other samples which indicates a broader polydispersity [28]. This observation coincides well with the statement that starch is degraded differently, and probably more strongly in the dynamic model. In vitro digestion of HAM led to slightly but not significantly higher enthalpies, especially when dynamically digested. Cooke et al. [27] suggested that the enthalpy of gelatinization primarily reflects the loss of the double helical order which could occur to a higher extent during dynamic digestion. Conclusions The results of this study show that both, a simple and a more sophisticated in vitro digestion model and the ileostomy study led to similar amounts of starch escaping digestion when using a RS3 containing carbohydrate source as substrate. In contrast, when the two in vitro digestion models were compared with a RS2 containing preparation, different amounts of indigestible starch were found. It is not possible to predict which model simulated the in vivo RS2 fraction more precisely. The more sophisticated in vitro digestion model attacked starch probably more strongly compared to the batch in vitro digestion model using an RS2 and an RS3 containing substrate. This was confirmed by differences in crystalline fractions and structural rearrangements as measured by X-ray and DSC measurements. These structural differences, caused by different experimental parameters influenced starch digestibility of HAM but not of RTmd. The behavior of other RS preparations in the two in vitro models described here can not be deduced from the obtained results and has to be examined case by case.

J_Med_Internet_Res-1-1-1761705

Can Examination of WWW Usage Statistics and other Indirect Quality Indicators Help to Distinguish the Relative Quality of Medical websites?

Background The Internet offers a great amount of health related websites, but concern has been raised about their reliability. Several subjective evaluation criteria and websites rating systems have been proposed as a help for the Internet users to distinguish among web resources with different quality, but their efficacy has not been proven. Introduction After the early enthusiasm generated by the potential use of the Internet in Medicine [1,2,3], concern has been raised about the quality of the resources available on the Internet compared to more academic media. It is technically very easy to publish on the Internet [4]. The lack of a review process of the documents on the Net, and the power of this media in transmitting the data has the risk of misinforming both lay people [5,6,7] and health care professionals [8]. However, only a few studies have tried to measure this risk of misinformation [9,10,11]. Nothing yet is known about the users' ability to discriminate between low and high quality resources. Several initiatives have been proposed which could be applied at different levels to improve the average quality of medical websites. For instance, we could apply certain basic methods for the websites to be correctly designed. In this sense, some academic organizations have proposed a set of basic information that every medical web site should provide about the author and sources of the web site contents, their potential conflicts of interest and funding, and the currency of the information [12]. But many of the available medical websites have been created without any quality control by a third party. How can Internet health care visitors distinguish between such different resources? Internet users can find health and medical related websites in several ways. World Wide Web search engines (e.g., AltaVista, Excite, Infoseek and many others) provide the users with a list of websites that match a given topic, with the results ordered by syntactic similarity with the query [13]. Unfortunately, the quality of contents is not guaranteed. On the other hand, certain websites indexes and review services, such as Medical Matrix (http://www.medmatrix.org/) and HealthAtoZ (http://www.HealthAtoZ.com/), offer systematic evaluations of medical resources on the Web [14], as a post publication editorial process. These rating systems could be an useful tool for guiding the visitors of medical websites [12]. However, authors who have reviewed these Internet resources, point out the variability of their evaluation criteria and their doubtful efficacy [14]. The quality of a given medical article on the Internet could be measured by the users opinion about it, for example by counting the number of times it is retrieved [15]. However, this idea has been criticized because it would replace the scientific peer review process with the opinion of the Internet users, whatever their qualification [6]. Despite the differences between the printed medical information and the Internet, several evaluation tools from the former could be useful if applied on the "Net." Similarly to printed medical journals, medical documents on the Internet could be ranked by their citation analysis [15,16], but no methods have been proven for use with medical websites. When an article is quoted in a paper, certain agreement among the authors may be supposed. Similarly, when a webmaster makes a link from his web site to another, certain credibility is given to the latter. In fact, the International Committee of Medical Journals Editors recommend caution when a link is made from a peer reviewed journal site to other sites [17]. If linking on the web can be equivalent to quoting in printed medical articles, a citation analysis on the web could be performed by the quantification of the links to a given medical web site. The ideal method for assessing the quality of medical websites should provide a means of rating great amounts of medical web resources while respecting the World Wide Web peculiarities, such as its multimedia capabilities and changing contents. At the same time, it should at least be as reliable as systematic reviews of those resources by editorial boards. In summary, it should be a method born in the Internet but with the efficacy of those used in the printed media. In this study, we evaluated the reliability of four websites characteristics as medical websites quality indicators. The four characteristis used: their authors' impact factor, their grade of updating, their daily visits and inbound links. The evaluations of a sample of pediatric websites by a number of Internet rating systems was the gold standard with which these websites characteristics were compared. Methods During March 1998 a subset of websites rating systems were compiled. From these, we selected a sample of websites that were studied during the first week of April 1998. Eight web rating systems, whose evaluations were offered as figures, were compiled from previous studies [13,14] (Table 1). One half of the selected rating systems gave the results of their evaluations by means of graphic analog scales, and the other half by numeric scales. Every web site evaluated by these rating systems that provided information about child health, whether for lay people or health professionals, was included in the study. Some of these rating systems (e.g., Lycos Top 5%) provides the visitors with a search tool by keyword. In these cases, the websites were selected using the keywords "Pediatrics", "Infancy", "Child health", and "Child Care." For the remaining rating systems, the pediatric websites were compiled manually. Those websites not accessible twice during the study period were excluded. Only three rating systems (Medical Matrix, Physician Choice, and Six Senses) gave information about their editorial boards. Most of their members were physicians. Two of the web rating systems only gave a global result of their websites evaluation (Medical Matrix and Magellan), while the rest (HealthAtoZ, Argus Clearinghouse, Lycos Top 5%, Sympatico Health, Physician Choice, and Six Senses) gave a result for each considered criterion. Content was a common criterion to all the eight ranking systems. Therefore, the results of the evaluation of each web site were divided in two categories, content and non-content (design) aspects. In order to make comparisons, the results of the evaluations of the websites supplied by each rating system were transformed to a one hundred scale. Table 1 Compiled web sites ranking systems. The results of evaluations are showed as two possible types of scales, graphic analog (A) or numeric (N) Rating systems (Included/excluded web sites) Uniform Resource Locator Type of scale Argus Clearinghouse Seal of Approval (16/1) http://www.clearinghouse.net/cgi-bin/chadmin/viewcat/Health___Medicine?kywd++ A HealthAtoZ (241/66) http://www.healthatoz.com * Lycos Top 5% (8/3) http://point.lycos.com/topics/Health_Overall.html N Magellan Internet Guide (40/11) http://www.mckinley.com/magellan/Reviews/Health_and_Medicine/index.magellan.html A Medical Matrix (75/11) http://www.medmatrix.org/SPages/Pediatrics.asp A Physician's choice (4/0) http://www.mdchoice.com/pcsites.htm N Six Senses Seal of Approval (4/0) http://www.sixsenses.com/winners.html N Sympatico Health (8/1) http://www1.sympatico.ca/Contents/Health/LISTS/D3-C03_all1.html A * Graphic analog scale developed in numeric When provided, the daily visits registered by the websites visits counters were recorded. In some websites the date from which the counter was started was not available. Thus, their webmasters were asked for this information by electronic mail, and it was included in the statistical study if provided before the end of the observation period, 15th April 1998. The websites authors and editors' names were searched in 1997 MEDLINE [18], and their articles were registered. Their impact factors of the journals wherein they were published were obtained by using the 1996 Science Citation Index (Institute for Scientific Information, Philadelphia, PA). The impact factor of a given web site author was the sum of the impact factors of his or her articles. For institutional websites only the name of the web editor was considered. When provided, the time since the last update was also recorded. Finally, by means of the Web search engine Infoseek [19], we calculated how many websites on the Internet linked to each web site of our sample. The searching strategy by syntax of this engine allows to know the websites that are linked to a given web site [20]. As a web site may be linked not only from external websites but also from websites of its own organization, we only considered external links. Although other search engines such as AltaVista, Excite or HotBot offer similar searching options, we chose Infoseek because it provided the results of the queries grouped by web site, which makes the exclusion of the internal links easier. Comparison of means was performed by Mann-Whitney U test, and correlation analysis by means of Spearman's correlation coefficient ( rS). P values equal or less than .05 were considered significant. All computations were made with SPSS for Windows 7.0 (SPSS Inc., Chicago, IL) statistical package. Results After excluding 93 non-accessible websites, a total of 363 pediatric websites were compiled. Table 2 Correlations among the number of daily visits to the web sites, the impact factor of their authors or editors, the grade of update, and the number of links that receive. NS means not significant Number of inbound links rS p Visits/day rS p Author's impact factor rS p Visits/day .46 .005 Author impact factor NS NS Weeks since the last update -.36 <.001 NS NS Table 3 Correlation among the number of links and visits to the web sites, the impact factor of their authors, and the time since the last update, and the results of their evaluation by HealthAtoZ and Medical Matrix. No significant correlations were demonstrated with the other systems. Medical Matrix only provides total results, does not specify results by contents and non-contents aspects Number of Inbound Links Visits/day Author impact factor Weeks since the last update rS p rS p rS p rS p HealthAtoZ Total Contents Non contents .29 <.001 .30 <.001 .24 <.001 NS NS -.19 .04 -.23 .00 NS Medical Matrix Total NS .79 .03 NS NS On average, the websites of our sample received links from 470 other sites on the Internet (range, 0 to 3574). In 48% of the websites, information on their last update was given. On average, they had been updated 47.5 weeks before (range, 0 to 395). Only 10% of the websites had a visit counter, and the average daily visits were 470 (range, 1.2 to 3145). Seven visit counters did not distinguish among different visitors, that is, they registered any visit to their websites. In 137 websites (38%) the editor/author's name was given, but only 60 of them had published at least one article since January 1997 in the journals included in MEDLINE database. Their average impact factor was 2.14. Figure 1 Weeks since the last update for the total of the sample, n=363, and for the websites evaluated at least by two rating systems, n=25 (median, 25th and 75th percentiles) Figure 2 Number of inbound links to websites for the total of the sample, n=363, and for the websites evaluated at least by two rating systems, n=25 (median, 25th and 75th percentiles) Table 4 Top 50 pediatric web sites of the sample (N= 363) by the number of their inbound links. The weeks since the last update, the number of daily visits to the web sites and their editor/author's impact factor are also provided. In parenthesis, the place that each web site would obtain if ranked by the two latter criteria. In italics, those web sites indexed at least by two rating systems. Missing values are due to the lack of visits counter, editor's name, or information about the last update, for many web sites Uniform Resource Locator Number of inbound links Daily visits to web sites Web site editor/author's impact factor Weeks since the last update 1 http://www.merck.com 3574 - - 13 2 http://www.ucalgary.ca/~dkbrown/index.html 2355 1620 (3º) 0 ( 360º) - 3 http://KidsHealth.org 1109 - - - 4 http://www.psych.med.umich.edu/web/aacap 927 - - 3 5 http://www.aap.org 896 - - 1 6 http://www.chadd.org 785 - - 4 7 http://www.castleweb.com/diabetes 767 - - - 8 http://www.medconnect.com 714 - - - 9 http://www.aaaai.org 677 - - - 10 http://www.aacap.org/web/aacap 612 - - 4 11 http://www.nas.com/downsyn 572 - 0 ( 360º) 1 12 http://www.childbirth.org 534 - - - 13 http://web.syr.edu/~jmwobus/autism 502 - 0 ( 360º) - 14 http://oncolink.upenn.edu/disease 487 - 10.1 (8º) 9 15 http://www.jdfcure.com/index.html 428 - - - 16 http://www.mic.ki.se/Diseases/index.html 423 1412 (5º) - - 17 http://www.asf.org 365 940 (6º) - 1 18 http://www.mdcc.com 365 - - 1 19 http://www.mc.vanderbilt.edu/peds 357 - - - 20 http://www.ama-assn.org/journals/standing/jama/jamahome.htm 330 - - - 21 http://www.med.jhu.edu/peds/neonatology/poi.html 322 253 (10º) 9.3 (13º) 2 22 http://www.wish.org 317 - - - 23 http://education.indiana.edu/cas/adol/adol.html 312 - 0 ( 360º) 52 24 http://www.kidsdoctor.com 297 - 0 ( 360º) - 25 http://www.xmission.com/~gastown/safe 297 94 (20º) - - 26 http://www.childquest.org 287 - - 6 27 http://www.uab.edu/pedinfo 284 - - - 28 http://www.childsecure.com 255 - - - 29 http://www.mc.vanderbilt.edu/peds/pidl 254 - 0 ( 360º) 1 30 http://www.stjude.org 251 - - - 31 http://www.nccf.org 249 70 (25º) - 8 32 http://www.mda.org.au 238 - 0 ( 360º) 12 33 http://www.peds.umn.edu 235 - 0 ( 360º) 2 34 http://www.csmc.edu/neonatology 232 117 (15º) - 1 35 http://med-aapos.bu.edu 225 - 0.4 (51º) 3 36 http://www.jhbmc.jhu.edu 220 - - 3 37 http://sids-network.org 214 3145 (1º) 0.3 (54º) 1 38 http://www.diabetes.com 212 - - - 39 http://sids-network.org/index.htm 208 3145 (1º) 0.3 (55º) 1 40 http://www.oneworld.org/scf 205 - - - 41 http://www.childmmc.edu 204 - - 13 42 http://www.os.dhhs.gov/hrsa/mchb 197 - - 7 43 http://www.wp.com/pedsrheum 197 81 (23º) 11.7 (6º) - 44 http://dem0nmac.mgh.harvard.edu/neurowebforum/neurowebforum.html 188 2441 (2º) 0 ( 360º) - 45 http://pedsccm.wustl.edu 179 145 (13º) 1.0 (39º) 2 46 http://www.drgreene.com 179 - - - 47 http://www.medsch.wisc.edu 162 - - - 48 http://www.blindcntr.org/bcc 150 - - - 49 http://home.coqui.net/titolugo 144 68 (26º) 0.2 (56º) 1 50 http://www.chmcc.org 141 - - 1 Only 25 websites of the sample were indexed and evaluated at least by two rating systems, and none by the eight. This subset of websites showed significantly better results of the evaluation of their contents and design by HealthAtoZ, and higher grade of updating (Figure 1) and higher number of inbound links (Figure 2). When the evaluations of these 25 websites by the different rating systems were compared, no significant correlations were found. Changes regarding the average impact factor of the authors of the websites or the number of daily visits could not be demonstrated in this subset of websites. Some interesting correlations between the results of the evaluations of the websites and the rest of study variables were found. The number of links received by the websites significantly correlated with their daily visits and with the time since the last update (Table 2). The number of inbound links also correlated with the results of the websites evaluation by HealthAtoZ (Table 3). The number of daily visits significantly correlated with the results of the websites evaluation made by Medical Matrix, and the grade of updating significantly correlated with the results of the contents and designs evaluation made by HealthAtoZ (Table 3). Finally, no correlation was demonstrated between the average impact factor of the websites authors and the other variables. The top fifty pediatric websites of the sample are shown in Table 4, ordered by the number of their inbound links according to the Infoseek indexing engine. More than a half of the 25 websites indexed by at least two rating systems may be found among these top fifty websites. Discussion In this study, certain websites characteristics that depend on the users' preferences have been compared with evaluations of pediatric resources on the Web by third parties. Although rating systems have been previously criticized because their editorial boards frequently do not employ uniform criteria [13], we have considered them as the standard method because it somewhat represents a post-publication review process. Some aspects of our method are open to discussion. Firstly, the reliability of the data regarding the daily visits and the updating frequency depends on the accuracy of the information that the websites editors offer in their sites. In this sense, we considered the grade of updating of the websites by the dates of their last changes. Clearly these changes could involve very different aspects and in different grades, and not necessarily provide more current contents. However, we believe that it could demonstrate the editor's efforts in maintaining or increasing the interest of his web site for the visitors. The results regarding the number of daily visits to the websites must be considered with caution when comparing one web site to another, because some visit counters were set to register every visit, instead of every distinct visitor. Nevertheless, both can be considered usage indexes of a given web site. On the other hand, quantification of links to the websites clearly depends on the power of the search engine we employ. By no means our results show the total number of links to the websites in our sample. In fact, a previous article states that it would be necessary to combine the databases from at least five large search engines to cover the most of the web [21]. Although all bibliometric indexes have limitations [22,23], we employed the impact factor as a measure of the webmasters' publishing capacity because it is a classical indicator of the quality of biomedical articles. Recently, it has been suggested that every medical web site should be evaluated following some basic criteria [24]. One of the more accepted criteria is that the authorship must be clearly stated, as a basic means for assessing the reliability of the web site contents. However, we could not demonstrate that the more highly evaluated, the most updated, or the most linked or visited pediatric websites, had the authors with the highest publishing capacity measured by their impact factor. In other words, some web quality standards do not correlate with classical quality standards from the printed media such as the impact factor of a given author's articles. We could not find statically significant correlations among the evaluations of the websites by the different rating systems. This is probably due to the small size of the subset of websites indexed and evaluated by all the systems, and their different evaluation criteria. However, some interesting data were found when we considered the correlations among the four websites characteristics and the evaluations. We found that the best websites for HealthAtoZ, the largest analyzed rating system, were the most updated and the most linked ones. On the other hand, the most valuable websites for Medical Matrix, the second rating system by size, were the most visited ones. In any case, both the number of daily visits and the time since the last update highly correlated with the number of inbound links. The lack of correlation among the four variables and the evaluations by the other rating systems could be due to their little contribution to our sample. Many efforts to establish quality criteria will have limited efficacy due to the dynamic behaviour of the Internet as a publishing medium. In fact, a recent article demonstrates the lack of consensus among the editorial boards of a large sample of evaluation and rating systems regarding the evaluation criteria they employ. The same authors pointed out that "... it may be difficult or even inappropriate to develop a static tool or system for assessing health related websites." [25] Therefore, the question could be to provide context to this issue. That is, to know how good a given medical web site is in comparison with the rest of medical websites. A democratic and feasible method for reaching this objective could be let the Internet community say which medical websites are the best ones, that is, which they usually visit or which they usually recommend by linking to them. Moreover, we believe that the fact that these usage indexes correlate with the evaluations by third parties, qualifies them as quality markers. Eysenbach and Diepgen [16] have recently proposed that an ideal quality control system for medical resources on the Internet should take in account the users opinion, and not only their evaluation by a third party, that is, a "downstream filtering" and not only an "upstream filtering" approach. More interestingly, our study demonstrates certain agreement among both approaches in identifying high quality resources. LaPorte et al [15] proposed an electronic publishing system in which the impact of a given resource on the Internet could be measured by counting how many times the document was retrieved or quoted. The introduction of the citation analysis of the medical resources on the web as a method to assess their quality has been recently proposed [16]. On the other hand, a very promising software system is being developed by Kleinberg [26,27]. This system would provide the users with a way of knowing the very best of the web on a given topic in a faster and more complete way than commercial human compiled directories. This system is based in the identification of two subsets of websites when a query on a given topic is made, those websites containing a lot of information about the topic (authoritative websites) and those which contain large amounts of links to the former (hub sites). Our work demonstrates that those authoritative websites, that is the more linked ones, are indeed the best ones regarding the evaluation of its contents and design by the editorial boards of some large web rating systems. The citation analysis of biomedical journals has been a classic tool in assessing their relative quality [28]. Similarly, medical web resources could be ranked by a "webcite index" [16], which is not yet defined. Linking in the World Wide Web could be equivalent to quoting in printed publications, and its quantification could be useful for measuring the relative quality of medical websites. Some indexes could be created to make more rational comparisons among websites with different sizes. For example, in the same way that the calculation of the impact factor of a given medical journal takes into account the number of articles published by that journal yearly, the size of a given domain could be considered to obtain some indexes that would express more accurately the grade of linkage of a medical web site. Moreover, Platform for Internet Content Selection (PICS) [29], an infrastructure that could be applied as a filtering system of the medical information on the Net [16], could incorporate these indexes as one of the meta- data assigned to every medical document as electronic labels. Then, these electronic labels could be checked automatically by an user's browser, bypassing those documents with a "webcite index" not high enough. A problem could be how to avoid false "self-labelling" by dishonest webmasters. In any case, more work is needed to give answers to these and other technical questions on the emerging field of Webometrics [30]. An evaluation system based on this quantification would bring advantages and risks. Rankings could be generated very quickly and in an objective way, because the Internet community by itself would evaluate great amounts of medical websites. However, this evaluation process would be made a posteriori, and the potential harmful effects of the diffusion of documents without enough quality could not be avoided. Therefore, this method could not replace previous editorial effort that warrants a minimal quality for each resource. Our work demonstrates that the visitors of pediatric websites and the editors of websites on the "Net," so called webmasters, show certain maturity when they have to identify the pediatric resources with high quality. We believe that the key point is how to augment the proportion of these resources. An important issue could be to establish a citation style not only for articles from peer reviewed electronic journals [31], but also for any medical document on the Net. The prestige that citation in a printed journal represents will stimulate high quality publishing on the Internet, and web site editors will employ enough review processes to obtain the necessary quality. A web site's ranking system based on the citation analysis on the web by the quantification of links would be an additional incentive. The more valuable resources will attract the Internet users' visits and the webmasters' links, and very likely the best funding and financial supports. In summary, although the Internet provides a very different publishing medium, traditional means borrowed from printed journals could also be used with this electronic media for achieving minimal levels of quality. These include certain peer review processes, that enhance the rigor of the documents submitted for publication taking in account the peculiarities of this media, and linking analysis as a measure of the citation on the World Wide Web.

Mol_Hum_Reprod-1-1-2408935

Regulation of spindle and chromatin dynamics during early and late stages of oocyte maturation by aurora kinases

Examination of factors regulating oocyte chromatin remodeling is crucial to circumvent embryonic aneuploidy and resulting defects. Aurora kinases (AURK) are involved in regulation of chromatin remodeling, however, little attention has been paid to AURKs in regard to oocyte maturation. Meiotically incompetent mouse oocytes contain transcripts for all three Aurk isoforms: A, B and C. Upon achieving meiotic competence, oocytes showed significant increases in transcript levels of all three Aurk isoforms and transcript levels remained unchanged as oocytes progressed through meiosis, with AurkA being the predominant isoform. Inhibition of oocyte AURKs during the prophase–metaphase I (MI) transition via inhibitor ZM447439 (ZM) had no effect on germinal vesicle breakdown. However, meiotic spindles were malformed, and microtubule organizing centers and chromatin were scattered. Chromosomal spreads of MI oocytes indicated AURK inhibition resulted in abnormal chromosome condensation. Furthermore, inhibition of AURK during prophase I–MII prevented completion of MII and extrusion of the polar body. Inhibition of AURKs during the MI–MII transition resulted in significantly fewer cells progressing to MII and induced aberrant chromatin remodeling. Further analysis indicated that inhibition of AURKs resulted in absence of histone-H3 phosphorylation at serine 10 and 28. These data suggest a ZM-sensitive AURK may be an oocyte histone-H3 kinase capable of regulating chromatin remodeling throughout oocyte meiosis, both pre- and post-MI. Introduction In vitro oocyte maturation offers immense potential for treatment of infertility, however, current systems are relatively inefficient (Tan et al., 2007). Additionally, mammalian oocytes are notorious for high rates of chromosomal abnormalities (Hassold and Hunt, 2001), resulting in subsequent embryonic aneuploidy, infertility and congenital defects. Unfortunately, components of successful oocyte maturation and regulation of these events remains enigmatic. Therefore, understanding regulatory mechanisms involved in oocyte meiotic maturation, especially those controlling chromatin remodeling, is imperative to establish therapies to improve current assisted reproductive technologies and circumvent oocyte-derived infertility and aneuploidy-induced congenital defects. Remodeling of chromosomes during oocyte meiosis begins when homologues initially pair and condense via actions of the synaptonemal complex during initiation of prophase I, accompanied by homologous recombination and crossing-over events (Vallente et al., 2006). Chromatin subsequently decondenses as oocytes enter a phase of quiescence prior to completing prophase I. In response to the pre-ovulatory gonadotropin surge, follicle-enclosed oocytes resume meiosis and homologues condense in preparation for a reductional division (Mehlmann, 2005). A bipolar meiotic spindle forms, consisting of polymerized microtubules, and attaches to homologues at their centromeres. Subsequently, physical contact between homologous pairs at chiasmata counteract forces pulling apart homologues, resulting in alignment of chromosomes along the metaphase plate, signaling completion of metaphase I (MI). The meiotic spindle then facilitates separation and segregation as homologues are pulled toward opposite spindle poles at the beginning of anaphase. Oocytes progress through telophase, resulting in disproportionate cytokinesis and extrusion of the first polar body signaling completion of meiosis I (Wang and Sun, 2006). Subsequently, oocytes forego DNA replication and re-arrest at MII. Reversible phosphorylation, controlled via actions of protein kinases and phosphatases, is an extremely important regulator of oocyte meiosis, and well-suited to the rapid changes required during the cell cycle (see reviews, Swain, 2007; Swain and Smith, 2007). Extensive research has been conducted on the role of kinases such as cyclin-dependant kinase (CDK1) and mitogen-activated protein kinase (MAPK) in mammalian oocytes (Motlik et al., 1998; Abrieu et al., 2001). Additionally, roles for protein phosphatases (PPP) in control of oocyte germinal vesicle breakdown (GVBD) (Alexandre et al., 1991; Gavin et al., 1991, 1994; Swain et al., 2003), spindle dynamics (Alexandre et al., 1991; Lu et al., 2002) and chromatin remodeling have been identified (Mailhes et al., 2003; Swain et al., 2007). However, relatively little attention has been paid to aurora kinases (AURK) in regard to mammalian oocyte maturation. AURKs are a family of serine/threonine protein kinases responsible for regulating several mitotic cell cycle events important to maintaining proper cellular ploidy, including chromosome condensation, spindle dynamics and cytokinesis (see review, Carmena and Earnshaw, 2003). Thus, considering high rates of aneuploidy associated with oocyte meiosis, AURKs may play an instrumental role in the female gamete was well. Based primarily on research from lower eukaryotes and mitotically dividing mammalian cells, three AURKs exist categorized by sequence differences and subcellular localization; AURKA, AURKB and AURKC (Nigg, 2001). Aurora-A has been identified in mammalian oocytes localized to nuclei in GV-intact oocytes; to the spindle, spindle poles and condensing chromatin during MI; and to spindle poles at MII (Yao and Sun, 2005). Neutralization of oocyte AURKA activity delays GVBD and distorts MI spindle organization (Yao et al., 2004). Aurora-B kinase is known as the equatorial kinase and localizes to condensed chromatin during mitosis and meiosis in lower eukaryotes, where it is thought to regulate chromosome condensation (Hsu et al., 2000; Murnion et al., 2001) and homologue separation (Kaitna et al., 2002; Rogers et al., 2002). Aurora-C kinase is the least studied AURK isoform, previously reported to be testes-specific with localization to centrosomes (Tseng et al., 1998; Kimura et al., 1999; Hu et al., 2000); though transcript for AurkC has now been identified in human cumulus–oocyte complexes (Assou et al., 2006) and reports indicate that it binds to chromosomal passenger proteins during mitosis (Li et al., 2004; Sasai et al., 2004; Yan et al., 2005). Objectives of this study were to determine which Aurk isoform transcripts are present in oocytes; examine functional roles of AURK activity during various meiotic stages of mouse oocyte maturation, and identify AURK phosphoprotein substrates and targets of actions, focusing primarily on regulation of chromatin remodeling during the first meiosis. Materials and Methods All procedures described within were reviewed and approved by The University Committee on Use and Care of Animals at the University of Michigan and were performed in accordance with the Guiding Principles for the Care and Use of Laboratory Animals. Mouse stimulation and oocyte collection Meiotically incompetent GV-intact oocytes were collected from 11-day-old female CF1 mice (Harlan, Indianapolis, IN). Meiotically competent GV-intact oocytes were collected from 20–23-day-old CF1 female mice, 42–44 h following injection with 10 IU eCG (Sigma, St Louis, MO). Oocytes were isolated by manual rupturing of antral ovarian follicles in Hepes-buffered human tubal fluid medium (HTFH; Irvine Scientific, Santa Ana, CA) supplemented with 0.3% w/v polyvinylpyrrolidone (Sigma). RNA isolation, reverse transcription and real-time PCR Oocyte total RNA was extracted from 50 oocytes at each development stage using Picopure RNA isolation kit (Arcturus Bioscience, Mountain View, CA) following manufacturer’s instructions. Oocyte cDNA was synthesized using 125 pmol random hexamer, 500 µM dNTP, 20 U RNase inhibitor and 62.5 U MultiScribeTM reverse transcriptase (ABI systems) in a final volume 50 µl. Primers for mouse AurkA, AurkB and AurkC were designed with no sequence overlap between isoforms (AurkA—forward primer: 5′ cactagcaaagagccaacca 3′, reverse primer: 5′ ggtggcttcaatagggtgtt 3′; AurkB—forward primer: 5′ cctgacctactgccacaaga 3′, reverse primer: 5′ gccaaagtctgcaatcttca 3′; AurkC—forward primer: 5′ ctgccatgagaagaaggtga 3′, reverse primer: 5′ gtccagagtcccacacattg 3′). Real-time PCR was performed on Applied Biosystems 7300 Real-Time PCR system. Each PCR was performed with 1.5 oocyte equivalents of cDNA added to SYBR Green PCR Master Mix (Applied BioSystems, Foster City, CA). In addition, control reactions were conducted consisting of no template with primers and master mix. Real-time PCR reactions were carried out for 40 cycles (95°C for 15 s, 60°C for 1 min) after initial 10 min incubation at 95°C. Following PCR, products were isolated and run on a 2% agarose gel for 60 min at 100 V to verify size of the amplified product. Additionally, DNA was isolated from gels using QIAquick Gel Extraction Kit (Qiagen, Chatsworth, CA) and subjected to DNA sequencing to verify identity of the product. Standard curve method was used to compare relative abundance of a single Aurk isoform between oocyte meiotic stages using normalization of β-actin levels. Data were collected over three replicates, with triplicate samples for each isoform and fold increases were based on meiotically incompetent GV-intact oocytes levels, which were normalized to 1. Statistical significance was determined using unpaired Student’s t-test, P < 0.05). To examine relative abundance of all three isoforms within a single time point, we ensured primer efficiency of all samples were within a 5% range of an internal β-actin control. We then analyzed data using comparative Ct method. Oocyte culture and AURK inhibition Aurora kinase inhibitor ZM447439 (ZM, Astra Zeneca, Wilmington, DE) was dissolved in dimethylsulphoxide (DMSO) to obtain a 10 mM stock. Stock solution was dissolved in HTF to obtain final concentrations of 0.625, 1.25, 2.5, 5, 10 and 20 µM. Control treatments contained DMSO. To assess effects of AURK inhibition on oocyte maturation, meiotically competent GV-intact oocytes (prophase I) were placed into culture in presence or absence of varying doses of ZM. Oocytes were assessed for GVBD and MII development at 2 and 16 h, respectively. Experiments were performed in triplicate and statistical differences in development were assessed using chi-square analysis with differences considered significant if P < 0.05. To determine effects of AURK on chromosome condensation and spindle formation during the prophase I to MI transition, prophase I oocytes were matured in vitro to MI (7 h) in the presence or absence of ZM (10 µM) then subjected to immunocytochemistry (ICC) or processed for chromosome spreading. Prophase-I oocytes were also cultured to a time point consistent with MII to assess spindle and chromatin characteristics following extended AURK inhibition. To assess effects of AURK inhibition on oocyte meiosis during the MI–MII transition, oocytes were matured for 7 or 9 h in the absence of any chemical manipulation to allow normal spindle formation and chromatin remodeling. Oocytes were then cultured to MII (an additional 9 or 7 h) in presence or absence of 10 µM ZM, followed by assessment of chromatin positioning and spindle configuration. All experiments were performed in triplicate and nonparametric parameters were analyzed for significant differences by Chi-square. Finally, to begin to determine substrates for oocyte AURK, histone-H3 phosphorylation at ser10 and ser28 was assessed following AURK inhibition at various time points utilizing ICC and western blot analysis. Immunocytochemistry To examine effects of AURK inhibition on spindle formation and metaphase chromosome positioning, MI and MII oocytes were attached to poly-lysine coated coverslips, and fixed in 2% (w/v) paraformaldehyde with 0.05% (v/v) Triton X-100 in phosphate-buffered saline (PBS) (pH = 7.3) for 30 min. Oocytes were then blocked overnight with 2% (w/v) bovine serum albumin, 0.1 M glycine and 5% (w/v) dry milk in PBS at 4°C. Oocytes were incubated with β-tubulin antibody (Sigma, 1:200) and pericentrin antibody (Abcam, 1:1000). Negative controls included non-immune mouse serum in place of primary antibody. After three 5 min washes with blocking solution, samples were reacted with the appropriate Alexa 568 and 488 conjugated secondary antibodies (Molecular Probes) at a 1:750 dilution for 1 h at 37°C. Following washing, slides were incubated with Hoescht 33 342 (1 µg/ml) in PBS for 20 min at 37°C. Coverslips were then mounted on glass slides with 90% glycerol in PBS for fluorescence microscopic visualization under ×1000 on a confocal microscope. Chromosomal spreading and analysis Following culture to MI in ZM (10 µM), oocytes were collected and prepared for chromosomal spreading (Hodges and Hunt, 2002). Briefly, zona pellucida were removed by exposure to 1% pronase in HTFH. Zona-free oocytes were then washed and fixed by carefully placing them onto a microscope slide dipped in a solution of 1% paraformaldehyde in distilled water (pH 9.2) containing 0.15% Triton X and 3 mM dithiothreitol. Slides were then placed into a humidified chamber overnight before being subjected to triplicate 5 min washes in PBS and air-dried at room temperature. To analyze chromosomal condensation, slides were placed into a 1% solution of Hoescht 33 342 in PBS for 10 min and subjected to three more washes in PBS. Glycerol mounting solution and a coverslip were added and slides were sealed. Chromosomal spreads were analyzed blind to treatment at ×1000 on a Leica DMR microscope. Statistical differences between treatment groups were analyzed using chi-square analysis. Electrophoresis and western blot analysis To assess effects of AURK inhibition on oocyte histone-H3 phosphorylation, groups of oocytes (n = 100) were prepared for western blot analysis. Oocytes were placed in 2× sodium dodecyl sulphate (SDS)–polyacrylamide gel electrophoresis (PAGE) sample loading buffer [80 mM Tris–HCl (pH = 6.8), 20% glycerol, 4% SDS, 4% β-mercaptoethanol, 0.04% bromophenol blue], vortexed and placed on ice for 15 min. Following sonication on ice for 10 s, samples were denatured at 90°C for 10 min and loaded for electrophoresis. Total protein from equal numbers of mouse oocytes was loaded in each lane and separated by one-dimensional SDS–PAGE. Resolving gels were cast using 12% acrylamide; stacking gels contained 5% acrylamide. HeLa cell histone lysate was used as a positive control for recognizing phospho-ser10. Gels were equilibrated and transferred to Hybond-P PVDF transfer membrane (Amersham Life Science, Little Chalfont Buckinghamshire, UK) by Semi-Dry Electrophoretic Transfer Cell (Bio-Rad Laboratories, Hercules, CA) according to the manufacturer’s instructions. Blots were blocked in 5% nonfat milk in Tris-buffered saline (TBS)+0.5% Tween (TBST) at room temperature for 1 h and incubated with the appropriate primary antibody diluted in TBST + 5% nonfat milk overnight at 4°C with agitation. Antibodies included anti-phospho-ser10-histone H3 (1:1000, Upstate) and anti-phospho-ser28-histone-H3 antibody (1:500, Upstate). After complete washing in TBST, blots were incubated with the appropriate horse-radish peroxidase-conjugated IgG secondary antibody (diluted 1:2000) at room temperature for 1 h, washed in TBST and developed with ECL Plus reagents (Amersham Life Sciences) according to the manufacturer’s instructions. To verify equal protein loading of lanes to allow densitometric analysis blots were stripped for 30 min in a 50°C water bath with agitation in a stripping buffer (62.5 mM Tris–HCl, pH 6.7, 100 mM β-mercaptoethanol and 2% SDS). Completely stripped blots were blocked in 5% nonfat milk in TBST for 1 h at room temperature, then incubated with histone-H3 antibody (diluted 1:1000, Chemicon) overnight at 4°C with agitation and processed further as described above. Band densities were assessed using NIH imaging software, Image J. Results AURK isoforms in mouse oocytes To determine specific AURK isoforms present in mouse oocytes, isoform-specific primers were designed for AurkA, AurkB and AurkC. Transcripts were amplified utilizing real-time PCR for all three isoforms in meiotically incompetent GV-intact oocytes. Significant increases in transcript levels for all three Aurk isoforms were identified in meiotically competent GV-intact and meiotically maturing oocytes, compared with incompetent oocytes (Fig. 1). No differences in transcript levels were identified for any isoform between meiotically competent GV-intact, GVBD, MI or MII oocytes. Additionally, AurkA appeared to be the predominant isoform transcript, displaying an approximate 12-fold increase compared with AurkB and AurkC, which displayed comparable levels, in meiotically competent GV-intact oocytes (data not shown). Single bands were present on agarose gels for each isoform, corresponding to predicted sizes (AurkA—90 bp, AurkB—100 bp, AurkC—145 bp). Sequencing verified that amplified gene products shared 100% homology with mouse Aurk isoforms (data not shown). Figure 1: Graphical representation demonstrating fold-increases of aurora kinase (Aurk) isoform levels between various oocyte types obtained from real-time PCR. Day 11 germinal vesicle intact (GVI) oocytes were used at controls and normalized to 1. (A) AurkA, (B) AurkB and (C) AurkC. Presence of an asterick represents statistical significance compared with groups without an asterick, P < 0.05. AURKs and oocyte meiotic progression To determine the influence of AURK on oocyte meiotic progression, meiotically competent GV-intact oocytes (prophase I) were matured in varying doses of the AURK inhibitor ZM and development was checked at 2 and 16 h. Though a significant reduction was observed with 2.5 μM, Inhibition of AURK had no effect on oocyte GVBD at 2 h at any other dose examined compared with control treatments (Table I). Furthermore, no differences were visually apparent at the light microscope level between ZM-treated and control oocytes at a time point consistent with MI development (7 h; Fig. 2, inset). However, concentrations of ZM at 2.5, 5, 10 and 20 µM prevented all oocytes from progressing to MII at 16 h (Fig. 3). Concentrations of 1.25 and 0.675 µM allowed a small portion of oocytes to extrude the first polar body at 16 h, which was significantly less than the percentage of MII oocytes obtained from control treatments, P < 0.01 (Table I). Additionally, culturing GV-intact oocytes for 7 h to MI in presence of 5 or 10 µM ZM, followed by thorough washing and 9 h of culture in the absence of the AURK inhibitor, indicated that, although ZM may be washed out from blocking the ATP binding pockets, defects caused by ZM treatment were not reversible, as oocytes were unable to complete meiosis and extrude the first polar body (data not shown). Based on these data, a dose of 10 µM ZM was selected for future experiments to ensure all AURK isoforms were inhibited. This dose has been used in other studies on mammalian oocytes (Jelinkova and Kulbeka, 2006), and is lower than the 20 µM used in other studies (Gadea and Ruderman, 2005). Figure 2: Representative micrographs demonstrating effect of aurora kinase (AURK) inhibition on oocyte meiotic spindle formation. Inhibition of mouse oocyte AURKs with 10 µM ZM447439 (ZM) for 7 h had no observable effect on oocyte development at the light microscope level (inset images). However, immunocytochemical examination demonstrated that morphology of the meiotic spindle of ZM-treated oocytes appeared abnormal, with polymerized microtubules (β-tubulin: green), microtubule organizing centers MTOCS (pericentrin: red/orange) and chromatin (blue) was scattered around the metaphase plate, compared with vehicle treated controls. Furthermore, 16 h treatment of oocytes with ZM resulted in arrest of oocytes at a MI-like stage (inset). Immunocytochemical examination demonstrated aberrant polymerized microtubules, MTOCs and condensed chromatin (blue) and MTOCs (red/orange). Polar body of control MII oocytes is indicated by PB. Figure 3: Representative micrographs of oocyte chromosome spreads demonstrating negative impact of AURK inhibition on oocyte chromatin remodeling. Treatment of oocytes with ZM447439 (ZM-10 µM) for 7 h resulted in metaphase I oocytes (MI) with significantly greater defects in bivalent formation, compared with controls (P < 0.0001). Table I. Development of mouse oocytes following treatment in varying doses of Aurora kinase inhibitor ZM447439 (ZM). ZM concentration (µM) 2 h GVBD 16 h MII 0 115/121 (95%)a 38/52 (73%)a 0.625 29/30 (97%) 8/30 (27%)b 1.25 30/35 (86%) 8/26 (31%)b 2.5 84/99 (85%)b 0/54 (0%)c 5 59/66 (89%) 0/28 (0%)c 10 55/64 (86%) 0/28 (0%)c 20 55/64 (85%) 0/28 (0%)c Significant differences in development between treatments within a time point are indicated by different superscripts, P < 0.01. AURKs and oocyte spindle morphology and chromatin remodeling Prophase–MI transition To begin to determine a temporal window when AURK inhibition may be conveying observed phenotypes, experiments examined the effects of AURK inhibition during the prophase I–MI transition. Aberrant spindle morphology and improper positioning of chromatin were observed in oocytes cultured for 7 h in presence of 10 µM ZM (19% normal, n = 32) compared with controls (91% normal, n = 35; P < 0.0001; Fig. 2). Similar patterns were also obtained from treatments containing 2.5, 5 and 20 µM ZM (data not shown). To examine effects of AURK inhibition on oocyte chromosome condensation in greater detail, chromosomal spreads of MI oocytes were examined following 7 h of ZM (10 µM) treatment from prophase I. Inhibition of AURKs resulted in oocytes with a significant reduction in normal chromosome condensation (0%, n = 21) compared with control treatments (95%, n = 22), as evidenced by the inability to resolve bivalents, P < 0.001 (Fig. 3). Prophase I–MII transition Culture of oocytes for 16 h in 10 µM ZM during the prophase I–MII transition indicated that arrest of oocytes was not due to inability of microtubule polymerization as β-tubulin staining indicated microtubules polymerized around chromatin (Fig. 2). Additionally, pericentrin staining indicated apparent microtubule organizing centers (MTOC) assembly. However, spindle formation and MTOC localization were disrupted compared with untreated controls. Furthermore, chromatin was scattered throughout the meiotic spindle, with the pattern of normal chromatin remodeling significantly reduced following ZM treatment (0%, n = 44) compared with controls (81%, n = 32), P < 0.001. MI–MII transition To determine effect of AURK inhibition following normal chromatin remodeling and spindle formation, oocytes were cultured for 7 or 9 h in the absence of ZM (cells typically at MI or AI, respectively). Subsequently, these oocytes were cultured to a time point consistent with MII (an additional 9 or 7 h) in presence or absence of ZM. Treatment of oocytes for 9 h with ZM resulted in significantly less cells progressing to MII (53%, n = 117) compared with 80% of control oocytes (n = 121), P < 0.05. Treatments of oocytes for 7 h with ZM also resulted in significantly fewer cells progressing to MII (58%, n = 88), compared with controls (73%, n = 78; P < 0.05). All MII oocytes obtained following ZM treatment displayed scattered chromatin around the metaphase plate and slightly irregular shaped spindles (Fig. 4). Figure 4: Representative micrographs of oocytes cultured to a time point consistent with MII in presence or absence of 10 µM AURK inhibitor ZM447439 during the MI-MII transition. Control MI oocytes developed normally to MII, extruding the first polar body and displaying condensed chromatin on the metaphase plate (blue) within the meiotic spindle with normally condensed β-tubulin (red) (A). However, following AURK inhibition, a portion of oocytes arrested prior to MII, displaying scattered chromatin within a malformed meiotic spindle (B). Those oocytes that did complete MII under AURK inhibition displayed disorganized chromatin within the meiotic spindle (C). Similar patterns of chromatin disorganization and inability to complete cytokinesis and progress to MII were also observed following culture of oocytes to MII (7 h) following AURK inhibition from a time point where cells are beginning to enter anaphase (E and F) compared with controls (D). AURKs and oocyte histone phosphorylation Because of reported roles for histone-H3 phosphorylation in chromatin remodeling, phosphorylation of histone-H3 following AURK inhibition was examined as a possible cause of aberrant condensation during the prophase I to MI transition. Western blot analysis and ICC utilizing phospho-ser10 histone-H3 antibody was performed. Inhibition of AURK for 2 or 7 h with 10 µM ZM resulted in a total lack of histone-H3 ser10 phosphorylation (Fig. 5A and B). Treatment of oocytes with ZM also completely inhibited phosphorylation of histone-H3 at ser28, as evidenced by ICC and western blot (Fig. 6A and B). These data raise the possibility that a ZM-sensitive AURK is an oocyte histone-H3 kinase and that histone-H3 phosphorylation may be influencing normality of oocyte metaphase chromatin condensation and subsequent separation and segregation of homologues. Figure 5: Representative micrographs and western blot demonstrating inhibition of oocyte aurora kinases (AURK) results in ser10-histone-H3 hypophosphorylation. (A) Culture of GV-intact oocytes for 2 h to allow germinal vesicle breakdown(GVBD) in presence of 10 µM ZM447439 resulted in a total lack of ser10-Histone-H3 phosphorylation (d), compared with controls (a). A similar reduction in ser10 phosphorylation was observed following 7 h (MI) of culture in the presence of ZM (j) compared with controls (g). Chromatin was stained with Hoescht and is pictured in blue (b,e,h,k). Overlays are also indicated (c,f,i,l). (B) Western blot analysis confirmed that ZM treatment inhibits ser10 phosphorylation (n = 100 oocytes/lane). Figure 6: Representative micrographs and western blot demonstrating inhibition of oocyte aurora kinases (AURK) results in ser28-histone-H3 hypophosphorylation. Chromatin is stained in blue, while phospho-ser28-histone-H3 is stained in red. (A) Culture of GV-intact oocytes for 2 h to allow germinal vesicle breakdown(GVBD) in presence of 10 µM ZM447439 (ZM) resulted in a total lack of ser28 phosphorylation (b), compared with controls (a). A similar reduction in ser28 phosphorylation was observed following 7 h (MI) of culture in the presence of ZM (d) compared with controls (c). (B) Western blot analysis confirmed that ZM treatment inhibits ser28 phosphorylation (n = 100 oocytes/lane). Discussion Maintaining integrity of chromatin remodeling is especially important in the oocyte considering its extreme susceptibility to aneuploidy, primarily during the first meiosis (Hassold and Hunt, 2001). AURKs are a family of serine/threonine kinases that regulate various structural elements and mechanistic events associated with the dynamic process of chromatin remodeling. Therefore, examination of AURKs during oocyte maturation is of interest when attempting to discern causative factors and molecular signaling pathways involved in aberrant oocyte chromosome modifications. We have determined that mouse oocytes contain transcripts for all three Aurk isoforms: AurkA, AurkB and AurkC and that levels of these transcripts increase significantly as oocytes gain meiotic competence, but do not change as oocytes progress through meiosis to MII. The predominant Aurk isoform transcript in meitocally competent and maturing oocytes appears to be AurkA. In agreement with these findings, during preparation of this manuscript, transcripts for AurkA, AurkB and AurkC were also reported in fully-grown immature bovine oocytes, with AurkA as the predominant isoform (Uzbekova et al., 2007). To determine effects of AURK on oocyte maturation, we utilized the highly selective pharmacological AURK inhibitor, ZM. Inhibition of AURKs via ZM occurs through blockage of the ATP binding site at an adjacent cleft not present in other kinases (Ditchfield et al., 2003). Although ZM is a selective inhibitor of AURKs, it does inhibit other kinases, including CDK1, MAPK and CDC25. However, concentrations of ZM much higher than those utilized in the majority of our studies (20 µM) had no effect on CDK1, CDC25 or MAPK activities in Xenopus egg extracts, indicating ZM did not directly affect these kinases, or affect any upstream regulatory kinases involved in their activation (Gadea and Ruderman, 2005). Furthermore, cellular characteristics and phosphorylation patterns following inhibition of CDK1 (Marchal et al., 2001; Kubelka et al., 2002; Swain et al., 2003; Bui et al., 2004) and MAPK (Tong et al., 2003; Yu et al., 2007) in mammalian oocytes are dramatically different than those observed following ZM treatment in this study; suggesting observed effects are indeed the result of AURK inhibition. It should be mentioned that ZM displays differential inhibitory action toward different AURK isoforms, demonstrating inhibition of AURKB ∼20 times more potently than AURKA (Girdler et al., 2006). Thus, differential phenotypes observed in our study using lower concentrations of ZM may indicate AURKB-specific functions. Future experiments will attempt to determine if lower doses of ZM, as well as utilization of other AURK inhibitors with differential inhibitory actions, can verify this and determine isoform-specific functions within the mammalian oocyte. AURK do not appear to play a role in oocyte meiotic resumption or regulation of oocyte nuclear envelope (NE) integrity during GVBD. Although ZM appeared to cause a slight delay in NE disassembly, no significant differences were apparent at any dose of ZM examined other than 2.5 μM. This is in agreement with findings that inhibition of bovine oocyte AURKs with VX680 has no effect on GVBD (Uzbekova et al., 2007). However, these findings are in contradiction to a study utilizing pig oocytes, which demonstrated lack of GVBD following exposure to elevated doses of ZM (7–10 µM; Jelinkova and Kubelka, 2006). Difference may be the result of varying experimental conditions or species-specific differences, as Jelinkova and Kulbelka (2006) utilized porcine cumulus–oocyte complexes from abattoir ovaries, whereas the current study utilized denuded oocytes obtained from gonadotropin stimulated mice. In our study, inhibition of AURKs during 16 h of culture inhibited progression to MII and polar body extrusion, similar to developmental observations in clam oocytes following treatment with another AURK inhibitor, Hesperadin (George et al., 2006), and bovine oocytes treated with VX680 (Uzbekova et al., 2007). Failure to complete cytokinesis has also been observed following AURK inhibition during mitosis (Ditchfield et al., 2003). Evidence exists suggesting AURKB may be the specific isoform responsible for failure of oocytes to extrude the first polar body, as specific inhibition of AURKB in Drosophila cells prevented cytokinesis (Adams et al., 2001; Giet and Glover, 2001). Interestingly, AURKB was localized to the region of the contractile ring in bovine MII oocytes; (Uzbekova et al., 2007). However, AURKA and AURKC were also found in the vicinity of the contractile ring in bovine oocytes, confounding interpretation. Whether failure to complete cytokinesis and extrude the polar body in our study is a primary effect of AURK inhibition on cytokinesis, or a secondary effect due to chromosome remodeling or spindle defects remains to be elucidated. To begin to determine possible causes for oocyte arrest prior to MII, we examined effects of AURK inhibition on chromatin remodeling and spindle formation during the prophase I–MI transition. Inhibition of AURKs during this time point resulted in improper positioning of chromatin, as evidenced by scattering throughout the meiotic spindle. This is in agreement with recent studies of AURK inhibition in mouse (Wang et al., 2006), pig (Jelinkova and Kubelka, 2006) and bovine oocytes (Uzbekova et al., 2007) reporting abnormal chromosome positioning at MI. Additionally, chromosomal spreading in our study indicates AURK inhibition negatively affects ability of MI oocyte chromosomes to condense properly and resolve bivalents. This may be due to premature decondensation, similar to that observed following AURK inhibition in Xenopus egg extracts (Gadea and Ruderman, 2005). It has been reported that AURKB inhibition via RNAi is responsible for chromosome misalignment in Drosophila cultured cells and results in amorphous chromatin (Adams et al., 2001) and only partial condensation (Giet and Glover, 2001). Thus, AURKB may be the AURK isoform responsible for chromatin defects observed in mammalian oocytes in this study. Indeed, a recent report demonstrates AURKB localized to condensed chromatin in MI and MII bovine oocytes (Uzbekova et al., 2007). Failure of ZM-treated oocytes to progress to MII does not appear to be the result of inability to polymerize microtubules or form MTOCs, as indicated by β-tubulin and pericentrin staining. This is in contrast to AURK inhibition studies in Xenopus egg extracts, where ZM treatment resulted in failure to form the mitotic spindle (Gadea and Ruderman, 2005). However, fidelity of spindle function in our study remains in question, as spindle morphology and MTOC localization was disrupted following AURK inhibition. AURKs, such as AURKA, regulate several components of the spindle apparatus and spindle poles (see review, Ducat and Zheng, 2004). Indeed, AURKA localized to spindle poles in mouse (Yao et al., 2004) and pig oocytes (Yao and Sun, 2005), but not bovine oocytes (Uzbekova et al., 2007) and specific neutralization of the kinase resulted in disorganization of the meiotic spindle. This disorganization is in agreement with initial reports of ZM influences on intact somatic cells (Ditchfield et al., 2003). Thus ZM inhibition of oocyte AURKA may account for observed defective spindle phenotypes. Alternatively, condensed chromosomes direct formation of the spindle apparatus via nucleation/stabilization of microtubules (Merdes and Cleveland, 1997; Khodjakov et al., 2000). Therefore, defects in meiotic spindle morphology and scattering of chromatin observed in these experiments may be the result of aberrant chromatin condensation, possibly controlled by AURKB (see review, Shannon and Salmon, 2002). Reports in Xenopus mitotic cell-free extracts indicate AURK inhibition does interfere with chromatin driven microtubules assembly (Gadea and Ruderman, 2005). To begin to determine if the inability of ZM-treated oocytes to reach MII following AURK inhibition was due only to defects incurred during the prophase–MI transition, or if AURKs had roles at other meiotic transition time points, we matured oocytes in vitro to time points consistent with MI and AI oocytes, thus allowing normal spindle formation and chromatin remodeling to occur. Subsequently, we then cultured oocytes in presence of ZM to a time point consistent with MII. Inhibition of AURKs during the MI-MII transition resulted in a portion of oocytes failing to segregate chromosomes and extrude the first polar body, while those that did complete MII displayed severely scattered chromatin. Thus, it appears as if AURKs may not only control spindle formation and chromatin condensation during early meiotic events, but also regulate separation and/or segregation of oocyte chromosomes during later meiosis. Effects may be directly on meiotic spindle components, possibly regulated by AURKA. Alternatively, defective chromosome remodeling could also be due to ZM inhibition of AURKB. Aurora B regulates kinetochores and their interactions with microtubules (Kaitna et al., 2002; Cimini et al., 2006). Another possible explanation for aberrant separation/segregation following AURK inhibition during late oocyte meiosis may be interferences with regulation of cohesion. Aurora B regulates release of chromosome cohesion during meiosis in Caenorhabditis elegans, apparently via phosphorylation of REC-8 (Rogers et al., 2002). Future studies will attempt to determine if defects include premature separation of sister chromatids due to premature release of cohesion, or if aberrant phenotypes are the result of failure to separate homologous chromosomes. To begin to determine possible targets of oocyte AURKs responsible for observed defects in chromosomal remodeling, we examined the phoshorylation state of histone-H3 at ser10 and ser28. It is thought that phosphorylation of histone-H3 may cause the histone to act as a receptor or recruitment factor for condensation factors (Hirano, 2000), or possibly reduce the affinity of histone-H3 for DNA and make the relatively compact chromatin fiber more readily accessible to remodeling factors (Hirano, 2000), such as the condensin complex. Condensin is a multi-subunit protein complex that play a central role in chromosome compaction and condensation and is reported to co-localize and bind with phosphorylated histone-H3 (Schmiesing et al., 2000; Ball et al., 2002). Interestingly, AURKB controls association of condensin with chromosomes during mitosis (Lipp et al., 2007; Takemoto et al., 2007), and may be functioning in a similar manner during mammalian oocyte meiosis, though this remains unknown. It should be mentioned that reports for the requirements of histone-H3 phosphorylation in chromatin condensation varies greatly (Van Hooser et al., 1998; Wei et al., 1998, 1999; Goto et al., 1999; Kaszas and Cande, 2000; Schmitt et al., 2002). These contradictions also appear to hold true in regard to oocyte meiosis, with differing reports on temporal and spatial localization and correlation with condensation (Jelinkova and Kubelka, 2006; Wang et al., 2006; Swain et al., 2007). However, contradictions may be explained by differences in experimental conditions or perhaps species-specific differences. Regardless, contradictions in the literature make it increasingly evident that differences in the role of histone phosphorylation exist depending on the organism and/or type of cellular division examined (Fuchs et al., 2006), and that the specific histone-H3 AURK within mammalian oocytes remains unknown. In summary, these studies indicate AURK plays a significant role in mouse oocyte maturation involving progression to MII, acting in both early (prophase–MI transition) and late meiotic events (MI–MII transition). When AURK is inhibited in early meiosis, microtubules polymerize and MTOCs form, but spindle morphology and MTOC localization is disrupted. Furthermore, metaphase chromatin does not condense or position normally. Inhibition of oocyte AURKs during late meiosis (MI–MII transition), following chromatin condensation and spindle formation, negatively affects separation and segregation of chromosomes. This aberrant chromatin remodeling following AURK inhibition in oocytes appears to be due, in part, to hypophosphorylation of histone-H3 at both ser10 and ser28. We report amplification and sequencing of all three Aurk isoforms transcripts: AurkA, AurkB and AurkC. Levels of these transcripts increase as oocytes achieve meiotic competence, with AurkA being the predominant isoform. Future studies will focus on determining protein expression levels of AURK isoforms, as well as specific functional roles and intracellular targets. Funding Funding for research by Gary D. Smith provided by NIH RO1 grant #HD046768-01A2. Support for Jason E. Swain provided by an NIH T32 grant. Partial support for Jingwen Wu provided from the Lyle C. Roll Research Fund to Professor James O. Woolliscroft, Dean of Medical School, University of Michigan.

Eur_Spine_J-2-2-1602205

Epithelioid sarcoma in the thoracic spine

Epithelioid sarcoma is a rare and highly malignant soft tissue tumor that is commonly found in the extremities and rarely in the trunk area. This malignant tumor often mimics granuloma or nodular fasciitis, which causes a delay in establishing the diagnosis. This type of cancer has a high recurrence rate. Surgical treatment requires wide radical resection. The objective of this case report is to highlight the unique location of a rare neoplasm and to illustrate the relentless course of epithelioid sarcoma despite initial radical resection. A 14-year-old boy was admitted to our facility with a soft tissue mass on the right lower thoracic spine. The large tumor mass had deeply penetrated into the muscles, infiltrated the neuroforamen of T9–T10 level, and compressed the dural sac. Immunohistological study of the biopsy was highly consistent with an epithelioid sarcoma. Wide excision of the mass, laminectomy and spine fusion with instrumentation was performed. The patient received chemotherapy and irradiation. The first recurrence of the neoplasm was seen as a contralateral metastasis 21 months after the resection. On the last follow-up, 3 years postoperatively, the patient was in a good general condition. However, further progression of the sarcoma had to be recognized. Our case encompasses multiple features that represent negative prognostic factors. Initial wide excision of the neoplasm and adjuvant therapy including chemotherapy and irradiation seem to slow down the relentless course of epithelioid sarcoma in the trunk. Introduction Epithelioid sarcoma is a rare type of a fibrohistiocytic tumor, which is not extensively documented in the literature. It is common at the extremities, and rare at the trunk area. The incidence of epithelioid sarcoma in the trunk is 4% with a range from 0 to 44% [2, 12, 14]. It affects mainly adolescent males [1–3, 11, 13, 14, 17]. To our knowledge there are only two cases reported of localization in the spine (lumbosacral junction and sacrum) [7, 14], and there are no reports on epithelioid sarcoma at the thoracic spinal level. Laskowski [9] has first described the unique histological appearance in 1961. It resembles a chronic inflammatory process, necrotizing granuloma, squamous cell carcinoma or other fibrohistiocytic tumors. Electron microscopic appearance suggests origin from synovial structures [6, 10]. There is 85% local recurrence rate and 30–50% likelihood for metastases predominately lymphatic or in the lungs [2, 4, 11]. The 5-year survival rate varies from 65 to 100% [1, 12, 13]. Factors adversely affecting the survival rate are [2, 12]:Size of the tumor;Infiltration of the tumor into muscular, neural or vascular structures;Primary location in the trunk;Pulmonary metastasis.Wide surgical resection is indicated, since epithelioid sarcoma frequently follows a relentless course with multiple recurrences [4, 17]. Chemotherapy and irradiation are recommended to decrease recurrence rate [2, 8]. Case report A 14-year-old overweight boy presented at an orthopaedist’s private practice with a painless, bulging mass on his back that was incidentally discovered by his grandmother. He was 183 cm tall and weighted 107 kg. Besides obesity no other medical problems were reported. There was no family history of a soft tissue mass. On the physical examination the tumor had a soft consistence and could not be shifted from the underlying tissue. Neurological examination did not reveal any deficit. Magnetic resonance imaging (MRI) with gadolinium of his thoracolumbar spine revealed a right paravertebral cystic mass. The soft tissue tumor extended from T7 to T11 level and measured 9.5×3×3 cm. The mass infiltrated via the neuroforamen at T9–T10 levels compressing one quarter of the spinal cord (Fig. 1). There was no osseous involvement. Open biopsy of the mass showed necrobiotic granuloma including epithelioid and fusiform cells. Immunohistochemical studies revealed a co-expression of vimentin and cytokeratin, with pathological diagnosis of epithelioid sarcoma grade III. During the tumor staging a craniolateral metastasis was depicted on the PET scan (Fig. 2). Consecutive MRI confirmed a soft tissue mass of 1 cm diameter, located about 10 cm distant from the main mass. Wide mass resection, including the locoregional metastasis, led to a sacrifice of large parts of the paravertebral thoracic muscles. In addition a hemilaminectomy of T9–T10 vertebra with resection of the adjacent pedicles and careful dissection from the dural sac was performed. There was no sparing of right ninth nerve root. The tumor resection was intralesional. Posterior transpedicular instrumentation from T6 to L1 was carried out. After surgery no new motor or sensory deficits were noted. Chemotherapy was initiated on the 6th postoperative week and irradiation on 10th postoperative week. The patient received nine courses of chemotherapy within 25 weeks (Table 1) and a total dose of 3-Gy irradiation over 4 weeks. He developed a wound dehiscence 2 weeks after initiation of the irradiation therapy. Under daily dressing no infection was observed. The wound was finally closed with a pedicle musculocutaneous flap using the Latissimus dorsi muscle 4 weeks after the completing the course of irradiation. Routine follow-up consisted of a physical examination, plain radiograms of the thoracic spine plus MRI Scans every 3 month (Fig. 3). Fig. 1Preoperative antero-posterior (a) and lateral (b) plain X-rays of the middle thoracic spine. On the ap view no irregular formation of the pedicles can be depicted—the lateral view of the thoracic spine show apophyseal endplate changes as a feature of Scheuermans disease. No signs of osteolytic lesion can be detected. Preoperative transverse (c) and sagittal (d) MRI scan (T2 sequences: TR=3,070, TE=120) demonstrating the tumor mass in the region of the paravertebral muscles, reaching and compressing the dural sac via the right neuroforamen Th9/10Fig. 2On the PET scan a craniolateral metastasis was detected (a) and the paravertebral tumor mass was confirmed (b). No further satellite lesions were foundTable 1Agents and dosage of the quadruple chemotherapyAgent (trade name)Total dose (mg)Ifosfamid (Holoxan®)100,440Vincristin (Onkocristin®)20Actinomycin D (Lyovac®)12Doxorubicin (Ribodoxo®)494Fig. 3Postoperative antero-posterior (a) and lateral (b) plain radiograph of the thoracic spine at final follow-up. Postoperative transverse (c) and sagittal (d) MRI scan (T2 sequences: TR=3,682, TE=120) at 1.5 year showing no sign of recurrence A metastasis at the fifth intercostal space of the chest contralateral to the initial tumor location was observed at 21 months follow-up. At this stage the patient denied further intervention. Two years postoperatively pulmonary metastases were detected on CT scans in addition to a local recurrence on the right thoracic spine. There was no significant progression of the tumor growth at the last follow-up 37 month after surgical treatment. The young male was in good general condition fully participating in social life. Discussion Epithelioid sarcoma is a rare soft tissue neoplasm with a high local recurrence rate. Epithelioid sarcoma presents histologic characteristics similar to inflammatory processes and other benign soft tissue tumors, which frequently delays the diagnosis and adequate treatment. Typical for this soft tissue sarcoma is an immunohistochemical co-expression of cytokeratin and/or epithelial membrane antigen (EMA) and vimentin, and in half of the cases an immunoreactivity for CD34, an occasional reactivity for smooth muscle actin (SMA), desmin, neuron specific enolase and S100 protein [2, 5]. In our specimen a strong positive immunoreaction with antibodies to cytokeratin, vimentin and EMA was registered (Fig. 4). Histomorphology and immunohistochemical pattern are typical and separates this tumor entity from other sarcomas. To differentiate from osteogenic sarcoma, which may also co-express cytokeratin and vimentin, and may demonstrate a immunoreactivity with SMA and CD99, epithelioid sarcoma, however, does not produce tumor osteoid, which is the key finding in osteogenic sarcoma. Fig. 4a Histomorphologic aspect of epithelioid sarcoma with solid and nodular growth pattern, H&E (10×). b The tumor cells are roundish, admixed with spindle shaped cells and are featuring an eosinophilic cytoplasm with vesicular, irregular nuclei (20×). c Immunohistochemical investigation with antibodies against cytokeratin shows a strong positive immunoreaction (20×). d The immunoreaction against epithelial membrane antigen (EMA) demonstrates a comparable picture (20×). e The immunoreactivity against smooth muscle actin (SMA) features a weak staining (20×). f The immunoreaction against CD 99 is negative (20×) Wide local resection is the recommended treatment of epithelioid sarcoma [1, 8, 11, 14]. In the spine however, total eradication of the sarcoma by an enbloc excision is compromised by the presence of neural structures [16]. Thus the concept of intralesional resection was followed both for the main tumor mass and the metastasis. Nerve sheaths may serve as a pathway for the spread of the sarcoma [1, 2, 8, 11, 17]. Enzinger [4] reported six cases with a mass infiltration along nerve sheaths or vascular structures. In our case there was involvement of the right ninth nerve root. The mass expansion into the right T9–T10 neuroforamen with compression of the dural sac necessitated a wide hemilaminectomy with resection of the facet joints and the adjacent pedicles. The wide paraspinal muscle resection required six levels segmental instrumentation and fusion to avoid scoliotic deformity due to muscular imbalance [15]. Pulmonary metastasis and local recurrence of the sarcoma were first diagnosed 24 month after the resection. In comparison, a survival period of 22 month after the initial surgery was described in a case report of epithelioid sarcoma in the lumbosacral spine [14]. Moreover, on first diagnosis, our case embodied many factors that promote recurrence of this neoplasm: large size, localization in the trunk and infiltration in both muscle and neural structures. Recurrence as the typical feature of epithelioid sarcoma cannot be prevented even in extremities, where a radical surgical regimen can be pursued [8]. In conclusion, epithelioid sarcoma of the spine is an extremely rare entity, which differs from other sarcoma subtypes in propensity for local recurrence and spread along the lymphatic pathways. Initial radical resection followed by adjuvant chemotherapy and irradiation can retard the relentless course of this malignancy.

Eur_Radiol-3-1-2077913

Dual-source computed tomography in patients with acute chest pain: feasibility and image quality

The aim of this study was to determine the feasibility and image quality of dual-source computed tomography angiography (DSCTA) in patients with acute chest pain for the assessment of the lung, thoracic aorta, and for pulmonary and coronary arteries. Sixty consecutive patients (32 female, 28 male, mean age 58.1±16.3 years) with acute chest pain underwent contrast-enhanced electrocardiography-gated DSCTA without prior beta-blocker administration. Vessel attenuation of different thoracic vascular territories was measured, and image quality was semi-quantitatively analyzed by two independent readers. Image quality of the thoracic aorta was diagnostic in all 60 patients, image quality of pulmonary arteries was diagnostic in 59, and image quality of coronary arteries was diagnostic in 58 patients. Pairwise intraindividual comparisons of attenuation values were small and ranged between 1±6 HU comparing right and left coronary artery and 56±9 HU comparing the pulmonary trunk and left ventricle. Mean attenuation was 291±65 HU in the ascending aorta, 334±93 HU in the pulmonary trunk, and 285±66 HU and 268±67 HU in the right and left coronary artery, respectively. DSCTA is feasible and provides diagnostic image quality of the thoracic aorta, pulmonary and coronary arteries in patients with acute chest pain. Introduction Acute chest pain represents one of the most difficult diagnostic challenges in emergency medicine. Chest pain history alone often cannot identify a group of patients who could be treated without further diagnostic testing [1], and triage decisions based on initial cardiac enzyme levels [2] and electrocardiography (ECG) [3] are often insufficient. Imaging may improve patient triage by decreasing delay in diagnosis and treatment and thus morbidity and mortality [4, 5]. Most studies [6–8] have focused on patients being suspected of having acute coronary syndrome, but did not include additional differential diagnoses such as aortic dissection or pulmonary embolism, conditions that may clinically mimic coronary syndromes [9]. Multi-detector row computed tomography (CT) angiography is widely accepted and routinely used as a primary tool in the emergency assessment of pulmonary embolism [10] and aortic dissection [11]. In addition, 64-slice CT coronary angiography has demonstrated the capability to diagnose and to rule-out coronary artery disease (CAD) [12, 13]. Some recent studies have shown that ECG-gated multi-detector row CT is logistically feasible [14] and yields promising results as a modality for evaluating chest pain patients with cardiac and non-cardiac disease in the emergency setting [15, 16]. Even though using 64-slice CT scanner technology, however, patients with elevated heart rates require medical heart rate control prior to CT coronary angiography by administering beta-blockers and/or benzodiazepines [8, 13, 17]. Dual-source CT (DSCT) represents the most recent scanner technology and is characterized by two tubes and two detectors that are mounted in orthogonal orientation onto the gantry. As compared to 64-slice CT, this scanner configuration maintains a consistently high spatial resolution while enabling ECG-gated imaging with an increased temporal resolution of 83 ms [18]. First studies have shown robust results of DSCT angiography (DSCTA) regarding image quality of cardiac structures even at high heart rates [19–21]. The purpose of this study was to investigate the feasibility and image quality of ECG-gated DSCTA of the chest as a tool to evaluate cardiac and non-cardiac causes of acute chest pain in patients presenting to the emergency department. Materials and methods Patient population Sixty consecutive patients (32 females, 28 males, mean age 58.1±16.3 years, age range 26–84 years) were prospectively included in this study. Intake was performed on weekdays from 7 am to 7 pm from August to October 2006. All patients suffered from acute chest pain and were referred to our department to diagnose or to rule out pulmonary embolism (n=56) or aortic dissection (n=4). Inclusion criteria were acute chest pain >5 min within the previous 24 h and/or elevated serum D-dimer levels. Dyspnea and hemodynamic instability were not considered exclusion criteria. Similarly, all patients irrespective of their mean or regularity of heart rate and irrespective of their ability to perform breath-hold were included. Exclusion criteria included pregnancy, previous adverse reaction to iodinated contrast agent, nephropathy (serum creatinine >1.3 mg/dl), elevated troponine-I or creatine kinase-MB level in the initial blood sample, initial diagnostic ECG changes indicating an acute coronary syndrome (i.e., ST elevation or depression >1 mm, T-wave inversion >4 mm in >2 anatomically contagious leads), and interference with standard clinical care of patients. The study was approved by the local ethics committee; informed consent was obtained. Scan protocol and data reconstruction All CT examinations were performed on a DSCT scanner (Somatom Definition, Siemens Medical Solutions, Forchheim, Germany). First, a single non-enhanced low-dose scan at the level of the aortic root was obtained. In this slice, a region of interest (ROI) was set in the lumen of the aorta for monitoring intraluminal contrast enhancement. The delay from start of contrast material injection to start of scanning was planned using the bolus-tracking technique. A total of 110 ml iodinated contrast material (iodixanol, Visipaque 320; 320 mg/ml, GE Healthcare, Buckinghamshire, UK) was administered at a flow rate of 4 ml/s via an 18-gauge needle placed into a superficial vein in the left antecubital fossa, followed by 30 ml saline solution at the same flow rate (4 ml/s). After reaching the preset contrast enhancement level of 80 Hounsfield Units (HU) in the ROI, a breath-hold signal was given and the scan was initiated automatically after a delay of 6 s. A topogram was used for planning the examination and determining the scan range. Data acquisition was performed using a dedicated biphasic chest pain protocol. As illustrated in Fig. 1, the lower chest including the heart was scanned with a tube current time of 320 mAs while the upper chest was scanned with a tube current time product of 160 mAs. The borderline between upper and lower scan range was set at approximately 2 cm below the carina. Data acquisition was performed in a cranio-caudal direction. Acquisition parameters were as follows: detector collimation 2×32×0.6 mm by using a z-flying focal spot for the simultaneous acquisition of 2×64 overlapping 0.6-mm slices, gantry rotation time 330 ms, and tube potential 120 kV. The pitch varied according to the patient's heart rate and ranged from 0.2–0.43, with higher pitch at higher heart rates. ECG-pulsing for radiation dose reduction [22] was applied in all patients. At mean heart rates below 60 bpm, full tube current was applied from 60 to 70%, at 61–70 bpm from 50 to 80%, and at heart rates above 70 from 30 to 80% of the R-R interval. Fig. 1Scan topogram illustrating planning of the chest pain protocol. The scan range covered the entire chest (red box). Premonitoring for bolus tracking was performed at the level of the aortic root (white line). The border for full tube current for the heart and half tube current for the upper lung is set approximately 2 cm below the tracheal bifurcation. It is delimited by a virtual horizontal line connecting the upper ends of the blue boxes on both sides Retrospective ECG-gating for phase synchronization was used. For the heart, CT data sets were reconstructed at 70% of the R-R interval with a slice thickness of 0.75 mm (increment 0.5 mm) by using a medium soft-tissue convolution kernel (B26f) (mean field of view, FoV: 151±17 mm, image matrix 512×512). If considered necessary, additional images were reconstructed in 5% steps using the same parameters within the time window of full tube current. Images of the mediastinum (mean FoV: 293±43 mm) including the aorta and pulmonary arteries were reconstructed with a slice thickness of 1 mm (increment 0.8 mm) by using a medium soft-tissue convolution kernel (B30f), and images of the lung were reconstructed with a slice thickness of 2 mm (increment 1.5 mm) by using a sharp convolution kernel (B60f, same FoV as for the mediastinum). All images were transferred to a second Wizard (Siemens) equipped with cardiac post-processing software (Syngo Circulation, Siemens). Data analysis All data were qualitatively evaluated regarding image quality and artifacts of different thoracic structures by two independent readers who are both experienced in cardiovascular radiology. This evaluation was performed on transverse source images, multi-planar reformations (MPR), curved MPR, and thin-slab maximum intensity projections. Image quality and artifacts-lung parenchyma Image quality of lung parenchyma was independently rated using a two-point scale, adapted from a previous publication [23]. Lung parenchyma allowing diagnostic assessment due to distinct anatomic details of bronchial and parenchymal structures without significant artifacts and noise was rated with a score of 1 (diagnostic). Lung parenchyma with artifacts or noise causing reduction of image quality and diagnostic value was rated with a score of 2 (non-diagnostic). Artifacts were rated to quantify the cranio-caudal distribution of artifacts within the lung parenchyma for the right and left lung separately in coronal MPR. Because of the different tube current at upper and lower parts along the z-axis (see Fig. 1), the following scores were separately applied for the apex and the basis of the lung, respectively: 1= no artifacts, 2= breathing artifacts (stair step artifacts), 3= ECG-gating (i.e., synchronization or interpolation artifacts), and 4= noise artifacts. If breathing and ECG-gating artifacts appeared at the same time in one patient, the artifact with the worst impact on image quality was noted. Image quality and artifacts-vascular structures Image quality of thoracic vascular structures was independently rated using the same two-point scale [23] as used for the rating of the lung parenchyma. Regarding the thoracic aorta, a score of 1 (diagnostic) indicated confident evaluation of the ascending aorta, the aortic arch, and descending aorta. Regarding pulmonary arteries, a score of 1 indicated confident evaluation of central, lobar, segmental, or subsegmental pulmonary arteries. Regarding coronary arteries, a score of 1 represented confident depiction (homogenous attenuation; no artifacts decreasing coronary analysis) of the right coronary artery (RCA), left main artery (LMA), left anterior descending artery (LAD), left circumflex artery (LCX), and their side branches. A score of 2 indicated decreased image quality of thoracic vascular structures with severe impairment of diagnostic value (non-diagnostic) due to breathing, motion, or ECG-gating artifacts. Image noise and attenuation-vascular structures Image noise was determined as the standard deviation of attenuation in a ROI placed in the ascending aorta [24]. Contrast attenuation was measured in each patient in the ascending aorta, pulmonary trunk, LMA, proximal segment of the RCA, and right and left ventricle. Image noise and contrast attenuation were assessed using a circular ROI positioned exactly within the vessel or ventricular lumen while avoiding superimposition or partial volume effects from the vessel wall or myocardium. Measurements of the ascending aorta and pulmonary trunk as well as of the right and left ventricle were performed on the same transverse image, while the ROI in the coronary arteries was individually placed on separate transverse images. Imaging findings Imaging findings indicating the possible underlying cause of acute chest pain were documented in each patient by both readers in consensus. Statistical analysis Statistical analysis was performed using commercially available software (SPSS 11.5, SPSS Inc., Il). Quantitative variables are expressed as mean ± standard deviation (SD) including 95% confidence intervals (CI) or range when appropriate. Categorical data were expressed as frequencies or percentages. Inter-observer agreement (kappa statistics) for image quality ratings was calculated. Two-tailed Student's t test for paired samples was used to explore significant differences in vessel attenuation among the ascending aorta, RCA, LMA, pulmonary trunk, and right and left ventricle. Intra-individual differences regarding vessel attenuation among the different vascular territories were performed using pairwise comparisons. Bonferroni correction for multiple comparisons was made, and a P-value <0.003 was considered statistically significant. Results All CT scans were well tolerated and were successfully performed in all 60 patients without complications. All patients had a sinus rhythm, and the average heart rate during data acquisition was 74.9±19.0 beats per minute (bpm) (range 45-130 bpm). No beta receptor antagonists or benzodiazepines were administered prior to CT; 13 patients (22%) took oral beta blockers as part of their baseline medication at the time of the scan. Mean scan time was 12.2±2.3 s (range 9.7–17.5 s) and mean scan length was 24.1±2.1 cm (range 19.3–29.7 cm). The reconstruction interval used for image reading was 70% of the R-R interval. In seven patients it was considered necessary to reconstruct additional data sets in 5% intervals within the window of full tube current to obtain images with diagnostic quality. Image quality and artifacts-lung parenchyma Image quality of lung parenchyma was rated as being diagnostic (score 1) by both readers in all 60 patients (100%; excellent inter-observer agreement, kappa =1.0), and pathology both at upper and lower lung parts could be diagnosed or excluded in all 60 patients. Lung parenchyma was rated by both readers as being artifact-free in 42 patients (70%). Breathing artifacts causing stair-step artifacts were encountered in nine patients (15%) by one reader and eight (13%) of these nine patients by the other reader (excellent inter-observer agreement, kappa =0.93). Breathing artifacts were exclusively found at the lung base, i.e., at the end of the breath-hold period. ECG-gating-related artifacts were found in ten patients (17%) by one reader, and in nine (15%) of these ten patients by the other reader (excellent inter-observer agreement, kappa =0.84). The two types of artifacts were not encountered in the same patient, and image quality was diagnostic despite the artifacts. Image quality and artifacts-vascular structures Image quality of the thoracic aorta was rated by both readers as being diagnostic (score 1) in all 60 patients (100%; excellent inter-observer agreement, kappa =1.0), and pathologies of the thoracic aorta could be diagnosed or excluded by both readers in all 60 patients. Both readers rated image quality of the pulmonary arteries in the one (i.e., the same) patient (2%) as being non-diagnostic (score 2) due to insufficient attenuation of segmental and subsegmental pulmonary arteries. In this patient, attenuation did not allow excluding segmental or subsegmental pulmonary embolism, while attenuation in the pulmonary trunk and right and left lobar artery was considered sufficient for diagnosis or exclusion of central and lobar pulmonary embolism. In this patient, image quality of the coronary arteries was considered as being diagnostic by both readers. Image quality of coronary arteries was rated by both readers as being diagnostic (score 1) in the same 58 patients (97%; excellent inter-observer agreement, kappa =1.0). In the remaining two patients (3%), image quality of the coronary arteries was considered by both readers as being non-diagnostic (score 2) due to ECG-gating-related artifacts (mean heart rate 71 bpm, ranging from 50 to 84 bpm during scanning) in one and severe image noise due to obesity (body mass index 40.4 kg/m2) in the other. In both patients, both readers rated image quality of the pulmonary arteries as being diagnostic. Image noise and attenuation-vascular structures Mean image noise in the ascending aorta (mean ROI size 3.6±0.2 cm2) was 29.8±5.6. Mean attenuation in the ascending aorta (same ROI and same ROI size as for measurements of image noise) was 291±65 HU (95% CI: 274–308 HU), mean attenuation in the pulmonary trunk (mean ROI size 2.4±0.2 cm2) was 334±93 HU (95% CI: 310–358 HU), mean attenuation in the proximal RCA (mean ROI size 0.14±0.02 cm2) was 285±66 HU (95% CI: 268–302 HU), and mean attenuation in the LMA (mean ROI size 0.12 cm2±0.01) was 286±67 HU (95% CI: 269–304 HU). Mean attenuation in the right ventricle (mean ROI size 2.6±0.2 cm2) was 313±89 HU (95% CI: 290–336 HU) and mean attenuation in the left ventricle (mean ROI size 2.6±0.2 cm2) was 277±66 HU (95% CI: 260–294 HU) (Fig. 2). Fig. 2Mean attenuation values (HU) within different vessels indicating a relatively homogenous contrast distribution between the different thoracic vascular territories. RCA = right coronary artery; LMA = left main coronary artery No significant differences regarding attenuation were found between the RCA and LMA (P=n.s.), and between the ascending aorta and both coronary arteries (P=n.s.). No significant differences in attenuation were present between the right and left ventricle (P=n.s.), between the right ventricle and ascending aorta (P=n.s.), between the right ventricle and pulmonary trunk (P=n.s.), and between the right ventricle and the RCA (P=n.s.) and LMA (P=n.s.). Similarly, there were no significant differences in attenuation between the left ventricle and ascending aorta (P=n.s.), and between left ventricle and LMA (P=n.s.) and RCA (P=n.s.). Significant differences were only found between the pulmonary trunk and ascending aorta (P <0.003), between the pulmonary trunk and the RCA and LMA (P<0.003), respectively, and between the left ventricle and pulmonary trunk (P<0.003). Pairwise comparisons of mean intra-individual differences regarding vessel attenuation were small and ranged between 1.18±5.86 HU comparing the right and left coronary artery and 56.45±8.96 HU comparing the pulmonary trunk and left ventricle (Table 1). Table 1Pairwise intra-individual comparisons between different thoracic vascular territories Mean differenceStandard deviation95% confidence intervalLower boundaryUpper boundaryAortaPulmonary trunk-4311-75-11AortaRCA65-1022AortaLCA54-817AortaRight ventricle-2313-6318AortaLeft ventricle145-330Pulmonary trunkRCA49111682Pulmonary trunkLCA47121283Pulmonary trunkRight ventricle2012-1657Pulmonary trunkLeft ventricle5792984RCALCA-16-1917RCARight ventricle-2813-6913RCALeft ventricle85-824LCARight ventricle-2714-7015LCALeft ventricle96-1028RCA = right coronary artery; LCA = left coronary artery Imaging findings No pathologic findings were found in 33 patients (55%), whereas various pathologies of the aorta, pulmonary and coronary arteries, lung, and mediastinum were present in 27 patients (45%) (Table 2, Figs. 3, 4, 5). Table 2Imaging findings in the study populationDiagnosisNumber of patients (n=60)No pathology33Pulmonary embolism11Aortic pathology, total5 Dissection type B3 Pseudoaneurysm of the aortic arch1 Plaque rupture1Coronary pathology, total3 Significant stenosis of the LAD2 Occlusion of the RCA1Pulmonary consolidation3Pericarditis/pericardial effusion2Seropneumothorax1Non-small cell lung cancer1Synovial cell carcinoma1LAD, left anterior descending artery; RCA, right coronary arteryFig. 3A 63-year-old female patient admitted to the emergency department with acute chest pain. (a) Curved multiplanar reformations along the centerline of the right coronary (RCA), left anterior descending (LAD), and the left circumflex artery (RCX) allow excluding significant coronary stenosis in this patient. Mean heart rate during DSCTA was 71 bpm. (b) Thin-slab transverse maximum intensity projection shows no evidence of pulmonary embolism. (c) Transverse image at the level of the pulmonary trunk demonstrates acute aortic dissection type B (arrow) with mild left-sided pleural effusionFig. 4A 58-year-old female patient admitted to the emergency department with acute chest pain. (a) Curved multiplanar reformations along the centerline of the right coronary (RCA), left anterior descending (LAD), and the left circumflex artery (LCX) demonstrate normal coronary arteries and no evidence of stenosis. Mean heart rate during DSCTA was 63 bpm. (b) Thin-slab transverse maximum intensity projection show bilateral pulmonary embolism (arrows). (c) Oblique-sagittal thin-slab maximum intensity projection demonstrates the thoracic aorta without evidence of diseaseFig. 5A 71-year-old male patient admitted to the emergency department with acute chest pain. (a) Curved multiplanar reformations along the centerline of the right coronary (RCA), left anterior descending (LAD), and the left circumflex artery (LCX) show occlusion of the proximal RCA (long arrow) and vessel wall calcifications without significant stenosis in the proximal and middle segment of the LAD and LCX (short arrows). Mean heart rate during DSCTA was 73 bpm. (b) Thin-slab transverse maximum intensity projection demonstrates normal opacification of pulmonary arteries with no evidence of embolism. (c) Oblique-sagittal thin-slab maximum intensity projection demonstrates the thoracic aorta with minimal atherosclerotic wall changes, but with no evidence of potential causes for acute chest pain Discussion Our study demonstrates that DSCT allows performing an ECG-gated chest examination for visualizing the different thoracic vascular territories at the same time-by employing an adjusted contrast media protocol-within a reasonable breath-hold period, and thus provides a diagnostic image quality in almost all patients. One of the most important findings of this study is that diagnostic data of coronary arteries could be obtained without foregoing heart rate control. Contrast media protocol and scan time With increasing gantry rotation times and faster volume coverage of newer CT scanners, scan times successively shorten and higher injection rates are required to achieve sufficient contrast in the vascular territory of interest. Similar to a previous 64-slice CT study [15], we have set the ROI for bolus tracking in the ascending aorta. With a relatively low threshold of 80 HU above which the scan was initiated, a homogenous attenuation in both aortic/coronary and pulmonary arteries could be achieved. By doing so, we aimed at the time interval where the contrast enhancement curve of the pulmonary circulation overlaps with enhancement of the aorta [25]. This has also resulted in comparable attenuation in left and right cardiac cavities. Recommended mean attenuation values for diagnostic image quality are approximately 300 HU for the pulmonary arteries [26] and 250–350 HU in the ascending aorta and coronary arteries [27, 28]. We obtained a mean attenuation of 291 HU in the ascending aorta, 334 HU in the pulmonary trunk, 285 HU in the RCA, and 286 HU in the LMA with only small intra- and inter-individual variations. These attenuation values were considered diagnostic in 98% of our patients and are comparable to previously published data [15, 29]. We encountered only one patient with non-diagnostic attenuation of segmental and subsegmental pulmonary arteries. However, this may have been also caused by other reasons negatively affecting contrast attenuation in pulmonary arteries such as Valsalva maneuver [30]. The other major issue for CT in acute chest pain represents the issue of scan duration. As ECG-gated CT requires low pitch [31], scan time is increased as compared to non-gated chest CT examinations. This, however, negatively affects image quality by making examinations more prone to breathing artifacts. This holds particularly true for patients with acute chest pain who often suffer from dyspnea. In our study, mean scan time was approximately 12 s, which represents a further improvement as compared to a mean scan time of 21 s with 64-slice CT [15]. Role of CT in the diagnostic pathway of acute chest pain Figure 6 represents a simplified flow-chart of a generally accepted clinical pathway for patients with cardiac and non-cardiac causes of acute chest pain [9, 32, 33]. While CT is the accepted reference modality for the diagnosis and exclusion of pulmonary embolism and aortic dissection [10, 11], the role of CT in patients with a suspected cardiac cause for chest pain is currently under investigation. Gallagher and colleagues [6] have shown that 64-slice CT has accuracy that is comparable to that of stress nuclear imaging for the detection of acute coronary syndrome in low-risk patients with negative serial ECG and biomarker results. Hoffmann and coworkers [8] have demonstrated that 64-slice CT has good performance characteristics for ruling out acute coronary syndromes in patients presenting to the emergency department with acute chest pain. The authors demonstrated in patients in whom initial triage was inconclusive that the absence of coronary artery plaque or significant stenosis on CT angiography had an excellent negative predictive value for the subsequent diagnosis of acute coronary syndrome. Furthermore, in those patients with CAD on CT, the extent of coronary atherosclerotic plaque provided incremental information to standard baseline patient variables and clinical risk assessment. Goldstein and colleagues [7] have shown that 64-slice CT coronary angiography is able to definitely establish or exclude CAD as the cause of acute chest pain. The high negative predictive value of the studies suggests that CT coronary angiography may be useful for facilitating and optimizing triage of patients with acute chest pain and/or inconclusive initial emergency department evaluation [7, 8]. Fig. 6Flow chart of a generally accepted clinical pathway for patients with cardiac and non-cardiac causes of acute chest pain. CT represents the standard of reference in patients with suspicion of acute aortic syndromes or pulmonary embolism (light gray boxes). The added value of CT in the evaluation of acute coronary syndrome has been already demonstrated [8] (moderate gray box). The potential future role of CT in patients with acute chest pain might be at an even earlier point of diagnostic work-up (dark gray boxes) to rule out life-threatening coronary, pulmonary, and aortic disease and to guide adequate therapeutic interventions Study limitations We included only a relatively small number of patients and thus a small number of pathologies. Second, our study lacks a standard end-point to assess the final diagnosis and misses comparison with the reference standard invasive coronary angiography for investigating CAD for which diagnostic accuracy could be calculated. However, only a minority of our patients underwent invasive coronary angiography because in our emergency department standard management in patients with suspected acute coronary syndrome directly undergoing invasive coronary angiography without prior CT. That is in line with most studies involving patients referred to the emergency department [34]. Third, we did not investigate the optimal contrast media technique for obtaining homogenous contrast attenuation of different thoracic vascular territories. It remains to be determined if the use of the test bolus technique would yield better results with regard to vessel opacification. Finally, it needs to be assessed whether caudo-cranial or caudo-cranial scan direction will result in better image quality. Conclusion First experience indicates that DSCTA is feasible in patients with acute chest pain and provides diagnostic image quality of the aorta, pulmonary arteries, and coronary artery system as well as of the lung parenchyma and mediastinum in a patient population without foregoing heart rate control. A dedicated contrast media protocol allows for homogenous attenuation of the different thoracic vascular territories.

Breast_Cancer_Res_Treat-3-1-2001217

The patient experience

The impact of improved treatments for the management of hormone-sensitive breast cancer extends beyond clinical responses. Thanks to appropriate literature and access to the internet, patient awareness of treatment options has grown and patients are now, in many cases, able to engage their oncologists in informed conversations regarding treatment and what to expect in terms of efficacy and safety. Indeed, patients realize that although there is no cure for metastatic disease, treatment can greatly reduce the risk of progression and in the adjuvant setting, where treatment is administered with a curative intent, current treatment options reduce the risk of relapse. The approval of letrozole throughout the breast cancer continuum has provided patients with many reassuring options. The improvement in outcome with letrozole is achieved without a detrimental effect on overall quality of life. Adverse events such as hot flushes, arthralgia, vaginal dryness, and potential osteoporosis are most significant from the patient’s perspective, and it is important that caregivers pay attention to patients experiencing these events, as they can impact compliance unless effectively explained and managed. The major benefits of letrozole are to improve prospects for long-term survivorship in the adjuvant setting and to delay progression and the need for chemotherapy in the metastatic setting. Introduction The diagnosis of breast cancer is a devastating blow for women and is associated with depression, anxiety, and a range of other psychological problems, such as self-blame and negative perception of body-image [1–5]. Newly diagnosed metastatic breast cancer (MBC) is also associated with high levels of psychological morbidity, particularly for younger women [6]. Prevention of relapse is a long-term therapeutic imperative, but the impact of therapy on quality of life (QOL) also needs to be taken into consideration when planning treatment strategies. All therapeutic modalities for early breast cancer (mastectomy or lumpectomy, radiation, chemotherapy, antibody therapy and endocrine therapy) can have a significant impact on patients’ QOL both in the short-term and in the transition period from primary treatment to long-term survivorship [7–9]. Receiving good quality information about prognosis, treatment options, side effects, and risks of breast cancer recurrence is, therefore, important for patients diagnosed with breast cancer [10–13]. The third-generation aromatase inhibitor (AI) letrozole was introduced a decade ago, but at that time, a limited amount of information on breast cancer and its treatment was available to patients. Since then, the international oncology community has made major advances in its knowledge about breast cancer biology, the individualization of treatment options, and communication with patients. Greater understanding of genetics and breast cancer risk has improved approaches for counseling individuals about their susceptibility [14]. In women who develop breast cancer, the individualization of treatment is becoming more sophisticated through the use of biomarkers as prognostic and predictive factors. These include human epidermal growth factor receptor (HER) 2 [15, 16], urokinase-type plasminogen activator and plasminogen activator inhibitor type 1 [17, 18], mRNA expression patterns [19, 20], and, more recently, genetic profiling techniques [21, 22]. Individualizing treatment is also made easier by the ability to more accurately determine the patient’s risk factors and prognosis; for example, by using the Adjuvant! Online program [23–25]. Recognizing breast cancer overtreatment has also contributed to tailoring treatment according to individual risk [26–30]. Oncology teams have facilitated patient involvement in selecting the most appropriate therapy [24, 31, 32] and have improved the provision of patient information, support, and counseling [33, 34]. At the same time, patients began to organize effectively and expand their access to knowledge of new drugs and treatment paradigms [35–37]. Patients have learned to be informed so that they can have constructive dialogues with their doctors and feel they understand and contribute to treatment decisions [38–42]. Patients who take an active role in the decision-making process perceive that they have a treatment choice, in contrast to those who prefer a shared or passive role [43]. They know they cannot be told at any time during the course of their disease that they are or are not cured; they recognize that breast cancer is a chronic illness. Patients are now empowered with hope and the knowledge that breast cancer can be managed. Moreover, increased survival rates have prompted greater interest in the QOL of breast cancer survivors [44–46]. Better education and improved access to screening has resulted in fewer patients being diagnosed with advanced breast cancer at their first presentation, and consequently, mortality has decreased [29, 47–49]. In addition, more effective adjuvant therapy with third-generation AIs has significantly decreased the risk of recurrence as compared with the previous standard of care based on tamoxifen [50–53]. However, there remains a long-term risk of breast cancer recurrence over time [54, 55]. Hormone receptor-positive (HR+) breast cancer is particularly challenging because of the persistent risk of recurrence with this chronic “smoldering” disease. The Early Breast Cancer Trialists Collaborative Group meta-analysis showed that in untreated women with the same nodal status, the breast cancer death rate is greater for patients with estrogen receptor-negative (ER−) versus ER+ tumors in the first 5–6 years, but substantially lower for ER− versus ER+ tumors over the next 10 years [56]. Similarly, Saphner reported that beyond 5 years, the risk of recurrence was higher for patients with HR+ tumors than for those with HR− tumors (P = 0.00002) [54]. Patients may experience stress after medical therapy is ended because they feel they have lost a safety net [57, 58]. This fear is understandable and justifiable in view of the persistent risk of recurrence for HR+ breast cancer. An analysis of the MA.17 trial of extended adjuvant therapy recently showed an increasing risk of disease recurrence over time in patients treated with placebo after discontinuing tamoxifen [59]. Not surprisingly, therefore, patients may be willing to receive long-term therapy (extended adjuvant) to prevent recurrence provided that the benefits outweigh the risks and QOL is maintained [60]. Patients will try to tolerate treatment-related adverse events if there is the prospect of achieving a “cure,” i.e. remaining free of relapse during their lifetime. Tamoxifen was the mainstay of breast cancer therapy, but its time-dependent efficacy and serious adverse events created a need for new therapies [61–63]. The suppression of estrogen was shown to be greater and more selective with third-generation AIs than with first- and second-generation compounds [64], and this has resulted in better clinical outcomes and improved tolerability [65]. This review examines the clinical use of the third-generation AI letrozole from the patient’s perspective and assesses how it has improved treatment outcomes across the breast cancer continuum, including advanced or MBC, extended and initial adjuvant therapy, and neoadjuvant therapy. Metastatic setting Endocrine therapy is the first-choice treatment for women with HR+ breast cancer without acute life-threatening symptoms and should be administered for as long as possible before switching to cytotoxic regimens. Endocrine therapy is preferred to cytotoxic chemotherapy because of its more favorable safety profile [66]. Thus, extending the time to the initiation of chemotherapy is important with endocrine therapy, because a considerable proportion of patients who progress to chemotherapy will experience toxic side effects without gaining benefit [67]. The most common acute side effects of chemotherapy, such as leukopenia, alopecia, and nausea and vomiting, are significantly increased in women receiving combination regimens compared with single agents [68]. Chemotherapy can also adversely affect certain aspects of QOL, notably increasing fatigue [69], and some regimens may be associated with severe or life-threatening complications such as cardiac failure [70]. A meta-analysis to review the evidence and determine whether chemotherapy or endocrine therapy has the most beneficial effect on treatment outcomes (survival, response rate, toxicity, and QOL) concluded that in women with HR+ MBC, a policy of treating first with endocrine therapy rather than chemotherapy is recommended except in the presence of rapidly progressive disease [71]. Prior to the introduction of third-generation AIs, postmenopausal patients with hormone-responsive MBC had few endocrine therapy options other than tamoxifen [62]. Second-line endocrine agents were limited by safety concerns, including weight gain or the risk of thromboembolism with megestrol acetate [72, 73] and cardiovascular toxicity with aminoglutethimide [65]. Randomized controlled trials demonstrated that letrozole is better tolerated and more effective than these second-line endocrine agents [65, 72, 73]. Furthermore, letrozole has also demonstrated superior early survival compared with tamoxifen as first-line therapy in postmenopausal patients with MBC [74]. Importantly, longer time to disease progression with letrozole versus tamoxifen was achieved without increased time with adverse events and resulted in more quality-adjusted survival for patients on letrozole [75]. Maintaining functional ability is an important goal for patients treated in this setting. A subanalysis by different sites of metastatic lesion and Karnofsky Performance Scores also showed the superiority of letrozole compared with tamoxifen in patients with nonvisceral metastases, with visceral metastases without liver involvement, and with liver metastases [76]. From the patient’s perspective, it is important to receive the most effective therapy first-line; therefore, letrozole represents a more attractive option than tamoxifen for postmenopausal women. Pharmacologic and clinical differences exist between third-generation inhibitors and should be considered in the selection of the most appropriate endocrine therapy [64, 77]. Data from a small randomized crossover trial in 72 postmenopausal women with HR+ MBC showed that overall QOL was significantly better with letrozole than with anastrozole (P = 0.002 for mean total Functional Assessment of Cancer Treatment-endocrine symptoms) [78]. Furthermore, letrozole was significantly better tolerated overall than anastrozole. Less nausea, fewer hot flashes, and less abdominal discomfort resulted in almost twice as many patients preferring letrozole to anastrozole [78]. While letrozole is an appropriate first-line therapy for the majority of patients with hormone-responsive MBC, a small subset of patients with HR+ HER2+ tumors have high-risk disease and are candidates for early treatment with chemotherapy plus the anti-HER2 monoclonal antibody trastuzumab [79]. A meta-analysis of 12 studies involving 2,379 patients with MBC demonstrated that HER2+ tumors are less responsive than HER2− tumors to endocrine treatment (overall relative risk 1.42; 95% confidence interval, 1.32–1.52; P < 0.00001) [80]. Evidence from preclinical models suggests that trastuzumab may overcome relative resistance to endocrine therapy [81], providing the rationale for combining anti-HER2 and endocrine therapies in the clinical setting. Recently, the first published results of an AI in combination with an anti-HER2 antibody (i.e. trastuzumab) show that the combination demonstrated durable responses for at least 1 year in 25% of patients [82]. Preliminary results from a phase 3 trial (TrAstuzumab in Dual HER2 ER-positive Metastatic breast cancer) demonstrated that the combination of trastuzumab with an AI (anastrozole) was more effective than anastrozole alone in postmenopausal patients with HR+ HER2+ MBC [83]. Another phase 3 trial, which is enrolling more than 1,200 patients in this setting and has just finished recruitment, is investigating letrozole in combination with lapatinib, a dual inhibitor of HER2 and HER1 tyrosine kinases [84]. The strategy of combining letrozole with trastuzumab (as investigated in the evaluation of Letrozole combined with Trastuzumab trial), or another HER2-directed therapy, may allow patients with HR+, HER2+ tumors to safely delay the initiation of cytotoxic chemotherapy. Adjuvant therapy Patients receiving adjuvant therapy expect their treatment to prevent breast cancer recurrence and offer the prospect of cure. However, even modest gains in survival are sufficient to make adjuvant endocrine treatment worthwhile for premenopausal women with early-stage breast cancer [85]. This benefit threshold also appears to apply to women with a higher risk of recurrence for whom adjuvant chemotherapy is indicated [86]. Adjuvant therapy should be tailored to suit the needs of individual patients based on their clinical risk factors, attitudes, and personal life circumstances [25]. However, it is important to recognize that individual preferences cannot always be fully explained on the basis of treatment-related determinants and patient or clinical characteristics [87]. The Anastrozole versus Letrozole: Investigation into Quality of Life study compared the effects of anastrozole and letrozole on estrogen levels, QOL, lipids, and bone health [88]. A total of 185 postmenopausal women with invasive breast cancer were randomized to receive adjuvant therapy with either 12 weeks of letrozole followed by 12 weeks of anastrozole or vice versa. Of the patients who have completed the 12 weeks of treatment (n = 146), 50 (34%) had a preference for neither drug, 50 (34%) preferred anastrozole, and 46 (32%) preferred letrozole. Both estradiol (E2) and estrone sulfate levels (E1S) were significantly lower on letrozole than on anastrozole (P < 0.000001). Thus, 2.5 mg of letrozole reduces circulating E2 and E1S levels to a significantly greater degree than 1 mg of anastrozole, with no significant difference in patient preference. The greater suppression of estradiol levels might translate into improved clinical efficacy, although further studies, such as the ongoing Femara Anastrozole Clinical Evaluation trial, are required to confirm these findings. Patients with HR+ tumors need to consider the prospect of life-long adjuvant therapy in view of the persistent risk of disease recurrence [54, 56, 89]. Following the introduction of third-generation AIs, there is much greater choice in endocrine therapy. The challenge for physicians is to select the most appropriate strategy to suit individual patient circumstances. Letrozole first entered the adjuvant setting as a treatment option for patients completing 5 years of tamoxifen [90] and has demonstrated clinical superiority over tamoxifen as initial adjuvant therapy [51] and neoadjuvant therapy [91]. Neoadjuvant therapy Neoadjuvant endocrine therapy with letrozole is an attractive option for some postmenopausal patients with HR+ early breast cancer [92], including elderly patients who are unable or unwilling to undergo chemotherapy or surgery, and patients with locally advanced HR+ tumors who wish to have breast-conserving surgery but are not suited for preoperative chemotherapy [91, 93]. In a randomized controlled trial, letrozole demonstrated a superior overall objective response rate and rate of breast-conserving surgery compared with tamoxifen as neoadjuvant therapy [91]. Currently, letrozole is the only AI approved in the neoadjuvant setting (in the United Kingdom and 16 other countries worldwide). It provides a reasonable therapeutic alternative to preoperative chemotherapy in postmenopausal women with HR+ disease in clinical situations where the low toxicity of the regimen is considered an advantage, e.g. in women older than 70 years [91]. However, patients need to be aware that a longer course of therapy may be required to achieve an objective response than is the case with neoadjuvant chemotherapy [94]. In one study, neoadjuvant letrozole was safely given over 12 months to postmenopausal women with large operable or locally advanced HR+ breast cancers [94]. A longer treatment course may suit some patients and give them more time to consider their options for surgery, radiotherapy, and chemotherapy. Neoadjuvant therapy with letrozole could be considered a sensitivity test of endocrine therapy that might be incorporated into strategies to individualize adjuvant treatment according to response [95]. This would provide reassurance to patients that they will be receiving an adjuvant endocrine therapy that has shown activity against their tumor. Objective assessment of the antitumor activity of neoadjuvant letrozole can be made on the basis of response rate and by analysis of predictive biomarkers. In one study, neoadjuvant letrozole was found to inhibit tumor proliferation (determined by the biomarker Ki67) more effectively than tamoxifen [95]. Letrozole was effective independently of HER2 expression status, although the greatest difference between letrozole and tamoxifen was seen in tumors that were HR+ and HER1/2+. Recent evidence has suggested that HER2+ tumors can continue to proliferate despite neoadjuvant letrozole or tamoxifen treatment, which could imply therapeutic resistance that may manifest later in the clinical course of the disease [96]. Biomarker studies may also reveal differences between AIs. In a randomized comparative trial, letrozole and anastrozole significantly reduced proliferation in HR+, HER2+ or HER2− tumors [97]. Decreased proliferation was seen at all Allred ER expression levels with both agents, but only letrozole showed a significant effect in the lower ER cases. In addition, more cases showed a reduction in progesterone receptor (PgR) expression following letrozole than anastrozole [97]. Another study showed evidence of a decrease in HER2 expression after neoadjuvant treatment with letrozole [98]. Although complex, biomarkers may become increasingly important from a patient’s perspective, because they can help to improve the individualization of treatment. Furthermore, biomarkers may be useful to predict the risk of resistance to endocrine therapy and the need to consider alternative approaches in the future, such as combination therapies or cytotoxic chemotherapy. Knowing what to expect next is important for patients. However, further work is necessary to validate the use of clinical markers and biomarkers in the neoadjuvant setting as surrogate end points for long-term outcomes [96]. Early adjuvant therapy Tamoxifen was the gold standard endocrine therapy for all women with HR+ breast cancer until recent results from large randomized trials challenged this paradigm for postmenopausal women [50, 51, 99]. Oncologists and patients now face an important choice when selecting adjuvant endocrine treatment [66, 100], whether to start with the most potent endocrine therapy (an AI) upfront or to start with tamoxifen and switch to an AI inhibitor after 2–5 years. The ongoing Breast International Group (BIG) 1-98 randomized trial is expected to provide more information on the benefits of switching to letrozole after 2–3 years on tamoxifen and will clarify which is the optimal strategy. The final results are expected to be released in 2008 [51], but until then, the most appropriate endocrine therapy will need to be selected on the basis of currently available evidence. The MA.17 trial has already demonstrated that letrozole treatment is beneficial after 5 years of adjuvant tamoxifen [90, 101, 102]. Patients will ideally wish to receive treatment that provides the greatest long-term efficacy with the lowest risk of adverse events. BIG 1-98 showed that letrozole was significantly more effective than tamoxifen as initial adjuvant endocrine therapy [51]. Moreover, tamoxifen is perceived to be more toxic than AIs, largely because of well-publicized “scares” about increased risk of uterine cancer and thrombosis [61, 103]. These side effects and vaginal bleeding were reported for tamoxifen in BIG 1-98, whereas letrozole was associated with more skeletal events. The overall rate of cardiovascular events was not significantly different between the groups. Letrozole was also shown to be extremely well-tolerated in comparison with placebo in the MA.17 trial [90, 101]. With the exception of adverse events related to suppression of estrogen, there was no difference in adverse events (Table 1). Table 1Safety profile of letrozole in comparison with placebo (A) and tamoxifen (B) reported in postmenopausal women with early breast cancerPlacebo Acute toxicities reportedTotal number (%), any grade101Letrozole (n = 2,572)Placebo (n = 2,577)Edema571 (22)542 (21)Hypertension130 (5)129 (5)Hot flushes1,486 (58)a1383 (54)Fatigue999 (39)998 (39)Sweating782 (30)760 (29)Anorexia142 (6)a110 (4)Constipation363 (14)382 (15)Diarrhea168 (7)176 (7)Nausea308 (12)314 (12)Vaginal bleeding145 (6)196 (8)aInfection without neutropenia124 (5)112 (4)Arthritis167 (6)137 (5)Hypercholesterolemia418 (16)411 (16)Dizziness458 (18)441 (17)Insomnia166 (6)135 (5)Depression143 (6)131 (5)Headache706 (27)685 (27)Arthralgia651 (25)a532 (21)Myalgia380 (15)a310 (12)Bone pain141 (5)149 (6)Dyspnea161 (6)163 (6)Alopecia126 (5)a89 (3)Vaginal dryness147 (6)129 (5)TamoxifenWorst grade adverse events recorded within first 28 daysTotal number (%), any grade51Letrozole (n = 3,975)Tamoxifen (n = 3,988)CVA or TIA39 (1.0)41 (1.0)Thromboembolic event61 (1.5)140 (3.5)aCardiac event (IHD, CF)162 (4.1)153 (3.8)Other CV event19 (0.5)a8 (0.2)Vaginal bleeding132 (3.3)263 (6.6)Hot flushes1332 (33.5)1516 (38.0)aNight sweats554 (13.9)647 (16.2)aFracture225 (5.7)a159 (4.0)Arthralgia806 (20.3)a491 (12.3)Myalgia254 (6.4)243 (6.1)CV cardiovascular, CVA cerebrovascular accident, TIA transient ischemic attack, IHD ischemic heart disease, CF cardiac failureaSignificant difference Based on the results from the BIG 1-98 and the Anastrozole and Tamoxifen Alone or in Combination (ATAC) trial, which compared upfront AIs with tamoxifen, it is clear that all postmenopausal women with HR+ breast cancer should be given the opportunity to receive adjuvant use of an AI, and this recommendation is now reflected in internationally recognized treatment guidelines [66, 100, 104, 105]. Patients with HR+ breast cancer considered eligible for adjuvant chemotherapy because of their increased risk for relapse should also be candidates for the most effective adjuvant endocrine strategy [100]. Of note, the BIG 1-98 trial showed that adjuvant letrozole provides significant disease-free survival (DFS) benefits for patients at increased risk of recurrence, specifically patients with node-positive tumors, large primary tumors (>2 cm), and recipients of chemotherapy and also demonstrated a significant reduction in the risk of distant recurrence, a well-known predictor of breast cancer death [51]. Additional analyses of data from BIG 1-98 to determine the predictive value of centrally tested ER, PgR, and HER2 status on the response to letrozole and tamoxifen indicate that there is no difference in subgroups by ER/PR or HER2 status regarding the superiority of upfront letrozole versus tamoxifen [106]. Low-risk patients with HR+ breast cancer who do not receive adjuvant chemotherapy are also candidates for AIs [104] because of the persistent risk of relapse [54, 56]. While it is known that tamoxifen loses effectiveness after 5 years [62, 63], the optimal duration of initial therapy with an AI remains to be determined [107]. The MA.17 randomized controlled trial of extended adjuvant endocrine therapy has already shown that letrozole treatment is beneficial after 5 years of tamoxifen in postmenopausal patients [90, 101], but there are no equivalent data on the use of tamoxifen after initial AI therapy in the adjuvant setting [100]. Furthermore, the efficacy of switching to another AI in patients who discontinue treatment with letrozole is not known. Therefore, patients starting upfront therapy with letrozole will need reassurance that according to present knowledge, treatment can be continued for up to 5 years, and other options will be available down the line. In addition, patients will require information about the potential long-term impact of AIs on bone and the cardiovascular system [107]. According to the current treatment guidelines [66, 100, 104, 105], patients already taking a course of adjuvant tamoxifen may wish to switch to anastrozole or exemestane, as these AIs have demonstrated efficacy in this setting [52, 108, 109]. Although recent data have shown that this sequential adjuvant strategy is associated with a survival advantage [53, 110, 111], it is important to note that these data were obtained from a selected population comprising patients at randomization who were disease-free after 2–3 years of tamoxifen. It is therefore not valid to make a direct comparison with an unselected population treated in trials of upfront AI therapy. The decision to switch endocrine therapy after 2–3 years, i.e. for a sequential tamoxifen–AI strategy, should be based on an individual’s risk for recurrence, risk for osteoporosis, and ability to tolerate tamoxifen. Importantly, BIG 1-98 is expected to provide more information on the benefits of switching to letrozole after 2–3 years of tamoxifen, and will clarify whether upfront or sequential AI use is the optimal strategy. The final results will be released in 2008 [51]. Extended adjuvant (including late extended adjuvant) therapy Many patients fear breast cancer recurrence [112] and are reassured by the “safety net” of continuing medical treatment and monitoring [9]. The MA.17 trial showed that extended adjuvant therapy with letrozole gives patients the opportunity of retaining the safety net for at least 5 more years after adjuvant tamoxifen [101], and these results led to its approval in this indication. Currently, letrozole is the only AI approved as an extended adjuvant therapy. The final analysis of MA.17 after a median follow-up of 30 months showed letrozole significantly improved DFS (42% reduction in risk vs. placebo), distant DFS (40% reduction in risk vs. placebo) and, in node-positive patients, overall survival (39% reduction in risk vs. placebo) [101]. A recent cohort analysis of the MA.17 trial data suggested that the longer patients are exposed to letrozole, the greater the benefit, at least out to 48 months [59]. In addition, an extension to the MA.17 trial (MA.17 Re-randomization) is being conducted to determine the benefits of continuing letrozole for a further 5 years [102]. When a planned interim analysis of MA.17 revealed a significant advantage for letrozole, the trial was unblinded and patients on placebo were given the option to switch to letrozole [90]. In this non-randomized comparison, women from the placebo arm who elected to switch to letrozole also experienced an improvement in outcome when compared with those who elected to have no treatment [102]. Recurrence risk persists beyond the completion of adjuvant therapy, even in patients at low risk of recurrence. One study shows that with systemic adjuvant therapy, patients with node-negative breast cancer have a ≥25% 10-year risk of relapse and a corresponding 10-year breast cancer death rate as high as ≥10%, depending on tumor grade and size [113]. Thus, the proven efficacy of letrozole given after a tamoxifen-free period means that physicians need to discuss the option of restarting endocrine therapy with almost all patients. Physicians have to consider how best to address this topic with patients who are up to 2–3 years out beyond their initial 5 years of tamoxifen, i.e. about 5–8 years after their initial diagnosis. This will be a major communication challenge and create a dilemma for patients who may feel well and have put their breast cancer behind them. Safety and compliance issues The clinical benefits of AIs and tamoxifen are generally achieved without a major detrimental effect on overall QOL [114]. Data from the MA.17 trial, which is the only large adjuvant trial comparing AI therapy not with tamoxifen but with placebo, showed that overall QOL was maintained during extended therapy with letrozole, and only a minority of patients experienced substantial changes in QOL, which were compatible with a reduction in estrogen synthesis [115]. It is important to recognize that patients may attribute such changes to their treatment, whereas they could in fact be symptoms of menopause [116]. However, side effects do occur, which not only affect patient adherence to endocrine therapy [117] but can also lead to additional morbidity and even serious or life-threatening complications in a small minority of patients [51, 52, 90, 99, 108, 109]. The main safety concerns with long-term estrogen deprivation include potential effects on bone health, cardiac health, lipid profile, cognitive functioning, and sexual health [100]. The MA.17 trial showed that letrozole is well-tolerated in comparison with placebo [101], and BIG 1-98 showed that letrozole was better tolerated than tamoxifen [51] (Table 1). A recent patient-reported outcomes study in 104 tamoxifen-intolerant women found that switching to letrozole was associated with a significant reduction in hot flushes (P = 0.001) and significant improvements in QOL (P = 0.001) and mood (P = 0.04). Furthermore, when given the choice of continuing therapy, 66% of women indicated that they preferred to remain on letrozole, while only 24% preferred to go back to tamoxifen [118]. Despite an increase in newly diagnosed osteoporosis with AIs, no significant difference in clinical fracture rate was seen between letrozole and placebo [101]. A companion study to MA.17 showed that there was only a modest increase in bone resorption and reduction in bone mineral density (BMD) in the spine and hip with letrozole compared with placebo [119]. Consequently, patients should be made aware of the risk of osteoporosis and given advice on lifestyle measures (e.g. exercise, diet, vitamins) to reduce risk. A baseline BMD measurement should be obtained for all patients before starting therapy with an AI. BMD should then be measured annually, and patients at high risk of osteoporosis should be considered for prophylactic use of a bisphosphonate [120, 121]. However, recent data from ATAC may be reassuring for patients facing upfront AI therapy, since they indicate that women with normal initial BMD did not develop osteoporosis during a 5-year AI treatment [122]. Letrozole has been associated with an increase in arthralgia compared with placebo or tamoxifen, and myalgia compared with placebo [51, 101]. Arthralgia and joint pain can be bothersome and may lead to impaired mobility. Patients need to be advised that constant exercising of the joints will alleviate this adverse event and that use of nonsteroidal anti-inflammatory drugs over a certain period of time may be helpful. Moreover, these complaints seem to be most frequent in the beginning of AI therapy [123]. Another potential consequence of endocrine therapy is a detrimental effect on sexual health [115, 116]. Vaginal dryness is a consequence of menopause and treatments that cause menopausal symptoms, and has an important bearing on sexual health and well-being of breast cancer survivors [124]. Vaginal dryness can be a significant problem that can interfere with the stability of relationships [125], and patients need to receive counseling with regard to this issue [126]. It has been shown that clinical assessment and an active intervention program for menopausal symptom management in breast cancer survivors can lead to an improvement in sexual functioning [127]. A recent report demonstrated that the use of local estradiol therapy may cause elevation of serum estradiol and may therefore be contraindicated in postmenopausal women on AI therapy [128]. So far, such data do not exist for estradiol-containing compounds; thus, local estradiol treatment may be indicated in individual cases. Androgen treatment may also improve sexual well-being in postmenopausal women [129], yet the oncological safety of this approach has not been validated in patients receiving AI therapy. Of note, the incidence of vaginal dryness in the MA.17 trial was similar in the letrozole and placebo groups (Table 1) [101]. Differential effects on lipid profiles and cardiac risk have been reported between tamoxifen and AIs [51, 130, 131], but data from the Letrozole, Exemestane, and Anastrozole Pharmacodynamics trial, directly comparing the effects of anastrozole, letrozole, and exemestane on lipid profiles in healthy postmenopausal women, suggest that the steroidal and nonsteroidal inhibitors have a similar impact on lipid profiles [132]. On the basis of evidence from large randomized trials, when comparing letrozole with tamoxifen or placebo, it appears that letrozole does not have clinically relevant adverse effects on lipids or cardiac risk during long-term adjuvant therapy, although further assessment is warranted [51, 101, 133]. While letrozole is well-tolerated, and any side effects that do occur can be managed, it is essential that patients are motivated to stay on therapy. This can be a particular challenge in the extended adjuvant setting. Thus, as patients now face the prospect of receiving endocrine therapy for 10 years or more, it is important to consider the overall life-long benefits and risks from the individual’s perspective. A low burden of adverse events and maintenance of QOL is important in maintaining adherence to long-term treatment [60]. In the MA.17 trial, the rate of patients choosing to discontinue therapy during the first year was similar for letrozole (n = 256/2,575) and placebo (n = 254/2,582) [90]. Letrozole is ingested orally and can be safely and conveniently taken at home. Physician contact with oral endocrine therapy can be much less frequent than, for example, with orally active chemotherapy where regular blood tests, side-effect monitoring, and resulting dose modifications make frequent physician contact mandatory. Patients with breast cancer prefer oral cancer therapy providing that it does not compromise treatment efficacy [134]. Conclusions Patients with breast cancer face bewildering choices at a time when they are experiencing highly stressful circumstances. Moreover, many of these women are not well-informed and thus are not able to take part in treatment decisions. The Gathering Information on Adjuvant Endocrine therapy initiative decided to delineate women’s knowledge and experience of adjuvant endocrine therapy. Results thus far indicate that only 22% of patients were fully or highly involved in the decision to start adjuvant endocrine therapy [135, 136]. Many of the women who took part in the survey were not satisfied with the degree to which they were involved in treatment decision-making, and women 60 years or older had the lowest levels of involvement [135, 136]. The results from this survey show that information provided to patients about adjuvant endocrine therapy is suboptimal and indicate the need for programs to raise patient awareness. The medical community has a responsibility to help patients understand their prospects for survival and make the right choices about treatment [24]. Doctor–patient discussions and programs to raise patient awareness will increase patients’ knowledge about the individualization of treatment and may increase the number of women who take an active role in treatment decisions. Patients with hormone-responsive breast cancer should be offered the opportunity of receiving the most effective endocrine therapy. Clinical evidence suggests that postmenopausal women should receive an AI rather than tamoxifen as their first option if tumor characteristics and individual side-effect profiles support this choice. Letrozole has consistently demonstrated superiority over tamoxifen in the metastatic and adjuvant treatment settings [51, 74, 91, 101]. As a result of its innovative clinical trial program [51, 74, 101], letrozole is approved for use in postmenopausal women throughout the breast cancer care continuum [137]. From the patient’s perspective, the major benefits of letrozole are improving prospects for long-term survivorship (“cure”) in the adjuvant setting and delaying progression and the need for chemotherapy in the metastatic setting. Endocrine therapy is very effective, has a generally favorable safety profile, and adds to the efficacy of chemotherapy. However, a major challenge for both physicians and patients is ensuring compliance with long-term daily therapy. This may be a particular problem in the extended adjuvant setting, where the patient may face the prospect of life-long therapy. As patients no longer feel sick, it is understandable that they may forget to take a dose and gradually lose interest in continuing with the treatment. Letrozole is well-tolerated and, as with all AIs, the majority of adverse events are secondary to the suppression of estrogen. The most important adverse events from the patient’s perspective are the “visible” ones, such as hot flushes, vaginal dryness, and arthralgias. To achieve optimal compliance, patients need to feel that physicians are taking their adverse events seriously and taking appropriate steps to alleviate any problems. Physicians who treat very few breast cancer patients may not have sufficient experience with AIs to satisfactorily manage individuals who are experiencing these adverse events. Letrozole is likely to continue to play a major role in the management of breast cancer in all settings (see the paper in this supplement by Drs. Ellis and Ma, on Femara® and the future). Considering its efficacy and favorable side-effect profile, it is the logical choice for inclusion in new regimens, including combinations with novel agents. As the future unfolds, the management of breast cancer is set for further change, and it is essential that patients are informed and educated so that they can actively participate in treatment decisions and thus derive the most benefit from treatment advances.

J_Hum_Genet-4-1-2413114

Variations in the FTO gene are associated with severe obesity in the Japanese

Variations in the fat-mass and obesity-associated gene (FTO) are associated with the obesity phenotype in many Caucasian populations. This association with the obesity phenotype is not clear in the Japanese. To investigate the relationship between the FTO gene and obesity in the Japanese, we genotyped single nucleotide polymorphisms (SNPs) in the FTO genes from severely obese subjects [n = 927, body mass index (BMI) ≥ 30 kg/m2] and normal-weight control subjects (n = 1,527, BMI < 25 kg/m2). A case-control association analysis revealed that 15 SNPs, including rs9939609 and rs1121980, in a linkage disequilibrium (LD) block of approximately 50 kb demonstrated significant associations with obesity; rs1558902 was most significantly associated with obesity. P value in additive mode was 0.0000041, and odds ratio (OR) adjusted for age and gender was 1.41 [95% confidential interval (CI) = 1.22–1.62]. Obesity-associated phenotypes, which include the level of plasma glucose, hemoglobin A1c, total cholesterol, triglycerides, high-density lipoprotein (HDL) cholesterol, and blood pressure were not associated with the rs1558902 genotype. Thus, the SNPs in the FTO gene were found to be associated with obesity, i.e., severe obesity, in the Japanese. Introduction Obesity is the most common nutritional disorder in developed countries, and it is a major risk factor for hypertension, cardiovascular disease, and type 2 diabetes (Kopelman 2000; Wilson et al. 2003). Genetic and environmental factors contribute to obesity development (Maes et al. 1997; Barsh et al. 2000; Rankinen et al. 2006). Recent progress in single nucleotide polymorphism (SNP) genotyping techniques has enabled genome-wide association studies on common diseases (Herbert et al. 2006; Frayling et al. 2007; Scuteri et al. 2007; The Wellcome Trust Case Control Consortium 2007; Hinney et al. 2007). Using a large-scale case-control association study, we found that secretogranin III (SCG3) (Tanabe et al. 2007) and myotubularin-related protein 9 (MTMR9) (Yanagiya et al. 2007) are involved in susceptibility to the obesity phenotype. Genome-wide association studies have shown that the fat-mass and obesity-associated gene (FTO) is also associated with the obesity phenotype (Frayling et al. 2007; Scuteri et al. 2007; Hinney et al. 2007). This association was also found in many Caucasian and Hispanic American populations (Frayling et al. 2007; Scuteri et al. 2007; Dina et al. 2007; Field et al. 2007; Andreasen et al. 2008; Wåhlén et al. 2008; Peeters et al. 2008), whereas it was not found in the Chinese Han population (Li et al. 2008). Among Japanese, body mass index (BMI) was higher in subjects who had the A allele of rs9939609, similar to that observed in Caucasians; however, this finding was not significant (Horikoshi et al. 2007). Another group reported that rs9939609 was associated with BMI in the Japanese (Omori et al. 2008). Thus, the association of SNPs in the FTO gene with obesity in the Japanese remains controversial. To investigate the relationship between the FTO gene and obesity in the Japanese, we performed a case-control association study using patients with severe adult obesity (BMI ≥ 30 kg/m2) and normal-weight controls (BMI < 25 kg/m2); we found that SNPs in intron 1 of the FTO gene were associated with severe adult obesity. Materials and methods Study subjects The sample size for severely obese Japanese subjects (BMI ≥ 30 kg/m2) was 927 (male:female ratio 419:508, age 48.7 ± 14.2 years, BMI 34.2 ± 5.4 kg/m2), whereas that for Japanese normal weight controls (BMI < 25 kg/m2) was 1,527 (male:female ratio 685:842, age 48.1 ± 16.5 years, BMI 21.7 ± 2.1 kg/m2). The severely obese subjects were recruited from among outpatients of medical institutes. Patients with secondary obesity and obesity-related hereditary disorders were not included, and neither were patients with medication-induced obesity. The normal-weight controls were recruited from among subjects who had undergone a medical examination for screening of common diseases. Clinical features of the subjects are illustrated in Table 1. Additionally, 1,604 subjects were recruited (male:female ratio 803:801, age 48.7 ± 16.9 years, BMI 22.66 ± 3.16 kg/m2) from the Japanese general population. Each subject provided written informed consent, and the protocol was approved by the ethics committee of each institution and that of RIKEN. Table 1Clinical characterization of obese and control subjectsObeseControlP valueGender (M/F)419/508658/842Age (year)49.1 ± 14.248.2 ± 16.50.049Body mass index (kg/m2)34.50 ± 5.3921.65 ± 2.08 <0.000001Glucose (mg/dl)129.2 ± 49.697.7 ± 23.9<0.000001HbA1c (%)6.5 ± 1.85.1 ± 0.6<0.000001Total cholesterol (mg/dl)209.9 ± 37.9201.2 ± 36.4<0.000001Triglycerides (mg/dl)153.2 ± 99.5104.0 ± 73.2<0.000001High-density lipoprotein cholesterol (mg/dl)53.1 ± 18.965.1 ± 15.7<0.000001Systolic blood pressure (mmHg)136.4 ± 18.1123.4 ± 17.8<0.000001Diastolic blood pressure (mmHg)83.8 ± 12.076.0 ± 11.1<0.000001P values were analyzed using Mann–Whitney U test. Data are mean ± standard deviation DNA preparation and SNP genotyping Genomic DNA was prepared from the blood sample of each subject by using the Genomix (Talent Srl, Trieste, Italy). We searched for dbSNPs with minor allele frequencies (MAF) > 0.10 in the FTO gene of Japanese people. We selected 90 SNPs and were able to construct Invader probes (Third Wave Technologies, Madison, WI) for them (Supplementary Table 1). SNPs were genotyped using Invader assays as described previously (Ohnishi et al. 2001; Takei et al. 2002). Nine SNPs (rs9937053, rs9939973, rs9940128, rs7193144, rs8043757, rs9923233, rs9926289, rs9939609, and rs9930506) reported in a previous genome-wide association study (Scuteri et al. 2007) were genotyped using TaqMan probes (C__29910458_10, C__11776771_10, C__29621384_10, C__29387650_10, C__29387665_10, C__29693738_10, C__30270568_10, C__30090620_10, and C__29819994_10; Applied Biosystems, Foster City, CA, USA). Statistical analysis Genotype or allele frequencies were compared between cases and controls in three different modes. In the first mode, i.e., the additive mode, χ2 test was performed according to Sladek et al. (Sladek et al. 2007). In the second mode, i.e., the minor allele recessive mode, frequencies of the homozygous genotype for the minor allele were compared using a 2 × 2 contingency table. In the third mode, i.e., the minor allele dominant mode, frequencies of the homozygous genotype for the major allele were compared using a 2 × 2 contingency table. A test of independence was performed using Pearson’s χ2 method. P values were corrected by Bonferroni adjustment and P < 0.00017 [0.05/99 (total SNP number)/3 (number of modes)] was considered significant. The odds ratio (OR) and 95% confidence interval (CI) were calculated by Woolf’s method. We coded genotypes as 0, 1, and 2, depending on the number of copies of the risk alleles. OR adjusted for age and gender was calculated using multiple logistic regression with genotypes, age, and gender as independent variables. Hardy–Weinberg equilibrium was assessed using the χ2 test (Nielsen et al. 1998). Haplotype blocks were determined using Haploview (Barrett et al. 2005). Simple comparison of the clinical data among the different genotypes was performed using one-way analysis of variance (ANOVA). Simple comparison of the clinical data between case and control groups was analyzed using Mann–Whitney U test. Difference in BMI between genotypes was analyzed using a multiple linear regression, with BMI as the dependent variable and genotype as the independent variable, and with gender and age as covariates for BMI. Statistical analyses were performed using StatView 5.0 (SAS Institute, Cary, NC, USA). Power was calculated by the Monte Carlo method. Results Case-control association studies We searched for dbSNPs with MAF > 0.10 in the FTO gene. By using Invader and TaqMan assay, we successfully genotyped 99 SNPs spanning the FTO gene (Supplementary Table 1). Using these SNPs, we performed tests of independence between the phenotype and genotypes of obesity at each SNP by using severely obese subjects (BMI ≥ 30 kg/m2) and normal weight controls (BMI < 25 kg/m2). For each SNP, the lowest P value among the three different modes was selected as the minimum P value. All SNPs, including rs1421084, were in Hardy–Weinberg equilibrium (P > 0.01) (Supplementary Table 1). The power of the test was calculated by Monte Carlo method with different MAFs and different effect sizes. Effect of the risk allele on penetrance was assumed to be multiplicative; i.e., the penetrances for three genotypes were assumed to be a, ar, and ar2, respectively, where a and r denote the lowest penetrance and genotype relative risk, respectively. Supplementary Table 2 shows the calculated values of the power of the test with different MAFs and different genotype relative risks (r). The lowest penetrance (a) was calculated for each gender by assuming the affection rates of 2.3% for men and 3.4% for women (Yoshiike et al. 2002). Genotype relative risk (r) was assumed to be the same for both genders. Supplementary Table 2 shows that the test has significant power at relative high risk allele frequency when genotype relative risk is >1.7. As shown in Fig. 1 and Supplementary Table 1, 15 SNPs demonstrated significant associations with the obesity phenotype; the threshold of significance using Bonfferoni correction was P < 0.00017. These SNPs included rs9939609 (Frayling et al. 2007) and rs1121980 (Hinney et al. 2007) that were reported to be significantly associated with the obesity phenotype in the Caucasian population, as determined by genome-wide association studies; rs9930506 (Scuteri et al. 2007) showed marginal association with obesity in the Japanese. Linkage disequilibrium (LD) analysis revealed that these 15 SNPs were in almost complete LD (D' > 0.98, r2 > 0.80) and were located within the same LD block of approximately 50 kb (Fig. 1). The most significant association was observed for rs1558902 [additive mode, P = 0.0000041 and allele-specific OR (95% CI) adjusted for age and gender was 1.41 (1.22–1.62)]. The minor alleles of rs9939609 (MAF = 0.24) and rs1121980 (MAF = 0.26) were significantly more frequent in the obese group than in the normal-weight control group (additive mode, P = 0.000012 and P = 0.000051, respectively), and ORs were 1.38 (95% CI = 1.20–1.59) and 1.33 (95% CI = 1.16–1.52), respectively (Table 2, Supplementary Table 1). The MAF of both SNPs in the control group was 0.18; this was consistent with data obtained from the haplotype map of the human genome (HapMap) (Supplementary Table 1). Our data indicated that the SNPs in the FTO gene were associated with severe obesity in the Japanese. Fig. 1Linkage disequilibrium (LD) mapping, polymorphisms, and P values obtained in the test of independence between the phenotype and genotypes of obesity at various single nucleotide polymorphisms (SNPs) in the fat-mass and obesity-associated gene (FTO) gene. P values are expressed as negative logarithm of the minimum P values obtained in the three models (additive, minor allele dominant, and minor allele recessive modes). LD coefficients (D') between each pair of SNPs were calculated and are displayed as a strand in the LD blocks. Minor allele frequencies of all SNPs used in this analysis are ≥10%. The genomic structure is shown in the upper. The gray bar marks the LD block associated with obesityTable 2Associations of single nucleotide polymorphisms (SNPs) in the fat-mass and obesity-associated gene (FTO) gene with obesity existing in the 50-kb linkage disequilibrium (LD) blockdbSNP IDAlleleGenotypeAdditive modeRecessive modeDominant modeCaseControl1/2111222Sum111222SumOR (95% CI)χ2P valueχ2P valueOR (95% CI)χ2P valueOR (95% CI)rs9937053A/G593604949136341477312501.31 (1.13–1.51)12.30.00047 2.00.161.30 (0.90–1.88)13.00.000311.37 (1.16–1.63)rs9939973A/G613674969247550494115201.32 (1.15–1.51)15.70.000077a3.00.0811.36 (0.96–1.93)16.10.000061a1.40 (1.19–1.66)rs9940128A/G603664989247550094115161.31 (1.15–1.50)15.20.00010a2.60.111.33 (0.94–1.89)15.90.000068a1.40 (1.19–1.65)rs1421085C/T4933853792457443101915191.38 (1.20–1.59)19.60.000011a3.30.0681.44 (0.97–2.12)20.00.0000078a1.47 (1.24–1.74)rs1558902A/T4834153692552449102115221.41 (1.22 -1.62)21.20.0000041a4.60.0321.55 (1.04–2.31)20.80.0000052a1.48 (1.25–1.75)rs1121980A/G613674999277350494715241.33 (1.16–1.52)16.50.000051a3.60.0591.40 (0.99–1.99)16.50.000050a1.41 (1.19–1.66)rs7193144C/T4933953292055447101415161.39 (1.21–1.61)20.40.0000067a4.00.0441.49 (1.01–2.22)20.30.0000067a1.47 (1.24–1.74)rs8043757T/A4831954190854436102715171.36 (1.18–1.57)17.40.000037a4.20.0401.51 (1.02–2.25)16.40.000052a1.42 (1.20–1.69)rs8050136A/C5133653892556450101815241.38 (1.20–1.59)19.40.000012a4.70.0311.53 (1.04–2.26)18.50.000017a1.45 (1.22–1.71)rs3751812T/G5134053492555458101315261.38 (1.20–1.59)19.60.0000098a5.10.0241.56 (1.06–2.31)18.50.000017a1.45 (1.22–1.71)rs9923233C/G5133553391955449101015141.38 (1.20–1.60)19.80.0000093a5.00.0251.56 (1.06–2.30)18.70.000015a1.45 (1.23–1.72)rs9926289A/G503235319045642599314741.37 (1.19 -1.58)18.70.000020a3.90.0471.48 (1.00–2.19)18.10.000021a1.45 (1.22–1.72)rs9939609A/T5133453491956443100515041.38 (1.20–1.59)19.50.000012a4.50.0341.52 (1.03–2.24)18.70.000015a1.45 (1.23–1.72)rs7185735G/A5134053692755455101415241.38 (1.20–1.59)19.90.0000089a5.00.0251.55 (1.05–2.30)18.80.000014a1.45 (1.23–1.72)rs9931494G/C643634949217150494215171.35 (1.18–1.55)18.40.000018a5.60.0181.52 (1.07–2.15)16.90.000039a1.42 (1.20–1.67)rs17817964T/C623615009236852493015221.30 (1.14–1.49)13.50.00022 5.80.0161.54 (1.08–2.19)11.40.000751.33 (1.13–1.57)rs9930506G/A673654889208252191315161.28 (1.12–1.46)12.80.00038 3.50.0611.37 (0.98–1.92)12.10.000511.34 (1.14–1.58)rs9932754C/T663684919257852591915221.29 (1.13–1.48)13.60.00023 4.20.0401.42 (1.01–2.00)12.60.000401.35 (1.14–1.59)rs9922619T/G663684899237852991915261.29 (1.13–1.48)13.50.00024 4.30.0381.43 (1.02–2.01)12.30.000441.34 (1.14–1.58)rs7204609C/T13441837392527371752915190.83 (0.73–0.93)9.680.0022 5.00.0250.77 (0.62–0.97)7.50.00630.79 (0.67–0.94)rs12149832A/G533495259276248098215241.33 (1.15–1.53)15.20.000098a3.50.0611.43 (0.98–2.08)14.80.00012a1.39 (1.17–1.64)The odds ratio (OR) for each SNP was adjusted simultaneously for age and gender using additive modelCI confidence interval, χ2 chi-squareaSignificant P value (P < 0.00017) Analysis of various quantitative phenotypes with rs1558902 To investigate whether the genotypes of SNP rs1558902 are associated with the phenotypes of metabolic disorders, we compared the following among the different genotypes in the cases, controls, and combined groups: ANOVA results, BMI, levels of fasting plasma glucose, hemoglobin A1c (HbA1c), total cholesterol, triglycerides, HDL cholesterol, and blood pressure. As rs1558902 showed the most significant association with obesity and its call rate was the highest, we analyzed various quantitative phenotypes by using this SNP. The quantitative phenotypes regarding BMI and the levels of fasting plasma glucose, HbA1c, total cholesterol, triglycerides, HDL cholesterol, and blood pressure were not found to be significantly associated with the genotypes at rs1558902 in either the case or control group (Table 3). Although there was no significant difference in BMI values among genotypes in either the control or case group, the direction of the difference (AA > AT > TT) was in accordance with the association between the qualitative obesity phenotype and the genotype shown. Table 3Comparison of various quantitative phenotypes among different genotypes at single nucleotide polymorphism (SNP) rs1558902 in obese and control subjectsObeseControlAA (n = 48)AT (n = 341)TT (n = 536)AA (n = 52)AT (n = 448)TT (n = 1022)Age (year)49.8 ± 15.349.6 ± 14.348.8 ± 14.146.9 ± 15.446.9 ± 16.748.8 ± 16.5P value0.640.098BMI (kg/m2)35.16 ± 5.7034.61 ± 5.4334.39 ± 5.3321.94 ± 2.23 21.62 ± 2.1021.65 ± 2.06P value0.580.56Glucose (mg/dl)142.8 ± 54.8125.4 ± 43.2130.8 ± 53.3101.7 ± 44.196.3 ± 18.198.2 ± 24.7P value0.0540.34HbA1c (%)6.9 ± 2.16.4 ± 1.76.5 ± 1.85.1 ± 1.25.0 ± 0.55.1 ± 0.7P value0.190.15Total cholesterol (mg/dl)215.1 ± 46.7211.3 ± 38.8208.6 ± 36.6195.6 ± 38.8201.4 ± 37.8201.4 ± 35.6P value0.370.53Triglycerides (mg/dl)171.7 ± 119.5151.3 ± 102.1153.2 ± 96.0111.7 ± 70.6102.0 ± 71.4104.4 ± 74.2P value0.420.63HDL cholesterol (mg/dl)53.2 ± 13.854.8 ± 24.052.0 ± 15.462.1 ± 14.265.1 ± 15.965.3 ± 15.6P value0.140.53SBP (mmHg)134.2 ± 20.4137.0 ± 17.8136.2 ± 18.2122.7 ± 17.3123.2 ± 18.8123.5 ± 17.5P value0.610.91DBP (mmHg)80.3 ± 11.784.1 ± 12.083.9 ± 12.075.5 ± 11.175.2 ± 11.776.3 ± 10.9P value0.140.22Data of each quantitative phenotype were compared among different genotypes at the rs1558902 in obese and control subjects. P values were analyzed using analysis of variance in each group of obese and control subjects. Data are mean ± standard deviation.HDL high-density lipoprotein, SBP systolic blood pressure, DBP diastolic blood pressure. Finally, we examined the BMI distribution of rs1558902 in the Japanese general population and found that rs1558902 genotype was significantly associated with BMI (Table 4). This result would confirm the association of rs1558902 with obesity. Table 4Association of body mass index (BMI) with rs1558902 genotypes in the Japanese general populationAAATTTP value (additive model)aBMI (kg/m2) (n)23.17 ± 3.20 (59)22.79 ± 3.26 (482)22.57 ± 3.11 (1063)0.041aThe difference in BMI according to genotypes was analyzed using a multiple linear regression, with BMI as the dependent variable and genotype as the independent variable and with gender and age as covariates for BMI. Data are represented as mean ± standard deviation Discussion Recent genome-wide association studies have shown that the FTO gene is associated with obesity (Frayling et al. 2007; Scuteri et al. 2007; Hinney et al. 2007). The associations between variations in the FTO gene and the obesity phenotype have been observed in many Caucasian subjects (Frayling et al. 2007; Scuteri et al. 2007; Dina et al. 2007; Field et al. 2007; Andreasen et al. 2008; Wåhlén et al. 2008; Peeters et al. 2008). However, these associations were controversial with regard to Asian subjects (Horikoshi et al. 2007; Li et al. 2008; Omori et al. 2008). BMI values did not significantly differ among the genotypes in the general population of Chinese and Japanese (Horikoshi et al. 2007; Li et al. 2008). We performed a case-control association study with regard to severe obesity and found that the SNPs in the FTO gene were significantly associated with severe obesity. Although the SNPs demonstrated the most significant association in the Japanese, which was different from that in Caucasians, the significantly associated SNPs existed in a similar block as that in Caucasians. Therefore, the FTO gene could also contribute to the development of severe obesity in the Japanese. BMI was modestly different among rs1558902 genotypes in the general population in this study; rs9939609 was not significantly associated with BMI in the general population (AA 23.22 ± 3.14 vs AT 22.79 ± 3.25 vs TT 22.58 ± 3.13, P = 0.063). In the Japanese population, rs1558902 may be more tightly associated with BMI than rs9939609. The National Nutrition Survey of Japan reported that the prevalence of subjects with a BMI of ≥30 kg/m2 is only 2.3% in men and 3.4% in women aged 20 years and older (Yoshiike et al. 2002), and the mean BMI was approximately 23 kg/m2 for ages 15–84 years (Yoshiike et al. 1998). Inconsistency in the results of effects of variations in the FTO gene on BMI between Japanese and Europians may be due to the relatively small mean and variance of BMI in the former than the latter. The significant SNPs were located in intron 1 of the FTO gene. The rs1558902 and other significant SNPs, for example, rs9939609 and rs1121980, would affect transcriptional activity of the FTO gene, although further investigation is necessary. The precise mechanism by which the FTO gene leads to obesity development is unclear (Gerken et al. 2007; Sanchez-Pulido et al. 2007). However, the FTO gene is expressed in the hypothalamus and regulated by fasting and leptin (Frayling et al. 2007; Gerken et al. 2007). Using large-scale case-control association studies, we determined that the SCG3 (Tanabe et al. 2007) and MTMR9 (Yanagiya et al. 2007) genes are involved in susceptibility to the obesity phenotype. These two genes are expressed in the hypothalamus. Genetic studies in mice have suggested that mutations in several genes, such as those encoding leptin, proopiomelanocortin, and melanocortin-4 receptor, are implicated in a monogenic form of inherited obesity (Barsh et al. 2000; Rankinen et al. 2006). Such mutations have also been reported in obese humans. As most such genes are expressed in the hypothalamus and have been indicated to play important roles in the regulation of food intake, genes expressed in the hypothalamus are likely to be good candidates for susceptibility to obesity. In summary, we have identified the genetic variations in the FTO gene that may influence the risk of severe obesity in the Japanese. Electronic supplementary material Below is the link to the electronic supplementary material. Summary of the association of SNPs between cases and controls (DOC 315 kb)

Cancer_Chemother_Pharmacol-4-1-2270367

Cisplatin-DNA adduct formation in patients treated with cisplatin-based chemoradiation: lack of correlation between normal tissues and primary tumor

Purpose In this study, the formation of cisplatin-DNA adducts after concurrent cisplatin-radiation and the relationship between adduct-formation in primary tumor tissue and normal tissue were investigated. Introduction Concurrent chemoradiotherapy is more effective than radiotherapy (RT) alone, both in in vitro studies [1, 2] as well as in clinical studies in many different tumor types, including advanced head and neck squamous cell carcinoma (HNSCC) and cervical cancer, leading to improvements in locoregional control and/or survival [11, 13, 21, 28]. In a metaanalysis on concurrent chemoradiation in HNSCC, the addition of concurrent single agent cisplatin to RT was the most effective treatment regime with the largest improvement on overall survival [5]. Concurrent cisplatin-based chemoradiation is now considered standard care in advanced-stage HNSCC and cervical cancer. In addition to the increased efficacy of the combined treatment, it was shown that the concurrent regimens are accompanied by higher acute toxicity rates compared to radiation alone [11, 21], with more severe mucositis and gastrointestinal toxicity. Since a substantial number of patients treated with concurrent chemoradiation still fail to respond to this toxic treatment, there is a need for an accurate predictive assay, based on which patients that are likely to respond to the therapy can be selected. This strategy may also provide a tool to individualize and tailor treatment, based on evaluation of the predictive assay, early during therapy. One potential predictive marker is the formation of cisplatin-DNA adducts, which are formed when cisplatin reacts with the cellular DNA by binding to nucleotides. The majority of adducts are either intrastrand adducts with cisplatin bound between two guanosine (GG) nucleotides or adenosine–guanosine (AG) nucleotides [8]. Cisplatin-DNA adducts can be measured in tumor and normal tissue. The level of adducts has been shown to correlate with cytotoxicity in vitro [34], and with response to therapy in patients [3, 17, 32, 35]. In most of these studies [3, 32, 35], adduct measurements were performed in the normal tissue, with the assumption that normal tissue can be used as surrogate marker for tumor. In our institute, study protocols with cisplatin-DNA adduct measurements are ongoing in patients with HNSCC and cervical cancer, who are all treated with concurrent cisplatin-based chemoradiation. The objectives of the current study are (1) to investigate the two major forms of cisplatin-DNA adducts (GG and AG adducts) after different schedules of cisplatin given concurrently with radiation and (2) to explore relationships between adducts in primary tumor and normal tissue. We specifically wanted to investigate whether the level of adducts in tumors are reflected by those in normal tissues. In studies focused on the predictive value of cisplatin-DNA adduct levels, this would then justify the use of more easily obtained normal tissues as a surrogate for tumor samples. Patients and methods Concurrent chemoradiation protocols This study on adduct formation was approved by the medical ethical committee of the participating hospitals. The main eligibility criteria were: patients scheduled for cisplatin chemoradiation, no previous treatment with cisplatin, and informed consent. Eligible patients were informed about the nature of the protocol and after written informed consent they were entered in the study. Patients were recruited from one of the following regimens. In advanced-stage HNSCC patients, two different cisplatin-based concurrent chemoradiation protocols (RADPLAT) were used. The RADPLAT 100 schedule, which is the most commonly administered schedule in HNSCC [11], consisted of cisplatin given intravenously (IV) at a dose of 100 mg/m2, as a 30 min infusion, 1–2 h before RT at days 1, 22, and 43 of treatment. This treatment was part of a randomized trial on IV vs. intra-arterial chemoradiation. In RADPLAT daily LD, low-dose (LD) cisplatin was given as a 1–2 min IV infusion at a dose of 6 mg/m2 daily, for a total number of 20 doses, 1–2 h prior to RT. This treatment was shown to be an effective alternative in HNSCC [6, 19]. Patients ineligible for or refusing the randomized trial on intra-arterial chemoradiation were treated with RADPLAT daily LD, since this treatment could be given on an outpatient basis. The RT target volumes for all schedules included the primary tumor and the bilateral neck at a dose of 46 Gy in 23 fractions. A boost was given to the macroscopic tumor extensions at the primary tumor site and lymph node metastases at a dose of 24 Gy in 12 fractions, resulting in a total dose of 70 Gy in 35 fractions. In patients with advanced-stage squamous cell cervical cancer, concurrent chemoradiation (CERVIX 40) consisted of weekly administration of cisplatin IV as a 4-h infusion at a dose of 40 mg/m2, followed by RT within 1–2 h. The total number of doses was 5–6, depending on external beam RT schedule and the number of intracavitary brachytherapy applications. The total radiation dose was usually 46 Gy to the cervical tumor, uterus, and pelvic lymph nodes, with a boost to the cervix tumor and other involved regions, to a total dose of 60–74 Gy, depending on treated volume and whether or not intracavitary brachytherapy was given. Cisplatin-DNA adducts Before and after chemotherapy, normal tissue samples [white blood cells (WBC) and buccal cells] were collected. In patients with an accessible primary tumor, a biopsy of the tumor was also taken. To avoid harvesting necrotic tissue, the biopsy was taken at the viable peripheral rim of the tumor. Samples were obtained at different times, due to logistic reasons: for the patients in the RADPLAT 100 study, this was done 23 h after the end of administration of the first cisplatin infusion (given on day 1 of treatment). In the patients in the CERVIX 40 study, samples were taken 20 h after the end of administration of the first weekly cisplatin infusion (given on day 1 of treatment). For the patients in the RADPLAT daily LD group, samples were taken 1 h after the 5th dose on day 5 of treatment. WBC were isolated from whole blood samples according to a previously published protocol [24]. Buccal cells were collected in phosphate-buffered saline by scraping the bilateral buccal mucosa using a cotton swab. In HNSCC patients, the buccal mucosa could be located within the RT treatment fields, depending on the tumor site. The harvested cells were centrifuged (5 min at 4°C, 1,000 rpm) and resuspended in a Tris–EDTA buffer and stored at −80°C until analysis. Tumor biopsies were taken and immediately frozen at −80°C until analysis. Quantification of GG- and AG-intrastrand adducts was performed by a 32P-postlabeling technique as previously described [29]. Internal standardization was incorporated in the present analysis method, by adding 300 fmol of TT nucleotides to each sample. From previous work, the reproducibility of the assay is known by analysis of duplicate specimens within the same experiment (within-run reproducibility) and by analysis of duplicate specimens in separate experiments (between-run reproducibility) [29]. The reproducibility was described for WBC and tumor samples and was within 10% for the within-run precision and between 2–20% for the between-run precision. It was also determined for buccal cells in the same way and similar reproducibility was obtained. The concentration of DNA present in the samples was measured spectrophotometrically at 260 nm with the Nanodrop ND-1000 (Nanodrop Technologies Inc, Wilmington, DE, USA). The cisplatin-DNA adduct levels were expressed as fmol/μg DNA. The lower limit of quantification for the Pt-GG and Pt-AG adducts was 0.087 and 0.053 fmol/μg DNA, respectively. Statistical analysis Analysis of data was performed in SPSS software (version 11.5, SPSS, Inc.). For quantitative comparison of numerical data between groups, the Student’s t-test was applied. The Pearson correlation coefficient and Spearman’s rank correlation coefficient were calculated for analysis of correlations between different samples (WBC, buccal cells, and primary tumor) on an intra-patient level. Results Samples for cisplatin-DNA adduct determination were obtained from 63 patients: 27 from RADPLAT daily LD, 15 from CERVIX 40, and 21 from RADPLAT 100. WBC samples were taken from 61 patients, buccal cells from 25, and tumor biopsies from 23 of these patients. The reasons for the missing data for the normal tissue samples were: no collection of samples due to logistics or (in minority of cases) not sufficient volume for analysis. The reason for missing primary tumor biopsy data were: tumor not accessible for direct outpatient-based biopsy (in HNSCC patients) or refusal (in cervix cancer patients). In WBC, all but three of the 60 available baseline samples were below the LLQ for the GG adducts and all but two below the LLQ for the AG adducts. This was probably due to some background signal inherent in the postlabeling method, since all patients had not been treated before with platinum chemotherapy. The yield of DNA, obtained from the buccal cell samples was rather low, ranging from 1–10 μg. Baseline samples of buccal cells were available from 16 of 25 patients, of whom posttreatment samples were also available. The baseline values of GG adducts in buccal cells ranged from 0.067 to 0.745 fmol/μg DNA (mean 0.282, SD 0.19) and baseline values of AG adducts in buccal the cells ranged from 0.087 to 1.538 fmol/μg DNA (mean 0.398, SD 0.38). All but two of the baseline GG-adduct values were above the LLQ and all the baseline AG-adduct values were above the LLQ. This was probably due to the low DNA quantities obtained from the buccal cell samples. The difference in adduct levels from baseline to post-infusion values was significant for the GG adducts, but not for the AG adducts. This implies that for measuring low quantities of AG adducts in the low amounts of buccal cell DNA available, we reached the limits of quantification with this postlabeling method. In Table 1, the results are presented for the GG- and AG-adduct levels in normal tissue and primary tumor for the three different cisplatin-chemoradiation regimes. Adduct levels in primary tumor were two to five times higher than those in WBC for all three treatment regimes for both GG- and AG-adduct formation (Student t-test, P < 0.001 for both adduct types). For the comparison of the adduct levels from the three different treatment protocols, the data from the RADPLAT daily LD were omitted, since the daily administration schedule and sampling time (1 h after the infusion) were different from the other two regimens. The adduct levels in the RADPLAT 100 schedule were statistically significantly higher than those after the CERVIX 40 schedule for both tumors (Student t-test, P = 0.01) and normal tissues (Student t-test, P < 0.01). Table 1Cisplatin-DNA adducts (in fmol/μg DNA) in normal tissue and primary tumor after different schedules of cisplatin-based chemoradiationTreatment scheduleWBCBuccal cellsTumorGGAGGGAGGGAGRADPLAT daily LDN2626111166 Mean0.340.050.840.210.660.10 SD0.100.190.390.140.370.05CERVIX 40N1414771010 Mean0.440.060.870.221.940.26 SD0.170.030.260.061.470.26RADPLAT 100N21217777 Mean1.0470.1241.5630.3403.8660.413 SD0.3770.0490.4340.0901.1010.089See text for explanation of treatment schedulesN number of patients, SD standard deviation, WBC white blood cells, GG GG-adducts, AG AG adducts A highly significant linear correlation (Pearson correlation, r = 0.93, P <0.001, n = 61) was observed between the level of GG and AG adducts in WBC (see Fig. 1a), with a mean ratio of GG/AG adducts of 7.6 ± 2.1 SD. Similar linear relationships and ratios were found for GG and AG adducts in primary tumor (r = 0.89, P < 0.001, n = 23, and ratio 8.5 ± 2.8; Fig. 1b) and buccal cells (r = 0.85, P < 0.001, n = 23, and ratio 3.9 ± 1.3; Fig. 1c). Fig. 1Correlation-plots of GG- and AG-adduct levels in white blood cells (WBC) (panel a), primary tumor biopsy (panel b), buccal cells (panel c). In each panel, the three different treatment groups are depicted: RADPLAT daily LD 5 × 6 mg (red squares), CERVIX 40 mg (green circles), and RADPLAT 100 mg (blue triangles) A trend was observed between GG-adduct levels in WBC and buccal cells, although not significant (r = 0.38, P = 0.07, n = 24). No significant correlations were found between tumor and normal tissue: tumor vs. WBC (r = 0.35, P = 0.13, n = 21) and tumor vs. buccal cells (r = −0.003, P = 0.99, n = 9). See Fig. 2 for scatter plots. Similar results were found for the AG adducts: no significant correlations were found between adducts in tumor vs. WBC (r = 0.14, P = 0.55, n = 21), tumor vs. buccal cells (r = 0.25, P = 0.58, n = 7) or WBC vs. buccal cells (r = 0.26, P = 0.24, n = 22). Fig. 2Correlation plots of GG adducts in white blood cells (WBC) vs. buccal cells (panel a), and normal tissue vs. tumor (panels b and c). In each panel, the three different treatment groups are depicted: RADPLAT daily LD 5 × 6 mg (red squares), CERVIX 40 mg (green circles), and RADPLAT 100 mg (blue triangles) Discussion There are two main conclusions from the 63 patients included in these analyses. First, intra-tumoral adduct levels were substantially higher than those in normal tissue (WBC) at all cisplatin-dose levels examined. Second, no positive correlations were evident between adducts in tumors and normal tissues. It should be noted that the various schedules, the cisplatin doses, and the duration of infusions differed, as well as the sampling times. However, all analyses on adducts were performed on paired samples, within the same patient. This eliminates variance of these factors since the normal tissue and tumor samples all were obtained at similar time points after the cisplatin infusion. In the RADPLAT daily LD, some accumulation from the previous four daily 6 mg cisplatin infusions would have occurred and affected the day 5 measurement after the 5th infusion. In the RADPLAT 100 and CERVIX 40 patients, no such accumulation would have occurred, since the sampling was done 20–23 h after the first infusion of cisplatin. Relatively little information is available regarding the in vivo formation of intra-tumoral cisplatin-DNA adducts in clinical series [17, 23]. Most studies focused on intra-tumoral platinum concentrations, both in HNSCC [12, 33] and cervical cancer patients [15, 22]. These studies are mostly characterized by relatively low numbers of patients, probably due to the invasive nature of the procedure. The data on correlations between adducts and platinum content are contradictory: In an experimental study [37], no relationship could be established between the intra-tumoral adduct levels and platinum content, although in one clinical study, a significant correlation was found [23]. Adduct levels in primary tumors were consistently two- to fivefold higher than in WBC. This was true for both GG and AG adducts. This finding was previously described in anecdotal clinical cases [9, 30]. Similar observations were made in platinum content studies in an experimental tumor model [18] and in HNSCC patients [33]. Adducts in tumor were also higher than in buccal cells in the CERVIX 40 and RADPLAT 100 group, but not in the RADPLAT daily LD group. We observed a linear relationship between the two major adduct forms (AG and GG) for both normal tissues and tumor (Fig. 1a–c), although GG-adduct formation was 5–12 times increased relative to the AG adducts, as reported earlier [8, 36]. The linear relationships between GG and AG adducts serve as a validation of the assay. Apparently within a sample, both types of adducts are present in an equal proportion, although the absolute amounts differ greatly. In previous studies, the type of adduct responsible for the cytotoxic effect of platinum compounds has been investigated. From two of these studies [7, 36] it was concluded that the AG adduct was responsible for the platinum cytotoxicity. From our study, such a conclusion cannot be made, since both types of adducts were present in equal proportions. Adduct formation in different tissue samples showed a lack of correlation between tumor and normal tissue. One might have expected that higher adduct levels in normal tissue would be accompanied by more adducts in tumor, although this was not the case. Similar results were found in an animal study [25] and in a clinical series of uterine cervix cancer patients [22], both demonstrating lack of correlation between tumor platinum and serum platinum concentrations. If adduct formation were merely a matter of cisplatin exposure, then a positive correlation would be expected. The reasons for the higher levels of adducts in tumor vs. normal tissue and the lack of correlation between them may be explained by differences between tumor and normal tissue in one of the following factors: Tumors are heterogeneous in terms of blood supply and perfusion, resulting in differences in cisplatin uptake and diffusion, drug-pumps may diminish intra-cellular cisplatin concentrations by active transmembrane transport of cisplatin, prohibiting adducts to be formed, and tumor cells may have less effective capacities to repair damage from cytotoxic agents. Adducts are formed rapidly and in a dose-dependent fashion within 1–2 h after cisplatin exposure, with a gradual decrease (repair) within the next 20–24 h [20, 32, 36]. The persistence of cisplatin-DNA adducts may therefore be regarded as a measure of repair and possibly be used as predictive assay. We therefore chose to measure adducts 20–23 h after the end of chemotherapy infusion. Ideally, more frequent measurements would have generated more information on the rate of adduct formation, repair, and total exposure to adducts (like an AUC analysis) [24]. However, obtaining repeated biopsies is not feasible in clinical practice. The results from the buccal cell samples need to be interpreted with caution, especially the AG-adduct levels, since uncertainties remain. With the low quantities of AG-adducts in the low amounts of buccal cell DNA we could extract, we reached the limits of quantification of the postlabeling method. The rationale for measuring cisplatin-DNA adducts is that it could be used as a predictive assay: higher levels of adducts would predict favorable treatment outcome. Many studies have been performed for this purpose, investigating adducts in normal tissue (WBC and buccal cells) [3, 4, 10, 14, 26, 31, 32, 35]. These study designs are based on the assumption that normal tissue can be used as a surrogate marker for tumor tissue with respect to cisplatin-DNA adduct formation. In our present study, however, we did not find such a correlation. This might explain why the results on the role of adduct formation to predictive outcome are heterogeneous and contradictory, as illustrated below. Several studies on adduct formation in WBC showed that the level of adducts was positively correlated with response to chemotherapy in patients with advanced disease in a variety of tumor sites [31, 32], while others showed no correlation [4, 26]. One study showed a positive correlation in one tumor site (ovarian cancer), but not in the other (breast cancer) [14], and in another study the level of adduct formation showed a negative association with survival for day-5 adducts, while there was no difference for day-1 adducts [10]. In studies on adduct levels in buccal cells, a positive correlation was found between adducts and either disease response [3] or better survival in non-small cell lung cancer (NSCLC) [35]. We recently showed that adduct formation in primary tumor appeared to be associated with better progression-free survival in HNSCC [17]. Differences were observed between the levels of intra-tumoral adducts for the three chemoradiation schedules, with lower adducts after lower dosages of cisplatin. Based on this, however, one cannot predict that the schedules with less adducts will result in less cytotoxicity, since not only the cisplatin dose, but also timing and schedule of cisplatin administration are crucial determinants of efficacy [1]. These factors may contribute to differences in the formation and rate of repair of adducts, resulting in different exposures to cisplatin-DNA adducts. These differences make it difficult to extrapolate from the observed adduct values in tumor and normal tissues to a prediction of superiority of one schedule over another. In future studies, immunohistochemistry on repair proteins like ERCC1 [27] or gene expression profiling studies on platinum resistance [16] could be used to improve the prediction of cisplatin sensitivity and prediction of therapy response. In conclusion, we have demonstrated that in concurrent chemoradiotherapy schedules, cisplatin adduct levels in tumors were significantly higher than in normal tissues (WBC). No evidence of a correlation was found between adduct levels in normal tissues and primary tumor biopsies. This lack of correlation may, to some extent, explain the inconsistencies in the literature regarding whether or not cisplatin-DNA adducts can be used as predictive test in anticancer therapy. Conflict of interest: none declared.

Int_Urol_Nephrol-4-1-2268719

Magnesium carbonate for phosphate control in patients on hemodialysis. A randomized controlled trial

Background Magnesium salts bind dietary phosphorus, but their use in renal patients is limited due to their potential for causing side effects. The aim of this study was to evaluate the efficacy and safety of magnesium carbonate (MgCO3) as a phosphate-binder in hemodialysis patients. Introduction Normalization of the serum phosphate level is the cornerstone of medical protocols aimed at preventing and treating secondary hyperparathyroidism. Moreover, it has been shown that elevated serum levels of phosphate and the calcium–phosphate product (Ca × P) play an important role in the development of extraosseous calcifications and are associated with increased mortality in hemodialysis patients [1, 2]. The ingestion of phosphate-binding agents in conjunction with dietary phosphate restriction and its adequate removal by dialysis are the cornerstones of serum phosphate control in end-stage renal disease (ESRD) patients. However, four decades after the introduction of chronic hemodialysis in early 1960s, we have not yet found the ideal phosphate binder(s) in terms of combined efficacy, safety and low cost. Aluminum and calcium salts and non-aluminum and non-calcium agents, such as sevelamer and lanthanum carbonate, all have advantages and disadvantages [3]. Magnesium-containing agents are aluminum- and calcium-free and are inexpensive phosphate-binders and, to date, their potential has not been explored extensively. In studies involving small series of both hemodialysis and peritoneal dialysis patients, magnesium carbonate (MgCO3) and magnesium hydroxide [Mg(OH)2] have been administered either alone or in combination with calcium salts with good results [4–10]. However, these compounds are not widely used in ESRD patients because nephrologists have an inordinate fear of hypermagnesemia and the belief that Mg administration frequently is accompanied by gastrointestinal disorders. We carried out this study in hemodialysis patients to evaluate the efficacy and safety of MgCO3 as a phosphate-binder when given with a concurrent low dialysate magnesium solution. The control of serum phosphate in the two groups of patients was the primary outcome, while secondary outcomes were changes in serum calcium, magnesium, Ca × P and PTH levels and changes in bowel movements. Subjects and methods Patients and study design Stable ESRD patients on maintenance hemodialysis in our renal unit participated in the study. The protocol of this project was approved by the Ethics Committee of the hospital. Exclusion criteria were: age <18 years, hemodialysis for less than 6 months, psychiatric or other disorders leading to non-compliance, unlikeliness to continue hemodialysis for more than 6 months in the same facility, critical illness at the time of recruitment, previous parathyroidectomy, severe hyperparathyroidism [serum intact parathyroid hormone (iPTH) > 500 pg/ml], normal serum phosphate (<5.5 mg/dl) without phosphate-binders, diseases resulting in diarrhea and the lack of informed consent. A total of 54 patients met the criteria and were approached for enrolment. The inclusion period lasted 2 months, between March 2 and April 29, 2006. Enrolled patients signed an informed consent and thereafter entered a 4-week washout period during which we withdrew all phosphate-binders as well as vitamin D medications. After the washout phase, three patients were removed because they had serum phosphate levels lower than 5.8 mg/dl; the remaining 51 patients were allocated to receive either MgCO3 tablets (MgCO3 group) or calcium carbonate tablets (CaCO3 group) for 6 months. We chose CaCO3 instead of calcium acetate because the latter is not commercially available in Greece. Four patients did not agree to consume MgCO3 but did agree to participate by taking their standard binder of CaCO3; we allocated these patients to the CaCO3 group, whereas the remainder of the patients were randomly allocated with a ratio 1:1 to either the CaCO3 (21 randomly assigned patients plus 4 = 25 patients) or to the MgCO3 (26) group. Each MgCO3 tablet contained 250 mg MgCO3, which is equivalent to 71 mg of elemental magnesium, and each CaCO3 tablet contained 420 mg of CaCO3 equivalent to 168 mg of elemental calcium. All patients were on the same standard dialysis schedule (three times weekly × 4 h), and in all of the patients the delivered dose of hemodialysis, as calculated by applying the single pool Kt/Vurea index using the second-generation formula of Daugirdas [11],, was ≥1.35. In both groups, calcium concentration in the dialysate bath was 1.50 mmol/l, whereas magnesium concentration was 0.48 mmol/l in the CaCO3 group and 0.30 mmol/l in the MgCO3 group. The National Kidney Foundation (NKF) suggests a bath calcium dialysate concentration of 1.25 mmol/l (based on opinion, not on evidence). We used a bath concentration 1.50 mmol/l because we wanted to avoid episodes of hypocalcemia, particularly in the MgCO3 group. The patients should be more susceptible to manifest hypocalcemia since they did not receive vitamin D. The starting dose of both phosphate-binders were three tablets of MgCO3 or CaCO3 daily; the dose was adjusted thereafter according to the serum phosphate values: weekly for the first month and then monthly. The dosage of the study drug was increased to one or two tablets per meal as required to achieve the target of serum phosphate level ≤5.5 mg/dl. Based on our clinical experience, to achieve adequate control of serum phosphate we set the maximum daily dose of CaCO3 to 3780 mg (nine tablets), which is equivalent to 1512 mg of elemental calcium, and the maximum daily dose of MgCO3 to 2250 mg (nine tablets), which is equivalent to 639 mg of elemental magnesium. Each patient’s full biochemical profile was obtained at baseline, then weekly during the first month and monthly thereafter. Parathyroid hormone was measured at baseline and then at monthly intervals. If a patient developed hypercalcemia (serum Ca > 10.5 mg/dl), the daily dose of CaCO3 was reduced by one or two tablets. The same was done in patients with severe hypermagnesemia, which is defined as a serum magnesium level >3.5 mg/dl. If severe hypermagnesemia persisted for more than 3 weeks, the administration of MgCO3 was stopped, and the patient was dropped from the study. No patient received vitamin D or a calcimimetic agent during the study. Serum total calcium values were corrected to the serum albumin values. Statistical analysis The laboratory values were expressed as mean [± one standard deviation (SD)]. Student’s two-tailed unpaired t test was used to compare values between the two groups of patients at baseline and at 6 months. We used repeated measures of analysis of variance (repeated ANOVA method) to test for differences between the two treatment groups in average values of calcium, phosphate, Ca × P, magnesium, alkaline phosphatase (ALP) and PTH over time (time-averaged mean value differences). The chi-square test was applied to determine differences in the prevalence of laboratory data. Significance was set at a P level of <0.05. Results Of the 26 patients enrolled in the MgCO3 group, one dropped out due to non-compliance. Of the remaining 25 patients, two (8%) discontinued ingestion of MgCO3 and dropped out: one (4%) because of persistent diarrhea, and one (4%) because of recurrent hypermagnesemia. Of the 25 patients in the CaCO3 group; five were removed from the study (two received a kidney transplant, one died (pneumonia), one suffered a stroke and was unable to swallow tablets and one moved to another hospital). The use of phosphate-binders and vitamin D by the patients before the washout period is given in Table 1. Patients’ mean serum values at baseline and at the end of the follow-up period are shown in Tables 2 and 3. The monthly follow-up of the mean biochemical parameters are shown in Figs. 1–4. The mean Kt/Vurea was 1.379 ± 0.026 in the MgCO3 and 1.381 ± 0.027 in the CaCO3 group. Table 1Patients’ data in terms of the use of phosphate-binders and vitamin D before the washout periodMgCO3 group (n = 25)CaCO3 group (n = 21)Vitamin D11/258/21Phosphate binders25/2521/21    CaCO31916    Sevelamer24    CaCO3 + sevelamer41Table 2Mean serum values at baselineMgCO3 groupCaCO3 groupP value (t test)Age (years)63.2365.32P = nsSD12.1911.68Calcium (mg/dl)a9.429.14P = nsSD0.540.43Phosphorous (mg/dl)6.636.58P = nsSD0.860.88Ca × P product (mg/dl)262.6260.08P = nsSD10.368.35Magnesium (mg/dl)2.382.36P = nsSD0.280.29ALP (IU/l)7686P = nsSD3735Intact PTH (pg/ml)316296P = nsSD182157SD, Standard deviation; ALP, alkaline phosphatase; PTH, parathyroid hormone; ns, not significantaSerum total calcium level corrected to serum albuminTable 3Mean values at 6 monthsMgCO3 groupCaCO3 groupP value (t test)Calcium (mg/dl)a8.979.72t = 2.16SD0.570.42P < 0.05Phosphorous (mg/dl)5.125.28t = 0.49SD0.700.74P = NSCa × P product (mg/dl)246.0451.38t = 0.30SD7.657.75P = NSMagnesium (mg/dl)2.592.40t = 1.59SD0.430.41P = NSALP (IU/l)8984t = 0.46SD2826P = NSIntact PTH (pg/ml)251212t = 0.42SD118198P = NSaSerum total calcium corrected to serum albuminFig. 1Monthly follow-up of serum phosphate (sP)Fig. 2Monthly follow-up of serum calcium (sCa)Fig. 3Monthly follow-up of serum Ca × P product (SCa × P)Fig. 4Monthly follow-up of serum PTH (siPTH) The mean daily dose of CaCO3 was 6.76 tablets (range 3–9) containing a total of 2839 mg (range 1260–3780) of CaCO3, which is equivalent to 1136 mg of elemental calcium (range 504–1512 mg) The mean daily dose of MgCO3 was 6.21 tablets (range 3–9) containing a total of 1552 mg (range 750–2,250) of MgCO3, which is equivalent to 441 mg of elemental magnesium (range 213–639 mg). A Shapiro–Wilk test did not indicate any evidence against the normality assumption for the distribution of calcium, phosphate, Ca × P, magnesium, ALP and PTH levels within the two groups (CaCO3 and MgCO3) at any time period during the follow-up. The significance level for the normality tests was set to a = 0.001. Average values of the biochemical data during months 1–6 are shown in Table 4. The initial mean serum phosphate levels were 6.63 mg/dl in the MgCO3 group and 6.58 mg/dl in the CaCO3 group (P = ns), while at the end of the study serum phosphate values were 5.13 and 5.26 mg/dl in the MgCO3 and CaCO3 groups respectively (P = ns; Tables 2, 3). Time-averaged mean serum phosphate values were 5.47 mg/dl in the MgCO3 group and 5.29 mg/dl in the CaCO3 group (P = ns; Table 4). Serum phosphate levels decreased by 23% in the MgCO3 group and by 19% in the CaCO3 group (P = ns). At the end of the study 17 of 23 (74%) patients in the MgCO3 and 13 of 20 (65%) of the CaCO3 group, (χ2 = 0.10, P = ns) had serum phosphate values within the range recommended by the Kidney Disease Outcomes Quality Initiative (K/DOQI) guidelines (upper threshold of 5.5 mg/dl) (Table 5). Table 4Average serum values during months 1–6 of the follow-up periodMgCO3 groupCaCO3 groupP value (t test)Calcium (mg/dl)a9.139.60P < 0.001SD0.530.45Phosphorous (mg/dl)5.475.29P = nsSD0.810.93Ca × P product (mg/dl)250.3550.70P = nsSD7.758.07Magnesium (mg/dl)22.572.41P = nsSD0.410.32ALP (IU/l)8880P = nsSD3227Intact PTH (pg/ml)285231P < 0.01SD161177aSerum total calcium corrected to serum albuminTable 5Patients with laboratory values within the Kidney Disease Outcomes Quality Initiative (K/DOQI) guideline range at 6 monthsMgCO3 group (n, %)CaCO3 group (n, %)χ2, P valueCalcium 17/23 (73.91)5/20 (25)χ2 = 8.42, P < 0.01Phosphorous17/23 (73.91)13/20 (65)χ2 = 0.10, P = nsCa × P product20/23 (86.95)14/20 (70)χ2 = 0.99, P = nsPTH11/23 (47.82)8/20 (40)χ2 = 0.79, P = ns Mean serum Ca values at baseline were 9.42 mg/dl in the MgCO3 group and 9.14 mg/dl in the CaCO3 group (P = ns) (Table 2). Time-averaged mean serum values were 9.13 mg/dl in the MgCO3 group and 9.60 in the CaCO3 group (P < 0.001) (Table 4). At 6 months, these values were 8.97 and 9.72 mg/dl, respectively (P < 0.05) (Table 2). Seventeen patients (17/23; 74%) who received MgCO3 and only 5/20 (25%) of those on CaCO3 had serum Ca values within the K/DQOI guidelines ( χ2 = 8.42, P < 0.01) (Table 5). We encountered six episodes of hypercalcemia, defined as a serum calcium level >10.5 mg/dl, in the CaCO3 group in comparison to one episode in the MgCO3 group. The two groups of patients had nearly equal values of serum Ca × P product at the beginning, at the end as well as during the period study (Tables 2–4), with the MgCO3 group having 62.6 mg2/dl2 and the CaCO3 group having 60.1 mg2/dl2 (P = ns) (Table 1). Mean levels of Ca × P during months 1–6 of the follow-up were similar – 50.35 mg2/dl2 in the MgCo3 group and 50.70 mg2/dl2 in the CaCo3 group (P = ns) (Table 4); at the end of the study, the corresponding values were 46.0 and 51.4 mg2/dl2 (P = ns) (Table 3), whereas 20/23 (87%) and 14/20 (70%) patients, respectively, had Ca × P product below the upper limits of K/DQOI recommendations (P = ns) (Table 5). Starting mean levels of serum iPTH were similar in both groups – 316 ± 182 pg/ml in the MgCO3 group and 296 ± 157 pg/ml in the CaCO3 group (P = ns) (normal range 10–65 pg/ml) (Table 2). During the 6-month follow-up period, average iPTH levels were significantly higher in the MgCO3-treated patients (285 pg/ml) than in those treated with CaCO3 (231 pg/ml) (P < 0.01) (Table 4). However, at the end of the study, those levels did not differ significantly – 251 ± 118 and 212 ± 198 pg/ml, respectively (P = ns) (Table 3). In terms of PTH levels, there was no statistically significant difference between the CaCO3 group and the MgCO3 group – 29 versus 21%, respectively (χ2 = 0.11, P = ns). Moreover, ten of 23 (43%) patients in the MgCO3 group and nine of 20 (45%) in the CaCO3 group had initial levels of iPTH above 300 pg/ml; in three of these ten patients (30%) of the MgCO3 group and five of the nine (56%) of the CaCO3 group, the iPTH decreased below 300 pg/ml [χ2 = 1.38, P = ns; 11/23 (48%)]. Eight of 20 patients (40%) in the CaCO3 group and six of 23 patients (26%) in the MgCO3 group (P = ns) had final values of iPTH below 150 pg/ml. The number of patients of both groups, patients who received MgCO3, and eight of 20 (40%) patients who received CaCO3 who had serum phosphorus, calcium, Ca × P product and iPTH values at 6 months that fell within the range recommended by the K/DOQI guidelines are shown in Table 5 (χ2 = 0.79, P = ns). During the follow-up period the average serum Mg levels were slightly – but not significantly – higher in the MgCO3 group than in the CaCO3 group: 2.57 vs. 2.41 mg/dl (P = ns) (Table 4). Similarly, at 6 months, these values were 2.59 and 2.40 mg/dl (P = ns) (Table 3). One patient of the 25 (4%) stopped taking MgCO3 because of recurrent high levels of serum magnesium (>3.5 mg/dl, the upper threshold according to our protocol). Two more patients manifested a transient elevation of serum magnesium – from 3.18 to 3.36 mg/dl. No patient in the CaCO3 group had a serum magnesium >3 mg/dl. Discussion This is the first study that compares MgCO3 with only one other phosphate-binder in hemodialysis patients. Our results show that MgCO3 administered for a period of 6 months has a good phosphate-binding ability, is well tolerated by most patients and is accompanied by a low incidence of side effects. O’Donovan et al. [4] described 28 patients on hemodialysis who were given magnesium chloride in place of oral aluminum hydroxide. These patients were also switched from a dialysate containing 0.85 mmol/l magnesium to one not containing any magnesium at all. After 24 months of treatment on this regimen, serum phosphate was effectively controlled in these patients. The researchers saw no evidence of increased secondary hyperparathyroidism. Delmez et al. [5] conducted a 10-week, prospective, randomized crossover study of 15 hemodialysis patients, who were on MgCO3 with a dialysate magnesium concentration of 0.25 mmol/l; with this regimen, these investigators were able to reduce the CaCO3 dose and use a higher dose of calcitriol. Mean serum phosphate levels in the patients were similar with the MgCO3/CaCO3 combination as with the CaCO3 treatment alone (5.7 ± 0.2 vs. 5.2 ± 0.2 mg/dl, respectively), and there were no adverse gastrointestinal effects. Parsons et al. [6] described 32 continuous cycling peritoneal dialysis (CAPD) in patients who were dialyzed with magnesium-free dialysate and used a mixture of CaCO3 and MgCO3 as a phosphate-binder for over 1 year. These patients achieved satisfactory control of hyperparathyroidism, with normal serum calcium, phosphorous and magnesium concentrations. Other investigators have also shown that the use of magnesium-containing agents as phosphate-binders can provide effective control of serum phosphate and hyperparathyroidism and that serum magnesium levels remain within acceptable ranges [7–10]. Spiegel et al. [12] recently compared a regimen of MgCO3 and CaCO3 versus one with calcium acetate and found that the two were equally effective as phosphate-binders. In our study, MgCO3 was well tolerated by most of our patients – only two of the 25 patients (8%) of this group were removed from the study, one for diarrhea and the other for hypermagnesemia. A third patient complained initially of mild diarrhea and abdominal discomfort, but this resolved after the first week. In comparison, five patients in the CaCO3 arm complained of constipation, but none dropped out of the study. Serum phosphate levels after the 4-week wash-out period were 6.63 mg/dl in the MgCO3 group and 6.58 mg/dl in the CaCO3 group. These values are lower than those reported by other investigators: 7.7 mg/dl by Qunibi et al. [13] in the CARE study, 7.4–7.7 mg/dl by Chertow et al. [14] in the Treat to Goal study and 6.6–7.2 mg/dl by Asmus et al. [15]. However, in the recently presented preliminary results of the CARE 2 study [16], patients’ mean serum phosphate values were 6.5–6.6 mg/dl. Our values may have been lower because our patients adhered more closely to their diet, they were receiving an adequate dose of hemodialysis or they had neither marked hyperparathyroidism (siPTH <500 pg/ml) nor did they receive vitamin D. The Mediterranean type of diet consumed by our population on the island of Crete is less likely to cause hyperphosphatemia. The serum phosphate level was reduced equally in both groups, and after the second month of treatment the mean values of serum phosphate in the two groups were comparable. (Table 3; Fig. 1). In terms of the biochemical findings, the main difference between the two groups was the levels of serum calcium. Hypercalcemia is the price one has to pay for adequate phosphate-binding control with calcium-containing binders; the absence of hypercalcemia is the major advantage of magnesium-based phosphate-binders. We observed that patients receiving MgCO3 were more likely to have serum calcium levels within the K/DQOI guidelines [17] than those on CaCO3. Both groups of patients achieved a mean serum Ca × P product of <55 mg2/dl2 (Tables 3, 4). Consequently, a significant proportion of patients had a Ca × P product within the K/DQOI recommendations (Table 5). The risk of severe hypermagnesemia is a major concern when magnesium salts are administered to hemodialysis patients. In such patients serum magnesium levels depend chiefly on the dialysate concentration of this ion, which standardly ranges between 0.45 and 0.50 mmol/l. In the MgCO3 group we used a low magnesium dialysate, 0.30 mmol/l, and this enabled us to avoid extreme hypermagnesemia and keep serum magnesium levels within acceptable ranges. Clinical symptoms and ECG disorders due to hypermagnesemia appear only when the level of serum magnesium >4 mg/dl [18]. The ionized fraction that represents approximately 60% of the total serum magnesium is the biologically active form of this element. Truttmann et al. [19] and Saha et al. [20] recently demonstrated that the ionized fraction of magnesium lower in HD patients than in individuals with normal renal function; therefore, the incidence of ‘real hypermagnesemia’ may be overestimated in these patients. Of note is that serum magnesium accounts for approximately 1% of the total body magnesium since it is mainly an intracellular cation. Estimation of total body magnesium is based on the determination of intracellular magnesium (IcMg) levels in skeletal muscles or in peripheral lymphocytes [21]. Calculations on the levels of magnesium in these tissues in uremic patients have produced conflicting results, with positive or neutral or even a negative correlation with serum magnesium being found [22, 23]. Although we did not examine the consequences of hypomagnesemia and high serum magnesium levels, various studies have shown that hypomagnesemia plays a significant role in the pathogenesis of cardiovascular diseases and that a high serum magnesium level may retard the development and/or acceleration of arterial atherosclerosis [24, 25]. Meema et al. [26] first showed that ‘hypermagnesemia might be associated with retardation or improvement of arterial calcifications in peritoneal dialysis patients’. Izawa et al. [27] described a hemodialysis patient in whom soft-tissue calcifications resolved after treatment with a dialysate with a high concentration of magnesium. Two studies by our team [28, 29] clearly demonstrated that the absence of mitral annular calcifications and a thinner carotid intima media thickness in hemodialysis patients both correlated with higher levels of serum and intracellular magnesium. In a previous study, we showed that both intracellular and serum magnesium concentrations were directly associated with 5-year survival in 94 hemodialysis patients of our unit [30]. One of the secondary objectives of the study was to investigate if MgCO3 effectively controls serum iPTH. Calcium carbonate was more effective than MgCO3 in decreasing serum iPTH. In the MgCO3-treated patients time-average mean values of iPTH during months 1–6 were significantly higher than in those treated with CaCO3 (Table 4). However, the final levels of serum iPTH at 6 months did not differ significantly between the two groups. Furthermore, at the end of 6 months, a slightly higher number of patients (48%) in the MgCO3 group than in the CaCO3 (40%) group achieved serum iPTH values in the range of 150–300 pg/ml. In the presence of almost equal serum P values in the two groups, we hypothesize that the relatively higher PTH reduction in the CaCO3 group was due to the significant higher concentration of serum calcium in this group. On the other hand, high levels of serum magnesium have a similar, but weaker, action in suppressing PTH secretion [31]. Figure 4, which shows the monthly follow-up of serum iPTH levels, shows that the curve of the change in mean iPTH values is different in the MgCO3-treated patients in that the mean values increase during the first month of the treatment, subsequently falling (but slower than in the CaCO3 group) and finally, in the sixth month, they approach the mean iPTH values of the CaCO3 group. We speculate that the shape of this curve indicates that suppressive action of serum magnesium on the parathyroid glands is weaker but delayed compared to that of serum calcium. The number of patient who had final serum iPTH values of <150 pg/ml, which could be associated with adynamic bone disease, did not differ in the two groups. However, the impact of the two regimens on the patients’ bone turn over was not included in the objectives of the study, so we did not obtain data associated with bone metabolism markers. Our experience suggests that a significant proportion of hemodialysis patients need more than one agent to achieve a satisfactory phosphate-binding. In such cases, MgCO3 could be if not the primary at least the second constituent of the phosphate-binding regimen, combined ideally with a calcium-containing salt. An additional advantage is that MgCO3 is much less expensive than the newest sevelamer HCL and lanthanum carbonate. The limitations of our study are: (1) it is a single-center study, although this has also a number of advantages; (2) the number of the participant patients was not large, although it is one of the largest groups reported to date; (3) the allocation of the patients to the two regimens was only partially random; (4) the relatively low starting serum phosphate level in conjunction with the no vitamin D use means that the patients were not typical of most HD patients; (5) we did not obtain data on the patients’ bone metabolism markers. In conclusion, our study showed that MgCO3 administered for a period of 6 months is an effective and inexpensive agent to control serum P levels in hemodialysis patients. Gastrointestinal disorders due to its use were minor, while its administration in combination with a low dialysate magnesium concentration reduces the risk of severe hypermagnesemia. Patients treated with MgCO3 had a mild suppression of PTH, an optimum regulation of Ca × P product, relatively low serum calcium and no episodes of hypercalcemia. We believe that this demonstration of the effectiveness of MgCO3 to bind phosphate warrants further investigation in a larger group of patients.

Behav_Genet-4-1-2226022

Heritability of Self-reported Phobic Fear

Twin studies on fear and phobia suggest moderate genetic effects. However, results are inconclusive regarding the presence of dominant genetic effects and sex differences. Using an extended twin design, including male and female twins (n = 5,465) and their siblings (n = 1,624), we examined the genetic and environmental influences on blood-injury, social, and agoraphobic fear and investigated their interaction with sex and age. Data of spouses (n = 708) of twins were used to evaluate assortative mating for the three fear dimensions. Results showed that there was no assortative mating for blood-injury, social and agoraphobic fear. Resemblance between biological relatives could be explained by additive and non-additive genetic effects for blood-injury and agoraphobic fear in all participants, and social fear in participants aged 14–25 years. For social fear in participants aged 26–65 only additive genetic effects were detected. Broad-sense heritability estimates ranged from 36 to 51% and were similar for men and women. Introduction Fears and phobias are extremely disruptive and distressing disorders that have a great impact on an individual’s life (Mendlowicz and Stein 2000; Saarni et al. 2007). The prevalence of anxiety disorders is high and anxiety disorders represent a serious public health concern (Kessler et al. 2005). However, the etiology of phobic fear is still poorly understood. Family studies of clinically defined phobias have consistently found evidence for the familial aggregation (Fyer et al. 1995; Stein et al. 2001). Twin family studies of clinically defined fears and phobias have also indicated familiarity (Torgersen 1983; Skre et al. 1993; Skre et al. 2000). Due to their relatively small sample sizes, clinical studies often are unable to establish whether this familiarity results from genetic or environmental similarities. Through community based twin and twin family studies, large samples can be assessed, resulting in sufficient power to disentangle genetic and environmental influences in the etiology of fears and phobias. Such studies have suggested an important role for genetic and unique environmental factors (e.g., aversive experiences) in the origin of fears and phobias, while shared environmental influences appeared to be of little etiological significance (Hettema et al. 2001; Sundet et al. 2003). The Virginia Twin Registry investigated the familial aggregation of phobia diagnoses in a large adult sample. In women, heritability estimates of 39% for agoraphobia and 30% for social phobia were found (Kendler et al. 1992). In men, these estimates were 37 and 20% (Kendler et al. 2001). Familial aggregation of blood-, needles-, hospital-, or illness-phobia was studied in female twins only and appeared to be due to shared environmental factors, although the statistical power to differentiate between additive genetic and common environmental influences was low (Neale et al. 1994a). An Australian study of self-reported fear of blood-injury-injection in female twins also found significant familial aggregation of blood-injury-injection fears. It was not possible to differentiate between genetic and shared environmental influences as the explanation for familial aggregation. Multivariate genetic analysis of fainting and blood-injury-injection fear with and without fainting indicated that the genetic variance in blood-injury-injection fear was attributable to additive genetic factors shared with fainting (55%) (Page and Martin 1998). Based on the correlation structure several studies suggested the presence of non-additive genetic effects in phobic fear, but due to insufficient statistical power, these effects could not convincingly be demonstrated (Skre et al. 2000; Neale et al. 1994a; Rose and Ditto 1983; Kendler et al. 1999; Page and Martin 1998). The heritability estimates reported in the Kendler et al. studies of female (1992) and male twins (2001) seem to indicate lower heritability estimates for fear in men than in women. Together with the fact that women display generally more phobias than men (Fredrikson et al. 1996; Furmark 2002; Bekker 1996; Curtis et al. 1998; Bijl et al. 2002; Middeldorp et al. 2005) this suggests that men and women differ in the extent to which genetic and environmental risk factors affect phobic fear. Sex differences in the genetic contribution to the etiology of phobic fear have only been formally tested in the Virginia Twin Registry, in the Netherlands Twin Register and in the Australian Twin Registry. In the Virginia Twin Registry, Kendler et al. (2002) found equal heritability in men and women for agoraphobia and blood-injury phobia. For social phobia, on the other hand, resemblance in men was explained by genetic factors while in women resemblance was explained by shared environment. Because earlier studies relying on a female sample (Kendler et al. 1992, 1999) indicated that familial resemblance was entirely due to genetic factors, the authors attributed this sex difference to stochastic factors. However, results from two multivariate studies using data from the Virginia Twin Registry (Hettema et al. 2005, 2006), indicated that around 10% of the variance in social phobia was explained by shared environmental factors. Qualitative sex differences on the liability for agoraphobia, social phobia and blood-injury phobia were suggested by Kendler et al. (2002) based on the dizygotic opposite sex (DOS) twin correlation, which was lower than the dizygotic (DZ) same sex correlations. A second study of the Virginia Twin Registry (Hettema et al. 2005) found no sex differences in the genetic and environmental risk factors for agoraphobia or social phobia. In a combined Dutch and Australian sample, Middeldorp et al. (2005) failed to find quantitative sex differences in the familial influences on agoraphobia and social phobia. For agoraphobia a low correlation in DOS twins suggested qualitative sex differences. To summarize, most studies report no sex differences in the heritability of fear and phobia while some do suggest qualitative sex differences based on the DOS and DZ correlation structure. Spousal resemblance is often found for psychiatric disorders and related phenotypic traits. The most cited explanation for spousal similarity is assortative mating, meaning that spouses are more similar for a trait or disorder than expected under random mating (Merikangas 1982; Garrison et al. 1968), although significant spousal correlation could also indicate, for example, marital interaction or social stratification. Spousal resemblance is an important issue to consider because increased phenotypic correlation between spouses increases genetic variance in the offspring generation. When non-random mating for a heritable trait is present but this is not specifically included in the twin model, heritability estimates may be downwardly biased. For depressive disorders, a meta-analysis reported marital resemblance for depression in 12 of 17 studies (Mathews and Reus 2001). An extensive study on spousal similarity for psychiatric disorders in a population-based sample was carried out by Maes et al. (1998). Several psychiatric diagnoses were examined, including generalized anxiety disorder, major depressive disorder, panic disorder and phobias. Moderate spousal correlations were seen for most psychiatric diagnoses, but for phobias spousal correlations were not significant. So far, age effects on the heritability of phobic fears have not received much attention in the literature. This is an omission because it has well been established that some specific fears decline from adolescence to later life (e.g., fear of blood, strangers or personal death) (Bienvenu and Eaton 1998; Hall 1897; Rose and Ditto 1983; Kessler et al. 1994; Henderson et al. 1998), while others increase or are not affected by age at all (e.g., fear of negative social interactions or small organisms) (Rose and Ditto 1983; Lapouse and Monk 1959). Rose and Ditto (1983) found evidence that with age there is a change in the genetic and environmental influences on fear of death. However, a more recent study reported that there were hardly any longitudinal changes in the genetic and environmental determinants of anxiety (Gillespie et al. 2004). In order to replicate and extend previous research on fears and phobias, we examined the heritability of three classes of phobic fears (i.e., blood-injury, social, and agoraphobic fear) using longitudinal data from Dutch twins and their siblings registered with the Netherlands Twin Register (Boomsma et al. 2006). The inclusion of siblings and repeated observations increases statistical power to discriminate between additive genetic and non-additive genetic effects and to detect sex specific genetic effects. Adding siblings to the sample also provides the opportunity to examine twin-specific effects on phobic fears (Posthuma and Boomsma 1999, 2000). Data were also available for a subsample of spouses of twins which were used to estimate the correlation between spouses to examine whether there is evidence for assortative mating. Since the determinants of phobic fear in adolescence and young adulthood may differ from those in later life, we divided the dataset into two age cohorts so as to test for differences in heritability between the two cohorts. In addition, we tested for qualitative and quantitative sex differences in the etiology of phobic fear. To summarize, in this report, we address the following questions. First, is there spousal similarity for blood-injury, social and agoraphobic fear? Second, what are the heritability estimates for these three subtypes of fear? Third, are there sex and age differences in the heritability estimates of these subtypes? Methods Participants This study is part of an ongoing study on health and lifestyle in twin families registered with the Netherlands Twin Register (Boomsma et al. 2006). Initially, adolescent and adult twins and their family members were recruited through City Councils in 1990–91 and in 1992–93. After 1993, an effort was made to recruit older twins through a variety of approaches. Siblings and spouses of twins were recruited in the study since 1995 and 2000, respectively. Parents did not participate in 1997 and 2000. All registered individuals gave their informed consent before they filled out questionnaires. Further details on response rates and demographic characteristics of the sample can be found elsewhere (Boomsma et al. 2002; Vink et al. 2004; Boomsma et al. 2006). The data on phobic fear were derived from surveys sent in 1997 and 2000. Table 1 shows the complete sample configuration of two age cohorts in 1997 and 2000. In total, there were 7,089 subjects (5,465 twins and 1,624 siblings) from 3,471 families, with longitudinal data available for 2,537 individuals. Based on the age of the twins, we subdivided the participating families into two age groups: a 14–25 year group (n = 4,275 individuals) and a 26–65 year group (n = 2,814 individuals). Families with siblings but without twins were assigned to one of the two age groups according to the age of the youngest sibling. There was a small overlap between the two age cohorts in age of the non-twin siblings. In the 14–25 year age group, there were 1,987 families of which 1,004 included one or more family members who participated at both time points. The 26–65 year age group involved 1,484 families of which 435 contained one or more family members who participated at both time points. In the youngest age group, mean age was 21.2 years (SD = 3.5) in 1997 and 23.2 years (SD = 3.6) in 2000. In the oldest age group, mean age was 36.6 years (SD = 10.5) in 1997 and 38.0 years (SD = 10.4) in 2000. For the 2000 survey, we asked spouses of all twins aged 25–30 years to participate. In total 708 spouses completed a survey. Table 1Family configuration in the sample according to zygosity, cohort and number of additional non-twin siblings of the two time points (separated by the slash; 1997/2000)Zero siblingsBrotherSisterBrother and SisterTotalAge 14–25MZMTwin pair70/10441/2935/279/4155/164Single twin21/574/610/111/336/77DZMTwin pair44/4533/1526/143/7106/81Single twin18/444/910/101/233/65MZFTwin pair145/23049/3562/5612/9268/330Single twin49/10610/1211/173/273/137DZFTwin pair84/10834/1937/277/7162/161Single twin33/8710/912/232/257/121DOSTwin pair101/10564/3768/4110/3243/186Single twin69/18311/2625/326/2111/243No twins23/4338/722/463/119Total age 14–25634/1069283/240334/33056/451307/1684Age 26–65MZMTwin pair16/608/67/115/836/85Single twin20/686/34/76/136/79DZMTwin pair11/313/63/77/324/47Single twin20/544/27/55/536/66MZFTwin pair64/22314/2223/539/12110/310Single twin30/1226/1010/162/448/152DZFTwin pair29/958/1411/153/651/130Single twin39/994/914/73/060/115DOSTwin pair27/869/710/154/750/115Single twin36/1206/810/226/558/155No twins21/4446/2512/879/77Total age 26–65292/95889/131145/18362/59588/1331Total number of families926/2027372/371479/513118/1041895/3015MZM = monozygotic males; DZM = dizygotic males; MZF = monozygotic females; DZF = dizygotic females; DOS = dizygotic opposite sex twins Zygosity A total of 1,322 twins were classified as DZ because they were of opposite sex. Zygosity of same sex twins was determined from DNA polymorphisms (n = 1,205), or from survey questions about physical twin resemblance and confusion of the twins (n = 2,938). Based on the answers from all available surveys, zygosity was determined. When there was an inconsistency over time or persons, the majority of the judgements determined the final outcome. The agreement between zygosity based on survey data and on DNA typing was 97% (Willemsen et al. 2005). Measures Participants completed the Dutch version of the Fear Questionnaire (FQ; Marks and Mathews 1979). The FQ is a widely used instrument that possesses good psychometric qualities (Van Zuuren 1988). The FQ contains 15 items that tap phobic avoidance in three domains: blood-injury fear (e.g., avoidance of hospitals), social fear (e.g., avoiding talking to people in authority), and agoraphobic fear (e.g., avoiding large, open spaces). Each domain is represented with five items. Subjects were instructed to indicate on a nine point scale (0 ‘would not avoid it’ to 8 ‘always avoid it’) how much they would avoid certain situations because of fear or other unpleasant feelings. Scores were summed across subscales to obtain scores for blood-injury fear, social fear, and agoraphobic fear. A score was only calculated if at least four out of five items of a subscale were answered. Missing answers and double entries were substituted by the mean item score. We checked the factor structure of the FQ in our sample. All items of the FQ were analyzed with PCA with promax rotation in SPSS. Analyses were done separately for the two time points with random selection of one person per family. At both time points we found the original three factor solution in our data which accounted for around 47% of the variance. Twin-family Studies Twin studies make use of the genetic relatedness of twins and their family members to address questions about the etiology of population variation. Monozygotic (MZ) twins are genetically (nearly) identical. Dizygotic (DZ) twins and siblings share on average 50% of their segregating genes. Additive genetic effects (A) are suggested if the correlation for a phenotype in MZ twins is larger than the correlation in DZ twins and siblings. When the DZ correlation is more than half the MZ correlation, this indicates environmental effects shared by members from the same family (C). When the DZ correlation is less than half the MZ correlation, this can be taken as evidence for non-additive genetic effects (D). Non-additive genetic effects can consist of interactions between alleles within a locus (dominance) or across different loci (epistasis). Differences in fear scores within MZ twin pairs are due to unique environmental influences (E), which also include measurement error. The observed variance in phobic fear can thus be decomposed in four possible sources of variance; A, C, D, and E. However, the observed variances and covariances only provide enough information to test either an ACE model or an ADE model. Based on the pattern of twin correlations (see results section), A, D, and E were modelled in this study. Statistical Analysis Since phobic fear is not normally distributed in the population we used a multiple threshold model (see Fig. 1) to estimate the genetic and environmental contributions to the liability of phobic fear. Liability to phobic fear is assumed to be continuous and normally distributed in the population. Derks et al. (2004) showed that this is the optimal approach when analyzing L-shaped distributed phenotypic data. We divided the liability to phobic fear in three categories separated by two thresholds so that all categories contained approximately equal numbers of subjects (see Fig. 2). The validity of the bivariate threshold model to the fear data divided in three categories, was assessed in PRELIS (Jöreskog and Sörbom 1996). The fit of the bivariate threshold model to the blood-injury, social and agoraphobic fear data was good; almost all RMSEA’s were <0.05 and all P values exceeded the 0.002 significance level (Bonferroni correction for multiple testing). Fig. 1Threshold model of the liability for phobic fear. Latent factors are symbolized in circles, observed phenotype as squares. E, unique environmental factor; D, non-additive genetic factor; A, additive genetic factor; L, liability for phobic fear. Latent factors can be correlated within families and influence the liability to phobic fearFig. 2Distribution of fear scores in 1997 for men and women age 14–25 and age 26–65 years. The Y-axis gives the percentage of males and females with each score. The dashed lines indicate the thresholds Genetic structural equation modelling was carried out in Mx (Neale et al. 2003). Testing of submodels was done by means of likelihood-ratio tests, by subtracting the negative log likelihood (−2LL) for the more restricted model from the −2LL for the more general model. This yields a statistic that is distributed as χ2 with degrees of freedom (df) equal to the difference in the number of parameters in the two models. If the χ2-test yields a P-value higher than 0.05, the constrained model is deemed not significantly worse than the previous model and is kept as the most parsimonious model to which the next model will be compared. We started with fitting saturated models to describe the correlational structure between twin pairs and between siblings in each zygosity group. We used repeated measures from two time points (Rebollo and Boomsma 2006). Twin and sibling correlations were constrained to be equal at the two time points. For each fear subtype (blood-injury, social, and agoraphobic fear) and within each age group (14–25 and 26–65 years), we estimated five twin correlations (one for each sex by zygosity group) and three sibling correlations (male–male, female–female, and male–female sib pairs). Two cross-time within person correlations (one for each sex) and eight cross-twin cross-time correlations (one for each of the five sex by zygosity and the three sib pair groups) were estimated for each trait (not reported in the results section). Thresholds were estimated separately for men and women at each time point. The spouse correlations (Spearman’s correlation coefficients) were calculated using SPSS 12.0 for windows. First, we tested whether the resemblance in sibling pairs was the same as in dizygotic twin pairs. Next, quantitative sex differences were investigated by constraining the correlations for same-sex male twin pairs and same-sex female twin pairs to be equal. Qualitative sex differences were investigated by constraining the correlations for DZ same-sex twin pairs and DZ opposite-sex twin pairs to be equal. In a genetic model, the influence of latent factors A, D and E on the phobic fear score in 1997 and 2000 was estimated by the parameters (factor loadings) a, d, and e. The estimates for a, d and e were constrained to be equal in 1997 and 2000 for all three fear subtypes because we assume that the same etiological mechanism influences phobic fear at both time points. Correlations between the latent A and D factors were fixed at 1 for MZ twins. For DZ twins and siblings, the correlations between the latent A factors were fixed at 0.5, because they share on average half of their segregating genes. Dominant genetic effects result from the interaction or combination of alleles at a particular locus. Offspring receive only one allele from each parent and not a combination of two alleles. Parents and offspring share on average half of their genes so the chance that two siblings receive the same allele is 0.5 × 0.5 resulting in a correlation of 0.25 between the latent D factor for DZ twins and sibling pairs. We first tested whether a, d and e in the two age groups could be constrained to be equal. Next, different models (ADE, AE and E) were fitted to the data to test which sources of variance contribute to individual differences in phobic fear. Results Figure 2 shows the distribution of fear scores in 1997 for males and females aged 14–25 and aged 26–65. The 2000 scores (not presented here) show a similar pattern. Women had significantly more often high scores on each fear dimension than men in each age group, except for blood-injury fear in the oldest age group for which there was no difference between men and women. There was no effect of age on the distribution of fear scores, except for women’s scores on blood-injury fear, which were significantly higher in the youngest age group. There was no significant correlation between twins and their spouses for blood-injury fear (rs = 0.05, P = 0.162), social fear (rs = 0.04, P = 0.324) or agoraphobic fear (rs = 0.04, P = 0.341). Table 2 displays the twin correlations and confidence intervals for blood-injury, social, and agoraphobic fear for both age groups. Correlations for same sex DZ and DOS twin pairs, and for sibling pairs could be constrained to be equal, suggesting that there is no evidence for a special twin environment. MZ male–male and female–female correlations were the same and DZ same-sex correlations were also the same, indicating that there are no quantitative sex differences in the influence of genetic effects on the three fear types. The DZ opposite-sex twin correlation was equal to the same-sex DZ correlation indicating that the same genes affect fear in men and women. After these constraints, MZ correlations were more than twice as large as the correlations for DZ twins and siblings suggesting that genetic effects contribute to individual differences in phobic fear and that these effects act in a partly non-additive manner. Table 2Twin and sibling correlations for blood-injury, social and agoraphobic fear for age group 14–25 years and age group 26–65 years (95% confidence intervals added in parentheses)Blood-injury fear Social fearAgoraphobic fearAge 14–25Age 26–65Age 14–25Age 26–65Age 14–25Age 26–65MZ males0.30 (0.17–0.42)0.36 (0.14–0.54)0.53 (0.41–0.63)0.29 (0.03–0.50)0.47 (0.34–0.56)0.49 (0.27–0.65)DZ males0.04 (−0.16–0.22)−0.04 (−0.32–0.25)0.15 (−0.05–0.34)0.49 (0.20–0.68)0.27 (0.09–0.43)0.22 (−0.10–0.48)MZ females0.34 (0.25–0.43)0.41 (0.32–0.50)0.48 (0.39–0.56)0.42 (0.31–0.52)0.40 (0.30–0.48)0.37 (0.26–0.47)DZ females0.18 (0.05–0.30)0.11 (−0.06–0.28)0.20 (0.06–0.33)0.37 (0.20–0.53)0.13 (−0.00–0.26)0.27 (0.08–0.44)DZ opposite sex0.23 (0.10–0.34)0.16 (−0.04–0.34)0.22 (0.10–0.34)0.25 (0.03–0.44)0.17 (0.04–0.29)0.22 (0.01–0.41)Brother–brother0.16 (0.03–0.28)0.28 (0.10–0.44)0.12 (−0.02–0.25)0.19 (−0.03–0.38)0.09 (−0.04–0.22)−0.01 (−0.24–0.22)Sister–sister0.10 (−0.01–0.20)−0.02 (−0.16–0.13)0.09 (−0.03–0.20)0.14 (−0.01–0.28)0.10 (−0.01–0.19)0.12 (−0.03–0.26)Brother–sister0.10 (0.02–0.18)0.06 (−0.05–0.17)0.15 (0.06–0.23)0.11 (−0.02–0.23)0.08 (−0.00–0.02)0.07 (−0.05–0.18)All MZa0.33 (0.26–0.40)0.39 (0.30–0.48)0.50 (0.43–0.56)0.40 (0.30–0.49)0.42 (0.35–0.49)0.39 (0.30–0.48)All DZ/siblingsa0.13 (0.08–0.18)0.09 (0.02–0.16)0.15 (0.10–0.20)0.20 (0.13–0.27)0.11 (0.07–0.16)0.12 (0.05–0.19)MZ = monozygotic; DZ = dizygoticaAfter constraining these correlations to be equal Model fitting results for blood-injury, social, and agoraphobic fear are shown in Table 3. The data from two age groups were simultaneously analysed in one model with different parameter estimates for the two age groups. For each fear subtype, a general ADE model (model 1) was fitted. Next, we tested if the A, D, and E components were the same for the two age cohorts (model 2). Finally, in model 3, and in the case of social fear in models 4 and 5, the significance of D and A in the two age groups was tested by constraining the relevant parameters at zero. For blood-injury and agoraphobic fear, this resulted in a model in which variance components did not differ between the two age groups. Non-additive genetic effects could not be left out of the model for blood-injury and agoraphobic fear. A, D and E explained respectively 9.9, 25.6, and 64.5% of the variance in blood-injury fear, and 10.7, 29.9, and 59.5% of the variance in agoraphobic fear. For social fear in the youngest age group, A, D and E contributed to the variance, and explained 7.8, 42.9, and 49.3% of the variance, respectively, while in the oldest age group D did not contribute to the variance. In fact, 40.3% of the variance in social fear in the oldest age group could be explained by additive genetic effects and 59.7% by unique environmental effects (Table 4). Table 3Model fitting results for blood-injury, social, and agoraphobic fear; comparisons of models are shownModelsvs−2LLdfΔdfΔχ2PBlood-injury fear1. ADE20254.5896082. ADE no age diff.120257.23961792.650.983. AE no age diff.220415.2796203158.030.00Social fear1. ADE19439.7296082. ADE no age diff.119468.909617929.180.003. AE youngADE old119454.059611314.330.004. ADE youngAE old119439.72961130.001.005. ADE youngE old419521.589614381.870.00Agoraphobic fear1. ADE 19967.6196142. ADE no age diff.119973.66962396.050.743. AE no age diff.219984.909626311.240.01vs = versus and indicates which model the sub model is compared to; −2LL = −2 log likelihood; df = degrees of freedomMost parsimonious solution per fear subtype is shown in boldTable 4Parameter estimates of the best-fitting models in age group 14–25 and age group 26–65 for blood-injury fear, social fear and agoraphobic feara2 (%)d2 (%)e2 (%)Blood-injury fearAge 14–659.925.664.5Social fearAge 14–257.842.949.3Age 26–6540.3–59.7Agoraphobic fearAge 14–6510.729.959.5a2 = additive genetic variance component; d2 = dominant genetic variance component ; e2 = unique environmental variance component Comments This is the first study that explored the genetic and environmental contributions to individual differences in blood-injury, social, and agoraphobic fear using an extended twin design. We tested whether there was evidence for assortative mating, for a specific twin environment, and if the impact of the genetic and environmental factors differed between men and women and between two age cohorts. We observed that more women than men scored high on all fear dimensions, except for blood-injury fear in the oldest age group. This pattern is consistent with the large majority of studies that looked at sex differences in fear levels: overall, women tend to have higher scores on self-report measures tapping various fear categories, the exception being blood-injury phobia, where the sex distribution is somewhat more balanced (Bijl et al. 1998; Kessler et al. 1994; Middeldorp et al. 2005). Prior studies showed that many fear types display a distinct developmental course, with, in general, a decline of fear levels from adolescence to later life. Our data suggest no large effect of age in fear scores: a pattern of declining fear scores was only seen for blood-injury fear, with younger participants having higher fear levels than older participants. This is in accordance with the literature (Bijl et al. 2002; Page 1994). However, many developmental changes in fear levels occur before the adolescent stage (Marks 1987), an underrepresented age group in our sample. Future research will have to determine whether changes in blood-injury, social and agoraphobic fear emerge before adolescence. Spouse similarity is an important issue to consider when studying genetic influences on population variation. In accordance with what was reported previously for phobias (Maes et al. 1998) and anxiety disorders (Low et al. 2007), we did not find a significant association between spouses, indicating that there is random mating for blood-injury, social and agoraphobic fears. Broad-sense heritability estimates for blood-injury, social, and agoraphobic fear were moderate ranging from 35.5% (blood-injury fear) to 50.7% (social fear in the youngest age group). These heritability estimates for blood-injury fear and agoraphobic fear are comparable to the estimates reported by Kendler et al. (1992, 2001). The heritability estimate for social fear (50.7% in the youngest age group and 40.3% in the oldest age group) was somewhat higher than that reported by Kendler et al. (30% for women and 20% for men) but comparable to the heritability estimate found by Middeldorp et al. (2005) (50%). Contrary to what was found by Hettema et al. (2005, 2006) shared environmental factors did not significantly contribute to the variance in social fear. All in all, these results confirm the conclusion of Kendler et al. (1992, p. 279) that “the estimated heritability of liability for phobias indicates that genetic factors play a significant, but by no means overwhelming role in the etiology of phobias.” In accordance with previous findings (Hettema et al. 2005; Middeldorp et al. 2005; Kendler et al. 2002), no differences in heritability between men and women were observed. Regarding qualitative sex differences, the results are somewhat different from previous studies. Whereas results of Kendler et al. (2002) and Middeldorp et al. (2005) suggested the presence of qualitative sex differences for agoraphobic fear, we did not find any evidence that different genes influence the liability of phobic fear in men and women. Since the sample in our study was enriched with siblings of twins, a large number of opposite sex first degree relatives was added to the study. These data do not indicate a lower familial correlation in opposite-sex relatives. For all three fear subtypes, MZ correlations were more than twice as high as the DZ/sibling correlations, indicating that non-additive genetic effects may contribute to the variance in phobic fears. Several previous studies suggested the influence of non-additive genetic effects on fears and phobia, which makes sense because fears are likely to be exposed to selection during evolution. Skre et al. (2000) investigated the genetic and environmental contributions to common phobic fears in a treatment sample of 61 twin pairs and found the twin correlations for social fear to fit best to a model including non-additive genetic effects. For blood-injection-injury fear, the correlational structure found by Skre et al. suggested the influence of non-additive genetic effects but model fitting showed that due to violation of equal variances between MZ and DZ twins all variance was accounted for by unique environmental effects. Neale et al. (1994b) also reported evidence for non-additive genetic effects in reanalysing the Rose and Ditto data (1983) of common social fears. In addition, Kendler et al. (1999) found evidence for non-additive genetic effects in agoraphobia, but not in social phobia and blood-injury phobia, while Page and Martin (1998) suggested the possibility of non-additive genetic effects on the variance in blood-injury phobia. In 2002, Kendler et al. (2002) did not model non-additive genetic effects explicitly, yet the correlation structure did suggest the presence of non-additive genetic effects for agoraphobia and blood-injury phobia. To summarize, many studies indicated the presence of non-additive genetic effects on the liability for fear and phobias but they often did not have sufficient statistical power to detect non-additive genetic effects. The large sample size of this study together with repeated measures enabled us to detect the influence of non-additive genetic effects on blood-injury fear, social fear (in the youngest age group), and agoraphobic fear, confirming the speculations discussed in previous studies. For social fear in the oldest age group, non-additive genetic effects were not detected. Thus, the mixture of genetic and environmental influences on social fear seems to differ between the two age cohorts, though the differences in correlation structure were small. These results could be relevant to gene mapping of anxiety and phobia. If many of the genes underlying anxiety and phobia have non-additive effects, as suggested by our results, then quantitative trait loci with little additive genetic effects may have been missed in linkage analysis (Purcell and Sham 2004). Explicitly modelling non-additive genetic effects may therefore be important in the linkage analysis of anxiety, fear and phobia. In general, our data are well in line with the multifactorial model of fears and phobias (Muris and Merckelbach 2001). This model assumes that a genetic vulnerability predisposes individuals to develop maladaptive fears and that discrete learning experiences (i.e., unique environmental factors) in combination with genetic vulnerabilities produce persistent fears. Results of this study should be interpreted in context of the following limitations. First, the present study relied on non-clinical groups, and, although there is no reason to assume that heritabilities increase at the extreme end of the fear continuum (Stevenson et al. 1992), it is important to replicate the findings in clinical samples. Second, data of our study were obtained with an instrument (i.e., the FQ) that primarily focuses on self-reported avoidance behavior related to a limited number of fear categories and our results may not be generalized to other specific fears such as fear of small animals. In summary, we found random mating for blood-injury, social and agoraphobic fear. Individual differences in blood-injury fear and agoraphobic fear in both age groups and social fear in the youngest age group (14–25 years) could be explained by additive and non-additive genetic factors and unique environmental factors, while individual differences in social fear in the oldest age group (26–65 years) could be explained by additive genetic and unique environmental factors. The heritability of the three fear subtypes was similar for men and women and no effects of sex-specific genes were detected.

Plant_Mol_Biol-3-1-2039854

Local coexpression domains in the genome of rice show no microsynteny with Arabidopsis domains

Chromosomal coexpression domains are found in a number of different genomes under various developmental conditions. The size of these domains and the number of genes they contain vary. Here, we define local coexpression domains as adjacent genes where all possible pair-wise correlations of expression data are higher than 0.7. In rice, such local coexpression domains range from predominantly two genes, up to 4, and make up ∼5% of the genomic neighboring genes, when examining different expression platforms from the public domain. The genes in local coexpression domains do not fall in the same ontology category significantly more than neighboring genes that are not coexpressed. Duplication, orientation or the distance between the genes does not solely explain coexpression. The regulation of coexpression is therefore thought to be regulated at the level of chromatin structure. The characteristics of the local coexpression domains in rice are strikingly similar to such domains in the Arabidopsis genome. Yet, no microsynteny between local coexpression domains in Arabidopsis and rice could be identified. Although the rice genome is not yet as extensively annotated as the Arabidopsis genome, the lack of conservation of local coexpression domains may indicate that such domains have not played a major role in the evolution of genome structure or in genome conservation. Introduction The fast-growing data sets on genome annotation and genome-wide gene expression facilitate the study and comparison of gene activity between and among genomes. The genomic context of genes is supposed to play an important role in the regulation of gene expression (van Drunen et al. 1997). Non-random clusters of similarly expressed (co-regulated, coexpressed, highly expressed and/or broadly expressed) genes have been described in almost all organisms, ranging from prokaryotes to eukaryotes. In eukaryotes, from yeast to Arabidopsis to human, both short-range co-regulated/coexpressed clusters of two to five genes (Cohen et al. 2000; Ren et al. 2005; Zhan et al. 2006) and longer-range coexpression domains of up to 30 genes spanning up to 100 kb and more (Lercher et al. 2003; Ma et al. 2005; Spellman and Rubin 2002; Zhan et al. 2006) have been described. Duplicated genes (Lercher et al. 2003), shared promoter regions (Kruglyak and Tang 2000), shorter gene distance (Cohen et al. 2000; Roy et al. 2002; Semon and Duret 2006; Williams and Bowles 2004) and/or functional relatedness (Cohen et al. 2000; Lee et al. 2004; Spellman and Rubin 2002; Williams and Bowles 2004) were found to account for only part of the coexpression between genes. Therefore, most studies postulate that the occurrence of coexpression domains, small or large, is regulated on the level of higher-order chromatin structure (Cohen et al. 2000; Hershberg et al. 2005; Ren et al. 2005; Spellman and Rubin 2002; Williams and Bowles 2004), although alternative views exist (Semon and Duret 2006). Previously we have defined and demonstrated the existence of local coexpression domains in the genome of Arabidopsis (Ren et al. 2005). A local coexpression domain was defined as any set of physically adjacent genes that are highly coexpressed with a pair-wise Pearson’s correlation coefficient larger than 0.7. It was shown that a small (5–10%) yet significant fraction of genes in the Arabidopsis genome is organized in such local coexpression domains. Genes in such local domains were for the major part not categorized in the same functional category (GOslim). Neither tandemly duplicated genes nor shared promoter sequence, nor gene distance, explained the occurrence of coexpression of genes in such chromosomal domains. This indicates that other parameters in genes or gene positions are important to establish coexpression in local domains of Arabidopsis chromosomes. Here it is analyzed whether a similar situation exists in the genome of the monocotyledonous model plant rice (Oryza sativa), for which an earlier study had shown the existence of longer-range domains (Ma et al. 2005). We combined the whole genome rice annotation data (TIGR version 3; www.tigr.com) with Massively Parallel Signature Sequencing (MPSS; mpss.udel.edu/rice) expression data as well as Affymetrix array expression data (GEO, GSE4438; www.ncbi.nlm.nih.gov/geo) for the rice cultivar japonica in a way similar to the analysis performed for the Arabidopsis genome (Ren et al. 2005). The results show that the characteristics of the two genomes with respect to the occurrence and configuration of local expression domains are remarkably similar. Also in the rice genome, a small yet significant fraction of genes is organized in local coexpression domains that predominantly consist of two, up to 4, genes that are not categorized in the same functional category, irrespective of the expression platform used for analyses. The presence of tandemly duplicated genes, shared promoter sequence or gene distance is not fully explaining the occurrence of coexpression of genes in such chromosomal domains. Therefore, the regulation of local coexpression domains is postulated to be at the level of higher-order of chromatin structure. Given the similarities in the characteristics and occurrence of coexpression domains between Arabidopsis and rice, we investigated whether the genes involved showed microsynteny between the two genomes. These analyses did not identify the presence of syntenic local coexpression domains between Arabidopsis and rice. Material and methods Genome data The rice (Oryza sativa) genome was obtained from the website of The Institute of Genomic Research (TIGR; www.tigr.org). The rice TIGR version 3 [Jan. 2005] annotation has 57,915 gene loci. In case of alternative splicing, the longest variant of the gene was used. The genes along each chromosome were sorted based on ascending start coordinates and were numbered consecutively. These rank numbers (rank ID) helped to eliminate any discontinuity in the unique Os gene identifiers of the annotated genes and facilitated analyzing physically adjacent genes. In case of overlapping gene loci, the smaller one of the overlapping genes was removed from the data set. This way the order of both gene and rank ID numbers was maintained. Expression data The MPSS expression data for rice (cv. Japonica) were obtained from the rice MPSS database (http://mpss.udel.edu/rice). Only the unique MPSS tags (mapping to the genome only once) and those mapping to unique gene identifiers in TIGR v3 were used in the analyses. The MA expression data were obtained from the Gene Expression Omnibus repository (GEO; www.ncbi.nlm.nih.gov/geo). The data with accession number GSE4438 were chosen for analysis, because they represent the only Affymetrix chip-based data set in the public domain with a reasonable coverage of the Oryza sativa cv. japonica transcriptome under four different experimental conditions (Walia et al. 2007). The earlier analysis in Arabidopsis also concerned Affymetrix chip data (Ren et al. 2005). The MA expression data were mapped to the TIGR v3 annotation based on gene accession numbers. The expression values in libraries representing biological replicates, i.e., the same tissues under the same experimental conditions, were averaged. This way, for the MPSS data 18 different libraries were generated, that cover expression in 9 tissues (callus, panicle, leaves, root, germinating seed and seedling meristem, ovary and stigma, pollen, stem) under different experimental treatments or in different developmental stages. From the microarray (MA) data, only the expression data from cv. japonica were used to allow comparison with the MPSS data. Here, 4 different libraries were generated: a sensitive japonica genotype under control conditions and under salt stress; and a tolerant japonica genotype under control conditions and under salt stress (Walia et al. 2007). The expression data is from crown and growing point tissue. Identification of local coexpression domains Pearson’s correlation coefficient (R) was calculated between all adjacent pairs (duplets) of genes using the expression data from all 18 libraries. If R was higher than 0.7, the gene pair concerned was considered to be coexpressed. The value of R > 0.7 is generally considered a rule-of-thumb threshold (see for example bbc.botany.utoronto.ca/affydb/BAR_instructions) and is used in various analyses (Cohen et al. 2000; Lee et al. 2004; Ren et al. 2005). The number of coexpressed adjacent pairs was counted. To evaluate the statistical significance of these numbers, they were compared with the number of coexpressed pairs from 100 randomizations of the population of expressed genes using the cumulative binomial distribution (Cohen et al. 2000). Previous analyses indicated that more than 100 randomizations did not result in significant changes in the numbers obtained (Ren et al. 2005). In each round of randomization, non-adjacent pairs of genes were randomly selected with replacement from the list of expressed genes that have expressed neighbors till the same total number of pairs was obtained. Similarly, coexpressed adjacent triplets, quadruplets and pentaplets were identified as series of genes with consecutive IDs in which all possible (that is, (n!/(n−2)!)*2; where n is the number of genes involved) pair-wise R’s should be above the cut-off of 0.7. The significance of results was evaluated with randomizations equivalent to the procedure used in case of duplets. Duplicated genes Duplicated genes were identified by local pair-wise protein BLAST (BLASTP 2.2.6 [Apr-09-2003]; (Altschul et al. 1997)), on all gene pairs in the rice genome. A gene pair was considered to be duplicated (dup) if BLASTP yielded an E-value < 0.2 (Fukuoka et al. 2004; Lercher et al. 2003; Williams and Bowles 2004). To determine duplicated triplets, quadruplets and pentaplets, it was required that any pair of the genes concerned had a BLASTP E-value <0.2. Analyses of gene orientation and gene distance Adjacent gene pairs were separated into tandemly, divergently and convergently transcribed pairs according to their relative direction of transcription. The number of coexpressed pairs in each orientation group was expressed as percentage relative to the total number of adjacent pairs in that group. Random pairs were made by randomly picking two non-adjacent genes from the list of expressed genes represented in pairs, analyzed for their orientation and compared with the real genome using a variant of the two-sample t test for proportions for determining the significance of a difference between two population proportions (Ott and Longnecker 2001). The test statistic is based on the z statistic from the normal distribution and is given by (p1−p2)/ √ (p1*(1−p1)/n1 + p2*(1−p2)/n2), with p1 and p2 the two sample proportions, n1 and n2 the two sample sizes, under the condition that n1*p1, n1*(1−p1), n2*p2 and n2*(1−p2) are all larger than 5. The z value is converted to a p value using standard normal tables. To determine the gene distance, the intergenic distance is used. This distance is defined as the length in nucleotides from the annotated end of one gene to the annotated start of the next gene, including the UTRs when known, otherwise the translation start and stop sites were taken. The data sets excluding the duplicated gene pairs were analyzed. For each data set, gene pairs were sorted based on gene distance from short to long and bins of 1,000 pairs were taken and analyzed, excluding the last bin with less than 1,000 pairs. The advantage of using equal pair bin is that it avoids unequal number of gene pairs in different distance categories. Per 1,000-pair bin, gene distance was calculated as the average over all 1,000 pairs. For each 1,000-pair bin, the fraction of coexpressed pairs relative to the total number of pairs in each orientation group in each bin was calculated and plotted. Functional categorization of genes TAIR’s GOslim, the Gene Ontology (GO) developed for plants (Berardini et al. 2004) was used to classify the genes present in local coexpression domains. The three aspects of GOslim, molecular function, biological process and cellular component, were analyzed in parallel. With Python scripts, the number of pairs of which both members could be classified in GOslim was determined, and the number of pairs of which both members fall into the same well-defined GOslim category was also determined. The GOslim categories of ‘unknown’ and ‘other’ were not included into well-defined categories, because they give less (or no) information about functional categorization. The percentage of coexpressed pairs falling into the same well-defined category was compared with that of non-coexpressed pairs to determine whether coexpressed genes are more enriched in the same functional category than non-coexpressed genes. Assessing synteny between Arabidopsis and rice The Inparanoid Eukaryotic Orthologous groups database (inparanoid.cgb.ki.se; O’Brien et al. 2005) was used to download all known orthologous and inparalogous clusters between Arabidopsis and rice. Inparanoid defines inparalogs as paralogs that arose through gene duplication after speciation. Inparalogs can form a group of genes that together are orthologous to a gene in another species. There are 9,044 orthologous clusters between Arabidopsis (from Ensemble) and rice (from the Model Organism database) and all of them were taken into account. These clusters were downloaded on Dec. 12, 2005. In the orthologous clusters, 15,544 sequences (proteins) from Arabidopsis are inparalogs and 14,807 sequences (proteins) from rice are inparalogs. More than half of Arabidopsis and rice inparalogs are many-to-many or many to-one orthology cases. Less than half of the cases are one-to-one orthology cases. The Ensembl protein IDs (for Arabidopsis) and the Model organism database protein IDs (for rice) in Inparanoid were first translated to their unique gene identifiers in the respective TIGR annotation by BLASTP using an E-value < e−20. This yielded 14,753 unique Arabidopsis genes and 12,428 unique rice genes as inparalogs. The pairs of genes in local coexpression domains were analyzed to determine which genes in a rice local coexpressed pair have orthologs in an Arabidopsis local coexpressed pair, and vice versa. Because of the larger coverage of genes and tissues, only the MPSS expression data were used for this analysis. As coexpressed triplets and quadruplets are always combinations of coexpressed pairs, they were not further analyzed. For comparison, the pairs of genes that are not coexpressed were analyzed to determine how many non-coexpressed pairs, or one of their member genes, have orthologs in the other plant species. The numbers were then compared between coexpressed pairs and non-coexpressed pairs to determine the significance of occurrence of syntenic local coexpression domains. Results Local coexpression domains consist of two to four neighboring genes The TIGR version 3 of the rice genome has 57,915 predicted genes. This is about twice the number of genes predicted for Arabidopsis (28,952 genes; TIGR5 annotation). The coverage of the MPSS expression data for the rice genome is 40%. The expression coverage in the Affymetrix array data set we used is 26% (Table 1). This is about half or less of the expression coverage for the Arabidopsis genome (72% in the TIGR5 update; Ren et al. 2005). This difference can reflect the more advanced annotation of the Arabidopsis genome at this time, and/or the more complex (duplicated) organization of the rice genome. The rice genome annotation has more genes that are physically overlapping than the Arabidopsis genome. Excluding the smaller overlapping genes from the analyses, we were able to identify 12,920 gene pairs with MPSS expression data and 6,032 pairs with MA expression data in rice (Table 1; see also Materials and methods). Of these, 584 (4.5%) in MPSS and 320 (5.3%) in MA were identified to represent a local coexpression domain as defined as being coexpressed with a pair-wise Pearson’s correlation coefficient larger than 0.7 (Table 1). This percentage is similar to what we have found previously for Arabidopsis (Ren et al. 2005) and agrees well with other findings that ∼3–5% of a genome is tightly coexpressed (Semon and Duret 2006). Table 1Description of rice expression data used for whole-genome local coexpression analysisMPSSMAGenes with expressionExcluding overlapping genes23,14614,789Without expressed neighbor(s)5,0815,438represented in pairs18,0659,351Adjacent pairsTotal12,9206,032Tandemly duplicated pairs (td) 1,663 (12.9%)a573 (9.5%)aCoexpressed584 (4.5%)b320 (5.3%)bTotal excluding td11,2575,459 Coexpressed excluding td 438 (3.9%)c288 (5.3%)cCoexpressed adjacent pairsTotal584320Tandemly duplicated pairs146 (25%)d32 (10%)dTandemly duplicated pairsTotal1,663573Coexpressed146 (8.8%)e32 (5.6%) eaPercentage of tandemly duplicated pairs relative to the total number of adjacent pairsbPercentage of coexpressed adjacent pairs relative to the total number of adjacent pairscPercentage of coexpressed adjacent pairs excluding td relative to the total number of adjacent pairs excluding tandemly duplicated pairsdPercentage of coexpressed tandemly duplicated pairs relative to the total number of coexpressed adjacent pairsePercentage of coexpressed tandemly duplicated pairs relative to the total number of tandem duplicated pairs Notably duplicated genes are supposed to influence coexpression statistics due to their common origin (Lercher et al. 2003), although a surprising finding for the Arabidopsis coexpression domains was that only a minor fraction of duplicated genes were actually coexpressed (Ren et al. 2005). The occurrence of duplicated pairs in the rice set was determined with pair-wise protein BLAST using a cut-off of E < 0.2 (Fukuoka et al. 2004; Lercher et al. 2003; Williams and Bowles 2004). This identified 1,663 (12.9%) duplicated gene pairs in the MPSS gene pair data set and 573 (9.5%) duplicated gene pairs in the MA gene pair data set. Of these, only 146 (8.8%) in the MPSS data and 32 (5.6%) in the MA data were coexpressed (Table 1). Although this percentage is somewhat higher than the percentage of coexpression in non-duplicated pairs (3.9%), the majority of all duplicated pairs (91.2% in MPSS data and 94.4% in MA data) are not coexpressed. This shows that also in rice gene duplication does not correlate well with coexpression and suggests that expression divergence is a common phenomenon after duplication (Williams and Bowles 2004). Excluding the duplicated pairs from the coexpressed sets, there are 438 gene pairs in the MPSS data and 288 gene pairs in the MA data coexpressed in rice. This accounts for 75% (=438/584) in MPSS and 90% (=288/320) in MA of all coexpressed pairs. Therefore, also in rice the occurrence of duplicated genes cannot explain the occurrence of local coexpression domains. Extending the size of the local coexpression domain to triplets, quadruplets, pentaplets and on, requiring that all pair-wise combinations of genes have a tightly correlated expression, shows that few larger local coexpression domains exist (Table 2). No quadruplet domains could be identified when tandemly duplicated genes were excluded (Table 2). To assess the significance of the occurrence of the various local coexpression domains, we compared the number of coexpressed pairs, triplets and quadruplets with the average of such domains in 100 randomly generated genomes using the cumulative binomial distribution (Cohen et al. 2000). Such comparisons revealed that local coexpression pairs occur in the rice genome significantly more often than expected by chance alone (Table 2). However, when excluding the duplicated genes, triplets and quadruplets do not occur significantly more often than by chance (at P < 0.05) in both expression datasets. Local coexpression domains therefore consist of at most 2 genes when duplicated genes are not taken into consideration. This number appears smaller than in the Arabidopsis genome (Ren et al. 2005), but this may reflect the lesser coverage of the annotation of the rice genome. Table 2Local coexpression domains in the rice genomeRice genomeRandom genome (100×)TotalaCoexpressedbAveragecP-valuedPairsMPSS + tde12,920584 (4.52%)408 ± 171.46 × 10−17MPSS-tdf11,257438 (3.89%)356 ± 212.17 × 10−6MA + tdg6,032320 (5.30%)301 ± 170.012MA-tdh5,459288 (5.28%)271 ± 160.014TripletsMPSS + td7,77523 (0.30%)8.78 ± 2.92.95 × 10−5MPSS-td6,83113 (0.19%)7.74 ± 3.00.025MA + td2,4615 (0.20%)6.54 ± 2.7n.s.MA-td2,1493 (0.14%)5.10 ± 2.4n.s.QuadrupletsMPSS + td4,8873 (0.06%)0.24 ± 0.471.81 × 10−3MPSS-td4,3180 (0%)0.18 ± 0.39n.s.MA + td1,0790 (0%)0.14 ± 0.37n.s.MA-tdndindndndaTotal number of pairs, triplets, quadruplets in each data setbCoexpressed pairs, triplets, quadruplets in each data set. Percentages in brackets are coexpressed relative to the totalcAverage plus/minus standard deviation from 100 randomizationsdP-value according to the cumulative binomial distribution (Cohen et al. 2000) for obtaining such a result by chance. P < 0.05 is considered significant; n.s.: not significanteMPSS data set including tandemly duplicated genesfMPSS data set excluding tandemly duplicated genesgMA data set including tandemly duplicated geneshMA data set excluding tandemly duplicated genesiNot determined Local coexpression domains seem randomly distributed over the genome (Fig. 1; MPSS data only). Only 10 coexpressed pairs are common between the MPSS and MA coexpressed sets, out of a total of 4,643 common pairs (excluding tandemly duplicated pairs). Detailed information about locations, orientations, expressions and gene distances (only for the pairs) for all pairs, triplets and quadruplets in both the MPSS and MA data sets are given in the supplementary data. All subsequent analyses were focused on domains consisting of non-duplicated gene pairs, unless stated differently. Fig. 1Distribution of local coexpression domains over all 12 rice chromosomes. Rectangles are schematic representation of chromosomes 1–12 from top to bottom. The numbers on the top show the scale in million bases along the chromosomes. Each gene in a local coexpression domain is depicted with a black bar. Only MPSS datasets excluding tandemly duplicated genes are shown. The orders of the drawings in each rectangle are: first lane, coexpressed pairs; second lane, coexpressed triplets; third lane, coexpressed quadruplets, fourth lane, partially syntenic coexpression domains (PSCDs) between Arabidopsis and rice Orientation and distance do not solely explain the occurrence of local coexpression In yeast, there are several examples that divergently transcribed promoter regions are the cause of co-regulated neighboring genes (Korbel et al. 2004; Kruglyak and Tang 2000). If promoter sharing is an important mechanism for coexpression in the rice genome, divergently transcribed gene pairs should be over-represented in the sub-population of coexpressed pairs, compared to coexpressed pairs that are tandemly or convergently transcribed. For all three-orientation groups, the number of pairs and the number of coexpressed pairs in the rice genome were determined (Table 3). For each orientation group, the fraction of coexpressed pairs relative to the total number of pairs in that group was calculated (Table 3). None of the fractions are significantly different from each other using a statistical test for comparing population proportions (Ott and Longnecker 2001). The fraction of coexpressed divergent pairs in the MPSS data is the lowest of the three groups (Table 3). Therefore, shared promoter regions cannot solely explain the coexpression of adjacent genes. Table 3Orientation of coexpressed gene pairsOrientation groupsaTotalbCoexpressedcMPSStan-td5,621239 (4.25%)div-td2,41882 (3.39%)con-td3,218117 (3.64%)MAtan-td2,707143 (5.28%)div-td1,22472 (5.88%)con-td1,52873 (4.78%)atan-td, div-td, con-td, respectively are the sub-groups of tandemly, divergently, convergently transcribed pairs excluding tandem duplicatesbTotal number of pairs in each direction groupcNumber of coexpressed pairs in each direction group. Percentages in the brackets are number of coexpressed pairs relative to the total number of pairs. None of the proportions are significantly different from each other according to the z test for comparing population proportions The physically closer two genes are, the higher the likelihood is that they are coexpressed due to either cis or trans-activation (Hershberg et al. 2005). Therefore, we determined the intergenic distance, defined as the sequence length in nucleotides from the annotated end of one gene to the annotated start of the neighboring gene, including UTRs when known, otherwise taking the start and stop site for translation. This distance was used to investigate whether it would explain the characteristics of local coexpression domains. In Fig. 2, the fraction of coexpressed pairs is plotted for each orientation and for each 1000-pair bin after sorting based on intergenic distance. The results show that the fraction of coexpressed pairs, irrespective of gene orientation, does not decrease with larger gene distance. When gene distance is defined as the sequence length from the start of one gene till the start of the next gene (Ren et al. 2005), the result is similar (data not shown). As a consequence, increasing intergenic distances do not seem to be a barrier for the occurrence of local coexpression and short intergenic distances do not favor coexpression. Therefore, intergenic distance does not solely explain local coexpression in the rice genome, as it did not in the Arabidopsis genome (Ren et al. 2005). Fig. 2Gene distance does not solely explain the occurrence of coexpression. Gene distance, defined as the length in nucleotides from the annotated end of one gene to the annotated start of the next gene relative to the strand the genome that is given, with annotated start always smaller than the annotated end. X-axis is the averaged gene distance (in base pair) in each 1,000-pair bin. The Y-axis depicts the number of pairs (A, D), number of coexpressed pairs (B, E) and the fraction of coexpressed pairs (C, F), relative to the total number of pairs in each orientation (tan: tandem pairs; div: divergent pairs; con: convergent pairs) in each 1,000-pair bin Functional categorization of coexpressed genes To characterize the kind of genes that are present in the rice coexpression domains, the gene ontology (GO) developed for plants (GOslim; Berardini et al. 2004) was used. The GOslim ontology provides a controlled vocabulary to describe gene and gene product attributes in plants, focusing on three aspects of annotation: molecular function, biological process and cellular component. Each aspect has 15–16 categories with 4–5 categories having terms like “unknown” or “other”. To all pairs of genes, each aspect of the GOslim annotation was assigned. For each aspect, the number of pairs was determined for which both member genes were covered by a GOslim assignment. In addition, the number of pairs for which both member genes fall into the same well-defined categories (excluding the “unknown” and “other” subcategories) was determined. The fraction of the latter was compared between coexpressed pairs and non-coexpressed pairs to determine whether coexpressed pairs were enriched in the same categories (Table 4). In the MPSS data, the GOslim annotation coverage for both member genes in a pair is 22% for molecular function, 12% for biological process and only 3.4% for cellular component. In the MA data, these 3 percentages are: 25%, 13% and 3.7%, respectively. Comparing these figures with the GOslim coverage of the Arabidopsis genes (data from Ren et al. 2005, but updated to the TIGR5 annotation), which is ∼94% for all 3 aspects, shows that currently the rice genome is considerably less well annotated than the Arabidopsis genome. When comparing the coexpressed and non-coexpressed pairs in rice for the fraction of gene pairs falling into the same well-defined GOslim category, there is no significant difference (Table 4). Therefore, also in the rice genome coexpressed gene pairs are not enriched for the same functional category. Table 4Distribution of gene pairs over GOslim categories (Non-duplicated pairs)AllaCoexpressedbNon-coexpressedcP-valuedMPSS    GO_func        coverede250210024020.42        sameKnCatf365 (14.6%)12 (12.0%)353 (14.7%)    GO_proc        Covered13665013160.47        sameKnCat144 (10.5%)7 (14%)137 (10.4%)    GO_comp        Covered383173660.60        sameKnCat113 (29.5%)6 (35.3%)107 (29.2%)MA    GO_func        coverede13658312820.13        sameKnCatf177 (13.0%)7 (8.43%)170 (13.3%)    GO_proc        Covered707436640.13        sameKnCat67 (9.48%)2 (4.65%)65 (9.79%)    GO_comp        Covered202101920.24        sameKnCat45 (22.3%)4 (40%)41 (21.4%)aNumber of neighboring pairs excluding td. All other pairs are all duplicate-free, unless stated otherwisebNumber of coexpressed pairscNumber of non-coexpressed pairsdP value from the standard normal tables of the z statistic for the difference of the population proportion between coexpressed pairs and non-coexpressed pairs in the rice genome; *, significant (two-tailed; P < 0.05). The P value is the probability under the null hypothesis that the two population proportions are the sameeNumber of pairs of which both members are assigned (covered) with GOslim categoriesfNumber of pairs of which both members fall into the same “known” GOslim category (excluding the categories with the indications ‘unknown’ and ‘other’). Percentage is the number of pairs relative to the number of pairs covered Microsynteny of local coexpression domains between rice and Arabidopsis The structural characteristics of local coexpression domains in rice and in Arabidopsis (Ren et al. 2005) are remarkably similar. This prompts the question whether such domains also share functional characteristics and possibly consist of the same or related genes. Microsynteny in local expression domains of these two genomes would reflect conservation of such domains. The Inparanoid Eukaryotic Orthologous database (O’Brien et al. 2005) was used to retrieve the current list of genes that are supposed to be orthologous between Arabidopsis (14,753 genes) and rice (12,428 genes), including all many-to-many relationships. The genes establishing coexpressed pairs based on MPSS expression data in either Arabidopsis (944 pairs including 116 duplicated pairs; data from Ren et al. 2005, but updated to the TIGR 5 annotation) or rice (584 pairs, including 146 duplicated pairs) were searched against these lists. This way, we aimed to identify the pairs of which both genes in the pair have an ortholog in the other plant and these orthologs are also coexpressed The analyses showed that there was not a single coexpressed pair in either Arabidopsis or rice of which both genes are orthologous to a gene of a coexpressed pair in the other species. Therefore, given the current annotation of the two genomes, there are no syntenic local coexpression domains between Arabidopsis and rice. Partially syntenic local coexpression domains can occur by chance In 34 cases though, one gene of a coexpressed pair in one plant species was orthologous to at least one gene of a coexpressed pair in the other plant. That is 3.6% of all (944) coexpressed pairs in Arabidopsis and 5.8% of all (584) coexpressed pairs in rice. We will refer to such a case as a partially syntenic coexpression domain (PSCD). To assess the significance of such partially syntenic domains, we evaluated all the genes in non-coexpressed pairs, comparing Arabidopsis (15,629 pairs including 617 duplicated pairs) and rice (12,336 pairs including 1,517 duplicated pairs) to establish whether PSCDs are more enriched in the genome than partially syntenic non-coexpressed domains (PSND). We identified 4,488 PSNDs (72 due to duplicated pairs) between all non-coexpressed pairs of genes in both plant genomes. This is 28.7% of all Arabidopsis non-coexpressed pairs and 36.4% of all rice non-coexpressed pairs. The percentage of PSNDs among non-coexpressed pairs is 6–8 times higher than that of PSCDs from coexpressed pairs. Therefore, PSCDs do not seem to occur more often than expected by chance alone. A complicating issue in the analysis of synteny is the occurrence of many-to-many orthologs. The Inparanoid database defines so-called inparalogs as paralogs arising through gene duplication after speciation. These can form a group of genes that together are orthologous to a gene in the other species. As a result, there can be many to many, many to one and one to one relationships. Individual member genes in many-to-many or many-to-one relationships may not be the main orthologs. Interestingly, there is one many-to-one case in which four Arabidopsis genes are all orthologs of the same single rice gene (Os07g43560.1). These 4 Arabidopsis genes are: At4g23140.2, At4g23150.1, At4g23230.1 and At4g23270.1. The first two, At4g23140.2, At4g23150.1, form a local coexpressed pair. The other two genes, At4g23230.1 and At4g23270.1, are not more than ten genes away from the previous two genes on the same chromosomal region. The latter two genes are separated from each other by a few genes. Further analysis shows that gene At4g23270.1 has a duplicated neighbor, At4g23280.1, but is not coexpressed with it. It is, however, coexpressed with its other neighbor At4g23260.1, but is not duplicated with it. Orthology is established between At4g23270.1 and the rice gene Os07g43560.1, but not between any of the neighbors of the Arabidopsis genes. The rice gene Os07g43560.1 is also coexpressed with one of its neighboring genes Os07g43540.1, but it is not a duplicate of it, while this rice gene itself is a duplicate of another neighboring gene (Os07g43570), but it is also not coexpressed with it. A schematic representation of the resulting gene configuration is given in Fig. 3. Such detailed analyses may reveal local microsynteny in the twilight zone of statistical significance and evolutionary relevance. Fig. 3Schematic representation of the chromosomal regions covering genes involved in a four-to-one orthology between Arabidopsis and rice. Top part of the figure is the chromosomal region from rice (from gene locus Os07g43540.1 to gene locus Os07g43570.1). Bottom part of the figure is the chromosomal region from Arabidopsis, representing 23 genes (from gene locus At4g23120 to At4g23340; the numbers in the picture do not carry “At4g”). Black arrows represent the four Arabidopsis and the one rice gene involved in this orthology, and dashed curved connecting lines show the orthology relationships. Black bracket-like lines depict duplication and genes connected and included within by black bracket line are duplicated to each other. Dotted lines depict coexpression relationship and genes connected and included by dotted line are coexpressed with each other Discussion Local coexpression domains represent only a small part of the genome Setting stringent criteria for coexpression using MPSS and microarray expression data, the rice genome was found to contain a small but significant number of local coexpression domains that range from predominantly two, up till 4, genes, irrespective of the expression platform used for analysis. This is similar to the situation in Arabidopsis (Ren et al. 2005). It shows that a genome is essentially not a random entity with respect to the occurrence of local coexpression domains. Our results agree with other coexpression studies where strong coexpression was shown to occur only within close proximity of several genes (Cohen et al. 2000; Hershberg et al. 2005; Lercher et al. 2003; Semon and Duret 2006). Although coexpression was shown to extend to regions covering up to 30 genes and more (Spellman and Rubin 2002) and to cover chromosomal regions up to 100 kb (Ma et al. 2005; Spellman and Rubin 2002; Williams and Bowles 2004) and more, there appears to exist a decrease in the strength of coexpression with increasing distances. The local coexpression domains described here represent ∼4–5% of the potential coexpression fraction in the whole genome as found in other studies (Semon and Duret 2006). Larger but looser coexpression domains might cover up to ∼10% (Cohen et al. 2000; Williams and Bowles 2004) till 20% (Spellman and Rubin 2002) of the genome. The difference in occurrence between local and longer-range weaker but still statistically significant coexpression domains is highly dependent on the method used (Semon and Duret 2006). Moreover, the expression platforms used focus on snapshots of gene expression at the RNA level. Such data ignore various post-transcriptional regulatory mechanisms that can for example result in poor correlations between RNA and protein levels. Although we have shown before that in a transgenic set-up the creation of an artificial local coexpression domain results in markedly improved RNA/protein correlations (Mlynarova et al. 2002), this does not need to be the case for all endogenous local coexpression domains now identified. The terms cluster or chromosomal domain and associated terms such as neighboring are generally based on a (much) more loose definition compared to the definition used here to identify local coexpression domains. Local coexpression domains require a pair-wise correlation between the expressions of ALL adjacent genes above 0.7. The larger domains are defined on the basis of the use of a sliding window of either a given sequence length (number of nucleotides) or of a given number of genes (Spellman and Rubin 2002; Williams and Bowles 2004). In such a window, the average correlation is calculated and compared with simulated sets. This allows for the presence of genes within a domain that are not strongly (co)expressed but are “carried along for a ride” in the open chromatin domain (Spellman and Rubin 2002). The two expression data sets here analyzed show very little overlap in coexpressed pairs (only 10 out of 4,643 all common pairs). The low number of shared pairs is thought to reflect the biological background of the data sets. The MPSS data concern a broad range of tissues and experimental conditions, while the MA data only cover crown and growing point tissue under control and salt stress condition (Walia et al et al. 2007). The number of libraries taken into account, as well as their biological background, obviously influences the possibility of identifying significant numbers of coexpression domains above the number of domains expected by chance. Two genes evaluated in four comparably similar conditions (MA data) are predicted to have a higher likelihood of exhibiting a similar expression pattern, than when examined in 18 different (MPSS) conditions. When considering genome-wide local coexpression of genes, a wide diversity of tissues and/or conditions should be taken for analysis. Highly tissue-specific coexpression may be masked in this approach and should be analyzed by other means. Parameters shaping local coexpression domains The existence of local coexpression domains in rice could not be explained solely by gene orientation, such as tandemly, divergently or convergently oriented gene pairs. No relative enrichment of the proportion of coexpressed pairs was seen. The fraction of coexpressed genes in the divergent orientation was even lower than for the other two orientations (Table 3). So shared promoter regions (for divergent pairs) and transcriptional read-through (for tandem pairs) do not explain the local coexpression domains in rice, similar to what we have concluded for Arabidopsis (Ren et al. 2005). This is in contrast to some other studies in which shared promoter region (for divergent pairs) and transcriptional read-through established coexpression domains (Semon and Duret 2006). Whereas we do not detect any preferred orientation to result in coexpression in rice, other studies show a higher degree of coexpression in divergent and tandemly oriented gene pairs (Williams and Bowles 2004; Zhan et al. 2006). The differences in conclusions are most likely due to the different methods used, such as the definition of coexpression of neighboring genes as well as the dataset and/or expression platform used. Gene distance is not the explanatory factor for the occurrence of local coexpression domains in rice. No significant decrease in the fraction of coexpressed genes was observed with increasing intergenic distance (Fig. 2). The fraction of coexpressed pairs does not decrease even with gene distances up to 12 kb. It shows that at a relatively large distance, neighboring genes can still be coexpressed. Another study reported that when genes >12 kb apart were taken into account, the negative correlation between coexpression and gene distance was gone (Williams and Bowles 2004). In a comparative study of 6 eukaryotic genomes, coexpression was shown to vary at chromosomal distances above 100 kb (Fukuoka et al. 2004). This suggests that considerable coexpression of neighboring genes can occur even at large gene distance, although the coexpression may not be related to the physical distance anymore. While gene distance itself is not predictive for coexpression (Cohen et al. 2000; Kruglyak and Tang 2000), the likelihood of coexpression would favor short gene distances (Hershberg et al. 2005; Hurst et al. 2002; Lercher et al. 2003; Semon and Duret 2006). It should be kept in mind that the rice data set now analyzed is far from complete in terms of its annotation, so what are now far-apart neighboring genes may be no longer directly neighboring the moment the annotation is improved. Because the characteristics of the local coexpression domains in Arabidopsis and rice are remarkably alike, the actual numbers and possibly even genes involved in local coexpression domains may be subject to change, but not the occurrence of such domains in these genomes. With the gene ontology developed for plants (GOslim), there is no evidence that coexpressed genes are more enriched in the same functional category in comparison to non-coexpressed genes (Table 4). Previous studies suggested that clustering of functionally related genes would occur in all metazoans (Cohen et al. 2000; Lercher et al. 2003). Recent studies demonstrated a significant enrichment for coexpressed genes in the same metabolic pathway (Williams and Bowles 2004) or the same biological processes (Zhan et al. 2006), although this appeared not to be the explanation for the coexpression of neighboring genes (Williams and Bowles 2004). In worm, clusters of similarly expressed genes cover similar biological functions (Roy et al. 2002). In human, coexpression over the whole genome was shown to correlate with functional relationships between the genes (Lee et al. 2004). Our study found no enrichment of coexpressed gene pairs in the same functional category than non-coexpressed pairs, suggesting that it is not necessarily true that the natural selection maintained regional coexpression by keeping genes with similar functions in adjacent positions (Cohen et al. 2000; Semon and Duret 2006). The genomic context of genes is supposed to play an important role in the regulation of gene expression (van Drunen et al. 1997). In a number of coexpression studies in various organisms, the occurrence of coexpression domains, whether small (local) or larger (global), sometimes independent of gene orientation and gene distance, were all supposed to be regulated at the level of higher-order chromosomal structure (Cohen et al. 2000; Hershberg et al. 2005; Ren et al. 2005; Spellman and Rubin 2002; Williams and Bowles 2004; Zhan et al. 2006). It will be difficult to formally exclude the possibility of transcriptional regulation as a cause of local coexpression. Co-regulated transcription could occur through shared promoter elements between the neighboring genes in coexpression domains. However, previous experience with transgene expression data indicated that the particular position of neighboring genes in a genome affects the expression of that gene considerably (Mlynarova et al. 1994, 1995). Two physically neighboring transgenes could only be made to into an artificial local expression domain, that is, show correlated expression, when chromatin-organizing elements were placed around the genes (Mlynarova et al. 2002). Although this experimental result indicates the importance of chromatin organization in establishing local coexpression in a plant genome, this result does not need to be the case for all coexpression domains identified. With this caveat, this study of local coexpression domains in rice that are independent of duplication, gene orientation, or gene distance strengthen the notion that the regulation of genes in such domains resides at the level of higher-order chromatin structures. Future studies using for example advanced fluorescence in situ hybridization (FISH) technology on interphase chromosomes (Walter et al. 2006), or the transgenic approach outlined above (Mlynarova et al. 2002), will generate further experimental support for the location and characteristics of local coexpression domains. In addition, the new approaches in and insights from epigenetics and epigenomics (Henderson and Jacobsen 2007), such as the genome-wide mapping and analysis of DNA methylation (Zhang et al. 2006; Zilberman et al. 2007) and future histone modification maps (Esteller 2007) of plants, possibly in combination with genetics (Jansen and Nap 2001), will be useful to get more insight into the occurrence and function of local coexpression domains in plant genomes. Lack of microsyntenic coexpression From an evolutionary point of view, syntenic regions between species reveal genes for conserved and important traits. Macrosynteny is generally not easily detectable after a long evolutionary time, as colinearity erodes by various mechanisms, such as transposon activity, intra or inter-chromosomal rearrangements, duplications, translocations, inversions and/or individual divergence after speciation (Salse et al. 2002). While macrosynteny may not be detectable any more for genomes that diverged more than 100 million years ago (mya), microsynteny, i.e., conservation of local gene order and orientation, may still exist and be informative (Devos et al. 1999; Salse et al. 2002). Arabidopsis and rice are thought to have diverged about 120–200 mya (Salse et al. 2002). Microsyntenic local coexpression domains between Arabidopsis and rice would indicate the importance of the evolutionary conservation of regulatory systems beyond sequence similarity after the divergence of dicotyledonous and monocotyledonous plants. Analyses show that there is not a single coexpressed pair in either Arabidopsis or rice of which both genes are orthologous to a gene in a coexpressed pair in the other species. Therefore, there are no syntenic local coexpression domains between Arabidopsis and rice. Although the analyses were performed for only one monocot and one dicot and should be extended to many more genomes, the results could be taken to suggest that maintenance of coexpression has not been an important driving force in genome conservation during or after the divergence of dicotyledonous and monocotyledonous plants. Although individual genes in local coexpression domains in either rice or Arabidopsis may have an ortholog in the other species, establishing so-called partially syntenic coexpression domains (PSCDs), this does not seem to occur above chance in the context of whole-genome configurations. Without statistical significance, the occurrence of such PSCDs is unlikely to have any evolutionary relevance on a genome-wide scale. Detailed analyses of individual cases and gene locations may suggest the occurrence of local microsynteny and point to chains of evolutionary events in which the conservation of coexpression could be involved. However, more detailed studies are required to assess the functional relevance, if any, of such genomic constitutions. Electronic supplementary material Below is the link to the electronic supplementary material. (XLS 231 KB)