Datasets:

Dhruvil47

/

bioarxiv_papers

like 0

The overall results confirmed that the genetic elements under 280 scrutiny (i. e., the plasmids derived from Ect313, Ect1012, Ect9605, Ect111 and Ect7816) 281 constitute self-mobilizable plasmids, similarly to the case of pLD209 (17). Thus, all them are 282 seemingly capable of spreading the blaVIM-2 containing-Tn402-like integrons they carry 283 among a wide range of bacterial species. 284 Plasmids present in the above Ect transconjugants were extracted and subjected to 285 restriction mapping using EcoRI. With the exception of Ect7816, the plasmids derived from 286 Ect1012, Ect9605, Ect313, and Ect111 showed very similar restriction profiles between them, 287 which were in turn very similar to those of Ect209 (data not shown, see also below). 288 Moreover, the obtained sizes corresponded closely to the EcoRI fragment sizes predicted in 289 silico from the pLD209 complete DNA sequence (Fig. 3e; 13). The plasmids purified from 290 Ect7816 (henceforth, pBA7816) and Ect111 (henceforth pLA111) were subjected to further 12 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 291 sequencing (see Materials and Methods for details) for a subsequent comparative analysis of 292 the obtained structures. pLA111 (GenBank accession number MT192131) was found to be 293 identical to pLD209 (Fig. 3d and 3e, respectively), in agreement with the restriction analysis 294 mentioned above. In turn, pBA7816 (Fig. 3c, GenBank accession number MN240297) was 295 found to be very almost identical (99%) to pLD209, including the replication, transfer, and 296 stability modules. The only main difference between these two plasmids was found in the 297 adaptive module, and consists in the absence of the aacA4 aminoglycoside resistance cassette 298 in the Tn402-like element (Fig. 2A and Fig. 3). It is worth noting that Tn6336 and Tn6335 are 299 positioned in equivalent positions in the plasmids, that they are bordered by the same 5-bp DR 300 (5’-GTTTT-3’), and that they are inserted within another potential mobile element as judged 301 by the external 34-bp inverted repeats (5’- 302 GGGGGTGTAAGCCGGAACCCCAGAAAATTCCGTC-3’, gray triangles facing inwards) 303 and accompanying direct repeats located upstream of the IRi and downstream of the IRt (Fig. 304 3). Our BLASTn search of the NCBI bacterial DNA database (as for October 9, 2020) using 305 as query the pLD209 sequence (KF840720.1) found homology (99% nucleotide identity) 306 between a composite fragment of 1,861 bp from this plasmid with a fragment extending 1,802 307 bp present in plasmid p3 from an environmental Pseudomonas koreensis strain, P19E3 308 (GenBank accession CP027480.1, positions 265,881 to 264,070). This region in P. koreensis 309 p3 covered exactly an element bordered by identical 34-bp inverted repeats, and also similar 310 hypothetical protein coding sequences, than those found in pLD209 after removing in silico 311 the Tn6335 insertion at the 5’-GTTTT-3’ direct repeat (122 bp from the IRi, from positions 312 891 to 1,012, and 1,739 bp from the IRt, from positions 8,651 to 10,389 in KF840720.1 (see 313 also 13).

These observations strongly suggest that a similar mobile external element was 314 collected by a pLD209 ancestor, probably as the result of a trans-mediated transposition 315 event, and subsequently targeted by a Tn402-like transposon thus generating the backbone of 13 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 316 the adaptive module now observed in pLD209 and related plasmids (Fig. 3). These plasmids, 317 which are amply disseminated among members of the P. putida group in clinical settings of 318 Argentina, are endowed with the worrying potentiality to disseminate carbapenem and other 319 antimicrobial resistance cassettes carried by Tn402-like integrons to co-existing pathogens 320 including members of the Enterobacteriaceae family or P. aeruginosa as shown above. 321 Concerning the other seven P. putida G isolates analyzed, i. e., P. asiatica HP613, P. 322 putida BA9115, P. putida BA7908, P. monteilii BA9713, P. monteilii HB157, P. putida G/I 323 HP813, and P. putida G/II LA1008, our repeated attempts to detect the presence of plasmids 324 by either conjugation or transformation assays were unsuccessful. This suggested a 325 chromosomal, rather than a plasmid, location of the genetic elements harboring blaVIM-2 (Fig. 326 2) in these P. putida G isolates. 327 328 Comparative analysis between pLD209 and related plasmids carried by nosocomial and 329 environmental Pseudomonas species 330 Our BLASTn search using the pLD209 sequence as a query (see above) detected two 331 plasmids, pKF715D and pMRVIM0812, showing high levels of nucleotide identity and 332 structural organization with pLD209 including the replication, stability, and transfer modules 333 (Fig. 3). Among them, pKF715D (Fig. 3a) was found in an environmental P. putida strain, 334 KF715, obtained from contaminated soils near a biphenyl manufacturing plant in Japan (43, 335 44), and pMRVIM0812 (Fig 3b) was isolated from a clinical Pseudomonas sp. in the U.S.A. 336 At their adaptive modules, pMRVIM0812 contains a typical class 1 integron 337 encompassing the intI1 gene at the 5’-CS, blaVIM-2 and aminoglycoside 6'-acetyltransferase 338 aacA27 gene cassettes in the variable region, and the 3’-CS including qacEΔ-sul1 genes (Fig. 339 3b). A similar class 1 integron, designated In984, was previously described in a clinical 340 Pseudomonas oleovorans isolate, M13320 (42, GenBank accession number KJ668596). In 14 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license .

341 pMRVIM0812 this class 1 integron is flanked by two copies of miniature inverted-repeat 342 transposable elements (MITEs), the whole structure bordered in turn by a 5-bp (5’-GATGA- 343 3’) DR (Fig. 3b). MITEs are non-autonomous mobile elements carrying inverted repeats, and 344 their mobilization is promoted by transposases encoded by adjacent transposons or by ISs 345 containing similar inverted repeats (45, 46). The assembly suggests that these MITE elements 346 captured the integron in a composite transposon-like structure, which was subsequently 347 mobilized to the present location in pMRVIM0812 by transposases provided in trans. Similar 348 capturing and transposing events of other class 1 integrons by flanking MITE elements have 349 been recognized as a mechanism for mobilizing antimicrobial resistance determinants (31, 350 47). 351 We noted that pMRVIM0812 and pKF715D, in contrast to other pLD209-type 352 plasmids (Fig. 3c-e), carry each a PAS-domain protein gene (Fig. 3 a,b). In pMRVIM0812 the 353 PAS-protein domain gene is bordered by a 50-bp inverted repeat, while in pKF715D is 354 adjacent to a MITE element with the whole arrangement flanked in turn by a 50-bp inverted 355 repeat. The inverted repeats in the above plasmids are flanked by 5-bp direct repeats in each 356 case, i. e., 5’-AGGAA-3’ in pKF715D and 5’-TGGAT-3’ in pMRVIM0812 (Fig. 3 a,b). 357 These analyses suggest that these PAS-domain coding sequences and associated elements 358 reached their present plasmid locations assisted in trans by factors provided by mobile 359 elements co-existing in the cells. 360 Finally, evidence exists that pLD209-related plasmids can additionally evolve not only 361 by gaining or loosing individual resistance cassettes in the Tn402 element, but also by losing 362 significant parts of their backbones as exemplified by plasmid pDCPR1 (18,182 bp; Fig. 3f) 363 isolated from clinical strains of both P. aeruginosa and Serratia marcescens in Argentina 364 (48). pDCPR1 shares high structural and nucleotide similarity to pLD209 at the adaptive, 365 replication, and stability modules, but lacks most genes involved in conjugal transfer (Fig. 3). 15 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 366 Still, the retention of oriT sequences suggests a mobilization potentiality for pDCPR1 in the 367 presence of a conjugative plasmid. The structural rearrangements found in pDCPR1 are most 368 likely associated to lateral transfer, and reinforce the role of pLD209-related plasmids as 369 efficient and plastic genetic platforms for the spreading of carbapenem and other 370 antimicrobial resistance genes among nosocomial pathogens (13; 48). 371 The above observations disclose a wide dissemination of conjugative plasmids sharing 372 similar backbones and structural organization among environmental and nosocomial members 373 of the Pseudomonas genus.

These plasmids are endowed with high adaptive significance, and 374 have most likely collected different mobile elements and resistant determinants during transit 375 through different bacterial hosts subjected to various selective conditions. Particular examples 376 are pLD209 and related plasmids, which provide platforms endowed with lateral transfer 377 ability to different class 1 integrons. These integrons apparently found their way to the 378 plasmid structure either in the form of a Tn402-like transposon or other entities capable of 379 transpose with the help of trans-acting factors. The plasticity inherent to these integron- 380 bearing adaptive modules in terms of exchanging gene resistance cassettes, exemplified by the 381 capturing of blaVIM-2 and other antimicrobial resistance genes (Fig. 3), has certainly 382 contributed to the adaptation of P. putida G species to the challenges of the clinical setting. A 383 worrying perspective thus emerges for the treatment of infections produced by MDR 384 nosocomial pathogens such as P. aeruginosa or members of the Enterobacteriaceae, 385 considering the ability of P. putida G species not only to serve as reservoirs but also 386 disseminate these wide-host range resistance plasmids by horizontal transfer. 387 388 blaVIM-2-containing Tn402-like transposons translocation to target sites in the genomes of 389 P. putida G members 16 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 390 Database searching analysis and experimental evidences with plasmid model systems 391 have identified a high selectivity of Tn5053/Tn402 family members for targets clustered in, or 392 close to, res regions upstream of tnpR genes of some members of the Tn3 family, and the 393 equivalent par regions associated to segregational mechanisms of particular plasmids (32, 33, 394 34, 35, 36, 37, 49). These studies also indicated that Tn5053/Tn402 transposons generally 395 insert on these target loci with the IRi boundary facing the resolvase gene, although in a very 396 few number of cases the opposite orientation (with the IRt end more close to the recombinase 397 gene) was also reported (26, 34, 35, 36). In addition, transposition events independent of the 398 presence of a res locus have also been found, albeit at very low frequencies and involving 399 random target selection and orientation (36, 50). 400 In six of the P. putida G strains analyzed here including P. asiatica (both PaA and PaB 401 clones), P. juntendi, P. putida G/II, and P. putida G/V, the Tn402-like transposons were 402 found in pLD209-type conjugative plasmids (Table 1). Our analysis above further indicated 403 that these transposons are inserted in these plasmids into a defective element bordered by 34- 404 bp inverted repeats, but no sequences resembling putative res or par target sites could be 405 identified in the vicinity of the insertion site.

This suggests either an unusual transposition 406 event or, alternatively, substantial sequence rearrangements near the site of insertion as 407 described in other cases (13, 33, 35, 51). 408 As noted above, Tn6335 was found in the pLD209 plasmid in P. asiatica LD209, but 409 apparently in another genomic location in isolate HP613 which is clonally related to LD209 410 (Table 1; Fig. 1 and Fig. S1). Different genomic locations for Tn6335 were also found in P. 411 putida G/II isolates HE1012 and LA1008 (Table 1; Fig. 1 and Fig. S1). Therefore, we decided 412 to characterize in further detail the genomic context of Tn6335 in both P. asiatica HP613 and 413 P. putida G/II LA1008. For this purpose, the cloning of the genomic region near the IRi 414 boundary of the transposon was attempted for both isolates, taking advantage of the presence 17 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 415 of a nearby blaVIM-2 gene conferring ceftazidime resistance (Table S4) and the single EcoRI 416 site within the tniB gene (Fig. 3). In line with this objective, we separately treated total 417 genomic DNA from these isolates with EcoRI, ligated the digested products into EcoRI- 418 digested pSU18 (52), transformed E. coli DH5α cells, and selected for colonies containing 419 inserts that, besides the chloramphenicol resistance provided by pSU18, additionally 420 conferred ceftazidime resistance (see Material and Methods for details). Different clones in 421 each case were then subjected to plasmid purification and sequencing analysis of the inserts. 422 In the case of P. asiatica HP613, we recovered a 9,031-bp EcoRI insert (designated 423 pSU18-HP613; Fig. S2A), showing the expected 5,307 bp region carrying blaVIM-2 and 424 extending from the IRi of Tn6335 to the EcoRI site located 406-bp of the 3´end of tniB (Fig. 425 S2A). The remaining 3,725 bp falling outside the IRi boundary and extending to a nearby 426 EcoRI site in the HP613 genome (Fig. S2A) exhibited 97.8% nucleotide identity to the 427 indicated equivalent res-tnpRA region present in a complete Tn501 element described in P. 428 aeruginosa plasmid pVS1 (GenBank accession number Z00027.1, nucleotide positions 4,620 429 to 8,343; Fig. S2B). Tn501 forms part of the Tn21 subgroup of the Tn3 transposon family (32, 430 33). All members of this subgroup show a tnpRA module composed of tnpR (serine 431 recombinase) and tnpA (transposase) genes transcribed in the same direction, preceded by a 432 res region composed of resI, resII and resIII subsites (32, 33, 53; 54; see also Fig. S2B). Our 433 comparative sequence analysis indicated that in the P. asiatica HP613 genome the resI subsite 434 of this Tn501-like element had in fact been impacted by Tn6335, with the IRi of the 435 transposon facing the recombinase gene (Fig.

S2B). A similar finding was observed by other 436 authors in P. aeruginosa Pavimgi1, where the res site was also impacted between the resI and 437 resII subsites (Fig. S2B) by another Tn402-like transposon (26). 438 A BLASTn search using this 4,442 bp genomic sequence as query identified almost 439 identical stretches of around 3,790 bp (99 % nucleotide identity) in plasmid RPL11 of a P. 18 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 440 aeruginosa isolate (GenBank accession AF313472) and in the chromosome of P. putida 441 H8234 (GenBank accession CP005976.1); positions 3,173,668 to 3,178,100). This fragment 442 includes a complete tnpRA module followed by a 38-bp IRr boundary (Fig. S2C) of a Tn1403 443 transposon (Stokes et al. 2007). Tn1403 is also included in the Tn21 subgroup of the Tn3 444 family (32). The remaining fragment towards the EcoRI site of around 650 bp also shows 445 almost complete nucleotide identity to the equivalent region of the P. putida H8234 446 chromosome, and included 124 bp of the 3´ coding region of sdhC encoding the cytochrome 447 b556 subunit of the succinate dehydrogenase (Fig. S2C). The complete sdhC gene (375 bp in 448 length, GenBank accession AGN79094.1) is actually present in this locus in the P. putida 449 H8234 chromosome, limited in turn at its 5´ region by another 38-bp inverted repeat 450 characteristic of Tn3 family transposons (not shown). This suggests that the whole element 451 may in fact represent a novel Tn1403-like transposon carrying a sdhC catabolic gene. Similar 452 genetic arrangements have in fact been described for other Tn21 subgroups members of the 453 Tn3 family, although the catabolic genes they carry differ from the above (32). 454 Concerning the target region impacted by Tn6335 in the P. putida G/II LA1008 455 genome, our comparative sequence analysis of the res regions of the Tn1403-like transposons 456 mentioned above (Fig. S2D) indicated that Tn6335 was inserted between the resI and resII 457 subsites of the target element. As above, Tn6335 was inserted with its IRi boundary facing the 458 tnpR gene of the Tn1403-like target. The tnpRA transposition modules of the Tn501-like and 459 Tn1403-like found above (Fig. S3) showed 86 % nucleotide identity between them, but 460 differed in the length of the corresponding tnpR genes which were 561- and 618-bp, 461 respectively. This situation has been described previously for Tn21 subgroup members of the 462 Tn3 family (32, 53). 463 Finally, in the case of P. monteilii HB157 our analysis of the immediate genomic 464 sequences in which the blaVIM-2-containing Tn402(tniABQ) element (Fig. 2C) was located 19 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020.

The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 465 indicated that it was also inserted within the res region of another Tn3 family transposon. 466 Using an inverse PCR approach followed by cloning and sequencing of the obtained 467 amplicons (see Materials and Methods for details), we could characterize a 732 bp fragment 468 of the HB157 genome corresponding to the insertion site of Tn402(tniABQ) in the 469 immediate vicinity of its IRt border (Fig. S2E). Comparative sequence analysis indicated that 470 this 732 bp fragment encompassed a complete tnpR resolvase gene (615 bp, 204 amino acids), 471 followed by the first 14 bp of an aminoglycoside O-phosphotransferase (aph(3'')-Ib) gene 472 (Fig. S2E). Moreover, a BLASTn search indicated that this fragment showed high identity 473 with equivalent segments located in the genomes of different Pseudomonas species including 474 (among others) the P. aeruginosa FDAARGOS_570 chromosome (Genbank CP033835.1, 475 positions 3,824,646 to 3,825,377), a plasmid carried by the same strain (CP033834.1, 476 positions 25,649 to 26380), the P. mosselii plasmid pMOS94 (MK671725.1, positions 21,091 477 to 21,822), the P. putida JBC17 chromosome (CP029693.1, positions 819327 to 823821), the 478 P. putida 15420352 plasmid p420352-strA (MT074087.1, positions 126666 to 131160), as 479 well as in the chromosomes and plasmids of other species of clinical and environmental 480 relevance including Klebsiella pneumoniae PMK1 plasmid pPMK1-C; Citrobacter freundii 481 RHBSTW-00444 plasmid pRHBSTW-00444_2; Stenotrophomonas maltophilia SM 866 482 chromosome; Aeromonas caviae WCW1-2 chromosome; Aeromonas salmonicida plasmid 483 pRAS2, etc. In all of the above genomes this fragment forms part of a transposon found in 484 Aeromonas salmonicida designated Tn5393c (55), which is essentially identical to the 485 originally-described Tn5393 found in Erwinia amylovora (56), an ubiquitously distributed 486 member of the Tn3 family carrying streptomycin resistance genes (32). In these Tn5393-like 487 transposons the tnpR gene is separated from an oppositely-oriented tnpA gene (55) by a 125 488 bp intergenic region in which the three res subsites recognized by the recombinase (32, 57) 489 could be inferred (Fig. S2F). Efficient transposition of Tn5053/Tn402 members generally 20 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 490 depend on externally coded accessory functions, namely a res site served by a cognate 491 resolvase, but each interaction system may target a different res subregion (or even lie outside 492 this region) (36).

In this context, our analysis indicated that Tn402(tniABQ) was inserted 493 into the intergenic region between the resII and resIII subsites of a Tn5393-like element 494 located in the P. monteilii HB157 genome (Figs. S2, E and F). Moreover, the IRt boundary of 495 Tn402(tniABQ) was found facing the tnpR resolvase gene of its Tn5393-like target (Figs. 496 S2E and S3C), an insertion orientation different to the most frequently found for 497 Tn5053/Tn402 members including Tn6335 on its Tn21targets described here (Fig. S2). This 498 insertion orientation with the IRt closer to the tnpR of their targets has however been reported 499 in other few cases for other members of the Tn5053/Tn402 family (35, 36). 500 It has been suggested that the predisposition of Tn5053/Tn402 family members for res 501 sites provides access to alternative/more efficient vehicles of dissemination within different 502 bacterial species sharing similar environmental niches, including those composing the human 503 microbiota (32, 34, 58). Compound elements between Tn402-like and Tn21 transposons, and 504 subsequent derivatives, have in fact been found widely distributed in the human microbiome 505 (58). The results presented here indicated that the res regions of Tn21 transposons present in 506 the genomes of different P. putida G species have been the preferential targets of Tn402-like 507 elements carrying blaVIM-2 such as Tn6335, which most likely arrived to these cells as 508 passengers of pLD209-related conjugative plasmids. Moreover, they also show that other 509 members of the Tn3 family such as Tn5393, of ubiquitous distribution among different 510 gammaproteobacteria families, can also accommodate Tn402-like elements. It follows that P. 511 putida G members can provide many elements to host Tn402-like integrons carrying 512 antimicrobial resistance cassettes. Carbapenem therapy appears as a main force behind the 513 selection of P. putida G clonal lineages in which transposition events relocated the incoming 514 blaVIM-2-containing Tn402-like elements from plasmids to other preferred genomic locations 21 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 515 such as some pre-existing members of the Tn3 family, and also of further rearrangements 516 occurring in the newly-generated hybrid structures. These chimeras combining Tn402-like 517 and selected Tn3 family elements could certainly play important roles in the dissemination of 518 blaVIM-2 genes to pathogenic species of Pseudomonas and other bacteria causing infections in 519 humans, food animals, and livestock (24, 26, 58, 59). 520 DISCUSSION 521 We characterized here in detail the genetic platforms harboring blaVIM-2 in a set of 522 carbapenem-resistant P. putida G clinical isolates obtained from different hospitals in 523 Argentina, which were collected along an extended time period.

Our study revealed notable 524 taxonomic and genetic features among these isolates, which belong to a group better known to 525 be composed of environmental organisms rather than nosocomial pathogens. The carbapenem 526 resistant phenotype of our local collection of P. putida G isolates could be generally ascribed 527 to the carriage of blaVIM-2 metallo--lactamase genes within Tn402-like integrons. The role of 528 class 1 integrons in the dissemination of antimicrobial resistance among bacterial 529 communities is well recognized (37), but the genetic elements that are exchanged between 530 nosocomial and/or environmental bacteria and underlying mechanisms of dissemination still 531 remain matters of debate and speculation. The detailed characterization conducted here of the 532 genetic contexts in which blaVIM-2 was located in our P. putida G isolates thus helps our 533 understanding of the events contributing to the spread of carbapenem resistance to pathogenic 534 species in the nosocomial setting, and sheds light on the role of this bacterial group as an 535 active environmental reservoir of antimicrobial resistance platforms. 536 The taxonomic characterization of the P. putida G isolates conducted first allowed us 537 a better definition of the different species involved in this study. We were able to distinguish 538 at the 13 P. putida G clinical isolates the species level, identifying among them 539 representatives of P. asiatica, P. putida sensu stricto, P. monteillii, P. juntendi, and other 3 22 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 540 species more recently described as forming part of this group (1, 4, 5). In particular, our 541 analyses (Table 1 and Fig. 1) could confidentially assign 4 of them including BA7816, 542 LD209, HB313 and HP613 to the recently described species P. asiatica (4). Also, our 543 comparisons indicated that the proposed species P. putida G/IV (2) can actually be ascribed to 544 P. asiatica (Table 1 and Fig. 1). Our results also emphasize the ability of different members of 545 the P. putida G to adapt and survive in the nosocomial habitat. 546 The searching of genetic elements containing blaVIM-2 conducted next among these P. 547 putida G isolates revealed that three Tn402-like class 1 integrons (Fig. 3) carry this 548 carbapenem resistance gene. In general, class 1 integrons are not mobile elements by 549 themselves, but two described here, In41 and In899 (Fig. 2A) were found embedded in 550 complete Tn402-like transposons (Tn6335 and Tn6336, respectively) carried by pLD209- 551 related plasmids (Fig. 3), and thus potentially capable of both intra- and inter--cellular 552 mobilization (Fig. 4). In this context, the finding of identical 5-bp DRs bounding these 553 transposons (indicated by black circles, Fig.

3, c-f) provides evidence that the original Tn402- 554 like transposon from which these elements derive reached their plasmid location by a 555 transposition event (32, 33, 34, 35, 36, 37, 49). 556 Besides their acquisition as passengers of pLD209-related conjugative plasmids, the 557 selection of transposition events from the incoming plasmid to pre-existing preferred 558 locations in the host genome, such as the res sites of Tn21 subgroup transposons, provides 559 another mechanism of blaVIM-2-containing Tn402-like integron dissemination among P. putida 560 G members (Fig. 4, a and c). The selection of clones that had lost resistance cassettes such as 561 aacA4, seemingly represents another genetic event occurring among P. putida G strains (Fig. 562 4b), as exemplified by the case of P. asiatica BA7816 harboring only blaVIM-2 into the 563 integron variable region carried by a pLD209-related plasmid (Fig. 3). Similar to Tn6335 564 above, clones may have been selected in which transposition of Tn6336 to other genomic sites 23 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 565 preserved the carbapenem resistance phenotype in the eventuality of pBA7816 loss (Fig. 4c), 566 as exemplified in the case of P. putida BA9115 (Table 1). Another possibility is represented 567 by the selection of plasmids that, with the exception of the replication and adaptive modules, 568 have lost substantial portions of the original structures (Fig. 4f). Such pLD209-derived 569 plasmids have in fact been recovered from clinical strains of S. marcescens and P. aeruginosa 570 in local hospitals (Fig. 3, 48), suggesting that members of the P. putida G group could have 571 been also their sources or reservoirs (Fig. 4, f and g). All these observations reinforce the 572 existence of both assisted intra- and inter-cellular mobilization events of blaVIM-2-containing 573 Tn402-like integrons among P. putida G species, increasing the threat of carbapenem 574 resistance dissemination to other pathogenic species co-existing in the clinical setting. In the 575 above context, we recently isolated a local P. aeruginosa clinical strain, PAE868 (Table S1), 576 harboring a plasmid (pPAE868) carrying a Tn6335 element. This plasmid has the ability to 577 replicate in different Pseudomonas species including P. aeruginosa PAO1 and in the 578 carbapenem-susceptible P. juntendi HPC451 strain characterized by us (Fig. 1, Table S3, and 579 data not shown), but not in E. coli. These observations suggest that Tn6335 could have been 580 acquired by a pPAE868 predecessor during their co-existence in a same cell (Fig. 4d), prior to 581 the horizontal transfer of the modified plasmid to P. aeruginosa (Fig.

4e). 582 The above observations reinforced our previous notion that self-mobilizable pLD209- 583 type plasmids are capable of spreading blaVIM-2-containing Tn402-like integrons not only 584 among a wide range of P. putida G species, but also to enterobacterial species (13, 48). In 585 addition, the persistence and/or expansion of particular P. putida G clonal lineages such as 586 PaA of P. asiatica (Table 1) certainly increases the possibilities of blaVIM-2 dissemination. In 587 the latter context, we observed not only the presence of clonally-related P. asiatica isolates 588 such as LD209 and HP613 in the same hospital at different dates (Table S1), but also an 24 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 589 almost simultaneous presence of clonally-related isolates of this species in different hospitals 590 such as HP613 and HB313 (Fig. S1 and Table S1). 591 In conclusion, our findings indicate that members of the P. putida G conform 592 nowadays active parts of the nosocomial microbiota, representing an important reservoir of 593 genetic determinants such as blaVIM-2 genes responsible of carbapenem resistance. Moreover, 594 the findings here that particular Tn402-like class 1 integrons have been disseminating among 595 P. putida G species in our clinical setting with the assistance of pLD209-type conjugative 596 plasmids over a period of 9 years support the postulated ability of these mobile genetic 597 elements to largely persist in the nosocomial habitat (60). Finally, the results presented here 598 provide evidences supporting the intra- and intergenomic mobilization of Tn402-like 599 integrons and derived composite transposons encompassing also Tn3 family members among 600 the components of this bacterial group, providing clues that may explain the vast 601 dissemination of resistance genes among Pseudomonas and other pathogens. 602 603 MATERIALS AND METHODS 604 Bacterial isolates and antimicrobial susceptibility testing 605 A total of 13 carbapenem-resistant clinical isolates initially identified as Pseudomonas putida 606 by the Vitek 2C System (bioMérieux, Marcy l'Etoile, France) were included in this study 607 (Table 1). These isolates were collected from inpatients of different hospitals of Buenos Aires 608 (B1-B3) or Rosario (R1-R5), Argentina, during the period 2006-2014 (Table S1). The isolates 609 designated as BA were obtained from Instituto Malbrán, Buenos Aires. The susceptibilities to 610 different antimicrobials including imipenem, meropenem, piperacillin-tazobactam, 611 ceftazidime, cefepime, amikacin, gentamicin and ciprofloxacin of the different strains (Table 612 S1) were evaluated usingns the Vitek 2C System (bioMérieux, Marcy l’Etoile, France).

The 25 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 613 interpretation of the MIC values shown in Table S1 was based on CLSI breakpoints 614 recommendations (61). 615 616 Assignments of Pseudomonas putida G clinical isolates to the species level and 617 phylogenetic relatedness between strains and isolates 618 The assignment of each of the clinical isolates originally characterized as belonging to 619 the P. putida group by phenotypic procedures (see above) to the species level was based on 620 multilocus sequencing analysis (MLSA) and sequence comparisons following described 621 procedures (1, 2). This approach is based on the percentage of nucleotide sequence identity 622 between alignments of the concatenated sequences here employed (2,647 bp in total), 623 corresponding to partial regions of the 16S rDNA (1,301 pb), gyrB (669 pb) and rpoD (677 624 bp) genes, and defined here as 97.5 % identity as the threshold value that separates species 625 within the P. putida group. For this purpose, genomic DNA of each clinical isolate was 626 purified using Wizard Genomic DNA Purification Kit (Promega, Madison, WI), and used as 627 templates for PCR reactions aimed to amplify the desired fragments of the 16S rDNA, gyrB 628 and rpoD genes used for concatenate construction (1, 2, 62; Table S2). The obtained 629 amplicons were then sequenced at the Sequencing Facility of the University of Maine (Orono, 630 ME, USA). The corresponding partial sequences of the mentioned genes from type strains 631 including 21 species of P. putida G, P. aeruginosa ATCC 10145, and P. oryzihabitans ATCC 632 43272, as well as those corresponding to representative members of 6 newly P. putida G 633 proposed species (2), were retrieved from the sequence data deposited on the NCBI database 634 (Table S3). Alignments of concatenated genes were done using ClustalW with default 635 parameters (https://www.genome.jp/tools-bin/clustalw). These alignments were also 636 employed for the construction of a Maximum-Likelihood (ML) phylogenetic tree using 637 MEGA7.0 (63). To determine the best-fit nucleotide substitution model, the tool included in 26 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 638 MEGA7.0 was employed resulting in the use of the GTR+G+I substitution model, taking into 639 account the Akaike information criterion (AIC). Only branches supported by bootstrap values 640 higher than 60% (1,000 replicates) are shown in the depicted tree.

641 642 Genomic relatedness among P. putida group isolates 643 The genomic relatedness among isolates assigned to the same species was evaluated by a 644 random amplification PCR assay employing degenerate oligonucleotides (DO-PCR) (39). 645 646 Detection of MβL by phenotypic and molecular methods 647 MβL-production was assayed by the EDTA-imipenem microbiological assay (EIM) 648 and EDTA disk synergy test (EDS) (38). The presence of blaVIM-like, blaIMP-like, blaSPM-1 or 649 blaNDM-like genes was evaluated by PCR using specific primers (Table S2). 650 651 Genetic environments of the blaVIM-2 genes in the P. putida G clinical isolates analyzed in 652 this work. 653 The association of blaVIM-2 genes with “unusual” class 1 integrons in the P. putida G 654 clinical isolates studied here was detected by PCR using the primer pair 5’-CS (forward) and 655 TniC-R2 (reverse) primers (Table S2) followed by sequencing analysis (17). The subsequent 656 characterization of the structures of the Tn402 structural elements in which the detected 657 integrons were embedded (Fig. 2) was done by PCR-overlapping assays using genomic DNA 658 from each isolate in each case and the appropriate pairs of primers (Table S2), followed by 659 sequencing and database searching analyses as described in detail previously (17). Also, the 660 left boundary sequences of the detected Tn402 transposons from the IRi to the blaVIM-2 gene 661 were determined by PCR using the primer combination IRHP/VIM-R; and the right boundary 662 sequences from the tniB gene to the IRt by using the primer combination TniB-F/IRHP (Table 27 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 663 S2). In the case of the Tn402-like element found in P. monteilii HB157, which lacks most of 664 the tni module except for a complete tniC gene and 44 bp upstream of this gene (Fig. 2B), we 665 completed the sequence of the whole element by an inverse PCR procedure (64). In short, 1 666 µg of genomic DNA from the corresponding isolates was digested by EcoRI (Promega, 667 Madison, WI), the amplified DNA fragments were subjected to ethanol precipitation (65), 668 resuspended in sterilized distilled water, and then ligated for 16 h at 4 ºC with T4 DNA ligase 669 (Promega). After a further purification step, PCR assays were performed in 25µl-reactions 670 containing as template 0.1 µg of this circularized DNA, 0.5 µM of the primers VIM-Rf and 671 IRHPr (Table S2), 200 μM of each dNTP, 2.0 mM MgSO4 and 1.0 U Platinum Taq DNA 672 polymerase High Fidelity (Invitrogen, Carlsbad, CA). The cycling protocol involved 5 min 673 denaturation at 94 °C, followed by 30 cycles of 30 s at 94 °C, 45 s at 53 °C, and 4 min at 68 674 °C, ending with a 10 min incubation at 68 °C.

Several attempts using this procedure resulted 675 in a discrete number of amplification bands ranging from around 2.5 to 4.5 kbp, probably as 676 the result of secondary hybridization sites recognized by the primers employed. The 677 amplification mixtures were thus subjected to ethanol precipitation and resuspension in 678 sterilized distilled water as above, ligated to pGEM-T Easy (Promega), and transformed into 679 E. coli DH5α cells by electroporation. After incubating the cells for 48 h at 37°C on LB agar 680 plates supplemented with 100 µg/ml ampicillin, 40 g/ml 5-bromo-4-chloro-3-indolyl-β-D- 681 galactopyranoside (X-gal) and 54 g/ml isopropyl β-D-1-thiogalactopyranoside (IPTG), 682 plasmids were extracted from different colonies using the Wizard Plus SV Minipreps DNA 683 Purification System, and analyzed by restriction mapping for the presence and size of inserts. 684 The DNA sequences of selected inserts in the vicinity of the pGEM-T Easy cloning site were 685 then determined to identify cloned fragments containing the desired sequences. We succeeded 686 by this procedure in cloning an approximately 2.7 kbp DNA fragment, which initiated at the 687 VIM-Rf primer (Table S2) and continued towards the tniC gene of the Tn402 element (Fig. 28 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 688 2B). The sequence was completed by primer walking with the sequential use of tniC-F and 689 IS6100-F primers (Table S2). This procedure allowed us to obtain the DNA sequence of a 690 2,758 bp fragment which not only covered the whole Tn402 element (Fig. 2B), but also 691 extended for an extra 732 bp into the P. monteilii HB157 genome in the immediate vicinity of 692 its IRt border (Fig. S2E). 693 694 Conjugation and transformation assays 695 Conjugation experiments were performed using the carbapenem-resistant P. putida G 696 clinical isolates analyzed here as donors, and rifampicin-resistant cells of E. coli DH5α or P. 697 aeruginosa PAO1 cells as recipients (17). Transconjugants carrying blaVIM-2-containing 698 plasmids were selected on LB agar containing 20 µg/ml ampicillin and 150 µg/ml rifampicin 699 in the former case, or 4 µg/ml ceftazidime and 150 µg/ml rifampicin in the latter. MIC values 700 towards different antimicrobials were determined on the obtained E. coli DH5α 701 transconjugants as described above. The subsequent self-transferability of the plasmids 702 present in the E. coli DH5α transconjugants was further tested by agar mating studies 703 employing as recipient the E. coli MC4100 strain harboring the chloramphenicol-resistance 704 plasmid pACYC184 (66). Transconjugants were selected in these cases using LB agar plates 705 containing 20 µg/ml ampicillin and 25 µg/ml chloramphenicol, and the loss of rifampicin 706 resistance was confirmed in all cases.

707 In the cases of P. putida G isolates in which no E. coli transconjugants could be 708 obtained by using the above procedures, plasmid DNA was isolated using the Wizard Plus SV 709 Minipreps DNA Purification System and used to transform E. coli DH5α which had been 710 made competent by chemical (CaCl2) procedures (65). Colonies were then selected on LB 711 agar plates containing 20 µg/ml ampicillin, after an overnight incubation at 37 °C. P. 712 aeruginosa PAO1 transformation was conducted by electroporation following described 29 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 713 procedures, followed by selection of resistant colonies in LB agar containing 4 µg/ml 714 ceftazidime (67). 715 The actual presence of the blaVIM-2 gene in the putative transconjugant or transformant 716 cells obtained as described above was confirmed by PCR and sequencing analysis (Table S2). 717 Plasmids from E. coli DH5α transconjugants were purified using the Wizard Plus SV 718 Minipreps DNA Purification System, and further characterized by EcoRI digestion followed 719 by agarose gel (0.7%) electrophoresis analysis of the obtained fragments (68). Selected 720 plasmids were subjected to further sequencing analysis (see below). 721 722 Plasmid sequencing and comparative sequence analyses 723 pLA111 and pBA7816 nucleotide sequences were determined on a 454 724 pyrosequencing platform (Roche Diagnostics Corporation) at the Instituto de 725 Agrobiotecnología Rosario (INDEAR). The obtained reads were assembled in silico using as 726 framework the structure previously determined for pLD209 (13). The circular structures of 727 these plasmids were confirmed by PCR procedures, in which remaining gaps between the 728 resulting contigs were closed using specifically designed primer pairs (Table S2). In the case 729 of pLA111 we employed a virB10-F/virB8-R primer combination, and for pBA7816 we used 730 mob-F/mob-R, 7816-F/VIM-R and TniA-F2/REPIR-T3 combinations (Table S2). 731 The DNA sequences of all the PCR amplicons and cloned inserts obtained in this work 732 were done at the University of Maine DNA Sequencing Facility, Orono, USA. 733 The Rapid Annotation using Subsystem Technology standard operating procedures 734 (RAST, http://rast.nmpdr.org/seedviewer.cgi) (69) and the National Center for Biotechnology 735 Information database (NCBI, U.S. National Library of Medicine, Bethesda MD, USA) were 736 used to annotate the open reading frames (ORFs). Searching for antimicrobial resistance 737 determinants was done using ResFinder 2.1 (https://cge.cbs.dtu.dk/services/ResFinder/; 70). 30 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020.

The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 738 The detection of IS was done with ISFinder (71) (https://www-is.biotoul.fr/) and ISsaga (72). 739 The sequences of pBA7816 and pLA111 were deposited at the GenBank nucleotide sequence 740 database under the accession numbers MN240297 and MT192131, respectively. 741 742 Target sites of Tn6335 in the chromosome of selected P. putida G isolates 743 Genomic DNA was purified from P. asiatica HP613 and P. putida G/II LA1008 using 744 the Wizard Genomic DNA Purification Kit, separately digested with EcoRI and ligated to 745 EcoRI-digested pSU18, a E. coli cloning vector conferring chloramphenicol resistance (52). 746 The ligation mixture was transformed into E. coli DH5α by electroporation, and transformants 747 carrying inserts containing complete blaVIM-2 genes were selected on LB agar plates 748 containing ceftazidime (4 µg/ml) and chloramphenicol (25 µg/ml) supplemented with 40 749 g/ml X-gal and 54 g/ml IPTG. After a 48 h incubation at 37 °C, plasmids were recovered 750 from selected colonies as described above, and the DNA sequences of the cloned EcoRI 751 fragments were determined employing first a primer hybridizing in the multiple cloning site 752 of pSU18 (pSU18-F, Table S2), followed by primer walking using the sequence information 753 obtained in each case. 754 755 756 757 758 31 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 759 ACKNOWLEDGMENTS 760 We are grateful to the personnel of the Bacteriology Service, Hospital Provincial, 761 Rosario, Argentina for kindly providing P. putida G clinical isolates used in this work. P.M. 762 and A.L. are Researchers of the National University of Rosario. A.M.V. and D.F. are Careers 763 Researchers of CONICET. M.B. is Fellow of CONICET. F.P. and A.C. are Researchers of the 764 Malbrán Institute, Buenos Aires. This work was supported by grants from the Agencia 765 Nacional de Promoción Científica y Tecnológica (ANPCyT; PICT 2012-0680 to A.L., and 766 PICT 2015-1072 to A.M.V. ); Consejo Nacional de Investigaciones Científicas y Técnicas 767 (CONICET); Secretaría de Ciencia, Tecnología e Innovación, Provincia de Santa Fe, and 768 Secretaría de Salud Pública, Municipalidad de Rosario. 769 770 REFERENCES 771 1. Mulet M, Lalucat J, García-Valdés E. 2010. DNA sequence-based analysis of the 772 Pseudomonas species. Environ Microbiol 12:1513-1530. 773 2. Mulet M, García-Valdés E, Lalucat J. 2013. Phylogenetic affiliation of Pseudomonas 774 putida biovar A and B strains.

Res Microbiol 164:351-359. 775 3. Peix A, Ramírez-Bahena MH, Velázquez E. 2018. The current status on the taxonomy of 776 Pseudomonas revisited: An update. Infect Genet Evol 57:106-116. 777 4. Tohya M, Watanabe S, Teramoto K, Uechi K, Tada T, Kuwahara-Arai K, et al. 2019. 778 Pseudomonas asiatica sp. nov., isolated from hospitalized patients in Japan and 779 Myanmar. Int J Syst Evol Microbiol 69:1361-1368. 780 5. Tohya M, Watanabe S, Teramoto K, Shimojima M, Tada T, Kuwahara-Arai1 K, et al. 781 2019. Pseudomonas juntendi sp. nov., isolated from patients in Japan and Myanmar. Int J 782 Syst Evol Microbiol 69:3377-3384. 32 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 783 6. Pang Z, Raudonis R, Glick BR, Lin TJ, Cheng Z. 2019. Antibiotic resistance in 784 Pseudomonas aeruginosa: mechanisms and alternative therapeutic strategies. Biotechnol 785 Adv 37(1):177-192. 786 7. Peña A, Busquets A, Gomila M, Mulet M, Gomila RM, Reddy TB, Huntemann M, Pati A, 787 Ivanova N, Markowitz V, García-Valdés E, Göker M, Woyke T, Klenk HP, Kyrpides N, 788 Lalucat J. 2016. High quality draft genome sequences of Pseudomonas fulva DSM 789 17717T, Pseudomonas parafulva DSM 17004T and Pseudomonas cremoricolorata DSM 790 17059T type strains. Standards in Genomic Sciences 11(1):55. 791 8. Loucif L, Cherak Z, Chamlal N, Bendjama E, Gacemi-Kirane D, Grainat N, Rolain JM. 792 2017. First Detection of VIM-2 Metallo-β-Lactamase-Producing Pseudomonas putida in 793 Blattella germanica Cockroaches in an Algerian Hospital. Antimicrob Agents Chemother 794 61(8):e00357-17. 795 9. Frasson D, Opoku M, Picozzi T, Torossi T, Balada S, Smits THM, Hilber U. 2017 796 Pseudomonas wadenswilerensis sp. nov. and Pseudomonas reidholzensis sp. nov., two 797 novel species within the Pseudomonas putida group isolated from forest soil. Int J Syst 798 Evol Microbiol 67(8):2853-2861. 799 10. Xiang W, Chen S, Tian D, Huang C, & Gao T. 2019. Pseudomonas hutmensis sp. nov., a 800 New Fluorescent Member of Pseudomonas putida Group. Current Microbiology 76(7): 801 872-878. 802 11. Seok Y, Shin H, Lee Y, Cho I, Na S, et al. 2010. First Report of Bloodstream Infection 803 caused by Pseudomonas fulva. J Clin Microbiol 48: 2656-2657. 804 12. Leneveu-Jenvrin C, Madi A, Bouffartigues E, Biaggini K, Feuilloley M, Chevalier S, 805 Connil N. 2013. Cytotoxicity and inflammatory potential of two Pseudomonas mosselii 806 strains isolated from clinical samples of hospitalized patients. BMC Microbiology 807 13:123. 33 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.

It is made available under a CC-BY-NC-ND 4.0 International license . 808 13. Marchiaro P, Brambilla L, Morán-Barrio J, Revale S, Pasteran F, Vila AJ, Viale AM, 809 Limansky A. 2014. The complete nucleotide sequence of the carbapenem resistance- 810 conferring conjugative plasmid pLD209 from a Pseudomonas putida clinical strain 811 revealed a chimeric design formed by modules derived from both environmental and 812 clinical bacteria. Antimicrob Agents Chemother 58:1816-1821. 813 14. Yamamoto M, Matsumura Y, Gomi R, Matsuda T, Nakano S, Nagao M, Tanaka M, 814 Ichiyama S. 2018. Molecular Analysis of blaIMP-1-Harboring Class 3 Integron in 815 Multidrug-Resistant Pseudomonas fulva. Antimicrob Agents Chemother 62(8):e00701- 816 18. 817 15. Yomoda S, Okubo T, Takahashi A, Murakami M, Iyobe S. 2003. Presence of 818 Pseudomonas putida Strains Harboring Plasmids Bearing the Metallo-β-Lactamase Gene 819 blaIMP in a Hospital in Japan. J Clin Microbiol 41(9):4246-4251. 820 16. Almuzara M, Radice M, Gárate N de, Kossman A, Cuirolo A, Santella G, Famiglietti A, 821 Gutkind G, Vay V. 2007. VIM-2–producing Pseudomonas putida, Buenos Aires. Emerg 822 Infect Dis 13:668-669. 823 17. Marchiaro P, Viale AM, Ballerini V, Rossignol G, Vila AJ, Limansky A. 2010. First 824 report of a Tn402-like class 1 integron carrying blaVIM-2 in Pseudomonas putida from 825 Argentina. J Infect Dev Ctries 4:412-416. 826 18. Juan C, Zamorano L, Mena A, Albertí S, Pérez JL, Oliver A. 2010. Metallo-β-lactamase- 827 producing Pseudomonas putida as a reservoir of multidrug resistance elements that can 828 be transferred to successful Pseudomonas aeruginosa clones. J Antimicrob Chemother 829 65:474-478. 830 19. Santos C, Caetano T, Ferreira S, Mendo S. 2010. Tn5090-like class 1 integron carrying 831 blaVIM-2 in a Pseudomonas putida strain from Portugal. Clin Microbiol Infect 16:1558- 832 1561. 34 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 833 20. Carvalho-Assef AP, Gomes MZ, Silva AR, Werneck L, Rodrigues CA, Souza MJ, Asensi 834 MD. 2010. IMP-16 in Pseudomonas putida and Pseudomonas stutzeri: potential 835 reservoirs of multidrug resistance. J Med Microbiol 59:1130-1131. 836 21. Scotta C, Juan C, Cabot G, Oliver A, Lalucat J, Bennasar A, and Alberti S. 2011. 837 Environmental microbiota represents a natural reservoir for dissemination of clinically 838 relevant metallo-β-lactamases. Antimicrob Agents Chemother 54:5376-5379. 839 22. Bogaerts P, Bouchahrouf W, Lissoir B, Denis O, Glupczynski Y. 2011. IMP-13- 840 producing Pseudomonas monteilii recovered in a hospital environment. J Antimicrob. 841 Chemother 66:2434-2440. 842 23. Ocampo-Sosa AA, Guzmán-Gómez LP, Fernández-Martínez M, Román E, Rodríguez C, 843 Marco F, Vila J, Martínez-Martínez L. 2015.

Isolation of VIM-2-producing Pseudomonas 844 monteilii clinical strains disseminated in a tertiary hospital in northern Spain. Antimicrob 845 Agents Chemother 59:1334-1336. 846 24. Zhang R, Liu Z, Li J, Lei L, Yin W, Li M, Wu C, Walsh TR, Wang Y, Wang S, Wu Y. 847 2017. Presence of VIM-positive Pseudomonas species in chickens and their surrounding 848 environment. Antimicrob Agents Chemother 61:e00167-17. 849 25. Peter S, Oberhettinger P, Schuele L, Dinkelacker A, Vogel W, Dörfel D, Bezdan D, et al. 850 2017. Genomic characterization of clinical and environmental Pseudomonas putida group 851 strains and determination of their role in the transfer of antimicrobial resistance genes to 852 Pseudomonas aeruginosa. BMC genomics 18(1):859. 853 26. Perez F, Hujer AM, Marshall SH, et al. 2014. Extensively Drug-Resistant Pseudomonas 854 aeruginosa Isolates Containing blaVIM-2 and Elements of Salmonella Genomic Island 2: a 855 New Genetic Resistance Determinant in Northeast Ohio. Antimicrob Agents Chemother 856 58(10):5929-5935. 35 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 857 27. Nordmann P, Poirel L. 2002. Emerging carbapenemases in Gram-negative aerobes. Clin 858 Microbiol Infect 8:321-31. 859 28. Hong DJ, Bae IK, Jang IH, Jeong SH, Kang HK, & Lee K. 2015. Epidemiology and 860 Characteristics of Metallo-β-Lactamase-Producing Pseudomonas aeruginosa. Infection & 861 chemotherapy 47(2):81-97. 862 29. Rådström P, Sköld O, Swedberg G, Flensburg J, Roy PH, Sundström L. 1994. Transposon 863 Tn5090 of plasmid R751, which carries an integron, is related to Tn7, Mu, and the 864 retroelements. J Bacteriol 176(11):3257-3268. 865 30. Kholodii GY, Mindlin SZ, Bass IA, Yurieva OV, Minakhina SV, Nikiforov VG. 1995. 866 Four genes, two ends, and a res region are involved in transposition of Tn5053: a 867 paradigm for a novel family of transposons carrying either a mer operon or an integron. 868 Mol Microbiol 17(6):1189-1200. 869 31. Gillings MR, Labbate M, Sajjad A, Giguère NJ, Holley MP, & Stokes HW. 2009. 870 Mobilization of a Tn402-like class 1 integron with a novel cassette array via flanking 871 miniature inverted-repeat transposable element-like structures. Appl Environ Microbiol 872 75(18):6002-6004. 873 32. Nicolas E, Lambin M, Dandoy D, Galloy C, Nguyen N, Oger CA, Hallet B. 2015. The 874 Tn3-family of Replicative Transposons. Microbiol Spectr 3:MDNA3-0060-2014. 875 33. Partridge SR, Kwong SM, Firth N, Jensen SO. 2018. Mobile genetic elements associated 876 with antimicrobial resistance. Clin Microbiol Rev 31:e00088-17. 877 34. Minakhina S, Kholodii G, Mindlin S, Yurieva O, Nikiforov V. 1999. Tn5053 family 878 transposons are res site hunters sensing plasmidal res sites occupied by cognate 879 resolvases.

Molecular Microbiol 33:1059-1068. 880 35. Kamali-Moghaddam M, Sundström L. 2000. Transposon targeting determined by 881 resolvase. FEMS Microbiol Lett 186:55-59. 36 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 882 36. Petrovski S, Stanisich VA. 2010. Tn502 and Tn512 are res site hunters that provide 883 evidence of resolvase-independent transposition to random sites. J Bacteriology 884 192:1865-1874. 885 37. Gillings MR. 2014. Integrons: past, present, and future. Microbiol Mol Biol Rev 78:257- 886 277. 887 38. Marchiaro P, Mussi MA, Ballerini V, Pasteran F, Viale AM, Vila AJ, Limansky AS. 888 2005. Sensitive EDTA-Based Microbiological Assays for the Detection of Metallo-β- 889 lactamases in Non-Fermentative Gram-Negative Bacteria. J Clin Microbiol 43:5648- 890 5652. 891 39. Limansky A, and Viale A. 2002. Can composition and structural features of 892 oligonucleotides contribute to their wide-scale applicability as random PCR primers in 893 mapping bacterial genome diversity? J Microbiol Methods 50:291-297. 894 40. Mustapha MM, Marsh JW, Ezeonwuka CD, Pasculle AW, Pacey MP, Querry AM, Muto 895 CA, Harrison LH. 2016. Draft Genome Sequences of Four Hospital-Associated 896 Pseudomonas putida Isolates. Genome Announcements 4(5):e01039-16. 897 41. Tansirichaiya S, Rahman A, Roberts AP. 2019. The Transposon Registry. Mobile DNA 898 10.40. 899 42. Faccone D, Pasteran F, Albornoz E, Gonzalez L, Veliz O, Prieto M, Bucciarelli R, Callejo 900 R, Corso A. 2014. Human infections due to Pseudomonas chlororaphis and 901 Pseudomonas oleovorans harboring new blaVIM-2-borne integrons. Infect Genet Evol 902 28:276-277. 903 43. Suenaga H, Yamazoe A, Hosoyama A, Kimura N, Hirose J, Watanabe T, Fujihara H, 904 Futagami T, Goto M, Furukawa K. 2017. Complete genome sequence of the 905 polychlorinated biphenyl-degrading bacterium Pseudomonas putida KF715 (NBRC 906 110667) isolated from biphenyl-contaminated soil. Genome Announc 5:e01624-16. 37 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 907 44. Suenaga H, Fujihara H, Kimura N, Hirose J, Watanabe T, Futagami T, et al. 2017. 908 Insights into the genomic plasticity of Pseudomonas putida KF715, a strain with unique 909 biphenyl-utilizing activity and genome instability properties. Environ. Microbiol. Rep. 9: 910 589-598. 911 45. Delihas N. 2011. Impact of small repeat sequences on bacterial genome evolution. 912 Genome Biol Evol 3:959-973.

913 46. Bardaji L, Pérez-Martínez I, Rodríguez-Moreno L, Rodríguez-Palenzuela P, Sundin GW, 914 Ramos C, et al. 2011. Sequence and Role in Virulence of the Three Plasmid Complement 915 of the Model Tumor-Inducing Bacterium Pseudomonas savastanoi pv. savastanoi 916 NCPPB 3335. PLoS ONE 6(10):e25705. 917 47. Domingues S, Toleman MA, Nielsen KM, & da Silva GJ. 2013. Identical miniature 918 inverted repeat transposable elements flank class 1 integrons in clinical isolates of 919 Acinetobacter spp. J Clin Microbiol 51(7):2382-2384. 920 48. Vilacoba E, Quiroga C, Pistorio M, Famiglietti A, Rodríguez H, Kovensky J, Centrón, D. 921 2014. A blaVIM-2 plasmid disseminating in extensively drug-resistant clinical 922 Pseudomonas aeruginosa and Serratia marcescens isolates. Antimicrob Agents 923 Chemother 58:7017-7018. 924 49. Betteridge T, Partridge SR, Iredell JR, & Stokes HW. 2011. Genetic Context and 925 Structural Diversity of Class 1 Integrons from Human Commensal Bacteria in a Hospital 926 Intensive Care Unit. Antimicrob Agents Chemother 55(8):3939-3943. 927 50. Shapiro JA, Sporn P. 1977. Tn402: a new transposable element determining trimethoprim 928 resistance that inserts in bacteriophage lambda. J Bacteriol 129:1632-1635. 929 51. Thorsted PB, Macartney DP, Akhtar P, Haines AS, Ali N, Davidson P, Stafford T, 930 Pocklington MJ, Pansegrau W, Wilkins BM, Lanka E, Thomas CM. 1998. Complete 38 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 931 sequence of the IncP beta plasmid R751: implications for evolution and organisation of 932 the IncP backbone. J Mol Biol 282(5):969-90. 933 52. Bartolomé B, Jubete Y, Martinez E, and de la Cruz F. 1991. Construction and properties 934 of a family of pACYC184-derived cloning vectors compatible with pBR322 and its 935 derivatives. Gene 102:75-78. 936 53. Rogowsky P, Halford SE, Schmitt R. 1985. Definition of three resolvase binding sites at 937 the res loci of Tn21 and Tn1721. EMBO J 4(8):2135-2141. 938 54. Stokes HW, Elbourne LD, & Hall RM. 2007. Tn1403, a multiple-antibiotic resistance 939 transposon made up of three distinct transposons. Antimicrob Agents Chemother 940 51(5):1827-1829. 941 55. L'Abée-Lund TM, & Sørum H. 2000. Functional Tn5393-like transposon in the R plasmid 942 pRAS2 from the fish pathogen Aeromonas salmonicida subspecies salmonicida isolated 943 in Norway. Appl Environ Microbiol 66(12):5533-5535. 944 56. Chiou CS, & Jones AL. 1993. Nucleotide sequence analysis of a transposon (Tn5393) 945 carrying streptomycin resistance genes in Erwinia amylovora and other gram-negative 946 bacteria. J Bacteriol 175(3):732-740. 947 57. Blake DG, Boocock MR, Sherratt DJ, & Stark WM. 1995. Cooperative binding of Tn3 948 resolvase monomers to a functionally asymmetric binding site.

Current biology 949 5(9):1036-1046. 950 58. Gillings MR, Paulsen IT, Tetu SG. 2015. Ecology and Evolution of the Human 951 Microbiota: Fire, Farming and Antibiotics. Genes (Basel) 6:841-857. 952 59. Tato M, Coque TM, Baquero F, Cantón R. 2010. Dispersal of Carbapenemase blaVIM-1 953 Gene Associated with Different Tn402 Variants, Mercury Transposons, and Conjugative 954 Plasmids in Enterobacteriaceae and Pseudomonas aeruginosa. Antimicrob Agents 955 Chemother 54(1):320-327. 39 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 956 60. Baquero F, Coque TM, and de la Cruz F. 2011. Ecology and evolution as targets: the need 957 for novel Eco-Evo drugs and strategies to fight antibiotic resistance. Antimicrob Agents 958 Chemother 55:3649-3660. 959 61. Clinical and Laboratory Standards Institute (CLSI). 2019. Performance standards for 960 antimicrobial susceptibility testing. Twenty-nine informational supplement. Document 961 M100-S24. CLSI, Wayne, PA. 962 62. Héritier C, Poirel L, and Nordmann P. 2004. Genetic and biochemical characterization of 963 a chromosome-encoded carbapenem-hydrolyzing Ambler Class D β-Lactamase from 964 Shewanella algae. Antimicrob Agents Chemother 48:1670-1675. 965 63. Kumar S, Stecher G, Tamura K. 2016. MEGA7: molecular evolutionary genetics analysis 966 version 7.0 for bigger datasets. Mol Biol Evol 33:1870-1874. 967 64. Ochman H, Gerber AS, & Hartl DL. 1988. Genetic applications of an inverse polymerase 968 chain reaction. Genetics 120(3):621-623. 969 65. Sambrok J, Fritsch EF, Maniatis T. 1989. Molecular cloning. A laboratory manual. Cold 970 Spring Harbor, NY: Cold Spring Harbor Laboratory Press. 971 66. Rose RE. 1988. The nucleotide sequence of pACYC184. Nucleic Acids Res 16(1):355. 972 67. Choi KH, Kumar A, Schweizer HP. 2006. A 10-min method for preparation of highly 973 electrocompetent Pseudomonas aeruginosa cells: application for DNA fragment transfer 974 between chromosomes and plasmid transformation. J Microbiol Methods 64:391-397. 975 68. Quinones-Falconi F, Galicia-Velasco M, Marchiaro P, Mussi MA, Ballerini V, Vila AJ, 976 Viale AM, Bermejo-Morales K and Limansky AS. 2010. Emergence of Pseudomonas 977 aeruginosa strains producing metallo-β-lactamases of the IMP-15 and VIM-2 types in 978 Mexico. Clin Microbiol Infect 16:126-31. 979 69. Aziz RK, Bartels D, Best AA, DeJongh M, Disz T, Edwards RA, et al. 2008. The RAST 980 server: rapid annotations using subsystems technology. BMC Genomics 9:75. 40 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.

It is made available under a CC-BY-NC-ND 4.0 International license . 981 70. Zankari E, Hasman H, Cosentino S, Vestergaard M, Rasmussen S, Lund O, Aarestrup 982 FM, & Larsen MV. 2012. Identification of acquired antimicrobial resistance genes. J 983 Antimicrob Chemother 67(11):2640-2644. 984 71. Siguier P, Perochon J, Lestrade L, Mahillon J, and Chandler M. 2006. ISfinder: the 985 reference centre for bacterial insertion sequences. Nucleic Acids Res 34:D32-36. 986 72. Varani AM, Siguier P, Gourbeyre E, Charneau V, and Chandler M. 2011. ISsaga is an 987 ensemble of web-based methods for high throughput identification and semi-automatic 988 annotation of insertion sequences in prokaryotic genomes. Genome Biol 12:R30. 989 73. Molina L, Bernal P, Udaondo Z, Segura A, Ramos JL. 2013. Complete Genome Sequence 990 of a Pseudomonas putida Clinical Isolate, Strain H8234. Genome Announcements 991 1(4):e00496-13. 992 74. Manchanda V, Rai S, Gupta S, Rautela RS, Chopra R, Rawat DS, Verma N, Singh NP, 993 Kaur IR, Bhalla P. 2011. Development of TaqMan real-time polymerase chain reaction 994 for the detection of the newly emerging form of carbapenem resistance gene in clinical 995 isolates of Escherichia coli, Klebsiella pneumoniae, and Acinetobacter baumannii. Indian 996 J Med Microbiol 29(3):249-53. 997 998 FIGURE LEGENDS 999 FIG 1. Phylogenetic analysis of the Pseudomonas putida G isolates analyzed in this work. 1000 A ML phylogenetic tree was constructed from alignments of the concatenated sequences of 1001 defined partial regions of 16S rDNA, and gyrB and rpoD genes corresponding to the different 1002 Pseudomonas spp. strains indicated in the figure. The P. putida G clinical isolates analyzed in 1003 this work are in bold. The analysis also incorporates the corresponding concatenated 1004 sequences of 21 P. putida G type strains which have received species assignation (2, 4, 5) as 1005 well as from 6 strains representing proposed novel species (P. putida G/I to P. putida G/VI, 2, 41 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 1006 indicated by b superscripts). In addition, the concatenate sequences of P. aeruginosa ATCC 1007 10145T and of P. orzihabitans ATCC 43272T type strains were included to serve as outgroup 1008 sequences of the P. putida group. For details on the corresponding nucleotide accession 1009 numbers of the sequences employed here see Table S3. Only bootstrap percentages higher 1010 than 60 % (1,000 replicates) are indicated at the nodes. In LD209a, the superscript indicates 1011 the representative isolate of the clonal lineage PaA among the P. asiatica isolates, which also 1012 includes HP613 and HB313 (not shown in the figure). These three isolates share identical 1013 sequences of core gene concatenate sequences.

The superscript T indicate the type strains. See 1014 Materials and Methods for details. 1015 1016 FIG 2. Genetic organization of blaVIM-2-containing Tn402-like class 1 integrons in the P. 1017 putida G clinical isolates analyzed. A. Schematic structure of the class 1 integrons In41 and 1018 In899 embedded into complete Tn402-like transposons designated Tn6335 (7,633 bp, 1019 GenBank accession number GQ857074) and Tn6336 (6,994 bp, GenBank accession number 1020 MN240297.1), respectively. B. Same, for the In528 class 1 integron embedded into an 1021 incomplete Tn402-like integron detected in P. monteilii HB157 and lacking most of the tni 1022 module (Tn402(tniABQ), 5,239 bp; GenBank accession number MT192132). The initial 1023 inverted repeats (IRi) and terminal inverted repeats (IRt) associated to the left and right 1024 boundaries, respectively, of each transposable element are indicated by oppositely-oriented 1025 closed arrows at the corresponding Tn borders. The individual genes are represented by boxed 1026 arrows (distinctively labeled in each case) that also indicate the corresponding directions of 1027 transcription. intI1, integrase (black and gray mosaics); blaVIM-2, VIM-2 MβL (black); aacA4, 1028 aminoglycoside acetyl transferase (gray); tniC, resolvase (light gray with black vertical 1029 stripes); tniQ/tniB, auxiliary transposition genes (light gray with black diagonal stripes); tniA, 1030 transposase (light gray with black horizontal stripes); dhfrB1, dihydrofolate reductase (light 42 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 1031 gray); IS6100 transposase (dark gray). In the cassette genes located within the integrons, the 1032 recombination sites are indicated by two halves of a circle located upstream of and 1033 downstream of the associated gene. The two open halves are from the original attI site, and 1034 halves exhibiting the same shade of gray are from the corresponding attC sequences originally 1035 associated to a distinctive resistance cassette. Regions on the transposons described in this 1036 work sharing significant sequence identity with described mobile elements (see main text for 1037 details) are indicated by lines above the corresponding structures. In B, the initial inverted 1038 repeat (IRi) and terminal inverted repeat (IRt) are indicated by black triangles facing inwards. 1039 A second 25-bp sequence identical to IRi, designated IRi´, was also located immediately 1040 upstream of the tniC gene (black triangle facing IS6100). The left inverted repeat (IRl) and 1041 right inverted repeat (IRr) of IS6100 are indicated by the oppositely-oriented open arrows 1042 located at the borders of the transposase gene (in dark gray).

The structures of Tn6335 and 1043 Tn6336 were determined by PCR using different pairs of primers (Table S2), followed by 1044 sequencing analysis of the amplicons and assembly of the overlapping segments, and further 1045 confirmed by complete plasmid sequencing (Fig. 3). The complete sequence of 1046 Tn402(tniABQ) was determined using an inverse PCR procedure. The figure is not drawn to 1047 scale. For details see Materials and Methods. 1048 1049 FIG 3. Comparative analysis of pLD209-type plasmids present in P. putida G strains. 1050 Linear representations of the structures of six circular plasmids including a) pKF715D 1051 (GenBank accession number AP015033.1), b) pMRVIM0812 (CP010893.1), c) pBA7816 1052 (MN240297, this work), d) pLA111 (MT192131, this work), e) pLD209 (KF840720.1), f) 1053 pDCPR1 (KJ577613). The direction of transcription of the genes are indicated by arrows, and 1054 the different colors delineate the replication, stability and transfer modules (see the lower part 1055 of the figure). For the different structural components of Tn6335 and Tn6336, including the 43 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 1056 adaptive gene cassettes present in the integrons they carry, see the legend to Fig. 2. The 1057 regions shaded in gray tones linking the different structures reflect percentages of nucleotide 1058 sequence identity ranging from 85% to 99% as detected in a BLASTn search, with the scale 1059 depicted at the lower right part of the figure. The positions of EcoRI restriction sites inferred 1060 from the corresponding DNA sequences are indicated below plasmids pBA7816, pLA111, 1061 and pLD209, with the fragment sizes predicted in silico shown below pLD209 only. In the 1062 case of pBA7816, the size of a differential EcoRI fragment when compared to the equivalent 1063 region in pLD209 (i. e., 5,065 versus 5,704 bp, respectively) is shown. A 5-bp 5’-GTTTT-3’ 1064 direct duplication (black circles) is present at the immediate outer borders of the 25-bp 1065 inverted repeats IRi and IRt (black triangles facing inwards accompanying the black circles) 1066 of both Tn6335 and Tn6336. These Tn402-like elements are flanked by an external 34-bp 1067 inverted repeat (5’-GGGGGTGTAAGCCGGAACCCCAGAAAATTCCGTC-3’, gray 1068 triangles facing inwards), with one extreme (IRie) located immediately upstream of IRi and 1069 its complementary (IRte) located immediately upstream of the repA gene. These external IRe 1070 sequences are bordered by a 5’-TATTC-3’ direct repeat (gray circles accompanying the gray 1071 triangles). The single MITE element in pKF715D is located between nucleotide positions 1072 20,221 to 20,482, and is flanked by a 5-bp direct repeat, 5’-AACTT-3’ (violet circles).

The 1073 two MITE elements in pMRVIM0812 are located between positions 27,596-27,858 and 1074 32,920-33,182, respectively, and the resulting composite transposon-like structure is flanked 1075 by the 5-bp direct repeat 5’-GATGA-3’ (light gray circles). The PAS-domain protein coding 1076 gene and adjacent MITE element in pKF715D are limited by a 50-bp inverted repeat, and 1077 flanked in turn by a 5-bp direct repeat, 5’-AGGAA-3’ (orange circles). In pMRVIM0812, the 1078 PAS-domain protein gene is also limited by a 50-bp inverted repeat flanked in turn by a 5-bp 1079 direct repeat, 5’-TGGAT-3’ (yellow circles). 1080 44 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 1081 FIG 4. Routes of intra- and inter-genomic dissemination of blaVIM-2-containing genetic 1082 platforms among P. putida G members, and between this and other co-existing bacterial 1083 groups in the clinical setting. Tn6335, the prevalent Tn402-like transposon found in the P. 1084 putida G isolates studied here, was most probably acquired as a passenger of a conjugative 1085 pLD209-related plasmid. The plasmid could either persist in the new host, or fail to 1086 accompany the host replication rate and consequently be lost. In the latter case, carbapenem 1087 pressure could select bacterial clones in which the blaVIM-2-containing Tn402-like element had 1088 transposed from the plasmid to pre-existing preferred locations located in the chromosome or 1089 in other plasmids such as the res sites of Tn21 subgroup transposons, exemplified in this work 1090 by the cases of P. asiatica HP613 and P. putida G/II LA1008 (a, c). Lack of aminoglycoside 1091 pressure could lead to the selection of clones in which the aacA4 gene cassette was lost from 1092 the Tn402-like integron, resulting in a pLD209-related plasmid now harboring a Tn6336 1093 element (b), exemplified here in the case of P. asiatica BA7816 (Table 1). As above, clones 1094 in which this element had transposed to other preferred sites on the genome could be selected 1095 by carbapenem pressure, a situation exemplified here for P. putida BA9115 (Table 1) (c). The 1096 transposition of blaVIM-2-containing Tn402-like elements to a co-habitant plasmid displaying 1097 an idiosyncratic host range (d) also carries the possibility of a further dissemination of these 1098 elements by horizontal transfer to pathogenic bacteria, exemplified here by the finding of such 1099 a plasmid in a local P. aeruginosa clinical isolate, PAE868 (e) (see Discussion for details). 1100 Finally, deletions on pLD209 may have resulted in the selection of related plasmids lacking 1101 self-transferability, exemplified by pDCPR1 (Fig. 3) (f). Since pDCPR1 still preserved the 1102 oriT region, the Tn402-like element could still be transferred by conjugation if appropriate 1103 mobilization functions are provided in trans (g), exemplified by the isolation of this plasmid 1104 from both P. aeruginosa and S. marcescens clinical strains (48).

1105 45 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 1106 FIG S1. Identification of different clonal lineages within particular species of the P. 1107 putida group assigned in this work by random PCR assays. Random PCR assays with 1108 degenerate oligonucleotide primers were conducted as described in Materials and Methods. 1109 Lanes 1 and 2: PpA and PpB clonal lineages identified in the two indicated P. putida sensu 1110 stricto isolates; lanes 3 and 4: PmA and PmB clonal lineages identified in the indicated P. 1111 monteilii isolates; lanes 5 and 6: PpG/IIA and PpG/IIB clonal lineages identified in the 1112 indicated P. putida G/II isolates; lanes 7-10: two different clonal lineages, PaB (lane 7) and 1113 PaA (lanes 8-10) identified in the indicated P. asiatica isolates. 1114 1115 FIG S2. Schematic representations of Tn402-like transposon insertions detected in this 1116 work into the res sites of Tn3 family members in the genomes of P. asiatica HP613, P. 1117 putida G/II LA1008, and P. monteilii HB157. A. Structural features of the 9,031 bp-EcoRI 1118 fragment containing a partial fragment of Tn6335 carrying blaVIM-2 cloned from P. asiatica 1119 HP613. The fragment encompasses the 5,307 bp region of Tn6335 spanning from the IRi to 1120 the EcoRI site present within the tniB gene (see Fig. 3), plus a 3,725 bp-fragment outside the 1121 IRi boundary to a nearby EcoRI site located in the HP613 genome. The latter fragment shows 1122 97.8% nucleotide identity to the indicated tnpA-tnpR module and partial res res), 1123 located in a complete Tn501 element described in P. aeruginosa plasmid pVS1 (GenBank 1124 accession Z00027.1) (see legend to Fig. S3 for sequence details and also the main text). B. 1125 Alignments of the ~200 bp homologous res regions (including the remaining resI as well as 1126 resII and resIII subsites) located upstream of tnpR recombinase genes of Tn501-like elements 1127 impacted by Tn6335 in HP613 (pSU18-HP613, this work) and by another Tn402-like 1128 transposon in P. aeruginosa Pavimgi1 (KJ463833.1, 26, included for comparison purposes). 1129 The equivalent complete res region of a Tn501 element present in plasmid pVS1 (see the 1130 main text) is also shown at the bottom, for a better appreciation of the res subsites locations 46 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 1131 and the precise sites of insertion (black arrowheads) of the Tn402 transposons in the above 1132 cases.

Conserved nucleotides between the three res regions are indicated by asterisks below 1133 the sequences, and the different res subsites are boxed. All sequences are shown in the 5’ to 3’ 1134 direction corresponding to the tnpR coding strand, and the position of the tnpR translation 1135 initiation codons is also indicated above the alignments. C. Structural features of the 9,749 1136 bp-EcoRI fragment containing a partial fragment of Tn6335 carrying blaVIM-2 cloned from P. 1137 putida G/II LA1008. The fragment encompasses the Tn6335 region from the IRi to the tniB 1138 EcoRI site as above, plus a 4,442 bp fragment outside the IRi boundary to a nearby EcoRI site 1139 located in the LA1008 genome. The latter fragment shows 99% nucleotide identity to the 1140 indicated region of a Tn1403-like transposon and neighbouring regions (see Fig. S3 for 1141 sequence details and also the main text) outside the IRr in the P. putida H8234 chromosome 1142 (GenBank CP005976.1; positions 3,173,668 to 3,178,100; 73). D. Alignments of the ~120 bp 1143 res regions located upstream of the tnpR genes of the Tn1403-like elements located in the 1144 genomes of LA1008 (pSU18-LA1008, this work), P. putida H8234, and P. aeruginosa 1145 plasmid RPL11 (AF313472). The insertion sites of the Tn402-like transposons in LA1008 1146 and RPL11 are indicated by black arrowheads. In plasmid RPL11, the resI subsite was 1147 reconstructed by fusing the 5’-AACTG-3’ DR sequence (labeled in dark gray) corresponding 1148 to the site where a Tn402-like element was inserted (AF313472). As above, all sequences are 1149 shown in the 5’ to 3’ direction corresponding to the tnpR coding strand, with the ATG 1150 translation initiation codon indicated above the sequences. Conserved nucleotides between 1151 sequences are also indicated by asterisks below the sequences. E. Structural features of the 1152 5,971 bp-region containing the Tn402(tniABQ) element inserted into the res region of a 1153 Tn5393-like element located in the P. monteilii HB157 genome. The complete sequence was 1154 derived from the combined data of PCR assays, inverse PCR, and cloning procedures (see 1155 Materials and Methods for details). The fragment encompasses the 5,239 bp Tn402(tniABQ) 47 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 1156 element spanning from the IRi to the IRt boundaries (Fig. 2B), plus a 732 bp-fragment of the 1157 HB157 genome located outside the IRt that included the resIII subsite, the complete tnpR 1158 gene, and the first 14 bp of an aph(3'')-Ib gene. The latter fragment shows complete 1159 nucleotide identity to the equivalent region of a Tn5393c element described in Aeromonas 1160 salmonicida subsp.

salmonicida plasmid pRAS2 (AF262622.1; 55). F. Alignments of the 125 1161 bp tnpR-tnpA res intergenic region containing the resI, resII and resIII subsites of the A. 1162 salmonicida pRAS2 Tn5393c transposon with the remnant equivalent region impacted by 1163 Tn402(tniABQ) in the HB157 genome. The precise site of insertion of Tn402Δ(tniABQ), 38 1164 bp upstream of the tnpR gene of a Tn5393-like element located in this genome, is shown by a 1165 black arrowhead. The Tn402 element is inserted immediately upstream of the resIII subsite of 1166 the Tn5393-like element, with the IRt facing the tnpR gene. Conserved nucleotides are 1167 indicated by asterisks below the sequences, and the different res subsites are boxed. All 1168 sequences are shown in the 5’ to 3’ direction corresponding to the tnpR coding strand, and the 1169 tnpR translation initiation codons are indicated above the alignments. The res subsites of the 1170 Tn501-like and Tn1403-like transposons were delineated as in ref. 54; and those of the 1171 Tn5393-like elements using the Tn3 subsites following ref. 57. 1172 1173 FIG S3. Nucleotide sequences and structural features of the target sequences 1174 corresponding to Tn3 family transposons in which the Tn402-like transposons described 1175 in this work were inserted. A. DNA sequence the EcoRI fragment cloned into pSU18 1176 (pSU18-HP613) corresponding to the tnpRA region and remaining res region of the Tn501- 1177 like element impacted by Tn6335 in the P. asiatica HP613 genome. B. Same, for the Tn1403- 1178 like element present in the P. putida G/II LA1008 genome (pSU18-LA1008). C. DNA 1179 sequence of the 732 bp-fragment of the P. monteilii HB157 genome in the immediate vicinity 48 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . 1180 of the IRt of the Tn402Δ(tniABQ) element. All sequences are shown in the 5’ to 3’ direction 1181 corresponding to the direction of transcription of the tnpR recombinase genes. 49 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . TABLE 1 Strains of P. putida group used in this study Percentage of concatenate 16S rDNA-gyrB-rpoD nucleotide similarity between the studied strain and the closest P. putida group species (between brackets)c Percentage of 16S rDNA- gyrB-rpoD concatenate nucleotide similarity between the studied strain and the closest type strain (between brackets)b Identified Tn402- like class 1 integron carrying blaVIM-2 Detection of plasmids harboring i blaVIM-2 Self- transferable resistance plasmidj Species assignationd Intraspecies similaritye Identified transposonh Straina Clonef g 99.58 (P. putida) P. putida In899 Tn6336 BA9115 PpB 99.81% 99.62 (P. putida) P. putida In41 Tn6335 BA7908 PpA 98.94 (P. monteilii) P. monteilii In41 Tn6335 BA9713 PmB 99.77% 98.79 (P. monteilii) P. monteilii In528 Tn402Δ(tniABQ) HB157 PmA 98.90 (P. asiatica) 97.88 (P. putida G/IV) P. asiatica In899 Tn6336 BA7816 PaB + + 99.92 (P. asiatica) 98.14 (P. putida G/IV) P. asiatica In41 Tn6335 LD209 PaA + + 98.82-100% 99.92 (P. asiatica) 98.14 (P. putida G/IV) P. asiatica In41 Tn6335 HP613 PaA 99.92 (P. asiatica) 98.14 (P. putida G/IV) P. asiatica In41 Tn6335 HB313 PaA + + 96.86 (P. monteilii) 99.05 (P. putida G/I) P. putida G/I In41 Tn6335 HP813 PpGI 96.75 (P. monteilii) 99.28 (P. putida G/II) P. putida G/II In41 Tn6335 HE1012 PpGIIB + + 99.32% 97.19 (P. monteilii) 99.81 (P. putida G/II) P. putida G/II In41 Tn6335 LA1008 PpGIIA BA9605 93.76 (P. plecoglossicida) 98.03 (P. putida G/V) P. putida G/V In41 Tn6335 PpGV + + 99.35 (P. juntendi) P. juntendi In41 Tn6335 LA111 Pj + + aAll P. putida G strains were isolated from patients in Hospitals of Buenos Aires City (BA strains) or Rosario City (HB, LD, HP, LA strains), Argentina.

For details of the source, year of isolation, and antimicrobial resistencia profiles see Table S1. bPercentages of nucleotide similarity between a concatenate of partial sequences of the 16S rDNA, gyrB and rpoD genes of the clinical strains under study and a similar concatenate of the closest type strain. The type strains in each case and the accession numbers of the corresponding sequences are shown in Table S3. cPercentages of nucleotide similarity between a concatenate of partial sequences of the 16S rDNA, gyrB and rpoD genes of the clinical strains under study and a similar bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . concatenate of the closest P. putida group type species ( G/I, G/II, G/IV or G/V) as defined by Mulet et al., 2013 (2). The accession numbers of the corresponding sequences are shown in Table S3. dThe assignment at a species or group level of the strains under study was based on the highest nucleotide similarity found between the corresponding 16S rDNA- gyrB-rpoD concatenates (see columns B and C). eRange of similarities between 16S rDNA, gyrB and rpoD genes contacatenates among strains assigned to a particular species or group (G/II) as indicated above. fClonal differentiation within different strains assigned to a given species or group as defined above. Total DNA extracted from strains belonging to a given species or group were subjected to random PCR amplification with degenerate oligonucleotides as described previouly (39), and the different profiles found in each case were alphabetically ordered using capital letters. gSee Figure 2 for the structural characteristics of the blaVIM-2-containing integrons. Integron assignation was provided by INTEGRALL (http://integrall.bio.ua.pt/). hTransposon assignation following Tn Number Registry (http://transposon.lstmed.ac.uk/). iThe presence of plasmids carrying blaVIM-2 in the indicated P. putida G isolates was determined by conjugation assays employing E. coli DH5a, or P. aeruginosa PAO1 as recipients followed by detection of blaVIM-2 in the transconjugants (see Materials and Methods for details). jThe self-transferability of the plasmids was tested by agar mating assays employing carbapenem and rifampicin-resistant E. coli DH5α transconjugants as donors and E. coli MC4100 harboring the chloramphenicol-resistant plasmid pACYC184 as recipient (see Materials and Methods for details).+: presence of transconjugants; -: no detection of transformants or transconjugants. bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.

It is made available under a CC-BY-NC-ND 4.0 International license . P. putida G/I HP813 P. putida G/I KT2440 P. putida G/II LA1008 100 b 96 100 P. putida G/II HE1012 P. putida G/II ATCC 23483 P. monteilii HB157 P. monteilii BA9713 P. monteilii ATCC 700476 P. putida G/III IFO 14671 P. asiatica BA7816 74 76 b 100 82 100 T b 61 100 a P. asiatica LD209 P. asiatica RYU5 P. putida G/IV CFBP 4966 P. putida BA7908 P. putida BA9115 P. putida ATCC 12633 P. juntendi HPC451 P. juntendi LA111 P. juntendi BML3 100 T 100 77 b 93 78 100 T 100 T 75 T P. hunanensis LV T P. taiwanensis DSM 21245 P. plecoglossicida ATCC 700383 P. putida G/V BA9605 62 T 61 100 b P. putida G/V W619 P. entomophila L48 P. mosselii ATCC BAA-99 P. soli LMG 27941 P. guariconensis PCAVU11 T 100 T 60 T T T P. parafulva DSM 17004 P. fulva ATCC 31418 100 92 T 68 T P. cremoricolorata DSM 17059 T 72 P. reidholzensis CCOS 865 T P. hutmensis XWS2 88 T P. alkylphenolia JCM 16553 100 T P. donghuensis HYS 100 93 T P. wadenswilerensis CCOS 864 P. vranovensis DSM 16006 93 T T P. japonica JCM 21532 b P. putida G/VI IFO 3738 T P. aeruginosa ATCC 10145 100 T P. oryzihabitans ATCC 43272 0.050 bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . S-d o m ain protein g e n e E re p A vir B 1 0 p ar B vir B 1 1 p ar A m o b C M I T tra C 4 vir D 2 vir D 4 vir B 8 vir B 5 vir B 3 vir B 4 vir B 7 vir B 2 vir B 9 vir B 6 trg oriV oriT P A a. pKF715D In984 Δ a a c A 2 7 E E sul1 q a c E M I T M I T bla V I M -2 intI1 b. pMRVIM0812 Tn6336 tni Q tni C tn p C tniB tniA c. pBA7816 5,065 bp Tn6335 A 4 c a a d. pLA111 e. pLD209 3,867 bp 5,704 bp 4,771 bp 7,828 bp 1,310 bp 11,532 bp 1,389 bp 2,002 bp f. pDCPR1 99% Direct repeats Inverted repeats 10 kpb 85% Conjugal transfer Stability Adaptive Unknown function Replication bioRxiv preprint doi: https://doi.org/10.1101/2020.12.23.424275 ; this version posted December 26, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license .