Complete Sequence of p07-406, a 24,179-Base-Pair Plasmid Harboring the bla_VIM-7 Metallo-β-Lactamase Gene in a Pseudomonas aeruginosa Isolate from the United States

Abstract

An outbreak involving a Pseudomonas aeruginosa strain that was resistant to all tested antimicrobials except polymyxin B occurred in a hospital in Houston, TX. Previous studies on this strain showed that it possesses a novel mobile metallo-β-lactamase (MBL) gene, designated bla_VIM-7, located on a plasmid (p07-406). Here, we report the complete sequence, annotation, and functional characterization of this plasmid. p07-406 is 24,179 bp in length, and 29 open reading frames were identified related to known or putatively recognized proteins. Analysis of this plasmid showed it to be comprised of four distinct regions: (i) a region of 5,200 bp having a Tn501-like mercuric resistance (mer) transposon upstream of the replication region; (ii) a Tn3-like transposon carrying a truncated integron with a bla_VIM-7 gene and an insertion sequence inserted at the other end of this transposon; (iii) a region of four genes, upstream of the Tn3-like transposon, possessing very high similarity to plasmid pXcB from Xanthomonas campestris pv. citri commonly associated with plants; (iv) a backbone sequence similar to the backbone structure of the IncP group plasmid Rms149, pB10, and R751. This is the first plasmid to be sequenced carrying an MBL gene and highlights the amelioration of DNA segments from disparate origins, most noticeably from plant pathogens.

In recent years reports of clinical isolates of Pseudomonas aeruginosa resistant to all β-lactams have become increasingly common. In some geographical regions this resistance has risen to approximately 40% to all antipseudomonal β-lactams, including carbapenems (1, 40). A number of these isolates have been shown to produce a metallo-β-lactamase (MBL) enzyme encoded by the transferable gene bla_IMP, bla_VIM, bla_SPM, or bla_GIM. (28, 39).

The VIM MBL family was first described in 1999 from a P. aeruginosa isolate in Italy. Subsequently, it was found in different species in Europe, Asia, and, more recently, in the United States (5, 7, 13, 14, 17, 19, 21, 22, 23-25, 27, 33). To date, more than 20 different VIM-type enzymes (http://www.lahey.org) have been identified; the dominant type is VIM-2, which has been found in more than 30 countries (3, 10, 14, 20, 21, 23, 26, 27, 30, 31, 39, 40). The bla_VIM gene is often carried on mobile gene cassettes inserted into class 1 integrons and is located either chromosomally or carried on plasmids.

P. aeruginosa strain 07-406 was isolated at a hospital in Houston, TX, from sputum of a cancer patient who presented with pneumonia. This isolate was resistant to all antimicrobials except polymyxin B, according to standard testing methods (4), and also gave a positive result with the MBL Etest strip (AB Biodisk, Solna, Sweden) (37).

The MBL gene (bla_VIM-7) from P. aeruginosa 07-406 and its immediate genetic context have been previously characterized (37). The encoded enzyme shares 77% identity with VIM-1 and 74% with VIM-2; it is the most divergent of the VIM MBLs characterized thus far and constitutes the third subgroup among the VIM-type β-lactamase family. bla_VIM-7 was shown to be located on a plasmid of approximately 24 kb (37).

Herein, we report the full nucleotide sequence of plasmid p07-406, including its complete annotation, and examine its functional characteristics.

MATERIALS AND METHODS

Bacterial strains and plasmids.

P. aeruginosa strain 07-406 carrying plasmid p07-406 was used for plasmid isolation. Escherichia coli strain DH5α [λ⁻ φ80dlacZΔM15 Δ(lacZYA-argF)U169 recA1 endA1 hsdR17(r_K⁻ m_K⁻) supE44 thi-1 gyrA relA1] was used as the recipient host for p07-406, and its fragments were cloned into pK18 as described previously (38).

Antimicrobials, reagents, and mercury susceptibility testing.

Antimicrobial agents used in this study were ceftazidime (GlaxoSmithKline, Worthing, United Kingdom) and kanamycin (Sigma Chemical Co., St. Louis, MO). Other general reagents were purchased from Sigma Chemical Co. or BDH (Poole, United Kingdom). Mercury resistance testing was carried out by plating 10⁵ CFU in 5 μl onto Muller Hinton agar containing serial dilutions of mercury (HgCl) in a manner similar to one previously described (18).

Plasmid subcloning construction, sequencing, and sequence analysis.

p07-406 plasmid DNA was isolated by an alkaline lysis method and the Qiagen maxi plasmid isolation kit (Qiagen Ltd., Cranley, United Kingdom). Plasmid DNA was restricted into four fragments (7.3 kb, 6.3 kb, 5.5 kb, and 5 kb) by EcoRI, which were subcloned into pK18 (38). Clones were sequenced on both strands by the dideoxynucleotide chain termination method. (ABC Sequencing Centre, Imperial College, London, United Kingdom). Sequence reads were assembled by Seqman (DNASTAR software [http://www.ebi.ac.uk/fasta33/nucleotide.html]), and finishing methods were included using the parent plasmid as a sequencing template for completing sequences across each junction. Sequencing analysis was performed with the Lasergene DNASTAR software package. The nucleotides were searched for potential open reading frames (ORFs) by using BLAST (http://www.ebi.ac.uk/blast2) against the EMBL prokaryotic database (http://www.ebi.ac.uk/embl/index.html).

Conjugations.

Donor (strain 07-406) and recipient bacterial cells (rifampin-resistant mutants of E. coli DH5α and P. aeruginosa strain PAO1) were grown separately to mid-log phase and harvested by centrifugation (at 12,000 × g), and the supernatant was discarded. The pellets were resuspended, and the cell suspensions were mixed together in a donor/recipient ratio of 1:1, spread on a nutrient agar plate without selective antibiotics, and incubated for 18 h at 37°C. The cell mixture was then plated onto selective medium (ceftazidime at 10 and 50 μg/ml and rifampin at 50 μg/ml) and incubated for 18 h at 37°C. E. coli DH5α and P. aeruginosa strain PAO1 carrying pUB6061 (kanamycin resistance) were used as positive controls for the mating experiments, as previously described (2).

Electroporation of p07-406.

Plasmid p07-406 was extracted using a Qiagen maxi kit and transformed by electroporation into rifampin-resistant mutants of either E. coli DH5α or P. aeruginosa PAO1. Electroporation was carried out at 2.5 kV, 25 μF, and 200 Ω with a Genepulser apparatus (Bio-Rad Laboratories, Corston, United Kingdom). Electrotransformants were selected on LB medium supplemented with ceftazidime (10 and 50 μg/ml) and rifampin (50 μg/ml).

Nucleotide sequence accession number.

The full sequence of plasmid p07-406 has been deposited under accession number AM 778842.

RESULTS AND DISCUSSION

General features of p07-406.

Plasmid p07-406 consists 24,179 nucleotides and possesses an overall GC content of 63.81%. It is predicted to contain a total of 29 ORFs possessing significant homology to functional proteins from databases. The deduced physical and genetic maps of p07-406 are shown in Fig. 1, and the genes are listed in Table 1. Numbering of this plasmid starts with the merR gene in the Tn501-like transposon. The main regions in this plasmid are the following: (i) a region of 5,200 bp having a Tn501-like mercuric resistance (mer) transposon upstream of replication region; (ii) a Tn3-like transposon carrying a truncated integron with the bla_VIM-7 gene and an insertion sequence inserted at the other end of this transposon; (iii) a region of four genes, upstream of the Tn3-like transposon, possessing very high similarity to a plasmid pXcB from Xanthomonas campestris pv. citri commonly associated with plants; and (iv) a backbone sequence similar to the backbone structure of the IncP group plasmids Rms149, pB10, and R751.

FIG. 1.

(A) Genetic map of plasmid p07-406 (accession number AM778842) showing the arrangement of the major DNA segments. Blue, Tra region; red, Mer; green, Rep; pink, ParA and ParC; yellow, class 1 integron containing bla_VIM-7; and aqua, ParE. (B) Physical maps of the regions containing transposons and transfer regions in plasmid p07-406.

TABLE 1.

Predicted ORFs in plasmid p07-406

Name	Position (bp)	No. of amino acids	Direction of transcriptiona	% GC content	Amino acid identity to informative database match		Reference or accession no.
					% Identity	Description of homolog (aa position)b
MerR	1-435	144	←	61.38	95.8	Mercuric resistance operon regulatory protein from Tn501 of P. aeruginosa plasmid pVS1 (1-144)	Z00027
MerT	507-857	116	→	61.82	93.1	Mercuric transport protein from Tn501 of P. aeruginosa plasmid pVS1	Z00027
MerP	873-1148	91	→	60.51	94.5	Mercuric transport periplasmic protein from Tn501 of P. aeruginosa plasmid pVS1	Z00027
MerA	1220-2905	561	→	65.72	99.8	Mercuric reductase from Tn501 of P. aeruginosa plasmid pVS1	Z00027
MerD	2923-3288	121	→	69.67	99	HTH-type transcriptional regulator merD (mercuric resistance protein merD) from Tn501 of P. aeruginosa plasmid pVS1	Z00027
MerE	3285-3521	78	→	64.56	100	Hypothetical mercuric resistance protein from Tn501 of P. aeruginosa plasmid pVS1	Z00027
Orf2	3518-4507	329	→	62.22	99.6	Orf2 from Tn501 of P. aeruginosa plasmid pVS1 (157-485, equivalent to aa 1-329)	Z00027
TnpR	4823-5200	125	→	60.05	88.4	Transposition resolvase from Tn501 of P. aeruginosa plasmid pVS1	Z00027
RepA	5853-7328	491	←	66.73	75.3	Replication protein from P. aeruginosa plasmid Rms149	NC007100
ParC	7527-7841	104	←	60.95	44.9	Putative partitioning-associated protein from plasmid Rms149 (6-92)	NC007100
Orf215	7925-8254	109	←	66.36	94.5	Hypothetical protein Orf215 from X. campestris pv. citri plasmid pXcB (1-109)	AY228335
ParA1	8247-8879	210	←	66.82	100	ParA1 from X. campestris pv. citri plasmid pXcB (1-210)	AY228335
Orf217	8993-9664	223	←	61.31	94	Hypothetical protein Orf217 from X. campestris pv. citri plasmid pXcB (3-223, equivalent to 44-264)	AY228335
ParA	9847-10758	303	←	68.2	84.5	Resolvase ParA X. axonopodis pv. citri (17-300)	XAC3375
p07-406.15	10913-11845	310	←	61.64	76	Transposase OrfL1 from Janthinobacterium spp.	AB095952
					68	Putative transposase TnpA from A. avenae subsp. citrulli
Vim7	12348-13145	265	←	54.51	100	blaVIM-7	37
IntI	13311-14324	337	→	61.14	100	Integrase/recombinase IntI-1 of P. aeruginosa	37
TnpM	14293-14550	85	→	60.47	98.7	Transposase TnpM	NC004989
p07-406.19	14690-14971	93	→	67.02	75.5	Hypothetical protein pB8_006 encoded by plasmid pB8	AJ863570
					43.9	YacA of X. campestris pv. citri plasmid pXcB
p07-406.20	15058-15249	63	→	64.06	80	Plasmid stabilization protein ParE from plasmid pB8	NC007502
					52	YacB of X. campestris pv. citri plasmid pXcB	AY228325
p07-406.21	15626-16234	202	→	67.32	74.6	Hypothetical protein Paer03001361 from P. aeruginosa UCBPP-PA14 (123-200)	AY257539
					75.5	Hypothetical protein TNCP25 (36-80)
p07-406.22	16449-16682	77	←	64.96	57	Hypothetical protein BPSL3261 from Burkholderia pseudomallei (1-75)	NC007502
TrbL	17006-18667	553	←	65.46	70	COG3846; type IV secretory pathway, TrbL components from P. aeruginosa (1-546)	AJ564903
TrbK	18664-18930	88	←	64.42	54	Entry exclusion protein	AJ564903
TrbJ	18946-19731	261	←	64.5	76	Mating pair formation protein (1-256)	AJ564903
					75	Conjugal transfer/entry exclusion protein precursor (1-256)
TraI	19971-21779	602	←	70.87	69.1	DNA relaxase (1-546)	AJ564903
TraJ	21797-22174	125	←	67.99	73.9	Conjugal transfer oriT-binding protein	AJ564903
TraK	22657-22956	99	→	64.33	49.5	Conjugal transfer oriT-binding protein TraK	AJ564903
KfrA	23091-23996	301	←	71.41	55.6	KfrA transcriptional regulator protein	AJ877225

^aIn accordance with Fig. 1: ←, leading strand; →, complementary DNA strand.

^baa, amino acids (aa position refers to the encoding sequence of that particular DNA structure); HTH, helix-turn-helix; A. avenae, Acidovorax avenae; B. pseudomallei, Burkholderia pseudomallei.

Replication and partition region.

The putative gene for replication (rep) of p07-406 was identified and was found to encode a 491-amino-acid protein. The highest level of similarity at the amino acid level is 75.3% identity to the putative replication protein from the IncP-6 plasmid, Rms149, characterized from P. aeruginosa strain Ps142 (accession no. NC007100). This Rep protein is also related (68% identity) to the replication protein encoded by an IncU plasmid, pFBAOT6 (accession no. CR376602), characterized from Aeromonas caviae (Fig. 2). The replicative origins of these two plasmids remain unclear (9, 29). Furthermore, we were unable to identify any of the promoter consensus sequences in the rep gene (36). Most notably, the series of identical direct repeat sequences proposed to be involved in the replication function in both Rms149 and pFBAOT6 (two copies in Rms 149 and five copies in pFBAOT6; accession numbers NC007100 and CR376602, respectively) were not found in the corresponding region of plasmid p07-406. Therefore, the rep region carried by p07-406 is clearly atypical.

FIG. 2.

Alignment of replication protein (RepA pro) of p07-406 with its closest relatives from IncP-6 plasmid Rms 149 and IncU plasmid pFBAOT6 (accession numbers NC007100 and CR376602, respectively). A black background indicates amino acids that are fully conserved.

The parC gene immediately next to the rep gene is most closely related to the parC gene (44.9%) carried on the IncP-6 plasmid Rms149 (accession no. NC007100). Interestingly, a cluster of three genes downstream of parC show very high identity (94.5%, 100%, and 94%) to orf215, parA1, and orf217, respectively, carried on plasmid pXcB from a X. campestris pv. citri strain, originating from South America and involved in the disease citrus canker (accession numbers NC005240 and AY228335).

The last gene of this region was predicted to encode a resolvase protein possessing 84.5% identity to ParA in Xanthomonas axonopodis pv. citri (accession number NC003921). There is a 16-bp inverted repeat flanking the orf215 and parA1 genes and a predicted stem-loop structure between the orf217 and parA genes. The parABC genes play a very important role in the inheritance of plasmid (6, 12, 34, 35); however, thus far, no definitive function has been attributed to either orf215 or orf217. It is unclear as to whether p07-406 possesses a ParB-type function and, if so, which ORF serves this role. It is possible that this is in part addressed by the function of KfrA.

In p07-406 the kfrA gene is separated from this section by insertion of a Tn501-like transposon between kfrA and repA. The predicted product of kfrA possesses 57% identity to the KfrA protein from plasmids R751 or pADP-1 (accession numbers AJ877225 and NC004956, respectively).

E. coli and the P. aeruginosa PAO1 strain were transformed to ceftazidime resistance by p07-406. The transformants appeared to be very stable, with 100% retention after 20 passages in nonselective growth medium; accordingly, it may be assumed that the kfrA gene from p07-406 is functional.

Nonfunctional transfer region.

To determine the transfer region of p07-406, conjugation experiments were undertaken to examine the possibility of transferring the plasmid from P. aeruginosa 07-406 (donor) to E. coli DH5α and P. aeruginosa PAO1 (recipients). p07-406 failed to transfer to the recipient strains using conjugation under laboratory conditions. However, p07-406 could be readily transferred by electroporation into both E. coli DH5α and P. aeruginosa PAO1.

The proposed conjugative region between bp 17006 and 22956 possesses a GC content of 66.3% and contains the loci trb (trbL, trbK, and trbJ) and tra (traI, traJ, and traK). These regions are very closely related to their corresponding counterparts of the IncP group plasmid pB10 (accession number AJ564903) isolated from a wastewater treatment plant and another IncP group plasmid pADP-1 (accession number NC 004956), as well as R751 (accession number AJ877225). The transfer region of p07-406 is approximately 6 kb. Comparisons with the functional transfer region of related plasmid R751 (Fig. 3) showed that there are many more genes present and involved in its conjugative transfer function, although in R751 the tra and trb regions are interrupted by insertion of a transposable element (accession numbers X5548 and U07618). The essential tra function genes traF and traG are missing in p07-406 (16). The trb region possesses genes (trbM, trbN, and trbP) which are known not to be essential for plasmid transfer and all are missing from p07-406 (14, 15). The largest segment normally associated with this locus and present in R751, trbA to trbI (accession number U07618) which is thought to play an unknown role in transfer, is also absent in p07-406.

FIG. 3.

Comparison of the transfer regions of plasmids p07-406 and R751. Genes are shown by arrows indicating the direction of transcription. The trb region is represented by gray fill; the tra region is shown as open arrows. The origin of transfer (oriT) in R751 and the predicted oriT are denoted by black circles. Genes encoding similar products are connected by broken lines. In the case of R751, the tra and trb regions are not adjacent to each other.

We attempted to identify the putative oriT of p07-406 by comparing it to the corresponding oriT sequence from plasmids R751 and RP4 (Fig. 4). The predicted oriT sequences are highly conserved and contain several inverted repeats that may be involved in target recognition during DNA processing. These data also suggest that the lack of transfer functions possessed by p07-406 may not relate to the oriT structure. These data would explain the lack of conjugation under experimental conditions.

FIG. 4.

Proposed oriT region of p07-406 aligned with the oriT region from plasmids R751 and RP4. The nucleotide numbers for the region in p07-406 are indicated on the right. A black background indicates residues that are fully conserved. seq, sequence.

Plasmid p07-406 contains a Tn501-like transposon.

p07-406 possesses a 5,200-bp region encoding the mercuric resistance (mer) transposon which has a GC content of 63% (Table 1). This mercuric resistance region of p07-406 is highly homologous to the transposon Tn501 even though it does not possess the tnpA gene required for transposition (accession number Z00027). The merA region of p07-406 displays 99.8% identity to the merA gene in Tn501. Adjacent to this region is the helix-turn-helix-type transcriptional regulator gene merD and another mercuric resistance gene, merE. Downstream of merD and merE is a large ORF, orf2, encoding a protein of 329 amino acids which is thought to play a role in the signaling cascade to MerR and MerD. This internal resolution site (res) located upstream of the tnpR gene is 127 bp long and homologous to the res site in Tn501. The transposase gene, tnpA, commonly found in Tn501 is not present in p07-406, presumably because it was deleted during the transposition event into this plasmid. In p07-406, there is a 25-bp inverted repeat sequence CGTGCTTTATTTTCCGTTTTCTGAG/CTCAGAAAACGGAAAATAAAGCACG immediately flanking the Tn501-like transposon.

The level or resistance to Hg ions was 8 μg/ml in E. coli carrying p07-406 and 16 μg/ml in the host strain (P. aeruginosa 07-406) compared to 0.25 μg/ml for the E. coli DH5α alone, indicating that the mer region is likely to be functional.

The Tn3 family transposon carries a truncated integron with the bla_VIM-7 gene and an insertion sequence.

There is a second transposition element present in this plasmid between bp 10913 and 14550 immediately downstream of the partition region and possessing a GC content of 59%. This region has a truncated class 1 integron, carrying the bla_VIM-7 gene, inserted into transposase gene tnpM. This gene is also truncated, only having 257 bp left from the N terminus, but the amino acid sequence showed 98% identity to other transposases from various plasmids including R478 and pAPEC-O1-R (8, 11). Another transposase gene, orfL1, possessing 70% identity to an insertion element from Janthinobacterium spp., is located downstream of the aacA4. Therefore, given the evidence, it is likely that orfL1 inserted into the aac gene cassette of the class 1 integron (also carrying bla_VIM-7) and that this region inserted into transposase gene tnpM.

Other unknown regions.

p07-406 possesses a large section of DNA (bp 14690 to 16682, between the Tn3-like transposon and the conjugative transfer region) having a GC content of 66%, but the encoded products from these ORFs have no defined function. Genes denoted p07-406.19 and p07-406.20 (downstream of the Tn3-like transposon) show homology to genes carried on the plasmid, pB8, from P. aeruginosa. p07-406.19 displays 75% identity to a unknown gene, orf3, and p07-406.20 encodes a protein possessing 80% identity to a plasmid stabilization protein, ParE (encoded on plasmid pB8) (32). Genes p07-406.21 and p07-406.22 (adjacent to the aforementioned p07-406.19 and p07-406.20) also encode proteins of undefined function. However, it is possible that this region is involved in either plasmid stabilization and/or transfer.

Conclusions.

Our molecular studies on plasmid p07-406 have revealed significant sections of the plasmid containing DNA related to the plant pathogen X. campestris pv. citri. The overall GC content of p07-406 is 64%, which suggests that it did not originate in Enterobacteriaceae but in environmental bacteria such as Pseudomonas or Xanthomonas (36). The two main segments of p07-406 (for conjugative transfer and mating-pair formation, on the one hand, and replication and stable inheritance, on the other) are derived from different ancestral IncP-type plasmids. Different sections of this plasmid suggest that p07-406 has been created from both environmental (plants) and clinical DNA segments. This is the first report of the complete nucleotide sequence of a plasmid harboring an MBL gene, in this case, bla_VIM-7, from the United States.