|Home | About | Journals | Submit | Contact Us | Français|
A carbapenem-resistant Pseudomonas stutzeri strain isolated from a Dutch patient was analyzed in detail. This isolate produced a metallo-β-lactamase (MBL) whose gene, with 43.5% GC content, was cloned and expressed in Escherichia coli. β-Lactamase DIM-1 (for Dutch imipenemase) was weakly related to other Ambler class B β-lactamases, sharing <52% amino acid identity with the most closely related MBL, GIM-1, and 45% identity with IMP-type MBLs. The β-Lactamase DIM-1 significantly hydrolyzed broad-spectrum cephalosporins and carbapenems and spared aztreonam. This MBL gene was embedded in a class 1 integron containing two other gene cassettes, encoding resistance to aminoglycosides and disinfectants, that was located on a 70-kb plasmid.
Metallo-β-lactamases (MBLs) are among the most challenging antibiotic resistance threats in Gram-negative organisms (8). These enzymes very efficiently hydrolyze all β-lactams, including carbapenems (with the exception of aztreonam), and are located most often on transferable genetic platforms (33). Different MBL-type enzymes have been described, with IMP and VIM derivatives being the most widespread (33). The blaIMP-like and blaVIM-like genes have been identified in most clinically relevant bacteria belonging to the Enterobacteriaceae family, in Pseudomonas spp., and in Acinetobacter spp. (33). Several other MBLs have been identified in specific geographical locations, including SIM-1 from Acinetobacter baumannii in Korea (13), KHM-1 from Citrobacter freundii in Japan (27), and NDM-1 from Klebsiella pneumoniae, Escherichia coli, and Enterobacter cloacae in India and the United Kingdom (8a, 35), although SPM-1 in Brazil (17, 20, 30), GIM-1 in Germany (5), and AIM-1 in Australia (36) were all identified in Pseudomonas aeruginosa.
The genetic vehicles that carry MBL genes vary, but most of them are found in the form of gene cassettes (blaIMP-like, blaVIM-like, blaSIM, and blaGIM-1) embedded into class 1 integron structures (23). In addition, a peculiar insertion sequence named ISCR4, belonging to the IS91 family and likely moving by rolling-circle transposition, has been identified at the origin of mobilization of blaSPM-1 (20, 29).
This study characterized a novel and clinically significant MBL whose gene was identified in a class 1 integron.
Pseudomonas stutzeri clinical isolate 13 was identified with the API-20 NE system (bioMérieux, Marcy l'Etoile, France) and confirmed by rRNA gene sequencing. E. coli TOP10 was the host for cloning experiments (19).
Antibiotic-containing disks were used for routine antibiograms by the disk diffusion assay (Sanofi-Diagnostic Pasteur, Marnes-la-Coquette, France), as recommended previously (6). The extended-spectrum β-lactamase (ESBL) double-disk synergy test was performed with disks containing ceftazidime or cefepime and ticarcillin-clavulanic acid on Mueller-Hinton agar plates, and the results were interpreted as described previously (11). MBL detection was performed by using Etest MBL strips (AB Biodisk, Solna, Sweden).
MICs were determined by an agar dilution technique with Mueller-Hinton agar (Sanofi-Diagnostic Pasteur), with an inoculum of 104 CFU per spot, as described previously (16). All plates were incubated at 37°C for 18 h at ambient atmosphere. MICs of β-lactams were determined alone or in combination with a fixed concentration of clavulanic acid (4 μg/ml) or tazobactam (4 μg/ml). MIC results were interpreted according to the guidelines of the CLSI (6).
Total DNA of P. stutzeri 13 was extracted as described previously (2). This DNA was used as a template under standard PCR conditions (25) with a series of primers designed for the detection of the class B β-lactamase genes blaIMP, blaVIM, blaSPM, and blaSIM (13, 22). Southern hybridizations were performed as described by Sambrook et al. (26), using an enhanced chemiluminescence (ECL) nonradioactive labeling and detection kit (GE Healthcare, Orsay, France).
Total DNA of the P. stutzeri 13 isolate was digested with the XbaI restriction enzyme, ligated into the XbaI site of plasmid pBK-CMV, and transformed into E. coli TOP10, as described previously (16). Recombinant plasmids were selected on Trypticase soy agar plates containing amoxicillin (50 μg/ml) and kanamycin (30 μg/ml). The cloned DNA fragments of several recombinant plasmids were sequenced on both strands with an Applied Biosystems sequencer (ABI 3100; Applied Biosystems, Foster City, CA). The entire sequence provided in this study was made of sequences of several plasmids that contained overlapping cloned fragments. The nucleotide and deduced amino acid sequences were analyzed and compared to sequences available at the National Center for Biotechnology Information website (http://www.ncbi.nlm.nih.gov).
Transformation experiments were performed with P. stutzeri 13 DNA and a P. aeruginosa PU21 recipient strain, as described previously (21). Plasmid DNA extraction from P. stutzeri 13 was attempted with a Qiagen Plasmid DNA Maxi kit (Qiagen, Courtaboeuf, France) by the Kieser method (12), and DNA was visualized and sized as described previously (21). Hybridization was performed with a 688-bp probe specific for the blaDIM-1 gene, generated with internal primers DIM-1A (5′-TCTATTCAGCTTGTCTTCGC-3′) and DIM-1B (5′-TGTTAGAGGCTGTCTCAGCC-3′).
Cultures of E. coli TOP10(pXD-1) were grown overnight at 37°C in 4 liters of Trypticase soy broth containing amoxicillin (100 μg/ml) and kanamycin (30 μg/ml). β-Lactamase was purified by ion-exchange chromatography. Briefly, the β-lactamase extract was obtained by sonication of the cells, resuspended in 100 mM sodium phosphate buffer (pH 7), cleared by ultracentrifugation, treated with DNase, and dialyzed against 20 mM diethanolamine buffer (pH 8.9). This extract was loaded on a Q-Sepharose column, and the β-lactamase-containing fractions were eluted with a linear 0 to 0.5 M NaCl gradient. The fractions containing the highest β-lactamase activity were again dialyzed against the buffer mentioned above, and the procedure was repeated by eluting the protein more slowly, with a linear 0 to 0.2 M NaCl gradient. The purity of the enzyme was estimated by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis analysis. The protein content was measured by the Bio-Rad DC protein assay.
IEF analysis was performed with an Ampholine polyacrylamide gel (pH 3.5 to 9.5) as described previously (15), using a purified β-lactamase extract from a culture of E. coli TOP10(pXD-1). The focused β-lactamases were detected by overlaying the gel with 1 mM nitrocefin (Oxoid, Dardilly, France) in 100 mM phosphate buffer (pH 7.0).
Purified β-lactamase was used for kinetic measurements performed at 30°C with 50 mM HEPES buffer (pH 7.5) supplemented with 50 μM ZnSO4, using an Ultrospec 2000 UV spectrophotometer (Amersham Pharmacia Biotech), as described previously (3). The specific activity of the purified β-lactamase from E. coli TOP10(pXD-1) was obtained as described previously, with imipenem as the substrate (2). One unit of enzyme activity was defined as the activity which hydrolyzed 1 μmol of imipenem per min per mg of protein. The total protein content was measured with a DC protein assay kit (Bio-Rad, Ivry-sur-Seine, France). In order to evaluate whether the zinc ion concentration might have an impact on DIM-1 hydrolytic activity, specific activities were measured using the purified enzyme and various concentrations of ZnSO4 (0, 20, 50, 100, 300, and 600 μM and 1 mM), with imipenem as the substrate.
The fifty percent inhibitory concentration (IC50) was determined for DIM-1 as the concentration of EDTA that reduced the hydrolysis rate of 100 μM benzylpenicillin by 50% under conditions in which DIM-1 was preincubated with various concentrations of EDTA for 10 min at 30°C before adding the substrate.
The nucleotide sequence data reported in this work have been deposited in the GenBank nucleotide database under accession no. GU323019.
P. stutzeri isolate 13 was isolated in June 2007 at the VU Medical Center, Amsterdam, Netherlands, from purulent exudate obtained from tibial osteomyelitis in a 55-year-old patient who did not have any history of recent travel or hospitalization elsewhere. P. stutzeri 13 was resistant to ticarcillin, piperacillin, piperacillin-tazobactam, and imipenem, had reduced susceptibility to ceftazidime, cefepime, and cefpirome, and remained fully susceptible to aztreonam. A double-disk synergy test was negative with clavulanate-ceftazidime and clavulanate-imipenem (data not shown) but was positive with the MBL Etest (MIC of IMP of 64 μg/ml versus MIC of IMP-EDTA of 2 μg/ml). This P. stutzeri isolate was also resistant to gentamicin, tobramycin, fluoroquinolones, rifampin, chloramphenicol, and tetracycline and remained susceptible to amikacin, netilmicin, and colistin.
Preliminary attempts to detect MBL-encoding genes by PCR failed (data not shown). Using total DNA of P. stutzeri 13 as a template in cloning experiments, several E. coli strains harboring recombinant plasmids, including pXD-1, were obtained. Sequence analysis of a ca. 10-kb cloned fragment of pXD-1 revealed a 756-bp open reading frame (ORF) encoding a 251-amino-acid preprotein corresponding to an Ambler class B β-lactamase designated DIM-1 (for Dutch imipenemase) (1). It possessed the conserved motifs characteristic of MBL enzymes (Fig. (Fig.1),1), including the consensus zinc binding motif HXHXD (residues 116 to 120), together with the critical residues for MBL activity, such as His116, His118, Asp120, His196, Cys221, and His263, according to the BBL nomenclature (9, 31-34) (Fig. (Fig.1).1). According to its amino acid sequence, it can be classified as a member of the subclass 1 MBLs, together with IMP- and VIM-type β-lactamases (25). DIM-1 was distantly related to other Ambler class B β-lactamases. Indeed, the highest percentages of amino acid identity were 52% with GIM-1 (5), 49% with a putative MBL identified in silico in the genome of Shewanella denitrificans (GenBank accession no. NC_007954), and 48% with KHM-1 (27). β-Lactamase DIM-1 shared 45% identity with the widespread IMP-type enzymes and only 30% identity with the VIM-type enzymes. The G+C content of the blaDIM-1 gene was 43.5%, a value which fits with those of many Gram-negative species but differs significantly from the G+C content of the P. stutzeri genome, which is 63% according to the GenBank database (accession no. NC_009434), thus suggesting horizontal acquisition of blaDIM-1.
MICs of β-lactams for E. coli TOP10(pXD-1) indicated the expression of an MBL that hydrolyzed expanded-spectrum cephalosporins (including cephamycins) together with carbapenems, that conferred reduced susceptibility to imipenem and meropenem, and that spared aztreonam (Table (Table1).1). The addition of β-lactamase inhibitors such as clavulanic acid or tazobactam did not restore the susceptibility to β-lactams.
IEF analysis showed that P. stutzeri 13 and E. coli TOP10(pXD-1) had β-lactamase activities with a pI value of 6.2, corresponding to that of DIM-1 (data not shown). The specific activity of the purified β-lactamase DIM-1 was 21 U mg of protein−1 with benzylpenicillin as the substrate. Its overall recovery was 80%, with a 45-fold purification. The purity of the enzyme was estimated to be >95% according to SDS gel electrophoresis analysis (data not shown). The kinetic parameters of DIM-1 showed its broad-spectrum activity against most β-lactams, including oxyimino-cephalosporins, cephamycins, and carbapenems but excluding aztreonam (Table (Table2).2). Analysis of the relative hydrolysis rates of DIM-1 showed that cefotaxime was hydrolyzed at a similar level to that of benzylpenicillin, and cefoxitin was also a good substrate. Ceftazidime was significantly hydrolyzed, with a Km value of 50 μM, reflecting a relatively good affinity of DIM-1 for that substrate, as commonly observed for many MBLs. On the other hand, the monobactam aztreonam was not hydrolyzed by DIM-1, like the case for all MBLs (4, 33). IC50 determinations performed with benzylpenicillin as a substrate showed that DIM-1 activity was inhibited by EDTA (175 μM).
In addition, we observed that the DIM-1 hydrolytic activity was correlated with the zinc ion concentration. Whereas the specific activity of DIM-1 for imipenem was 0.08 U mg of protein−1 in the absence of ZnSO4, it was 0.12, 0.15, 0.18, 0.28, 0.39, and 0.54 U mg of protein−1 in the presence of increased concentrations of ZnSO4, i.e., 0, 20, 50, 100, 300, and 600 μM and 1 mM, respectively, thus indicating a significant correlation.
Sequence analysis of the recombinant plasmid pXD-1 harboring the blaDIM-1 gene revealed that it was in the form of a gene cassette, which was inserted at the attI1 recombination site (7), similar to the other acquired MBL-encoding genes identified in Gram-negative organisms, such as blaIMP, blaVIM, and blaSIM genes. Analysis of the 5′-end sequence of the integron showed that the gene cassettes' expression was under the control of the PC promoter, but the secondary promoter (P2) was not under its active form (14). Thus, the gene cassettes located in that integron are under the control of weak promoter sequences.
The dim-1 gene cassette possessed imperfect core (GTTAGAG) and inverse core (CGCTAAC) sites, with the latter being located inside the blaDIM-1 coding sequence (23 bp from the 3′ end of the gene). The length of its 59-be sequence was only 31 bp, and the usually conserved 2R and 2L regions of 59-be sites were not detected, indicating that this 59-be was likely truncated and therefore suggesting that the dim-1 cassette may no longer be functional for recombination. A second gene cassette was identified as containing the aadB gene, encoding resistance to aminoglycosides (Fig. (Fig.2).2). The third gene cassette contained the qacH gene, encoding resistance to disinfectants. Inside the qacH gene cassette, the ISKpn4 insertion sequence that targeted the 59-be was identified (24), as previously noticed with other members of the IS1111 family that preferentially target the gene cassette 59-be sites (18, 28).
Analysis of the right extremity of this integron, named In124, showed that the usually identified 3′-conserved segment made of the qacEΔ1 and sul1 genes was absent, but the tniC gene of transposon Tn5090, encoding a 207-amino-acid-long resolvase, was identified. The tnpA gene of transposon Tn1403 was identified (a Tn3-like transposon), as well as its right inverted repeat (IRR) extremity (Fig. (Fig.22).
The left extremity of class 1 integrons, defined by an inverted repeat structure, was identified downstream of the int1 gene, truncating a tnpA gene (lacking 266 bp at its 3′ extremity) that encodes the transposase of transposon Tn1721. The IRR of Tn1721 was also identified and was preceded by a novel insertion sequence element named ISPst10 (Fig. (Fig.2).2). ISPst10 is 1,113 bp long and belongs to the IS30 family, and its transposase shares 90% amino acid identity with that of ISPstI, also identified from a P. stutzeri isolate (http://www-is.biotoul.fr). Its transposition had generated a 3-bp duplication.
Electrotransformation experiments did not result in the transfer of blaDIM-1 to either P. aeruginosa PU21 or E. coli TOP10 as a recipient strain. However, analysis of the plasmid content of P. stutzeri 13 identified a single 70-kb plasmid that harbored the blaDIM-1 gene, as confirmed by Southern hybridization (data not shown). However, the negative mating-out result prevented us from knowing which other antibiotic resistance markers were plasmid associated along with the blaDIM-1 gene.
In conclusion, we identified a novel MBL sharing weak amino acid identity with other MBLs but sharing similar biochemical properties. The dissemination of the blaDIM-1 gene among other Gram-negative isolates, especially among the Enterobacteriaceae, remains to be evaluated. Indeed, several pieces of evidence have indicated that environmental species such as Pseudomonas sp. may act as intermediate reservoirs for capturing antibiotic resistance genes from other environmental species and then exchanging those genes with genes of the Enterobacteriaceae.
This work was mostly funded by the INSERM, France, and by grants from the Ministère de l'Education Nationale et de la Recherche (UPRES-EA3539), Université Paris XI, France, and from the European Community (DRESP2 [LSHM-CT-2005-018705] and TROCAR [HEALTH-F3-2008-223031]). J.-M.R.-M. was funded by a postdoctoral grant from the Ministerio de Educación y Ciencia from Spain (2007/0292).
Published ahead of print on 22 March 2010.