Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, May 2003, p. 1522-1528, Vol. 47, No. 5
0066-4804/03/$08.00+0 DOI: 10.1128/AAC.47.5.1522-1528.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Dipartimento di Biologia Molecolare, Sezione di Microbiologia Università di Siena, I-53100 Siena,1 Laboratorio di Microbiologia, Ospedale di Circolo and Università dell'Insubria, I-21100 Varese,2 Dipartimento di Scienze e Tecnologie Biomediche, Università di L'Aquila, I-67100 L'Aquila, Italy3
Received 26 August 2002/ Returned for modification 18 November 2002/ Accepted 25 January 2003
|
|
|---|
|
|
|---|
Molecular characterization of clinical isolates producing these enzymes revealed the existence of several IMP variants. Currently, 11 variants have been reported: IMP-1 from Japan (1, 26) but also from Singapore (18), Korea (38), and the United Kingdom (34); IMP-3 (14), IMP-6 (37), IMP-10 (13), and IMP-11 (EMBL/GenBank accession no. AB074437) from Japan; IMP-2 from Italy (27); IMP-4 from Hong Kong (2) and China (9); IMP-5 from Portugal (4); IMP-7 from Canada (8) and Malaysia (10); IMP-8 from Taiwan (36); and IMP-9 from China (EMBL/GenBank accession no. AY033653). Some variants are relatively divergent from each other (85 to 96% identity at the amino acid sequence level), namely, IMP-1, IMP-2, IMP-4, IMP-5, IMP-7, IMP-9, and IMP-11 (2, 4, 8, 26, 27; EMBL/GenBank accession no. AY033653 and AB074437), while others appear to be single or double point mutants of one of the above variants (IMP-3, IMP-6, and IMP-10 of IMP-1; IMP-8 of IMP-2) (13, 14, 36, 37).
The very broad substrate specificity of IMP enzymes, including carbapenems, oxyiminocephalosporins, and serine ß-lactamase inhibitors (21), accounts for their notable clinical relevance. That is further enhanced by the fact that blaIMP determinants are carried on mobile gene cassettes inserted into chromosome- or plasmid-borne integrons (1, 4, 8, 14, 15, 20, 27, 36, 37; EMBL/GenBank accession no. AY033653 and AB074437), a location which facilitates their horizontal dissemination among different replicons and, eventually, among different strains.
In this paper we report the discovery and characterization of a new plasmid-encoded IMP variant, named IMP-12, produced by a clinical isolate of Pseudomonas putida from a hospital in northern Italy. IMP-12 is quite divergent from any other known IMP variant and exhibits significant functional differences from IMP-1.
|
|
|---|
(Gibco Life Technologies, Gaithersburg, Md.) and P. aeruginosa PAO1 (11) were used as hosts in electroporation experiments. DH5
was also used as a host for recombinant plasmids. E. coli BL21(DE3) (Novagen, Inc., Madison, Wis.) was used as a host for overexpression of the metallo-ß-lactamase gene. Bacteria were grown aerobically at 37°C unless otherwise specified. Plasmids pBC-SK (Stratagene, Inc., La Jolla, Calif.) and pET-9a (Novagen) were used as cloning and expression vectors, respectively, for the blaIMP-12 gene. In vitro susceptibility testing. Antibiotics were from commercial sources as described previously (21). MICs of ß-lactams were determined by a macrodilution broth method (24) by using Mueller-Hinton (MH) broth (Difco Laboratories, Detroit, Mich.) and a bacterial inoculum of 105 CFU per tube. Results were recorded after incubation for 18 h at 37°C and were interpreted according to the guidelines of the National Committee for Clinical Laboratory Standards (25).
ß-Lactamase assays. Carbapenemase activity in crude cell extracts was assayed spectrophotometrically as described previously (22) by using 150 µM imipenem as the substrate. Inhibition by EDTA was assayed as described previously (22) by using a final EDTA concentration of 5 mM. Protein concentrations in solution were assayed by the method of Bradford by use of a commercial kit (Bio-Rad [Richmond, Calif.] protein assay) with bovine serum albumin as a standard. Analytical isoelectric focusing (IEF) for detection of ß-lactamases was performed as described previously (22) by using crude extracts prepared from cultures grown in antibiotic-free MH broth.
DNA analysis methodology.
Basic recombinant DNA analysis was carried out as described by Sambrook and Russell (28). Genomic DNA was extracted from P. putida as described previously (16). Multiplex PCR for detection of the blaIMP and blaVIM genes was carried out with primers IMP-DIA (IMP-DIA/f, 5'-ggAATAgAgTggCTTAATTCTC; IMP-DIA/r, 5'-gTgATgCgTCYCCAAYTTCACT) and VIM-DIA (VIM-DIA/f, 5'-CAgATTgCCgATggTgTTTgg; VIM-DIA/r, 5'-AggTgggCCATTCAgCCAgA), designed to amplify internal regions of the blaIMP (361 bp) and blaVIM (523 bp) genes, respectively. PCR was carried out in a 50-µl volume by using 50 pmol of each primer, 200 µM deoxynucleoside triphosphates, 10 ng of genomic DNA of the test strain, and 3.5 U of the Expand High Fidelity PCR system (Roche Biochemicals, Mannheim, Germany) in the reaction buffer provided by the manufacturer, with the following cycling parameters: initial denaturation at 94°C for 240 s; 25 cycles of denaturation at 94°C for 60 s, annealing at 52°C for 60 s, and extension at 72°C for 90 s; and a final extension at 72°C for 600 s. Plasmid DNA was extracted by the alkaline lysis method (28). Southern blot analysis was carried out on dried agarose gels, as previously described (33), using a probe labeled with 32P by the random priming technique (28). The probe was made of a mix (molar ratio, 1:1) of blaIMP-1 and blaIMP-2 amplicons obtained with IMP-DIA primers (see above). Mating experiments were performed on MH agar plates. The initial donor/recipient ratio was 0.1. Mating plates were incubated at 30°C for 14 h. E. coli transconjugants were selected on MH agar containing 25 µg of ceftazidime/ml plus 300 µg of rifampin/ml. P. aeruginosa transconjugants were selected on MH agar containing 50 µg of ceftazidime/ml plus 300 µg of rifampin/ml. The detection sensitivity of the assay was
1 x 10-8 transconjugants/recipient with either recipient. Electroporation of E. coli and P. aeruginosa was performed by using a Gene Pulser apparatus (Bio-Rad) according to the manufacturer's instructions (for E. coli) or as described previously (5) (for P. aeruginosa). Transformants were selected on MH agar containing ceftazidime at the same concentrations used for selection of transconjugants. PCR amplification of the variable region of type 1 integrons was carried out using primers INT/5CS and INT/3CS, designed on the basis of the 5'-conserved segment (5'-CS) and 3'-CS of type 1 integrons, and the Expand High Fidelity PCR system (Roche Biochemicals), as described previously (27). DNA sequences were determined on PCR amplicons or on plasmid templates as described previously (27). Both strands were sequenced. Similarity searches against sequence databases were performed by using an updated version of the BLAST program at the National Center for Biotechnology Information server (http://www.ncbi.nlm.nih.gov/). The BBL numbering scheme (7) is used throughout this paper.
Overexpression of the blaIMP-12 gene and purification of the IMP-12 enzyme. Primers IMP-12/f (5'-CCgAATTCATATgAAgAAATTATTTgTTTTATgC-3'), containing EcoRI (underlined) and NdeI (boldfaced) restriction sites, and IMP-12/r (5'-CCggATCCTTAgTTACTTggCAgTAATgg-3'), containing a BamHI restriction site (boldfaced), were used to amplify the blaIMP-12 open reading frame by using a genomic DNA preparation of P. putida VA-758/00 as a template. The reaction was performed as described for multiplex PCR (see above) with the following cycling conditions: initial denaturation at 95°C for 240 s; 30 cycles of denaturation at 95°C for 60 s, annealing at 55°C for 90 s, and extension at 72°C for 120 s; and a final extension at 72°C for 600 s. The purified amplicon, digested with EcoRI and BamHI, was cloned into pBC-SK to produce recombinant plasmid pJD758-M, which was subjected to confirmatory sequencing. The 0.74-kb NdeI-BamHI insert of pJD758-M was then subcloned into pET-9a to produce expression vector pET-IMP12. The IMP-12 enzyme was purified from E. coli BL21(DE3)(pET-IMP12) as follows. Bacteria were grown in Buffered Super Broth (20 g of yeast extract/liter, 35 g of tryptone/liter, and 5 g of NaCl/liter, buffered with 50 mM sodium phosphate buffer [pH 7.0]) supplemented with kanamycin (50 µg/ml) at 37°C. When the culture reached an A600 of 0.7, isopropyl-ß-D-thiogalactopyranoside (IPTG) was added (final concentration, 1 mM). After 18 h, cells were collected by centrifugation (at 6,000 x g and 4°C for 15 min), resuspended in 50 mM HEPES (pH 7.5) containing 50 µM ZnSO4 (HB buffer) (1/20 of the original culture volume), and disrupted by sonication (5 cycles, for 20 s each, at 45 W). The sample was clarified by centrifugation (at 10,000 x g and 4°C for 1 h) and loaded (flow rate, 2 ml/min) onto an HR 16/5 column packed with 10 ml of Source 15S gel (Amersham Biosciences, Uppsala, Sweden) preequilibrated with HB buffer. Elution was performed using a linear NaCl gradient (0 to 1 M in 100 ml; flow rate, 2 ml/min). Fractions containing ß-lactamase activity were pooled, concentrated 20-fold using a Centriplus concentrator (YM10 membrane; Millipore, Bedford, Mass.), and loaded onto a Superdex 75 HR column (Amersham Biosciences) preequilibrated with HB supplemented with 0.15 M NaCl (HBS buffer). Elution was performed with the same buffer at a flow rate of 0.8 ml/min. The purified ß-lactamase (final concentration, 0.87 mg/ml) was stored in HBS buffer at -80°C until use.
Protein electrophoretic techniques. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis was performed according to the work of Laemmli (19) by using final acrylamide concentrations of 12 and 5% (wt/vol) for the separating and stacking gels, respectively. After electrophoresis, protein bands were stained with Coomassie brilliant blue R-250.
Determination of kinetic parameters. The kinetic parameters of the IMP-12 enzyme were determined by using essentially the same methodology previously adopted for characterization of IMP-1 and IMP-2 (21, 27). Hydrolysis of ß-lactams was monitored by using a Cary 100 UV-Vis spectrophotometer (Varian Instruments, Walnut Creek, Calif.) equipped with thermostatically controlled cells. The enzyme concentration in the reaction mixture was in the range of 0.17 to 350 nM.
Nucleotide sequence accession number. The nucleotide sequence reported in this paper has been submitted to the EMBL/GenBank/DDBJ sequence databases and assigned accession no. AJ420864.
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. MICs of various ß-lactams for P. putida VA-758/00 and for P. aeruginosa PAO1 harboring plasmid pVA758a
|
Agarose gel electrophoresis of a plasmid DNA preparation from VA-758/00 revealed the presence of plasmid DNA that was recognized by a blaIMP-1/2 probe mix in a Southern blot hybridization (Fig. 1). The blaIMP-containing plasmid, named pVA758, was purified and estimated to be approximately 50 kb, based on restriction analysis (Fig. 1).
![]() View larger version (73K): [in a new window] |
FIG. 1. (A) Agarose gel electrophoresis of a plasmid DNA preparation from P. putida VA-758/00, either undigested (lane U) or digested with EcoRI (lane RI). Identical plasmid profiles (not shown) were observed with ß-lactam-resistant transformants obtained by electroporation of the plasmid preparation into P. aeruginosa PAO1. (B) Results of Southern blot analysis of the digested plasmid preparation by using a blaIMP-1/2 probe mix. DNA size standards (in kilobases) are shown on the left.
|
by electroporation were unsuccessful, suggesting that the plasmid was unable to replicate in that host. Structure of the blaIMP determinant and of its genetic environment. PCR mapping using primers INT/5CS and INT/3CS, designed on the basis of the 5'-CS and 3'-CS of type 1 integrons (24), and pVA758 as a template yielded a 2.3-kb amplicon which was recognized by the blaIMP-1/2 probe mix in Southern blot hybridization (data not shown). Amplicon sequencing revealed the presence of two gene cassettes bounded by a 5'-CS and a 3'-CS typical of the sul1-associated type 1 integrons (31).
The first cassette carries a new blaIMP allele, named blaIMP-12, which, at the nucleotide sequence level, is 10% divergent from blaIMP-2 and blaIMP-8 (the closest homologues) and 15% divergent from blaIMP-1. The attC recombination site of the blaIMP-12-containing cassette is 125 bp long and is very similar to those of the cassettes carrying blaIMP-7 and blaIMP-5 (94 and 91% sequence identity, respectively) and also clearly related to those of the cassettes carrying blaIMP-11 (83% sequence identity) and blaIMP-1 or closely related alleles (80 to 81% sequence identity). On the other hand, it is apparently unrelated to the attC recombination site of the blaIMP-2- and blaIMP-8-containing cassettes (Fig. 2).
![]() View larger version (89K): [in a new window] |
FIG. 2. Sequence alignment of the attC recombination sites of blaIMP-containing gene cassettes. Descriptions and sources of the sequences are as follows: IMP-1, attC site of the blaIMP-1-containing cassette of integron In31 from P. aeruginosa 101/1477 (20); IMP-1/6/10, attC site of the blaIMP-1-containing cassette from Serratia marcescens AK9373 (1), which is identical to that of the blaIMP-6-containing cassette from S. marcescens KU3838 (37) and to those of the blaIMP-10-containing cassettes from P. aeruginosa PAI97 and Alcaligenes xylosoxidans AXI2 (13); IMP-11, attC site of the blaIMP-11-containing cassette from P. aeruginosa PAI112 (EMBL/GenBank accession no. AB074437); IMP-5, attC site of the blaIMP-5-containing cassette from Acinetobacter baumannii 65FFC (4); IMP-7, attC site of the blaIMP-7-containing cassette from P. aeruginosa 98/P/6327 (8); IMP-12, attC site of the blaIMP-12-containing cassette from P. putida VA-758/00 (this report); IMP-2/8, attC site of the blaIMP-2-containing cassette of In42 from A. baumannii AC-54/97 (27), which is identical to that of the blaIMP-8-containing cassette from Klebsiella pneumoniae KPO787 (36). The inverse core site is boxed; the positions of the 2L and 2R core sites (32) are indicated by arrows. Conserved residues in the first group of sequences are shaded; in the IMP-2/8 sequence, residues identical to those conserved within the first group are shaded.
|
Comparison of IMP-12 with other IMP-type enzymes. The IMP-12 protein is quite divergent from other members of the IMP family. Its closest relatives are IMP-8 and IMP-2 (88.6 and 88.2% amino acid sequence identity, respectively), while the most distant IMP variant is IMP-9 (81.6% sequence identity).
Multiple sequence alignment analysis including the 12 IMP variants revealed that IMP-12 carries unique residues at several positions where invariant residues are found in the other IMP enzymes, namely, at positions 54 (Leu versus Val), 62 (Ser versus Asn), 89 (Asn versus Ala), 102 (Gly versus Glu), 105 (Phe versus Tyr), 113 (Val versus Ile), 149 (Asn versus Asp), 240 (Lys versus Glu), 279 (Asn versus Asp), and 295 (Lys versus Asn) (Fig. 3). Compared to the other IMP enzymes, IMP-12 also carries unique residues at positions where variability was already detected, namely, at positions 38 (Val versus Ser, Arg, Ala, or Pro), 68 (Thr versus Pro or Ser), 78 (Asn versus Ala or Thr), 97 (Ala versus Thr or Asn), 106 (Thr versus Lys or Arg), 301 (Leu versus Ser, Thr, or Asn), and 302 (Leu versus Lys, Gln, or Thr) (Fig. 3).
![]() View larger version (70K): [in a new window] |
FIG. 3. Amino acid alignment of the sequence of the IMP-12 protein with those of other IMP-type enzymes. Stars indicate residues involved in the coordination of zinc ions. Substitutions observed in other variants, compared to IMP-1, are shown on a solid background. Secondary-structure elements (H, helices; S, strands) of IMP-1 (3) are also indicated below the sequences. Numbering is according to the BBL scheme (7). References for the various sequences are as follows: references 26 (IMP-1), 27 (IMP-2), 14 (IMP-3), 2 (IMP-4), 4 (IMP-5), 37 (IMP-6), 8 (IMP-7), 36 (IMP-8), and 13 (IMP-10); EMBL/GenBank accession no. AY033653 (IMP-9) and AB074437 (IMP-11); and this study (IMP-12).
|
IMP-12 was capable of hydrolyzing several ß-lactam substrates including penicillins, narrow- to expanded-spectrum cephalosporins, and carbapenems. No hydrolysis of aztreonam or temocillin was detected (Table 2). The best substrates were cephalosporins and carbapenems (kcat/Km ratios, >105 M-1 · s-1), while penicillins were uniformly poorer substrates (kcat/Km ratios, around 104 M-1 · s-1) (Table 2). The poor activity on piperacillin is consistent with the relatively low MICs of this agent for both VA-758/00 and PAO1(pVA758).
|
View this table: [in a new window] |
TABLE 2. Kinetic parameters of the purified IMP-12 enzymea
|
|
|
|---|
Considering the epidemiological distribution of IMP variants, and data from comparative structural analysis of their genes and cognate genetic elements, some conclusions can be drawn concerning the origin and evolution of the blaIMP resistance determinants. (i) There is a blaIMP gene pool which is geographically widespread and includes several allelic variants. The original source of this gene pool remains unknown, but it is likely represented by some environmental species or group of species from which blaIMP genes eventually escape to opportunistic pathogens that colonize the hospital environment, such as pseudomonads, acinetobacters, and some Enterobacteriaceae. (ii) In the latter hosts blaIMP genes have always been found on integron-borne gene cassettes, and they are most likely acquired when already part of these elements. This allows exploitation of the integron recombination system for rapid dissemination in the clinical setting, under the selective pressure generated by antimicrobial agents. The finding of the same gene cassette in different isolates and in different integrons (1, 15, 20) underscores this potential for dissemination. (iii) Gene cassettes carrying blaIMP variants are equipped with attC recombination sites that belong to two different lineages: a "long" type, including members that are 119 to 129 bp long and clearly related to each other, found in the blaIMP-1, blaIMP-5, blaIMP-6, blaIMP-7, blaIMP-10, blaIMP-11, and blaIMP-12 cassettes, and a "short" type, 72 bp long, found in the blaIMP-2 and blaIMP-8 cassettes (Fig. 2). blaIMP variants located within similar cassette frameworks could originate from the same ancestral cassette, while the occurrence of notably divergent blaIMP variants located in different cassette frameworks (e.g., IMP-1 and IMP-2) points to a different phylogeny of the cassettes, either at the time of assembly or due to shuffling of recombination sites. (iv) The appearance of quite divergent IMP variants carried on different cassette frameworks, in epidemiologically unrelated clinical isolates, most likely reflects independent acquisition of the corresponding resistance genes and suggests that recruitment of similar genes by opportunistic gram-negative pathogens could be a relatively common and widespread phenomenon.
Biochemical characterization of IMP-12 revealed both common features with, and notable functional differences from, IMP-1 and other IMP-type enzymes (IMP-2, IMP-3, IMP-4, IMP-6, and IMP-10) for which kinetic data are available (2, 13, 14, 21, 27, 37). Common features of these enzymes include an overall preference for cephalosporins and carbapenems rather than for penicillins and a lack of activity toward temocillin. The functional differences concern individual kinetic parameters with various substrates, which can eventually affect the hydrolytic efficiency for the corresponding substrate. In particular, IMP-12 is less efficient than IMP-1 and other IMP variants at hydrolysis of penicillins (kcat/Km values were around 104 M-1 · s-1 for all compounds tested). Another notable feature of IMP-12 is the very high Km (almost 1 mM) for imipenem. A similar feature was previously reported for IMP-3, which differs from IMP-1 by Glu-to-Gly and Ser-to-Gly substitutions at positions 126 and 262, respectively (11), and for which the critical role of the latter substitution in drastically lowering the affinity for imipenem was confirmed through site-directed mutagenesis experiments (14). Interestingly, the S262G substitution, which is not present in IMP variants with higher imipenem affinities, was also present in IMP-12, where it might, at least in part, contribute to that behavior. However, it should be noted that kinetic data reported for IMP-6 (37), a natural IMP-1 mutant harboring only the S262G substitution, were at variance (Km for imipenem, 110 µM) with those reported by Iyobe et al. (14). Investigation of this aspect is currently under way. The poor affinity for imipenem exhibited by IMP-12 is likely the cause for the relatively low increase in the imipenem MIC (compared to that of meropenem) for P. aeruginosa PAO1 producing the IMP-12 enzyme, and it underscores the notion that enzyme affinity for the substrate can be critical to expression of the resistance phenotype in species of low outer membrane permeability, such as P. aeruginosa, even though the hydrolytic efficiency is relatively high (in this case the kcat/Km ratio was around 2 x 105 M-1 · s-1).
A unique structural feature of IMP-12 is a substitution (Asn-62 to Ser) in the "flap" region (positions 58 to 67 [3]), which is highly conserved in all IMP variants (Fig. 3) and was found to be a sensitive structural component of IMP-1 in codon randomization and selection experiments (23). In that work, only two substitutions (Pro or Ala) were found to be tolerated, at position 62, if enzyme activity toward ampicillin was to be maintained (23). Therefore, it might be speculated that the presence of a different residue (Ser) at that position could contribute to the reduced activity toward ampicillin observed for IMP-12 relative to IMP-1. Further investigation of these aspects is under way.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»