Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, May 2005, p. 1973-1980, Vol. 49, No. 5
0066-4804/05/$08.00+0 doi:10.1128/AAC.49.5.1973-1980.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Dipartimento di Biologia Molecolare, Laboratorio di Fisiologia e Biotecnologia dei Microrganismi, Università di Siena, I-53100 Siena,1 Dipartimento di Scienze Morfologiche, Eidologiche e Cliniche, Sezione di Microbiologia, Università di Pavia, I-27100 Pavia, Italy,3 Centre d'Ingénierie des Protéines & Laboratoire d'Enzymologie, Université de Liège, B-4000 Liège, Belgium2
Received 19 August 2004/ Returned for modification 31 October 2004/ Accepted 19 January 2005
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
A number of OXA-type ß-lactamases are encoded by chromosomal genes that appear to be resident in some microbial genomes (such as in those of some Aeromonas spp., of some Shewanella spp., of Ralstonia pickettii, and of Pseudomonas aeruginosa) (12, 15, 27, 31, 32). On the other hand, several OXA-type enzymes are encoded by genes associated with mobile elements, which are often represented by gene cassettes inserted into integrons (24). These secondary OXA-type ß-lactamase (blaOXA) genes have been reported to occur in isolates of several pathogenic species, including members of the family Enterobacteriaceae, Pseudomonas aeruginosa, Acinetobacter spp., and Burkholderia cepacia, where they can variably contribute to acquired ß-lactam resistance (4, 20, 24).
OXA-type ß-lactamases are resistance determinants of increasing clinical importance, due to their potential activity on oxyimino-cephalosporins and carbapenems, their overall poor susceptibility to ß-lactamase inactivators, and the ability that some blaOXA genes exhibit to disseminate among major gram-negative pathogens. Moreover, these enzymes represent interesting models for enzymology and protein chemistry, since their mechanism exhibits notable differences from the mechanisms of other classes of serine-ß-lactamases (13, 22, 29) and since their structure-function relationships are still poorly understood.
In this paper, we report on the identification and characterization of OXA-46, a new OXA-type enzyme belonging to the OXA-2 lineage, encoded by a gene cassette inserted in an integron from a multidrug-resistant clinical isolate of P. aeruginosa.
| MATERIALS AND METHODS |
|---|
|
|
|---|
(36) was used as the host for recombinant plasmids. E. coli BL21(DE3) (Novagen Inc., Madison, Wis.) was used as the host for overproduction of the OXA-46 enzyme. Plasmid pBC-SK (Stratagene Inc., La Jolla, Calif.) was used as a cloning vector. Plasmid pET-9a (Novagen Inc.) was used as a T7-based expression vector for overexpression of the blaOXA-46 gene.
ß-Lactamase assays.
Analytical isoelectric focusing (IEF) for detection of ß-lactamases was carried out using Ampholine PAGplate (pH range, 3.5 to 9.5) (Amersham Biosciences, Uppsala, Sweden) as described previously (19), using nitrocefin (Oxoid, Basingstoke, United Kingdom) at 0.5 mM as the chromogenic substrate. ß-Lactamase activity in crude E. coli extracts and during the purification procedure was assayed spectrophotometrically by measuring the hydrolysis of 200 µM nitrocefin at 482 nm (change in
M, +15,000 M1 · cm1) in 50 mM sodium phosphate buffer (pH 7.0) at 25°C.
DNA analysis and manipulation methodology. Basic procedures for DNA analysis and manipulation were performed as described by Sambrook and Russell (36). Characterization of the variable region of the blaVIM-1-containing integron from PPV-97 was carried out by a PCR mapping and sequencing approach as described previously (34) using primers INT/5CS and INT/3CS, designed on the 5'- and 3'-end-conserved segments (5'-CS and 3'-CS, respectively) of class 1 integrons, in combination with primers VIM-DIA/f and VIM-DIA/r, designed on conserved regions of blaVIM genes, to obtain partially overlapping PCR products (Fig. 1 and Table 1) (this strategy was adopted since the INT/5CS and INT/3CS primers preferentially amplified the variable region of another integron, which was shorter than that of In80 and did not contain the blaVIM cassette). Nucleotide sequences were determined on both strands directly on PCR products, as described previously (34). Plasmid DNA and total genomic DNA were extracted from P. aeruginosa isolates as described previously (19). Southern blot analysis was carried out directly on dried gels (44) using a 32P-labeled blaVIM-1 probe as described previously (21). Construction of plasmid pBC-OXA-46 was carried out as follows. The blaOXA-46 open reading frame (ORF) was amplified by PCR with primers OXA-46/fwd, which added SacI and NdeI restriction sites at the 5' end of the ORF, and OXA-46/rev, which added a BamHI restriction site after the blaOXA-46 stop codon (Table 1). Amplification was carried out in a 100-µl volume using 50 pmol of each primer and the Expand PCR system (Roche Biochemicals, Mannheim, Germany), under the conditions recommended by the manufacturer, and the following cycling parameters: initial denaturation at 94°C for 3 min; denaturation at 94°C for 1 min, annealing at 56°C for 1 min, and extension at 72°C for 1 min, repeated for 30 cycles; and a final extension step at 72°C for 10 min. Genomic DNA of PPV-97 (10 ng) was used as a template. The amplification product was digested with SacI and BamHI and cloned into the plasmid vector pBC-SK digested with the same enzymes, resulting in recombinant plasmid pBC-OXA-46. Construction of the expression plasmid pET-OXA-46 was carried out as follows. Plasmid pBC-OXA-46 was digested with NdeI and BamHI, and the 0.8-kb fragment containing the blaOXA-46 ORF was subcloned into the expression vector pET-9a digested with the same enzymes, resulting in recombinant plasmid pET-OXA-46. The authenticity of the cloned DNA inserts was always verified by confirmatory sequencing.
|
|
(pBC-OXA-46) was determined by a macrodilution broth method (26) using cation-supplemented Mueller-Hinton broth (Difco Laboratories, Detroit, Mich.) and a bacterial inoculum of 106 CFU per tube. Results were recorded after incubation at 37°C for 18 h. Antimicrobial agents were from Sigma Chemical Co. (St. Louis, Mo.) unless otherwise specified. Mezlocillin, ceftazidime, cefepime, and meropenem were from commercial sources. Imipenem was from Merck (Rome, Italy).
Production and purification of OXA-46.
Plasmid pET-OXA-46 was transformed in E. coli BL21(DE3). The OXA-46 enzyme was purified from E. coli BL21(DE3)(pET-OXA-46) as follows. The strain was grown aerobically in 1 liter of buffered Super Broth (8), containing 50 µg/ml kanamycin, at 28°C (at this temperature, ß-lactamase production was found to be significantly higher than at 37°C in preliminary experiments). Isopropyl-ß-D-thiogalactopyranoside (IPTG) (Sigma Chemicals Co., St. Louis, Mo.) was added to a final concentration of 0.5 mM when the culture reached an A600 of 0.8, and incubation was continued for an additional 18 h. Cells were collected by centrifugation (10,000 x g for 40 min at 4°C), resuspended in 40 ml of 20 mM triethanolamine-NaOH buffer (pH 7.5) containing 1 mM MgCl2, and disrupted by sonication (10 times, for 30 s each time, at 45 W). Cell debris were removed by centrifugation (10,000 x g for 60 min at 4°C). The clarified supernatant was loaded (flow rate, 5 ml/min) on an XK 26/20 column packed with DEAE Sepharose Fast Flow (bed volume, 75 ml) (Amersham Biosciences) equilibrated with 20 mM triethanolamine-NaOH buffer (pH 7.5). Under these conditions the enzyme was obtained in the flowthrough. The ß-lactamase-containing fractions were pooled, desalted with a HiPrep desalting 26/10 column (Amersham Biosciences) against a 20 mM Tris-H2SO4 buffer (pH 9.5) (buffer A) and then loaded (flow rate, 2 ml/min) on an HR 16/5 column packed with Source Q (bed volume, 10 ml) (Amersham Biosciences) equilibrated with the same buffer. After the columns were washed with buffer A, the enzyme was eluted (flow rate, 2 ml/min) with a pH gradient obtained by mixing 20 mM Tris-H2SO4 buffer (pH 6.5) with buffer A (from 0 to 100% in 100 ml). The enzyme-containing fractions, eluted at a pH of
8.5, were pooled and stored at 80°C. All the chromatography steps were performed using an Äkta Purifier platform (Amersham Biosciences).
Protein analysis techniques. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis was performed by the method of Laemmli (16) using acrylamide concentrations of 12% and 5% (wt/vol) for the resolving and stacking gels, respectively. Electrospray mass spectrometry was carried out as described previously (8), using a Finnigan LTQ mass spectrometer equipped with an ion spray source (Thermo Electron Co., Shaumberg, Ill.). The data were analyzed with the software delivered with the instrument. Size exclusion chromatography to analyze the quaternary structure of the native OXA-46 enzyme was carried out with 0.12 mg of purified protein on a prepacked Superdex 75 HR 10/30 column (Amersham Biosciences) preequilibrated with 100 mM Tris-H2SO4 buffer supplemented with 300 mM K2SO4 (pH 7.0) and eluted with the same buffer at a flow rate of 0.4 ml/min. The column was calibrated with a mixture containing bovine serum albumin (67 kDa), ovalbumin (43 kDa), chymotrypsinogen A (25 kDa), and RNase A (13.7 kDa) (Amersham Biosciences), using the same buffer and flow rate conditions. The same experiment was also performed in the presence of 5 mM EDTA (both in the protein sample and in the elution buffer) after preincubation of the enzyme with the various compounds as described below (see inhibition assays). The retention volumes of the various standards were not affected by the presence of EDTA. Protein concentration in solution was assayed by the method of Bradford using a commercial kit (protein assay; Bio-Rad, Richmond, Calif.) and bovine serum albumin as the standard. Theoretical prediction of the leader peptide size was carried out at the SignalP 3.0 server (2). Theoretical calculation of protein molecular mass and pI was carried out using the software available at the ExPASy proteomic server (http://ca.expasy.org/). Multiple amino acid sequence alignments and unrooted tree construction were made with the help of the ClustalX program (42).
Determination of kinetic parameters and of the effect of various compounds on enzyme activity. Kinetic parameters were determined by measuring substrate hydrolysis by the purified enzyme using a Cary 100 UV-visible-light spectrophotometer (Varian, Walnut Creek, Calif.). The wavelengths and changes in the extinction coefficients used in the spectrophotometric assays were 260 nm and 7,400 M1 · cm1 for cefazolin, 260 nm and +470 M1 · cm1 for oxacillin, and as described previously for other substrates (10, 18). The steady-state kinetic parameters (Km and kcat) were determined under initial-rate conditions using the Hanes-Woolf plot (38). Km values lower than 20 µM were measured as Ki using 0.1 mM nitrocefin as the reporter substrate, as described previously (11). Enzyme reactions for kinetic measurements were carried out in 50 mM sodium phosphate buffer (pH 7.0) at 25°C, in a total volume of 500 µl, using an enzyme concentration ranging from 3 to 830 nM. Inhibition by chloride ions was assayed under the same conditions, using 100 µM nitrocefin as the reporter substrate, after a 5-min preincubation of the enzyme with various NaCl concentrations (10 to 250 mM). The effect of EDTA and of divalent cations on the OXA-46 activity was assayed, using 100 µM nitrocefin as the substrate at 25°C after a 20-min preincubation at 25°C of a 2.2 µM enzyme solution containing the various compounds (1 mM for EDTA and 0.5 mM for the divalent cations) in the buffer system used by Paetzel et al. (29) for studying the effect of cations on OXA-10 (100 mM Tris-H2SO4 [pH 7.0] containing 0.3 M K2SO4). The inactivation rates for carbapenems and mechanism-based inactivators (clavulanic acid and tazobactam) were measured at 25°C using 200 µM cephalothin as the reporter substrate. Individual inactivation parameters (k+2, K, and k+3) were calculated as previously described (7).
Nucleotide sequence accession number. The nucleotide sequence data reported in this paper have been submitted to the EMBL/GenBank database and assigned the accession number AF317511.
| RESULTS |
|---|
|
|
|---|
Analytical IEF analysis of a crude extract of PPV-97, prepared from a culture grown in antibiotic-free medium, revealed three ß-lactamase bands with pIs of 5.1, 7.8, and 8.7. The pI 5.1 band likely corresponded to the VIM-type enzyme, while the most alkaline band was likely contributed by the resident P. aeruginosa enzymes (AmpC and/or OXA-50) (12).
Analysis of the variable region of the blaVIM-containing integron of PPV-97, by means of a PCR mapping and sequencing approach as described in Materials and Methods, revealed an original array of four gene cassettes (Fig. 1), including a blaVIM-1 cassette, two tandemly repeated aacA4 cassettes, and a cassette encoding a new OXA-type ß-lactamase. This new OXA-type enzyme was assigned the name OXA-46, and the integron was named In80. The blaVIM-1 cassette of In80 was identical to that previously described as found in other integrons from VIM-1-positive clinical isolates from northern Italy (19, 21, 33). The two aacA4 cassettes are identical to each other, except for a silent A
T transversion at position 384 of the gene cassette, and encode an AAC(6')-Ib' aminoglycoside acetyltransferase (17).
Plasmid DNA was not detectable in PPV-97, and in a Southern blot experiment, a blaVIM-1 probe yielded a hybridization signal corresponding to the band of chromosomal DNA (data not shown). These results suggested a chromosomal location of In80.
The OXA-46 protein presents all the conserved motifs typical of class D enzymes, including S67XXK, S115XV, and K205TG (29), and a predicted 21-amino-acid signal peptide whose removal would yield a mature polypeptide with a calculated molecular mass of 28,583 Da and a pI of 8.7. Comparison of OXA-46 with other class D ß-lactamases at the level of primary structure revealed that it belongs to the OXA-2 sublineage (or group II according to the classification of Sanschagrin et al. [37]). Within this sublineage, OXA-46 exhibits the highest similarities (92.5% and 91.7% sequence identity, respectively) with an OXA-like enzyme encoded by a plasmid from unidentified bacteria from a wastewater treatment plant in Germany (41) (henceforth indicated as OXA-PMW) and with an OXA-like enzyme from Burkholderia cepacia clinical isolates from Ireland (4) (henceforth indicated as OXA-Bce), which are almost identical to each other (Fig. 2). OXA-46 shares 77.8% amino acid sequence identity with OXA-2, with OXA-46 being shorter by 9 residues at the carboxy terminus, and exhibits identical residues at positions 150 and 164 (Fig. 2), which are mutated in OXA-2 variants (OXA-15 and OXA-32) with extended-spectrum activities (6, 30). OXA-46 exhibits a similar degree of divergence (77.5% sequence identity) with OXA-53, an OXA-2-related extended-spectrum enzyme recently described to occur in a Salmonella enterica clinical isolate (23) that is even more divergent (72% sequence identity) from OXA-20 (25). Overall, members of the OXA-2 lineage appear to be clustered in three major sublineages (Fig. 3).
|
|
|
transformed with plasmid pBC-OXA46, a pBC-SK derivative carrying a copy of the blaOXA-46 gene cloned downstream of the Plac promoter, produced a pI 7.8 ß-lactamase activity, as determined by analytical IEF (data not shown).
Compared to E. coli DH5
(pBC-SK), DH5
(pBC-OXA46) exhibited a decreased susceptibility to penicillins and narrow-spectrum cephalosporins, while susceptibility to oxyimino-cephalosporins, cefsulodin, aztreonam, and imipenem was unaffected (Table 2). The ampicillin MIC was decreased in the presence of mechanism-based inactivators. Tazobactam was more active than clavulanic acid (Table 2), as already reported for other class D enzymes (3, 24).
|
The pI of the purified protein, evaluated by analytical IEF, was 7.8 ± 0.2, a value somewhat lower than the predicted one but in agreement with that of the ß-lactamase band detected in the crude extract of E. coli DH5
(pBC-OXA46) (see above). This also suggested that the pI 7.8 ß-lactamase band detected in P. aeruginosa PPV-97 corresponded to OXA-46.
The protein mass, determined by electrospray mass spectrometry, was 28,582 ± 3 Da, in excellent agreement with the calculated value after removal of the putative 21-amino-acid leader peptide (28,583 Da).
The Mr of the native protein, determined by size exclusion chromatography, was estimated to be 50 ± 6 kDa, suggesting that the native protein is found as a dimer. This value was apparently unaffected in the presence of 5 mM EDTA.
Kinetic parameters of OXA-46.
The kinetic parameters of OXA-46 were determined with several ß-lactam substrates. The enzyme was able to hydrolyze most penicillins and narrow-spectrum cephalosporins, while no hydrolysis was detected with temocillin, oxyimino-cephalosporins, aztreonam, or carbapenems (Table 3). Biphasic kinetics were observed with some substrates, including nitrocefin, carbenicillin, and cefazolin. The parameters for the burst phase could be determined only for the last two compounds (the burst time observed with nitrocefin was shorter than 15 s). With those compounds, the steady-state phase was essentially characterized by a drop of the turnover rate, which was more evident with carbenicillin. The highest acylation efficiencies (kcat/Km values close to 106 M1 · s1) were observed with oxacillin and nitrocefin, kcat/Km values being lower with other penicillins and narrow-spectrum cephalosporins. Turnover rates were low overall (kcat,
20 s1), except with oxacillin (Table 3).
|
Clavulanic acid was a poor inactivator of OXA-46, while tazobactam exhibited an acylation efficiency 2 orders of magnitude higher (Table 4), in agreement with susceptibility data (see above). Carbapenems efficiently inactivated OXA-46, showing an acylation efficiency even higher than that of tazobactam. In inactivation experiments, a steady state was observed with all compounds, and deacylation rates were measurable (Table 4).
|
Effect of EDTA and metal ions on OXA-46's activity. The effect of EDTA and of divalent metal ions on the activity of OXA-46 was investigated under the same experimental conditions used in previous studies of class D ß-lactamases (9, 29). The activity of OXA-46 was apparently unaffected by 1 mM EDTA or by 0.5 mM divalent cations (Mg2+, Zn2+, Mn2+, Cd2+, and Cu2+), except for a slight decrease observed with Ca2+ (residual activity, 85%).
| DISCUSSION |
|---|
|
|
|---|
OXA-46 is a new member of the OXA-2 lineage which exhibits a narrow substrate specificity, limited to penicillins and narrow-spectrum cephalosporins, that is similar to the specificities of several other narrow-spectrum oxacillinases (3, 24). Given these functional features and the repertoire of ß-lactamases present in P. aeruginosa PPV-97 (also including a VIM-1 metallo-enzyme), it is hardly conceivable that OXA-46 could provide a significant contribution to the ß-lactam resistance phenotype of that strain. Most likely, the acquisition of OXA-46 was antecedent to that of the metallo-enzyme (as also suggested by the relative positions of the two ß-lactamase gene cassettes in the integron) and could reflect a previously acquired resistance mechanism against antipseudomonas penicillins.
Within the OXA-2 lineage, at least three different sublineages could be identified. OXA-46 belongs to one of these lineages, as does another integron-encoded class D enzyme with two minor variants, one of which was recently described to occur in B. cepacia clinical isolates (4) and the other of which is encoded by a plasmid from unidentified bacteria from a wastewater treatment plant (41). Another sublineage includes OXA-2 (and its close allelic variants OXA-15 and OXA-32), OXA-3 (and its close allelic variant OXA-21), and OXA-53, while the third lineage includes OXA-20 (and its close allelic variant OXA-37). At the genetic level, the similarity between members of each lineage does not appear to be limited to the coding sequences but extends to the attC recombination sites of the gene cassettes, suggesting a common ancestry for these blaOXA cassettes. Conversely, a substantial diversity (both in sequence and in length) is observed among the attC recombination sites of the gene cassettes encoding members of different sublineages, pointing to different phylogenies of those cassettes.
OXA-46 was apparently a dimeric enzyme, and its quaternary structure was not affected by the presence of EDTA. Moreover, neither EDTA nor divalent metal ions affected the activity of the enzyme. This behavior was different from that observed with OXA-10 (29) and OXA-29 (10) and suggests a considerable heterogeneity in the effect of divalent cations on the quaternary structure and function of class D enzymes. Compared to other class D enzymes, most of which have been reported to be fully inhibited by a NaCl concentration of 100 mM (24), another peculiar feature of OXA-46 was represented by its relatively low susceptibility to NaCl inhibition. In conclusion, the biochemical characterization of OXA-46, a new member of the OXA-2 lineage, further underscores the notion that considerable functional and structural diversity can be encountered among class D ß-lactamases. Additional investigation of enzymes of this class, which exhibit an increasing clinical importance, will be important to clarify their structure-function relationships which remain poorly understood.
| ACKNOWLEDGMENTS |
|---|
J.-D.D. is a postdoctoral fellow of the Belgian Fonds National de la Recherche Scientifique.
| FOOTNOTES |
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Clin. Vaccine Immunol. | Clin. Microbiol. Rev. |
|---|---|
| J. Clin. Microbiol. | ALL ASM JOURNALS |