Biochemical Characterization of the Naturally Occurring Oxacillinase OXA-50 of Pseudomonas aeruginosa

Abstract

The bla_OXA-50 gene (formerly known as the PA5514 gene) is an oxacillinase gene identified in silico in the genome of Pseudomonas aeruginosa PAO1. By using a mutant strain of P. aeruginosa PAO1 that had an inactivated bla_AmpC cephalosporinase gene, the bla_OXA-50 gene was shown to be expressed constitutively in P. aeruginosa. This β-lactamase gene was cloned onto a multicopy plasmid and expressed in P. aeruginosa and Escherichia coli. It conferred decreased susceptibility to ampicillin and ticarcillin and, interestingly, to moxalactam and meropenem in P. aeruginosa but not in E. coli. Overexpression and purification enabled us to determine the molecular mass (25 kDa), the pI value (8.6), and the hydrolysis spectrum of the OXA-50 β-lactamase. It is a narrow-spectrum oxacillinase that uncommonly hydrolyzes imipenem, although at a low level. Very similar oxacillinase genes were identified in all P. aeruginosa isolates from various geographical origins tested. The weak variability of the nucleotide sequence of this gene (0 to 2%) corresponded to that found for the naturally occurring bla_AmpC cephalosporinase gene of P. aeruginosa. The study indicated that P. aeruginosa harbors two naturally encoded β-lactamase genes, one of which encodes an inducible cephalosporinase and the other of which encodes a constitutively expressed oxacillinase.

Pseudomonas aeruginosa is an opportunistic pathogen that mainly causes severe pneumonia in immunocompromised hosts and may colonize the lungs of patients with cystic fibrosis (31). A mutant strain of P. aeruginosa PAO1 that had an inactivated cephalosporinase gene was constructed in a previous study (20). The resulting β-lactam resistance profile was oversusceptibility to all β-lactam antibiotics, demonstrating the main role of cephalosporinase in naturally occurring resistance in that species. However, in silico analysis of the P. aeruginosa PAO1 genome (32) identified a putative β-lactamase gene, named PA5514, whose sequence was consistent with those of several oxacillinase genes. Thus, the purpose of the present work was to clone and overexpress the gene, characterize its biochemical properties, and analyze its diversity among P. aeruginosa isolates.

MATERIALS AND METHODS

Bacterial strains and plasmids.

P. aeruginosa PAO1 was used as the template (32) for PCR amplification and cloning experiments (29). P. aeruginosa KG2505 (20), an AmpC-deficient PAO1 mutant (ampC::ΩSm mexA::resΩSm), was used as the host in the cloning experiments with the multicopy shuttle vector pBBR1MCS.3 (13). A series of clinical P. aeruginosa isolates was analyzed, i.e., P. aeruginosa COL-1 (23) and P. aeruginosa ED-1 (24), which were obtained from the Hospital Bicêtre, Paris, France; P. aeruginosa GW-1 (26), which was from South Africa; P. aeruginosa Ka.209 (27), which was from Spain; P. aeruginosa strain 1, which was from Thailand (8); and P. aeruginosa IND, which was from India (this study). Escherichia coli reference strains DH10B and BL21(DE3), as well as the TOPO plasmid (Invitrogen, Life Technologies, Cergy-Pontoise, France), were used for the cloning experiments; and pET9a (Stratagene, Amsterdam, The Netherlands) was used as an expression vector.

Susceptibility testing.

The antimicrobial agents and their sources have been referenced elsewhere (22). Antibiotic-containing disks were used to detect antibiotic susceptibility with Mueller-Hinton agar plates and by a disk diffusion assay (Sanofi Diagnostics Pasteur, Marnes-la-Coquette, France). The results were interpreted according to the guidelines of the French Society for Microbiology (http://www.sfm.asso.fr/nouv/general.php?pa=2). The MICs of selected β-lactams were determined by an agar dilution technique, as described elsewhere (22). MIC results were interpreted according to the guidelines of the National Committee for Clinical Laboratory Standards (18).

PCR amplification and cloning experiments.

Whole-cell DNAs of P. aeruginosa PAO1 and clinical isolates were extracted as described previously (22, 29) and were used as templates for PCR amplification. PCR products of 869 bp, including the complete sequence of the bla_OXA-50 gene (previously named PA5514), were generated with primers S (5′-AATCCGGCGCTCATCCATC-3′) and AS (5′-GGTCGGCGACTGAGGCGG-3′), the sequences of which are located at each end of the bla_OXA-50 gene of the P. aeruginosa PAO1 genome (32). The diversity of the oxacillinase genes in P. aeruginosa was studied by sequencing the PCR products obtained with whole-cell DNA of unrelated P. aeruginosa strains, as described previously (8).

The PCR amplicon of the entire bla_OXA-50 gene of P. aeruginosa PAO1 was then cloned into the pPCRBluntII-TOPO plasmid, as recommended by the manufacturer (Invitrogen, Life Technologies), and expressed in E. coli DH10B. The cloned insert was then removed by restriction with SpeI and XbaI (Amersham Pharmacia Biotech, Orsay, France) and subcloned into the SpeI- and XbaI-digested shuttle vector pBBR1MCS.3 (13), which replicates in P. aeruginosa and E. coli. The recombinant plasmid, named pBB-1, was introduced into P. aeruginosa strains PAO1 and KG2505, as described previously (13).

The following primers were used to clone the bla_OXA-50 gene into the expression vector pET9a: primer O₅₀-ATG, which contained the initiation codon of the bla_OXA-50 gene; ATG, along with an NdeI restriction site, which is underlined (5′-AAAAACATATGCGCCCTCTCCTCTTCAGT-3′); and primer O₅₀-TGA, which contained the termination codon of the bla_OXA-50 gene, TGA, along with a BamHI restriction site, which is underlined (5′-AAAAGGATCCATCAGGGCAGTATCCCGAGAGC-3′). Whole-cell DNA of P. aeruginosa PAO1 was used as the template. After PCR amplification, a 788-bp NdeI-BamHI-restricted fragment was ligated into the NdeI-BamHI-restricted pET9a expression vector and electroporated into E. coli BL21(DE3). E. coli BL21(DE3) containing recombinant plasmid pET-1 was selected on kanamycin-containing agar plates (30 μg/ml).

DNA sequencing and protein analysis.

Both strands of the PCR products corresponding to the bla_OXA-50 genes of unrelated P. aeruginosa isolates and all the insert region of the recombinant plasmids were sequenced with an automated sequencer (ABI 3100; Applied Biosystems, Les Ulis, France). The nucleotide and protein sequences were analyzed with software available over the Internet at the National Center for Biotechnology Information website (http://www.ncbi.nlm.nih.gov).

IEF analysis.

β-Lactamases were submitted to isoelectric focusing (IEF), as described previously (22), with culture extracts of strains P. aeruginosa PAO1, P. aeruginosa KG2505, and E. coli BL21(DE3) with and without recombinant plasmids (i.e., pBB-1 for the P. aeruginosa strains or pET-1 for the E. coli strain) and with the purified β-lactamase OXA-50.

β-Lactamase purification, relative molecular mass determination, and N-terminal sequencing.

Induction of an exponentially growing culture of E. coli BL21(DE3)(pET-1) with 0.4 mM isopropyl β-d-thiogalactopyranoside (IPTG) was performed at 37°C for 5 h in Trypticase soy (TS) broth. Four liters of this culture was pelleted and resuspended in 30 ml of 20 mM Tris H₂SO₄ buffer (pH 9.0). The protein extracts obtained were purified as described previously (24). Briefly, the extracts were subjected to several purification steps, including ion-exchange chromatography on a Q-Sepharose column equilibrated with a 20 mM Tris-H₂SO₄ buffer (pH 9.0), followed by chromatography on an S-Sepharose column equilibrated with 25 mM malonate buffer (pH 5.6). Elution of the β-lactamase was performed with a linear K₂SO₄ gradient (0 to 500 mM). Peaks of β-lactamase activities were dialyzed overnight against 50 mM phosphate buffer (pH 7.0). The protein content was measured by the Bio-Rad DC protein assay, and the specific activities of the crude extract and the purified β-lactamase from E. coli BL21(DE3)(pET-1) were compared. The protein purification rate and the relative molecular mass of OXA-50 β-lactamase were estimated by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) analysis (22). The proteins were then transferred onto a polyvinylidene difluoride (PVDF) membrane (Problott; Applied Biosystems) by passive absorption, as described by Messer et al. (16), with some modifications. After the destaining step, the bands of interest were excised and dried in a Speed-Vac dryer for 30 min, and then the gel pieces were reswollen in 136 μl, which corresponded to the initial volume of the excised band (2% SDS in 0.2 M Tris-HCl [pH 8.5]), for 60 min. After the gel pieces were swollen, a fivefold volume of high-pressure liquid chromatography-grade water was added, and then a piece of prewetted (methanol) PVDF membrane (4 by 4 mm; Problott; Applied Biosystems) was added to the solution. The protein was transferred to the membrane by gentle shaking for 2 days at room temperature (23°C). Subsequently, the membrane was washed five times with 1 ml of 10% methanol with vortexing. N-terminal Edman sequencing was performed on an Applied Biosystems Procise 494HT sequencer with the reagents and by the methods recommended by the manufacturer.

Kinetic studies.

Purified β-lactamase was used for kinetic measurements at 30°C in 100 mM Tris-H₂SO₄-300 mM K₂SO₄ (pH 7.0) (21, 24). The k_cat and K_m values were determined by analyzing β-lactam hydrolysis under initial-rate conditions with a UV spectrophotometer, as described previously (24). The 50% inhibitory concentrations (IC₅₀s) of clavulanic acid, tazobactam, sulbactam, and NaCl were determined (24). Various concentrations of these inhibitors were preincubated with purified enzyme for 3 min at 30°C to determine the concentrations that decreased the rate of hydrolysis of 100 μmol of nitrocefin by 50%. The effect of carbon dioxide on the modulation of the enzymatic properties of OXA-50 was investigated by adding NaHCO₃ to the reaction buffer at a final concentration of 10 mM. Specific activities and k_cat and K_m values were determined for most substrates in the presence and in the absence of bicarbonate (9, 15, 21). The specific activities of the protein extracts and purified β-lactamase from culture of E. coli BL21(DE3)(pET-1) were determined with 100 μM nitrocefin as the substrate. One unit of activity was defined as the amount of enzyme that hydrolyzes 1 μmol of nitrocefin per min (24). The inducibility of the β-lactamase content from the P. aeruginosa KG2505 culture was tested in TS broth at 37°C by the induction protocol described previously (19) with cefoxitin and imipenem as β-lactam inducers at various concentrations (0.1, 0.5, and 1 μg/ml).

Nucleotide sequence accession numbers.

The nucleotide sequences of the OXA-50 variants reported in this paper have been submitted to the EMBL/GenBank nucleotide sequence database and assigned accession numbers AY306130 to AY306135.

RESULTS AND DISCUSSION

Antibiotic susceptibility and β-lactamase activity.

P. aeruginosa KG2505 (ampC::ΩSm mexA::resΩSm [20]), an AmpC-deficient derivative of wild-type strain P. aeruginosa PAO1, was hypersusceptible to all β-lactams compared to the susceptibilities of any of the wild-type P. aeruginosa strains (Table 1), underlining an important role of the cephalosporinase in the natural resistance of P. aeruginosa to β-lactams. The multicopy plasmid pBB-1, which contains the bla_OXA-50 gene from P. aeruginosa PAO1, provided a slight additional level of resistance to amoxicillin, ticarcillin, and, interestingly, meropenem in P. aeruginosa KG2505. Specific β-lactamase activities of 0.2 and 20 mU/mg of protein were determined for P. aeruginosa KG2505 and P. aeruginosa KG2505(pBB-1), respectively, with 100 μM nitrocefin as the substrate. This result suggests that the bla_OXA-50 gene is expressed in P. aeruginosa KG2505 and that it induces only minor phenotypic changes when expressed from a multicopy plasmid in P. aeruginosa KG2505. The introduction of plasmid pBB-1 into parental strain P. aeruginosa PAO1, which had wild-type levels of AmpC and MexA-OprM expression, provided additional β-lactam resistance, i.e., resistance to moxalactam and meropenem (Table 1).

TABLE 1.

MICs of β-lactams for P. aeruginosa PAO1 and its isogenic mutant, P. aeruginosa KG2505 (AmpC⁻, MexA⁻, OmpF⁻), with and without recombinant plasmid pBB-1

β-Lactam	MIC (μg/ml)
	P. aeruginosa PAO1	P. aeruginosa PAO1(pBB-1)a	P. aeruginosa KG2505	P. aeruginosa KG2505(pBB-1)b
Ampicillin	>512	>512	0.25	1
Ticarcillin	8	8	0.25	0.5
Ticarcillin + CLAc	8	8	0.25	0.5
Piperacillin	2	2	0.12	0.12
Piperacillin + TZBd	4	4	0.12	0.25
Cephaloridine	>512	>512	4	4
Cephalothin	>512	>512	2	2
Cefsulodin	2	2	0.5	0.5
Cefuroxime	>512	>512	2	2
Ceftazidime	4	4	0.5	0.5
Ceftazidime + CLA	2	2	0.5	0.5
Cefotaxime	16	32	0.25	0.25
Cefepime	2	2	0.5	0.5
Cefpirome	2	2	1	1
Moxalactam	8	32	1	2
Aztreonam	4	4	0.12	0.12
Imipenem	1	1	0.5	0.5
Meropenem	0.5	0.5	<0.06	0.12

^aP. aeruginosa PAO1 harboring plasmid pBB-1, which encoded OXA-50.

^bP. aeruginosa KG2505 harboring plasmid pBB-1, which encoded OXA-50.

^cCLA, clavulanic acid (2 μg/ml).

^dTZB, tazobactam (4 μg/ml).

Plasmid pBB-1 was electroporated into E. coli DH10B. Antibiotic susceptibility testing revealed that E. coli DH10B(pBB-1) was susceptible to all the β-lactams tested, even though β-lactamase production could be demonstrated by nitrocefin hydrolysis (data not shown). The bla_OXA-50 gene of P. aeruginosa PAO1 was subsequently cloned into expression vector pET9a and expressed in E. coli BL21(DE3). Similarly to E. coli DH10B (pBB-1), E. coli BL21(DE3)(pET-1) was fully susceptible to all the β-lactams tested whether the bacterium was grown in the presence or in the absence of IPTG (data not shown). Culture extracts of E. coli BL21(DE3)(pET-1) contained strong β-lactamase activity, as assessed by nitrocefin hydrolysis activity (see Table 4).

TABLE 4.

Purification steps of β-lactamase OXA-50 produced by E. coli BL21(DE3) harboring recombinant plasmid pET-1

β-Lactamase source	Total amt of protein (mg)	Sp act (U/mg of protein)a	Total activity (U)	Yield (%)	Puri- fication (fold)
Crude extract	780	1.6	1,248	100	1
Purified	2.4	75.5	181	14.5	47

^aSpecific activity was determined with 100 μM nitrocefin as the substrate. One unit of activity is defined as the amount of enzyme that hydrolyzes 1 μmol of nitrocefin per min.

IEF analysis of P. aeruginosa strain KG2505 culture extracts revealed a single β-lactamase band with a pI value of 8.6 focusing at the same pI value as the chromosomally encoded cephalosporinase of parental strain P. aeruginosa PAO1 (data not shown). The identical pI values of the cephalosporinase and oxacillinase of P. aeruginosa might explain why this oxacillinase was unidentified for decades. IEF of IPTG-induced culture extracts of E. coli BL21(DE3) harboring recombinant plasmid pET-1 had β-lactamase activity with a pI value of 8.6, which corresponds to that of the cloned oxacillinase OXA-50; the chromosome-encoded cephalosporinase AmpC from E. coli had a pI value of 9.2.

Induction studies with cefoxitin or imipenem as the β-lactam inducers at various concentrations failed to demonstrate any induction of β-lactamase expression in P. aeruginosa KG2505 (data not shown).

Detailed structural analysis of the bla_OXA-50 gene and its surrounding DNA sequences.

The G+C content of the bla_OXA-50 gene open reading frame was 65%, which lies within the expected range of the G+C contents of P. aeruginosa genes, suggesting that the bla_OXA-50 gene is naturally occurring in that species. The deduced amino acid sequence of the β-lactamase OXA-50 contained a motif found in serine β-lactamases (11), with a serine-threonine-tyrosine-lysine (S-T-Y-K) motif replacing the most commonly found serine-threonine-phenylalanine-lysine (S-T-F-K) motif at positions 70 to 73 (class D β-lactamase [DBL] numbering) (4) (Fig. 1). The five structural elements characteristic of class D β-lactamases were found: S-X-V at DBL positions 119 to 120, W-X-X-X-X-L-X-I-X at DBL positions 164 to 172, Q-X-X-X-L at DBL positions 176 to 190, K-T-G at DBL positions 216 to 219, and Y-G-N at DBL positions 144 to 146 (4). The amino acid sequence of OXA-50 was compared with those of known oxacillinases (Fig. 1; Table 2). A K-T-G motif (DBL positions 216 to 219) was found in OXA-50, as in the carbapenem-hydrolyzing β-lactamases OXA-23 and OXA-27, whereas it was replaced by a K-S-G motif in the second group of carbapenem-hydrolyzing oxacillinases, made up of OXA-24, OXA-25, OXA-26, and OXA-40. The fifth structural element of oxacillinases, Y-G-N, was not replaced by an F-G-N motif in OXA-50, as in the most of the carbapenem-hydrolyzing oxacillinases (10). As illustrated in Fig. 1, the oxacillinase genes exhibit many different structural features, and little is known about the relevance of these differences for explaining the variabilities of the hydrolysis spectra (7). Indeed, OXA-50 is the only oxacillinase containing an S-T-Y-K motif that replaces the classical S-T-F-K motif of oxacillinases. Further investigations may demonstrate whether this structural feature plays a role in the biochemistry of OXA-50. OXA-50 was weakly related to other oxacillinases; the highest identity was with the carbapenem-hydrolyzing β-lactamases OXA-23 (44%) (6) and OXA-27 (43%) (1) from Acinetobacter baumannii, OXA-SHE (41%) from Shewanella algae, and β-lactamase all2480 (40%) (12) from a Nostoc cyanobacterial sp. A phylogenetic tree constructed with the known oxacillinases and based on amino acid sequence identity (2) showed that OXA-50 (formerly PA5514) could not be included in any of the five defined groups (30). However, OXA-50 might share a common ancestor with the group of enzymes that includes OXA-10 and OXA-5, referred to as “group I” by Sanschagrin et al. (30).

FIG. 1.

Comparison of the OXA-50 sequence (in boldface) with the sequences of other class D β-lactamases representative of the principal lineages of this group of β-lactamases. The previously proposed class D β-lactamase consensus sequence and numbering (DBL) (4) is indicated below the alignment (uppercase letters are the currently recognized invariant residues, and lowercase letters are the others). Boxed amino acid sequences are those conserved for oxacillinases. The vertical arrow indicates the cleavage site for the leader peptide of OXA-50. The enzymes included in the alignment are described in detail in Table 2.

TABLE 2.

Sources of oxacillinases for protein sequence comparison

β-Lactamase	Original host	GenBank no.a
OXA-1	E. coli K10-35	J02967
OXA-2	Salmonella enterica serovar Typhimurium 1a	M95287
OXA-3	Klebsiella pneumoniae	L07945
OXA-5	P. aeruginosa 7607	X58272
OXA-9	K. pneumoniae JHCK1	M55547
OXA-10	P. aeruginosa POW151	U37105
OXA-12	Aeromonas jandei AER14M	U10251
OXA-18	P. aeruginosa Mus	U85514
OXA-20	P. aeruginosa Mus	AF024602
OXA-22	R. pickettii PIC-1	AF064820
OXA-23	Acinetobacter baumannii 6B92	AJ132105
OXA-24	A. baumannii RYC 52763/97	AJ239129
OXA-27	A. baumannii I-16	AF201828
OXA-29	Legionnella gormanii ATCC 33297	AJ400619
OXA-40	A. baumannii CLA-1	AF509241
OXA-48	K. pneumoniae 11978	AY236073
OXA-50	P. aeruginosa PAO1	AE004964
OXA-SHE	S. algae N-511	AY066004
all2480	Nostoc sp. PCC 7120	AP003589
LCR-1	P. aeruginosa 229E	X56809

^aGenBank nucleotide sequence database accession number.

The nucleotide sequences of the flanking regions of the bla_OXA-50 gene did not show any inverted or repeated sequences indicating the presence of a transposable element. In addition, the bla_OXA-50 gene was not inserted into a classical integron, as indicated by the absence of 59-base elements linked to the β-lactamase gene. No open reading frame that shared significant sequence identity with known β-lactamase regulator genes was found immediately upstream of the bla_OXA-50 gene.

A bla_OXA-50-like gene was identified in all the P. aeruginosa isolates tested (Table 3). Comparison of the nucleotide sequences of these genes from unrelated P. aeruginosa isolates in different geographical locations showed 2 to 7 nucleotide changes compared to the nucleotide sequence of the bla_OXA-50 gene (32). Most of these nucleotide changes remained silent, and only a few led to amino acid changes (Table 3). The deduced amino acid sequences of these oxacillinases showed 98 to 100% identity (Table 3). The low nucleotide substitution rate of the bla_OXA-50-like genes corresponded to that found for ampC-like β-lactamase genes in that species (99.6% identity) (5, 14, 31).

TABLE 3.

Comparison of amino acid sequences of the chromosome-encoded oxacillinases of several P. aeruginosa isolates

P. aeruginosa isolate	Date of first isolation (mo-yr)	Reference or geographical origin	Amino acid at DBL positiona:
			19	28	87	113	178
PAO1		ATCC BAA-47	T	Q	K	D	R
GW-1	3-2000	South Africa	—	—	—	—	—
COL-1	3-1999	France	A	R	—	—	—
IND	10-2000	India	A	R	—	—	—
Ka. 209	Unknownb	Spain	A	—	R	—	—
ED-1	12-1998	France	—	—	—	E	H
Strain 1	8-1999	Thailand	—	—	—	E	H

^aAmino acid residues that differed from the amino acid sequence of β-lactamase OXA-50 from P. aeruginosa PAO1 are indicated, whereas identical amino acid residues are represented by a dash.

^bFrom 1996 to 2001.

Biochemical properties of OXA-50.

After purification, the specific activity of the β-lactamase OXA-50 against nitrocefin was 75.5 U/mg of protein, and its purification factor was 47-fold (Table 4). On SDS-PAGE, the purified protein appeared as single band of ca. 25 kDa and was estimated to be >95% pure (Fig. 2). N-terminal amino acid sequencing of the mature protein revealed that the cleavage site for the leader peptide is between amino acid positions 18 and 19 (A-SEWND). OXA-50 had a narrow-spectrum hydrolysis profile that included ampicillin, benzylpenicillin, cephaloridine, cephalothin, nitrocefin, piperacillin, and, surprisingly, imipenem (Table 5). OXA-50 did not significantly hydrolyze moxalactam or meropenem, although parental strain P. aeruginosa PAO1 harboring the multicopy plasmid pBB-1 had decreased susceptibilities to moxalactam and meropenem (Table 1).

FIG. 2.

SDS-PAGE analysis of purified enzymes from E. coli BL21(pET-1) (lane 1) and the culture extract (lane 2). A total of 25 μg of each sample was submitted to electrophoresis. Lane M, molecular mass markers, with sizes indicated on the left.

TABLE 5.

Kinetic parameters for purified β-lactamase OXA-50^a

Substrate	kcat (s−1)	Km (μM)	kcat/Km (mM−1 s−1)
Benzylpenicillin	110	800	140
Ampicillinb	>3c	>1,000	<3
Ticarcillin	>0.1	>1,000	<0.1
Piperacillinb	>2	>1,000	<2
Nitrocefin	100	200	500
Cephalothin	0.2	900	<1
Cephaloridine	>2	>1,000	<2
Moxalactam	>0.01	>1,000	<0.01
Cefsulodinb	>0.1	>1,000	<0.1
Imipenem	0.1	20	5
Meropenem	>0.02	>1,000	<0.02
Oxacillin	>0.2	>1,000	<0.2
Cloxacillin	—d	—	—

^aData are the means of three independent experiments. Standard deviations were within 10% of the means.

^bFor these substrates, as the time course for hydrolysis showed biphasic behavior, k_cat and K_m were determined at the steady-state rate.

^cSince the K_m values were very high, k_cat values could not be determined precisely. The k_cat value was determined for a K_m value of 1 mM (corresponding to a minimal value).

^d—, no hydrolysis could be detected, despite the use of 1 μM pure enzyme. Similar results were obtained with cefepime, cefotaxime, cefpirome, ceftazidime, cefuroxime, and aztreonam.

Hydrolysis of oxacillin or cloxacillin by OXA-50 was not detected, as is the case for β-lactamase OXA-24 (3). The k_cat/K_m values were low for all β-lactams except benzylpenicillin and nitrocefin; this was mostly due to the low affinity (high K_m) of this enzyme for these substrates. The best affinity was obtained for imipenem (K_m, 19 μM). The level of imipenem hydrolysis was very low. Compared to carbapenem-hydrolyzing oxacillinases, the catalytic activity (k_cat) for imipenem was twofold lower for OXA-50 than for OXA-40 and the affinity of OXA-50 for imipenem was similar to those of OXA-27 (1) and OXA-24 (3). Surprisingly, hydrolysis of meropenem was very weak, although the MICs were significantly increased for P. aeruginosa PAO1(pBB1) and P. aeruginosa KG2505(pBB1) (Table 1). Hydrolysis of meropenem was detected at a very low level, as is the case for OXA-25, OXA-26, and OXA-27; and the weak affinity of OXA-50 for this substrate did not allow determination of a k_cat value. OXA-50 was found to have similar weak affinities for other substrates, such as ampicillin, ticarcillin, piperacillin, cephaloridine, cefsulodin, moxalactam, and oxacillin (Table 5). Biphasic kinetics were seen for ampicillin, cefsulodin, and piperacillin. For these substrates, k_cat and K_m were determined for the steady-state part of the kinetics (the second part of the curve). As described for OXA-10 (9, 15, 21), CO₂ may strongly influence the kinetics of oxacillinases. In the case of OXA-50, NaHCO₃ transformed the hydrolysis to a linear curve only for cefsulodin, whereas biphasic turnover kinetics were still observed for ampicillin and piperacillin. Kinetic parameters were also determined for cefsulodin in the presence of 10 mM NaHCO₃. Since the K_m value was still very high for these substrates, the k_cat/K_m value could not be determined more precisely. The specific activities of all β-lactams with and without sodium bicarbonate were compared. Addition of NaHCO₃ did not significantly modify the catalytic efficiency of OXA-50 except for catalytic efficiencies for imipenem and meropenem, for which the specific activities were increased by 4.5- and 2.5-fold, respectively (data not shown). As for cefsulodin, k_cat and K_m values for imipenem were both increased after bicarbonate addition, resulting in a k_cat/K_m value that remained almost unchanged (data not shown).

Studies of activity inhibition, as measured by determination of IC₅₀s, showed that OXA-50 is weakly inhibited by clavulanic acid (500 μM), tazobactam (350 μM), and sulbactam (>2 mM), as is found for most of the oxacillinases (17). OXA-50 was inhibited by NaCl (IC₅₀, 50 mM), as is observed for most oxacillinases (17).

This study indicates that P. aeruginosa harbors two chromosomally and naturally encoded β-lactamase genes; the first one encodes an inducible cephalosporinase (class C) (14), whereas the second one encodes a constitutively expressed oxacillinase (class D). Both β-lactamase genes display minor strain-to-strain variations (5, 31), suggesting their possible use in combination for PCR-based P. aeruginosa identification. The oxacillinase may not contribute significantly to the overall β-lactam susceptibility pattern of P. aeruginosa except for that to moxalactam. Furthermore, as opposed to most oxacillinase genes, the bla_OXA-50-like genes were not part of class 1 integrons, and unlike the chromosomally located and naturally occurring oxacillinase genes from Aeromonas spp. (17, 28) and Ralstonia pickettii (19), bla_OXA-50 gene expression was expressed constitutively. Finally, this work underlines the fact that an important reservoir of oxacillinase genes is likely gram-negative environmental species such as R. pickettii (19), Aeromonas spp. (28), Shewanella spp. (25), and now P. aeruginosa.