The OXA-1 β-lactamase is one of the few class D enzymes that has an aspartate residue at position 66, a position that is proximal to the active-site residue Ser67. In class A β-lactamases, such as TEM-1 and SHV-1, residues adjacent to the active-site serine residue play a crucial role in inhibitor resistance and substrate selectivity. To probe the role of Asp66 in substrate affinity and catalysis, we performed site-saturation mutagenesis at this position. Ampicillin MIC (minimum inhibitory concentration) values for the full set of Asp66 mutants expressed in Escherichia coli DH10B ranged from ≤8 μg/ml for cysteine, proline and the basic amino acids to ≥256 μg/ml for asparagine, leucine and the wild-type aspartate. Replacement of aspartic acid by asparagine at position 66 also led to a moderate enhancement of extended-spectrum cephalosporin resistance. OXA-1 shares with other class D enzymes a carboxylated residue, Lys70, that acts as a general base in the catalytic mechanism. The addition of 25 mM bicarbonate to Luria–Bertani-broth agar resulted in a ≥16-fold increase in MICs for most OXA-1 variants with amino acid replacements at position 66 when expressed in E. coli. Because Asp66 forms hydrogen bonds with several other residues in the OXA-1 active site, we propose that this residue plays a role in stabilizing the CO2 bound to Lys70 and thereby profoundly affects substrate turnover.

INTRODUCTION

β-Lactam antibiotics kill bacteria by inhibiting the D-alanyl-D-alanine transpeptidase responsible for cross-linking cell-wall components [1]. Regrettably, the efficacy of these agents has been diminished by the action of β-lactamase enzymes [EC 3.5.2.6], which hydrolyse the β-lactam bond of penicillins and cephalosporins, thereby rendering them ineffective [2]. There are several hundred β-lactamases, which have been grouped into four main classes (A–D) [3]. Classes A, C, and D of the Ambler classification system use covalent catalysis, whereas class B enzymes contain a metal ion (usually Zn2+) to help polarize the target carbonyl group of the β-lactam [4].

OXA family β-lactamases, so named because of their preference for oxacillin as a substrate, are members of class D. Recognition of the clinical relevance of class D lactamases, the least studied of the four classes, has increased for a number of reasons. First, most OXA family members are resistant to inactivation by mechanism-based inhibitors such as clavulanate, tazobactam and sulbactam [3]. Secondly, although many enzymes have a narrow substrate spectrum, there has been a recent emergence of OXA β-lactamases capable of hydrolysing oxyiminocephalosporins and carbapenems [58]. Lastly, the presence of blaOXA genes on mobile genetic elements has allowed these enzymes to disseminate widely [9]. Variants of the OXA family, now numbering more than 100 members, have been described in (among others) Shigella flexneri [10], Klebsiella pneumoniae [11] and, particularly, Acinetobacter baumanii [12].

Atomic structural analysis of class D β-lactamases in both the apo and the acylated forms has enhanced our understanding of their catalytic mechanisms [1318]. Despite a generally weak sequence identity among the OXA family members and even weaker homologies with class A and C enzymes, all three classes share the same topological fold [19]. Moreover, all three classes hydrolyse substrates using a similar serine-nucleophile-mediated acylation/deacylation mechanism. One feature unique to the OXA family is a catalytically important carboxylated lysine residue in the active site [1315]. The carboxylated lysine residue, which forms from the reaction of Lys70 with CO2, is stabilized by a network of hydrogen bonds [13,15]. It has been proposed that this ionized carbamyl group serves as the general base in both the acylation step (deprotonating Ser67) and the deacylation step (deprotonating the catalytic water molecule) [1315]. This argument is supported by mutagenesis studies of Lys70 in OXA-10; alteration of this residue to alanine completely eliminates catalysis by this enzyme [15]. Moreover, there is an absence in all class D enzymes of any counterpart to Glu166, the general base in class A enzymes. Because of the reversible nature of the carboxylation, the binding of CO2 to Lys70 can be affected by pH and bicarbonate/CO2 concentration [4,15].

The catalytic serine residue at position 67 and carboxylated lysine residue at position 70 are part of a conserved patch with the sequence (using the one-letter code) STFK. The sequences YGN (residues 144–146) and KTG (residues 212–214) are also nearly invariant in class D enzymes. Ser115, which is homologous with the class A residue Ser130, is completely conserved in all OXA members. Sun et al. [13] suggested that this residue is involved in proton transfer to substrate during catalysis.

The focus of the present study, OXA-1, has several notable structural differences compared with other class D enzymes. Unlike OXA-10, OXA-1 is a monomer [13] and possesses an omega loop insertion of six amino acids (residues 149–154) not found in OXA-10 and OXA-13 [13]. It is one of only a few class D enzymes (along with minor variants OXA-4, OXA-31 and OXA-47) known to possess an aspartate residue at position 66; most OXA enzymes have an alanine or a methionine residue at this site. Asp66 helps form the floor of the active site in OXA-1 and is partially buried under the B9 β-strand (Figure 1). Its location and position are very similar to that seen for the homologous residue (Met69) found in the class A enzymes TEM-1 and SHV-1 [20,21]. Interestingly, the main-chain atoms of the residue at position 66 adopt highly strained Φ,Ψ torsion angles in OXA-1 (Asp66), OXA-10 (Ala66), OXA-13 (Ala66), SHV-1 (Met69) and TEM-1 (Met69). Met69, along with other residues near it, plays a critical role in substrate specificity and resistance to mechanism-based inhibitors in class A β-lactamases [22,23]. In SHV-1 for instance, it has been shown that phenylalanine, tyrosine and lysine at position 69 lead to ceftazidime resistance, while replacement with isoleucine, valine or leucine results in inhibitor resistance [23].

Interactions between Asp66 and the omega loop of OXA-1 (1M6K)

Figure 1
Interactions between Asp66 and the omega loop of OXA-1 (1M6K)

The crystal structure co-ordinates available at the RCSB (Research Collaboratory for Structural Bioinformatics) Protein Data Bank were used to represent the interactions of key residues in OXA-1. The Figure was constructed using Pymol.

Figure 1
Interactions between Asp66 and the omega loop of OXA-1 (1M6K)

The crystal structure co-ordinates available at the RCSB (Research Collaboratory for Structural Bioinformatics) Protein Data Bank were used to represent the interactions of key residues in OXA-1. The Figure was constructed using Pymol.

The aspartate residue at position 66 in OXA-1 is thought to share a hydrogen atom with Glu162 in the omega loop, possibly helping to stabilize this structure (see Figure 1) [13]. On the basis of this observation, we hypothesized that Asp66 serves to orient Ser67 and carboxy-Lys70 for proper acylation and deacylation. Thus, we performed site-saturation mutagenesis at this position to explore the unique nature of this aspartate at position 66 in the OXA family and the potential role that it may play in substrate specificity and catalysis. Our data show that this aspartate residue plays a major role in β-lactam catalysis in OXA-1.

MATERIALS AND METHODS

Materials

Plasmids pOXA-1 [used for mutagenesis and MIC (minimum inhibitory concentration) determination] and pETKMOXA-1 (used for native OXA-1 β-lactamase expression) were generously given by Professor James Knox and Dr Michiyoshi Nukaga (Department of Molecular and Cell Biology, University of Connecticut, Storrs, CT, U.S.A.). pET-28a(+) was purchased from EMD Biosciences (San Diego, CA, U.S.A.). Antibiotics were obtained from the following companies: ampicillin, cephalothin and cefotaxime (Sigma, St. Louis, MO, U.S.A.); cefepime (Bristol Meyers–Squibb, Princeton, NJ, U.S.A.); and ceftazidime (Abbot Laboratories, North Chicago, IL, U.S.A.). CM-cellulose (carboxymethylcellulose) and iminodiacetic acid Sepharose 6B were obtained from Sigma. CENTA {(6R,7S)-3-[(3-carboxy-4-nitrophenyl)sulfanylmethyl]-8-oxo-7-[(2-thiophen-2-ylacetyl)amino]-5-thia-1-azabicyclo[4.2.0]oct-2-ene-2-carboxylic acid}, a chromogenic cephalosporin, was synthesized using a published method [24].

Site-saturation mutagenesis

The complete set of amino acid substitutions at position 66 was prepared in the blaOXA-1 gene background of plasmid pOXA-1. Degenerate oligonucleotides (see Table 1) were used with the QuikChange® II Mutagenesis Kit (Stratagene, La Jolla, CA, U.S.A.). All mutant plasmids were transformed into E. coli strain DH10B. Mutant plasmids were sequenced with an ALF Express automated DNA sequencer (Amersham Pharmacia Biotech, Piscataway, NJ, U.S.A.) by using the Thermo Sequenase fluorescence-labelled primer cycle sequencing kit [25].

Table 1
Degenerate oligonucleotides used in the present study

The first two oligonucleotide primers were used in site-saturation mutagenesis at position 66. The third and fourth primers were used to subclone mutant blaOXA-1 genes into the NdeI and BamHI sites (underlined) of pET-28a(+).

Oligonucleotide Sequence 
D66X forward 5′-GCAACGCAAATGGCACCANNNTCAACTTTCAAGATCGC-3′ 
D66X reverse 5′-GCGATCTTGAAAGTTGANNNTGGTGCCATTTGCGTTGC-3′ 
OXA-1 forward 5′-CGTGCATACCATATGTCAACAGATATCTCTACTG-3′ 
OXA-1 reverse 5′-AATTAACCGGATCCTTATAAATTTAGTGTGTTTAG-3′ 
Oligonucleotide Sequence 
D66X forward 5′-GCAACGCAAATGGCACCANNNTCAACTTTCAAGATCGC-3′ 
D66X reverse 5′-GCGATCTTGAAAGTTGANNNTGGTGCCATTTGCGTTGC-3′ 
OXA-1 forward 5′-CGTGCATACCATATGTCAACAGATATCTCTACTG-3′ 
OXA-1 reverse 5′-AATTAACCGGATCCTTATAAATTTAGTGTGTTTAG-3′ 

Western-blot analysis

Steady-state expression levels for WT (wild-type) and variant OXA-1 β-lactamases in E. coli DH10B were measured as previously described [25,26], with the exception of the detecting antibody. The polyclonal anti-OXA-1 antibody was raised in rabbits, purified on Protein G columns, and characterized as reported for anti-SHV-1 and anti-CMY-2 antibodies [27]. The Western blot was probed with 1 μg/ml anti-OXA-1 antibody and then with horseradish-peroxidase-linked Protein G.

Antibiotic susceptibility

MICs for various antibiotics were determined using a Steers Replicator that delivered 10 μl containing 104 CFU (colony-forming units)/spot, with Luria–Bertani broth agar prepared at pH 7.2. For those plates requiring bicarbonate supplementation, stocks of 500 mM NaHCO3 adjusted to pH 7.4 with NaH2PO4 were added to the agar to give a final concentration of 25 mM. The MIC values listed represent the median of results obtained with a minimum of three repetitions.

β-Lactamase purification

WT OXA-1 β-lactamase was purified after overexpression from the vector pETKMOXA-1 in E. coli BL21(DE3) cells [28]. Cells were thawed in 20 mM NaH2PO4/1 mM EDTA, pH 7.0, and lysed with a final concentration of 1.0 mg/ml lysozyme (Sigma). The lysate was supplemented with 2 mM MgCl2 and adjusted to 2.5 μg/ml DNAse I to eliminate chromosomal DNA. After centrifugation (6750 g for 20 min 4 °C; Sorvall SS-34 rotor), the supernatant was dialysed against 4 litres of 5 mM NaH2PO4, pH 5.8. The dialysis residue was applied to a CM-23 CM-cellulose (Sigma, St. Louis, MO, U.S.A.) column (1.5 cm×3 cm) equilibrated with 5 mM NaH2PO4, pH 5.8. The protein was eluted with a linear gradient of 5 mM NaH2PO4, pH 5.8 to 50 mM NaH2PO4, pH 7.0. Column fractions containing OXA-1 β-lactamase were combined and dialysed against 50 mM NaH2PO4/25 mM NaHCO3, pH 7.4. The dialysis residue was then concentrated using centrifugal filtration (Centricon; 10 kDa molecular-mass cut-off). OXA-1 purified by this technique was >95% pure (as assessed by SDS/PAGE analysis) and was stored at 4 °C.

Mutant blaOXA-1 genes were amplified from a pOXA-1 background using Pfu polymerase (Promega, Madison, WI, U.S.A.) and subcloned into NdeI and BamHI sites in pET-28a(+). This step created a short fusion peptide containing a hexahistidine tag at the N-terminus, thereby allowing quick and facile purification of all enzymes. The primers used for this amplification are shown in Table 1. blaOXA-1 variant genes that had been subcloned into pET-28a(+) were expressed and purified using nickel affinity chromatography. Cells were grown to an attenuance (D600) of about 0.80, and protein production was induced with 100 μM IPTG (isopropyl β-D-1-thiogalactopyranoside) for 2 h. The cells were harvested by centrifugation (9500 g for 15 min) and frozen at −20 °C. Lysates were prepared as described above and applied to an iminodiacetic acid–Sepharose 6B column (1.5 cm× 3 cm) charged with nickel. The protein was eluted with a linear gradient of 5–200 mM imidazole in 20 mM Tris/HCl/500 mM NaCl, pH 7.5. Pure fractions (>95% as assessed by SDS/PAGE) were dialysed against 50 mM NaH2PO4/25 mM NaHCO3, pH 7.4, and stored at 4 °C.

Bicarbonate titration

To decarboxylate Lys70, purified WT and select variant β-lactamases were dialysed against 50 mM sodium acetate, pH 4.5, under vacuum at 4 °C [15]. Samples of each enzyme were diluted into 50 mM NaH2PO4 buffer, pH 7.4, containing various concentrations (0–100 mM) of NaHCO3, pH 7.4. Buffers devoid of bicarbonate were degassed under vacuum prior to use. After an incubation of at least 30 min at room temperature (20 °C), aliquots of diluted enzyme were added to 500 μM ampicillin in the same buffer used for equilibration to determine the rate of hydrolysis. Bicarbonate concentrations were converted into [CO2] values using the Henderson–Hasselbach equation and a pKa adjusted for ionic strength [29]. Each assay was carried out in triplicate and the results averaged. Both data sets were fitted to the following single-binding-site model, and then normalized to Vmax:

 
formula
(1)

Kinetic assays

Michaelis–Menten kinetic analysis was carried out in 50 mM NaH2PO4/25 mM NaHCO3, pH 7.4, at room temperature in a Beckman DU-800 spectrophotometer. Measurements were carried out using the following Δϵ [molar absorption coefficient (M−1·cm−1)] values: ampicillin, −900 (λ=235 nm); cephalothin, −6500 (λ=260 nm); cefepime, −10000 (λ=260 nm); cefotaxime, −7500 (λ=260 nm); CENTA, 6400 (λ=405 nm). The average of three measurements of initial velocity was plotted as a function of substrate concentration, and the Km and kcat were determined by non-linear regression to the Michaelis–Menten equation.

RESULTS

Mutagenesis

The pOXA-1 plasmid was derived from the chloramphenicol-resistant pHSG398 vector and contains the full-length coding region for OXA-1 attached to the native promoter from plasmid RGN238 [30,31]. Site-saturation mutagenesis was used to successfully create the complete set of amino acid residues at position 66, all of which were transformed into E. coli strain DH10B.

Expression profile of Asp66 variants

Before determining the consequences of site-saturation mutagenesis at position 66, we determined the level of variant OXA-1 expression by Western blotting. The antibody that was used was prepared against native OXA-1 and was shown to bind to epitopes derived from amino acid residues 18–30, 39–48, 143–152 and 167–176 (numbering includes amino terminal leader sequence). Therefore determination of expression levels was not affected by mutations at position 66. Figure 2 shows that nearly all substitutions at position 66 result in only minor changes in steady-state protein levels. The only exception to this is D66C; the expression of this variant was diminished relative to that of the other proteins.

Western blot showing the relative expression levels of all replacements for Asp66

Figure 2
Western blot showing the relative expression levels of all replacements for Asp66

Replacements at position 66 are noted using the single-letter amino acid code. The D66W and D66Y proteins were repeated along with DH10B cells containing pBCSK(−) vector alone to demonstrate antibody specificity.

Figure 2
Western blot showing the relative expression levels of all replacements for Asp66

Replacements at position 66 are noted using the single-letter amino acid code. The D66W and D66Y proteins were repeated along with DH10B cells containing pBCSK(−) vector alone to demonstrate antibody specificity.

MICs

Antibiotic-susceptibility tests showed that almost all Asp66 mutants in E. coli DH10B were more susceptible to ampicillin; we observed that more than half of the 19 variants were susceptible at ≤32 μg/ml compared with 2048 μg/ml for the WT (see Table 2).

Table 2
MIC values for Asp66 mutants for several antibiotics in the absence (−) or presence (+) of 25 mM bicarbonate

Mutants are listed using the one-letter amino acid code.

  MIC value (μg/ml) 
 Antibiotic … Ampicillin Cephalothin Cefotaxime Cefepime Ceftazidime 
Mutant 25 mM HCO3 … − Ratio +/− − − − − 
 32 1024 32 16 0.06 0.06 0.06 0.12 0.5 
 32 <0.03 <0.03 <0.03 <0.03 0.25 0.25 
D (WT)  2048 8192 16 0.5 0.5 0.25 0.5 
 64 512 0.06 0.06 <0.03 <0.03 0.25 0.5 
 32 2048 64 32 0.06 0.06 <0.03 <0.03 0.5 
 32 1024 32 0.06 0.06 <0.03 <0.03 0.25 0.5 
 256 64 16 0.06 <0.03 <0.03 <0.03 0.25 0.25 
 256 32 16 0.06 0.06 <0.03 <0.03 0.25 0.5 
 32 16 0.06 0.06 <0.03 <0.03 0.25 0.5 
 512 8192 16 16 0.12 0.5 0.25 0.5 
 128 4096 32 16 0.06 0.06 0.06 <0.03 0.5 
 2048 16384 16 0.5 0.5 0.5 
 16 16 0.06 0.06 <0.03 <0.03 0.25 0.5 
 128 4096 32 16 0.06 0.25 <0.03 0.5 0.25 0.5 
 32 16 0.06 0.06 <0.03 <0.03 0.25 0.5 
 128 2048 16 16 0.06 0.12 0.06 0.25 0.25 0.5 
 32 1024 32 16 0.06 0.06 <0.03 <0.03 0.25 0.5 
 32 1024 32 16 0.06 0.12 <0.03 0.12 0.25 0.5 
 16 2048 128 16 0.06 0.06 <0.03 <0.03 0.25 0.5 
 256 8192 32 16 0.06 0.06 0.12 0.12 0.5 
DH10B  <1  <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 
  MIC value (μg/ml) 
 Antibiotic … Ampicillin Cephalothin Cefotaxime Cefepime Ceftazidime 
Mutant 25 mM HCO3 … − Ratio +/− − − − − 
 32 1024 32 16 0.06 0.06 0.06 0.12 0.5 
 32 <0.03 <0.03 <0.03 <0.03 0.25 0.25 
D (WT)  2048 8192 16 0.5 0.5 0.25 0.5 
 64 512 0.06 0.06 <0.03 <0.03 0.25 0.5 
 32 2048 64 32 0.06 0.06 <0.03 <0.03 0.5 
 32 1024 32 0.06 0.06 <0.03 <0.03 0.25 0.5 
 256 64 16 0.06 <0.03 <0.03 <0.03 0.25 0.25 
 256 32 16 0.06 0.06 <0.03 <0.03 0.25 0.5 
 32 16 0.06 0.06 <0.03 <0.03 0.25 0.5 
 512 8192 16 16 0.12 0.5 0.25 0.5 
 128 4096 32 16 0.06 0.06 0.06 <0.03 0.5 
 2048 16384 16 0.5 0.5 0.5 
 16 16 0.06 0.06 <0.03 <0.03 0.25 0.5 
 128 4096 32 16 0.06 0.25 <0.03 0.5 0.25 0.5 
 32 16 0.06 0.06 <0.03 <0.03 0.25 0.5 
 128 2048 16 16 0.06 0.12 0.06 0.25 0.25 0.5 
 32 1024 32 16 0.06 0.06 <0.03 <0.03 0.25 0.5 
 32 1024 32 16 0.06 0.12 <0.03 0.12 0.25 0.5 
 16 2048 128 16 0.06 0.06 <0.03 <0.03 0.25 0.5 
 256 8192 32 16 0.06 0.06 0.12 0.12 0.5 
DH10B  <1  <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 

The OXA-1 β-lactamase possesses a catalytically important and reversibly bound CO2 on the active-site lysine residue, thus we next tested whether supplementation of bicarbonate would affect ampicillin resistance. The addition of 25 mM bicarbonate, pH 7.4, to agar plates increased the MIC values of all Asp66 variants expressed in E. coli DH10B (Table 2). The MICs of WT, D66C and D66P were enhanced 4-fold (2048 to 8192, 8 to 32 and 4 to 16 μg/ml respectively), whereas the MICs of variants with large hydrophobic residues, such as tryptophan, phenylalanine, tyrosine and methionine, were increased between five and seven dilutions (32–128-fold). The ratio of the MIC value in the presence and absence of bicarbonate is listed in a separate column in Table 2.

MIC values were also determined against four cephalosporin β-lactam antibiotics in the absence or presence of 25 mM bicarbonate. WT OXA-1 β-lactamase expressed in E. coli provided little resistance to cephalosporins, reflecting the penicillinase activity of this class D enzyme (Table 2). Examining the results of substitutions at Asp66 in the presence of added bicarbonate, a gain in resistance against cephalothin was observed for all variants (4–16 μg/ml). In addition, an increase in cefepime and cefotaxime resistance was noted for the substitution D66N (0.5 to 2 μg/ml and 1 to 4 μg/ml respectively).

In the absence of added bicarbonate, the D66L and D66M mutants expressed in E. coli DH10B showed a rise in resistance against ceftazidime compared with the WT. Likewise, the D66A mutant showed an 8-fold increase in MIC for ceftazidime compared with the WT (0.25 to 2 μg/ml).

Kinetics

The kinetic behaviour of WT OXA-1 (purified by CM-cellulose chromatography) and His6 (hexahistidine)-tagged WT OXA-1 (purified by nickel-affinity chromatography) were compared. The results in Table 3 show that the Km and kcat for WT OXA-1. and His6-tagged OXA-1 are comparable for cefotaxime and ampicillin. Thus the fusion peptide minimally affects substrate binding and catalytic activity.

Table 3
Kinetic properties of native and His6-tagged OXA-1
  Value 
Antibiotic Kinetic parameter WT OXA-1 His6-tagged WT OXA-1 
Cefotaxime Km (μM) 29±2.2 36±2.0 
 kcat (s−16.7±0.2 5.3±0.1 
Ampicillin Km (μM) 14±2.3 21±1.0 
 kcat (s−1440±15 520±6.0 
  Value 
Antibiotic Kinetic parameter WT OXA-1 His6-tagged WT OXA-1 
Cefotaxime Km (μM) 29±2.2 36±2.0 
 kcat (s−16.7±0.2 5.3±0.1 
Ampicillin Km (μM) 14±2.3 21±1.0 
 kcat (s−1440±15 520±6.0 

Steady-state kinetic analyses were next carried out on His6-tagged WT and five variant enzymes for several different substrates (Table 4). In all cases, assays were performed in buffer supplemented with 25 mM sodium bicarbonate, to ensure full carboxylation of Lys70. Comparison of Km values for WT and D66N showed little or no difference for cefotaxime (WT, 36±2 μM; D66N, 24±1 μM), ampicillin (WT, 21±1 μM; D66N, 23±2 μM), cephalothin (WT, 9±0.7 μM; D66N, 11±1 μM) or cefepime (WT, 170±14; D66N, 150±13 μM). Owing to reduced kcat values, catalytic efficiencies were slightly lower.

Table 4
Steady-state kinetics of His6-tagged WT OXA-1 and several His6-tagged Asp66 variants
  Value 
Mutant Kinetic parameter Ampicillin Cefotaxime Cefepime Cephalothin CENTA 
WT Km (μM) 21±1 36±2 170±14 9±0.7 2.4±0.1 
 kcat (s−1520±6 5.3±0.1 34±1 2.7±0.1 1.6±0.1 
 kcat/Km (μM−1·s−125±1 0.15±0.01 0.20±0.02 0.30±0.03 0.7±0.05 
D66N Km (μM) 23±2 24±1 150±13 11±1 2.0±0.2 
 kcat (s−1480±9 3.5±0.1 27±1 1.6±0.1 1.2±0.1 
 kcat/Km (μM−1·s−121±2 0.15±0.01 0.18±0.02 0.15±0.02 0.6±0.08 
D66Y Km (μM) 2.7±0.3 57±4 230±36 37±2 7±0.7 
 kcat (s−110.6±0.1 0.71±0.02 3.8±0.4 0.28±0.01 0.25±0.01 
 kcat/Km (μM−1s−13.9±0.4 0.012±0.001 0.017±0.003 0.008±0.001 0.034±0.003 
D66W Km (μM) 21±3 ND 210±14 59±3 6±0.6 
 kcat (s−156±1 ND 0.008±0.003 0.15±0.01 0.080±0.001 
 kcat/Km (μM−1s−12.7±0.4 ND 4 ×10−5±1×10−5 0.003±0.0002 0.014±0.001 
D66A Km (μM) 6±1 70±8 27±2 100±13 13±2 
 kcat (s−113±1 0.37±0.02 0.55±0.01 0.27±0.01 0.19±0.01 
 kcat/Km (μM−1s−12.1±0.4 0.005±0.001 0.020±0.002 0.003±0.0003 0.015±0.002 
D66M Km (μM) 12±1 36±1 35±1 10±0.6 3.2±0.3 
 kcat (s−143±1 0.31±0.01 0.43±0.01 0.38±0.01 0.17±0.03 
 kcat/Km (μM−1s−13.5±0.3 0.009±0.0004 0.012±0.001 0.040±0.003 0.05±0.011 
  Value 
Mutant Kinetic parameter Ampicillin Cefotaxime Cefepime Cephalothin CENTA 
WT Km (μM) 21±1 36±2 170±14 9±0.7 2.4±0.1 
 kcat (s−1520±6 5.3±0.1 34±1 2.7±0.1 1.6±0.1 
 kcat/Km (μM−1·s−125±1 0.15±0.01 0.20±0.02 0.30±0.03 0.7±0.05 
D66N Km (μM) 23±2 24±1 150±13 11±1 2.0±0.2 
 kcat (s−1480±9 3.5±0.1 27±1 1.6±0.1 1.2±0.1 
 kcat/Km (μM−1·s−121±2 0.15±0.01 0.18±0.02 0.15±0.02 0.6±0.08 
D66Y Km (μM) 2.7±0.3 57±4 230±36 37±2 7±0.7 
 kcat (s−110.6±0.1 0.71±0.02 3.8±0.4 0.28±0.01 0.25±0.01 
 kcat/Km (μM−1s−13.9±0.4 0.012±0.001 0.017±0.003 0.008±0.001 0.034±0.003 
D66W Km (μM) 21±3 ND 210±14 59±3 6±0.6 
 kcat (s−156±1 ND 0.008±0.003 0.15±0.01 0.080±0.001 
 kcat/Km (μM−1s−12.7±0.4 ND 4 ×10−5±1×10−5 0.003±0.0002 0.014±0.001 
D66A Km (μM) 6±1 70±8 27±2 100±13 13±2 
 kcat (s−113±1 0.37±0.02 0.55±0.01 0.27±0.01 0.19±0.01 
 kcat/Km (μM−1s−12.1±0.4 0.005±0.001 0.020±0.002 0.003±0.0003 0.015±0.002 
D66M Km (μM) 12±1 36±1 35±1 10±0.6 3.2±0.3 
 kcat (s−143±1 0.31±0.01 0.43±0.01 0.38±0.01 0.17±0.03 
 kcat/Km (μM−1s−13.5±0.3 0.009±0.0004 0.012±0.001 0.040±0.003 0.05±0.011 

The other four mutations caused much more extensive alterations of kinetic properties. D66Y, D66W, D66A and D66M generally showed significant decreases in turnover rate, with kcat reduced 8-fold or more for all substrates. Interestingly, for ampicillin, D66Y demonstrated the largest decrease in kcat (WT, 520 s−1; D66Y, 10.6 s−1), whereas for cefepime it was the D66W mutant that had the largest decrease (WT, 34 s−1; D66W, 0.008 s−1). Moreover, D66W exhibited a very low kcat for the two advanced-generation cephalosporins used (>1000-fold decrease), but only an ∼20-fold decrease for cephalothin. A clear pattern of Asp66 substitution with regard to Km values was not seen; there was a large increase in some cases (e.g. D66A with cephalothin) and a large decrease in others (e.g. D66Y with ampicillin).

On the basis of the bicarbonate-sensitivity of many of the Asp66 mutants, we decided to determine the effect of bicarbonate concentration on OXA-1 catalysis. To do this, we chose to compare His6-tagged WT OXA-1 and the D66A, D66M and D66W mutants. We were particularly eager to test the D66W mutant because, when comparing ampicillin MIC values in the absence or presence of 25 mM bicarbonate, this variant showed the largest bicarbonate-dependant increase (from 16 to 2048 μg/ml). Using the kinetic analysis of OXA-10 as a model [15], the pure β-lactamases were stripped of their carboxylation on Lys70 through dialysis at pH 4.5 under vacuum. For all variants tested, this treatment caused an ∼80–90% reduction in ampicillin hydrolytic activity compared to that seen before treatment. Figure 3 shows that, for all β-lactamases tested, there was a hyperbolic increase in activity as bicarbonate (and thus CO2) concentration increases. It is apparent, however, that it takes a much higher concentration of bicarbonate to bring the variant β-lactamases (particularly D66W) to the same level as WT enzyme. The data were fitted to a single-site binding model, which showed that the apparent affinity of CO2 is >100-fold lower for the D66W mutant compared with the WT (Kd=320±28 μM for D66W; 2±0.6 μM for WT). The other two mutants showed intermediate CO2 affinities (Kd=31±3 μM for D66M; 80±13 μM for D66A).

Hydrolytic activity of WT OXA-1 and three variants as a function of CO2 concentration

Figure 3
Hydrolytic activity of WT OXA-1 and three variants as a function of CO2 concentration

Aliquots of each protein were dialysed at pH 4.5 to remove CO2 from Lys70, and then equilibrated in 50 mM NaH2PO4, pH 7.4, supplemented with various amounts of bicarbonate. Samples were then removed and added to 500 μM ampicillin in the same buffer, and hydrolysis was monitored at 235 nm. Bicarbonate concentrations were converted into [CO2] using the Henderson–Hasselbach equation.

Figure 3
Hydrolytic activity of WT OXA-1 and three variants as a function of CO2 concentration

Aliquots of each protein were dialysed at pH 4.5 to remove CO2 from Lys70, and then equilibrated in 50 mM NaH2PO4, pH 7.4, supplemented with various amounts of bicarbonate. Samples were then removed and added to 500 μM ampicillin in the same buffer, and hydrolysis was monitored at 235 nm. Bicarbonate concentrations were converted into [CO2] using the Henderson–Hasselbach equation.

DISCUSSION

Similarly to observations on the TEM-1 and SHV-1 enzymes, our data clearly demonstrate that replacements of the homologous residue in OXA-1, namely Asp66, also profoundly influence catalysis. Replacement of Asp66 with most other amino acids causes a significant reduction in ampicillin resistance. The most notable exception to this is asparagine, which closely resembles the WT residue's size, shape and hydrogen-bonding capability. Variant OXA-1 β-lactamases which possess a positive charge at position 66 (arginine, histidine and lysine) or the structurally confined proline showed the greatest decrease in ampicillin resistance. Major changes in β-lactamase expression were not seen, with the exception of D66C, which is produced at a lower level. Moreover, preparations of WT and five variant enzymes with widely varying MIC values and kinetic properties did not differ in solubility or yield during purification and concentration. We therefore conclude that the changes in resistance observed by MIC analysis are based predominantly on alteration of kinetic properties.

An unexpected finding of our studies was that the addition of bicarbonate to MIC assays can partially restore some of the loss of catalytic activity for a number of the Asp66 mutants. The results shown in Figure 3 indicate that three different replacements of Asp66 (methionine, alanine and tryptophan) result in lower affinity between CO2 and Lys70. Interestingly, the Kd values showed a correlation with the magnitude of the MIC increase observed for those three mutations upon bicarbonate supplementation. The loss of the carbamate removes the general base that is thought to be critical for the activation of the nucleophile in both the acylation and deacylation reactions. The carboxylysine residue is stabilized by a number of hydrogen bonds, most notably from side-chains of Trp160 and Ser67 (Figure 1). Although Asp66 is more than 8 Å (0.8 nm) away from the carbamate moiety, there are many indirect effects by which its substitution might alter interactions, and thereby decrease CO2 affinity. First, the aspartate residue is thought to hydrogen-bond to both the side-chain and the main-chain amide hydrogen of Glu162. As noted in the OXA-1 crystal structure, this linkage pulls the omega loop closer to the B9 strand compared with the interaction seen in the OXA-2 or OXA-10 structures [13]. When Asp66 is replaced, the lack of these hydrogen bonds may destabilize the omega loop and disrupt the stabilizing interaction between Trp160 and the carbamate. Secondly, any perturbation of Asp66 might affect the ability of its neighbor, Ser67, to form the optimal hydrogen bond with the general base, as has been suggested for the homologous residue, Met69, of TEM-1 [22]. It is known that the residue on the N-terminal side of the catalytic serine residue adopts highly strained torsion angles in both class A and class D enzymes, and therefore may be especially sensitive to structural perturbation. Many examples of β-lactamase residues affecting catalysis indirectly over large distances have been noted previously [32,33], and it is likely that other mutations altering stability or positioning of the omega loop will similarly affect carbamate formation and thereby substrate turnover in OXA-1.

As noted above, OXA-1 is one of only a handful of class D enzymes to have an aspartate residue at position 66. Almost all other OXA enzymes have an alanine residue at this position, although there are two enzymes that have a methionine residue (OXA-18 and -45) and one each of a cysteine residue (OXA-9) and a glutamate residue (OXA-29). Of those with alanine at position 66, the site equivalent to Glu162 in OXA-1 is typically occupied by glutamic acid, asparagine or valine. The crystal structures of OXA-10 and -24, however, show that, in these β-lactamases, the omega loop has a different conformation that results in the side-chains of these residues pointing out towards the solvent [16,34]. It is therefore likely that position 162 plays a unique role in OXA-1 β-lactamase.

It is very notable that several substitutions at position 66 in OXA-1 β-lactamase resulted in modest increases in resistance to extended-spectrum cephalosporins. These observations are very reminiscent of previous analyses in SHV-1; changes in SHV-1 at position 69 (M69F, M69Y and M69K) resulted in an increase in resistance against ceftazidime [23]. In OXA-1, an asparagine replacement at position 66 resulted in a 4-fold increase for the MICs of both cefotaxime and cefepime, while having no affect on ampicillin resistance. The D66N replacement in OXA-1 may cause a slight increase in affinity for these two cephalosporins. Although the D66N OXA-1 β-lactamase expressed in E. coli demonstrates MIC values for cefotaxime (2 μg/ml) and cefepime (4 μg/ml) that are below the level considered clinically resistant, there are examples in the literature where such values are associated with poor outcomes during treatment with those two cephalosporins [35]. It is therefore possible that changes such as D66N could contribute toward acquisition of a fully resistant phenotype, particularly in combination with other mutations [36].

We further observe the contrast between the effect of bulky side-chain substitutions at position 66 of OXA-1 compared to similar variants at the equivalent position (Met69) of SHV-1. In the latter, phenylalanine, tyrosine and lysine result in ceftazidime resistance [23]. For OXA-1 expressed in E. coli, the replacement of bulky amino acids at position 66 generally results in lower MICs for all antibiotics. It is possible that, although substrate access to the active site increases in these variants, catalytic efficiency is lowered because of a loss of CO2 binding to Lys70. The structural features required for stable carbamate formation in class D enzyme active sites may confer a lower level of mutational tolerance on these enzymes compared to their counterparts in classes A and C.

Asparagine is the most conservative replacement for aspartate, and its side-chain amide retains the ability to hydrogen-bond to Glu162 (part of the omega loop). Although there are no known class D β-lactamases that naturally possess an asparagine at position 66, a related protein family does contain several examples that do. The BlaR and MecR β-lactam sensor proteins, though not capable of full β-lactam hydrolysis, share a common fold with class D β-lactamases and also use a serine nucleophile for acylation [4,3740]. Several studies of the BlaR1 and MecR1 from Staphylococcus aureus have implicated an N-carboxylysine residue in the acylation of β-lactams, with a subsequent decarboxylation of the lysine residue preventing deacylation [37,39,4143]. Other crystal structure studies of the BlaR1 sensor domain from S. aureus and Bacillus licheniformis failed to reveal a carboxylated lysine residue in the active site [38,44]. Despite the crystal structure discrepancies and the mechanistic differences between the sensor proteins and the class D β-lactamases, it is still possible to gain insight from a comparison of the basic architectural features of these two classes of proteins. Interestingly, the sensor proteins from several species including S. aureus, S. haemolyticus and Clostridium difficile contain an asparagine residue at the position equivalent to Asp66 in OXA-1. The crystal structure of the sensor protein of the BlaR from S. aureus shows that this asparagine (Asn388) does in fact form a hydrogen bond with Glu477, the omega-loop residue that occupies the same position as Glu162 in OXA-1 [39] (Figure 4). We therefore propose that the D66N variant of OXA-1 maintains the interactions that stabilize the omega loop and thereby the carboxylation of Lys70 (via Trp160). This is supported by the data showing that D66N activity for ampicillin is affected little by the addition of bicarbonate. The amide group of the asparagine residue, however, may impart slight conformational changes to the active-site structure, including the position of strand B9. These changes, in turn, may give rise to the alterations in cefotaxime and cefepime resistance.

Overlay of active-site–omega loop interaction region of OXA-1 and BlaR1

Figure 4
Overlay of active-site–omega loop interaction region of OXA-1 and BlaR1

The PDB (Protein Data Bank) co-ordinates of OXA-1 (green; 1M6K) and BlaR1 from S. aureus (grey; 1XKZ) were aligned using the MagicFit feature of Swiss-PDBViewer.

Figure 4
Overlay of active-site–omega loop interaction region of OXA-1 and BlaR1

The PDB (Protein Data Bank) co-ordinates of OXA-1 (green; 1M6K) and BlaR1 from S. aureus (grey; 1XKZ) were aligned using the MagicFit feature of Swiss-PDBViewer.

In summary, our mutational analysis of Asp66 reveals unique insights into the relationship of amino acids that are proximal to the active-site Ser67 in OXA-1 β-lactamase. The novel effect on Lys70 carboxylation and alteration in substrate specificity is reminiscent of how critical amino acid changes in other serine β-lactamases have profound impact on catalysis.

R. A. B.'s laboratory is supported by the Veterans Affairs Merit Review Program and the National Institutes of Health (grant RO1 AI063517-01).

Abbreviations

     
  • CENTA

    (6R,7S)-3-[(3-carboxy-4-nitro-phenyl)sulfanylmethyl]-8-oxo-7-[(2-thiophen-2-ylacetyl)amino]-5-thia-1-azabicyclo-[4.2.0]oct-2-ene-2-carboxylic acid

  •  
  • CM-

    carboxymethyl

  •  
  • His6

    hexahistidine

  •  
  • IPTG

    isopropyl β-D-1-thiogalactopyranoside

  •  
  • MIC

    minimum inhibitory concentration

  •  
  • PDB

    Protein Data Bank

  •  
  • WT

    wild-type

References

References
1
Tipper
D. J.
Strominger
J. L.
Mechanism of action of penicillins: a proposal based on their structural similarity to acyl-D-alanyl-D-alanine
Proc. Natl. Acad. Sci. U.S.A.
1965
, vol. 
54
 (pg. 
1133
-
1141
)
2
Livermore
D. M.
Woodford
N.
The β-lactamase threat in Enterobacteriaceae, Pseudomonas and Acinetobacter
Trends Microbiol.
2006
, vol. 
14
 (pg. 
413
-
420
)
3
Bush
K.
Jacoby
G. A.
Medeiros
A. A.
A functional classification scheme for β-lactamases and its correlation with molecular structure
Antimicrob. Agents Chemother.
1995
, vol. 
39
 (pg. 
1211
-
1233
)
4
Fisher
J. F.
Meroueh
S. O.
Mobashery
S.
Bacterial resistance to β-lactam antibiotics: compelling opportunism, compelling opportunity
Chem. Rev.
2005
, vol. 
105
 (pg. 
395
-
424
)
5
Aubert
D.
Poirel
L.
Chevalier
J.
Leotard
S.
Pages
J. M.
Nordmann
P.
Oxacillinase-mediated resistance to cefepime and susceptibility to ceftazidime in Pseudomonas aeruginosa
Antimicrob. Agents Chemother.
2001
, vol. 
45
 (pg. 
1615
-
1620
)
6
Walther-Rasmussen
J.
Hoiby
N.
OXA-type carbapenemases
J. Antimicrob. Chemother.
2006
, vol. 
57
 (pg. 
373
-
383
)
7
Danel
F.
Hall
L. M.
Livermore
D. M.
Laboratory mutants of OXA-10 β-lactamase giving ceftazidime resistance in Pseudomonas aeruginosa
J. Antimicrob. Chemother.
1999
, vol. 
43
 (pg. 
339
-
344
)
8
Poirel
L.
Girlich
D.
Naas
T.
Nordmann
P.
OXA-28, an extended-spectrum variant of OXA-10 β-lactamase from Pseudomonas aeruginosa and its plasmid- and integron-located gene
Antimicrob. Agents Chemother.
2001
, vol. 
45
 (pg. 
447
-
453
)
9
Dubois
V.
Arpin
C.
Quentin
C.
Texier-Maugein
J.
Poirel
L.
Nordmann
P.
Decreased susceptibility to cefepime in a clinical strain of Escherichia coli related to plasmid- and integron-encoded OXA-30 β-lactamase
Antimicrob. Agents Chemother.
2003
, vol. 
47
 (pg. 
2380
-
2381
)
10
Siu
L. K.
Lo
J. Y.
Yuen
K. Y.
Chau
P. Y.
Ng
M. H.
Ho
P. L.
β-lactamases in Shigella flexneri isolates from Hong Kong and Shanghai and a novel OXA-1-like β-lactamase, OXA-30
Antimicrob. Agents Chemother.
2000
, vol. 
44
 (pg. 
2034
-
2038
)
11
Poirel
L.
Heritier
C.
Tolun
V.
Nordmann
P.
Emergence of oxacillinase-mediated resistance to imipenem in Klebsiella pneumoniae
Antimicrob. Agents Chemother.
2004
, vol. 
48
 (pg. 
15
-
22
)
12
Brown
S.
Amyes
S.
OXA β-lactamases in Acinetobacter: the story so far
J. Antimicrob. Chemother.
2006
, vol. 
57
 (pg. 
1
-
3
)
13
Sun
T.
Nukaga
M.
Mayama
K.
Braswell
E. H.
Knox
J. R.
Comparison of β-lactamases of classes A and D: 1.5-Å crystallographic structure of the class D OXA-1 oxacillinase
Protein Sci.
2003
, vol. 
12
 (pg. 
82
-
91
)
14
Maveyraud
L.
Golemi-Kotra
D.
Ishiwata
A.
Meroueh
O.
Mobashery
S.
Samama
J. P.
High-resolution X-ray structure of an acyl-enzyme species for the class D OXA-10 β-lactamase
J. Am. Chem. Soc.
2002
, vol. 
124
 (pg. 
2461
-
2465
)
15
Golemi
D.
Maveyraud
L.
Vakulenko
S.
Samama
J. P.
Mobashery
S.
Critical involvement of a carbamylated lysine in catalytic function of class D β-lactamases
Proc. Natl. Acad. Sci. U.S.A.
2001
, vol. 
98
 (pg. 
14280
-
14285
)
16
Paetzel
M.
Danel
F.
De Castro
L.
Mosimann
S. C.
Page
M. G.
Strynadka
N. C.
Crystal structure of the class D β-lactamase OXA-10
Nat. Struct. Biol.
2000
, vol. 
7
 (pg. 
918
-
925
)
17
Pernot
L.
Frenois
F.
Rybkine
T.
L'Hermite
G.
Petrella
S.
Delettre
J.
Jarlier
V.
Collatz
E.
Sougakoff
W.
Crystal structures of the class D β-lactamase OXA-13 in the native form and in complex with meropenem
J. Mol. Biol.
2001
, vol. 
310
 (pg. 
859
-
874
)
18
Maveyraud
L.
Golemi
D.
Kotra
L. P.
Tranier
S.
Vakulenko
S.
Mobashery
S.
Samama
J. P.
Insights into class D β-lactamases are revealed by the crystal structure of the OXA10 enzyme from Pseudomonas aeruginosa
Structure
2000
, vol. 
8
 (pg. 
1289
-
1298
)
19
Massova
I.
Mobashery
S.
Kinship and diversification of bacterial penicillin-binding proteins and β-lactamases
Antimicrob. Agents Chemother.
1998
, vol. 
42
 (pg. 
1
-
17
)
20
Kuzin
A. P.
Nukaga
M.
Nukaga
Y.
Hujer
A. M.
Bonomo
R. A.
Knox
J. R.
Structure of the SHV-1 β-lactamase
Biochemistry
1999
, vol. 
38
 (pg. 
5720
-
5727
)
21
Jelsch
C.
Mourey
L.
Masson
J. M.
Samama
J. P.
Crystal structure of Escherichia coli TEM1 β-lactamase at 1.8 Å resolution
Proteins
1993
, vol. 
16
 (pg. 
364
-
383
)
22
Wang
X.
Minasov
G.
Shoichet
B. K.
The structural bases of antibiotic resistance in the clinically derived mutant β-lactamases TEM-30, TEM-32, and TEM-34
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
32149
-
32156
)
23
Helfand
M. S.
Hujer
A. M.
Sonnichsen
F. D.
Bonomo
R. A.
Unexpected advanced generation cephalosporinase activity of the M69F variant of SHV β-lactamase
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
47719
-
47723
)
24
Bebrone
C.
Moali
C.
Mahy
F.
Rival
S.
Docquier
J. D.
Rossolini
G. M.
Fastrez
J.
Pratt
R. F.
Frère
J. M.
Galleni
M.
CENTA as a chromogenic substrate for studying β-lactamases
Antimicrob. Agents Chemother.
2001
, vol. 
45
 (pg. 
1868
-
1871
)
25
Hujer
A. M.
Hujer
K. M.
Bonomo
R. A.
Mutagenesis of amino acid residues in the SHV-1 β-lactamase: the premier role of Gly238Ser in penicillin and cephalosporin resistance
Biochim. Biophys. Acta
2001
, vol. 
1547
 (pg. 
37
-
50
)
26
Hujer
A. M.
Hujer
K. M.
Helfand
M. S.
Anderson
V. E.
Bonomo
R. A.
Amino acid substitutions at Ambler position Gly238 in the SHV-1 β-lactamase: exploring sequence requirements for resistance to penicillins and cephalosporins
Antimicrob. Agents Chemother.
2002
, vol. 
46
 (pg. 
3971
-
3977
)
27
Hujer
A. M.
Page
M. G.
Helfand
M. S.
Yeiser
B.
Bonomo
R. A.
Development of a sensitive and specific enzyme-linked immunosorbent assay for detecting and quantifying CMY-2 and SHV β-lactamases
J. Clin. Microbiol.
2002
, vol. 
40
 (pg. 
1947
-
1957
)
28
Sun
T.
Nukaga
M.
Mayama
K.
Crichlow
G. V.
Kuzin
A. P.
Knox
J. R.
Crystallization and preliminary X-ray study of OXA-1, a class D β-lactamase
Acta Crystallogr. Sect. D Biol. Crystallogr.
2001
, vol. 
57
 (pg. 
1912
-
1914
)
29
Butler
J. N.
Carbon Dioxide Equilibria and their Applications
1991
Chelsea, MI
Lewis Publishers
30
Yamamoto
T.
Tanaka
M.
Nohara
C.
Fukunaga
Y.
Yamagishi
S.
Transposition of the oxacillin-hydrolyzing penicillinase gene
J. Bacteriol.
1981
, vol. 
145
 (pg. 
808
-
813
)
31
Ouellette
M.
Bissonnette
L.
Roy
P. H.
Precise insertion of antibiotic resistance determinants into Tn21-like transposons: nucleotide sequence of the OXA-1 β-lactamase gene
Proc. Natl. Acad. Sci. U.S.A.
1987
, vol. 
84
 (pg. 
7378
-
7382
)
32
Bonomo
R. A.
Dawes
C. G.
Knox
J. R.
Shlaes
D. M.
β-Lactamase mutations far from the active site influence inhibitor binding
Biochim. Biophys. Acta
1995
, vol. 
1247
 (pg. 
121
-
125
)
33
Knox
J. R.
Extended-spectrum and inhibitor-resistant TEM-type β-lactamases: mutations, specificity, and three-dimensional structure
Antimicrob. Agents Chemother.
1995
, vol. 
39
 (pg. 
2593
-
2601
)
34
Santillana
E.
Beceiro
A.
Bou
G.
Romero
A.
Crystal structure of the carbapenemase OXA-24 reveals insights into the mechanism of carbapenem hydrolysis
Proc. Natl. Acad. Sci. U.S.A.
2007
, vol. 
104
 (pg. 
5354
-
5359
)
35
Paterson
D. L.
Ko
W. C.
Von Gottberg
A.
Casellas
J. M.
Mulazimoglu
L.
Klugman
K. P.
Bonomo
R. A.
Rice
L. B.
McCormack
J. G.
Yu
V. L.
Outcome of cephalosporin treatment for serious infections due to apparently susceptible organisms producing extended-spectrum β-lactamases: implications for the clinical microbiology laboratory
J. Clin. Microbiol.
2001
, vol. 
39
 (pg. 
2206
-
2212
)
36
Bethel
C. R.
Hujer
A. M.
Hujer
K. M.
Thomson
J. M.
Ruszczycky
M. W.
Anderson
V. E.
Pusztai-Carey
M.
Taracila
M.
Helfand
M. S.
Bonomo
R. A.
Role of Asp104 in the SHV β-lactamase
Antimicrob. Agents Chemother.
2006
, vol. 
50
 (pg. 
4124
-
4131
)
37
Thumanu
K.
Cha
J.
Fisher
J. F.
Perrins
R.
Mobashery
S.
Wharton
C.
Discrete steps in sensing of β-lactam antibiotics by the BlaR1 protein of the methicillin-resistant Staphylococcus aureus bacterium
Proc. Natl. Acad. Sci. U.S.A.
2006
, vol. 
103
 (pg. 
10630
-
10635
)
38
Wilke
M. S.
Hills
T. L.
Zhang
H. Z.
Chambers
H. F.
Strynadka
N. C.
Crystal structures of the Apo and penicillin-acylated forms of the BlaR1 β-lactam sensor of Staphylococcus aureus
J. Biol. Chem.
2004
, vol. 
279
 (pg. 
47278
-
47287
)
39
Birck
C.
Cha
J. Y.
Cross
J.
Schulze-Briese
C.
Meroueh
S. O.
Schlegel
H. B.
Mobashery
S.
Samama
J. P.
X-ray crystal structure of the acylated β-lactam sensor domain of BlaR1 from Staphylococcus aureus and the mechanism of receptor activation for signal transduction
J. Am. Chem. Soc.
2004
, vol. 
126
 (pg. 
13945
-
13947
)
40
Zhu
Y.
Englebert
S.
Joris
B.
Ghuysen
J. M.
Kobayashi
T.
Lampen
J. O.
Structure, function, and fate of the BlaR signal transducer involved in induction of β-lactamase in Bacillus licheniformis
J. Bacteriol.
1992
, vol. 
174
 (pg. 
6171
-
6178
)
41
Golemi-Kotra
D.
Cha
J. Y.
Meroueh
S. O.
Vakulenko
S. B.
Mobashery
S.
Resistance to β-lactam antibiotics and its mediation by the sensor domain of the transmembrane BlaR signaling pathway in Staphylococcus aureus
J. Biol. Chem.
2003
, vol. 
278
 (pg. 
18419
-
18425
)
42
Cha
J.
Mobashery
S.
Lysine Nζ-decarboxylation in the BlaR1 protein from Staphylococcus aureus at the root of its function as an antibiotic sensor
J. Am. Chem. Soc.
2007
, vol. 
129
 (pg. 
3834
-
3835
)
43
Cha
J.
Vakulenko
S. B.
Mobashery
S.
Characterization of the β-lactam antibiotic sensor domain of the MecR1 signal sensor/transducer protein from methicillin-resistant Staphylococcus aureus
Biochemistry
2007
, vol. 
46
 (pg. 
7822
-
7831
)
44
Kerff
F.
Charlier
P.
Colombo
M. L.
Sauvage
E.
Brans
A.
Frère
J. M.
Joris
B.
Fonze
E.
Crystal structure of the sensor domain of the BlaR penicillin receptor from Bacillus licheniformis
Biochemistry
2003
, vol. 
42
 (pg. 
12835
-
12843
)