L1 Enzyme Research Articles

QM/MM investigation of substrate binding of subclass B3 metallo-β-lactamase SMB-1 from Serratia marcescents: insights into catalytic mechanism.

Metallo-β-lactamases (MβLs) can hydrolyze and deactivate lactam-containing antibiotics, which are the major mechanism to cause drug resistance in the treatment of bacterial infections. This has become a global concern due to the lack of clinically approved inhibitors so far. SMB-1 from Serratia marcescents is a novel B3 subclass MβL, which could inactivate nearly all β-lactam-containing antibiotics, e.g., cephalosporins and carbapenems. It represents a new round of worrisome bacterial resistance. In this work, the Michaelis model of SMB-1 in complex with ampicillin was simulated using combined quantum mechanical and molecular mechanical method. Similar with other dizinc MβLs, a Zn-bridged hydroxide ion was simulated as the nucleophile for the hydrolysis reaction assisted by D120. The protonation of D120 could lead to the loss of Oδ2-Zn2 coordination bond, whereas the C3 carboxylate group moves down to become a new ligand to Zn2. The initial β-lactam ring-opening reaction leads to a conserved nitrogen anionic intermediate, which forms a new ligation between the resulted nitrogen anion and Zn2. The corresponding reaction free energy barrier for the first step of lactam ring-opening reaction was calculated to be 19.2kcal/mol. During the reaction, Q157 serves as the putative "oxyanion hole" rather than Zn1 in L1 enzyme, which was confirmed via the site-directed mutagenesis simulation of Q157A. Our theoretical studies showed some insights into the substrate binding and catalytic mechanism of the SMB-1 metallo-β-lactamase.

Structural and biochemical analysis of the metallo-β-lactamase L1 from emerging pathogen Stenotrophomonas maltophilia revealed the subtle but distinct di-metal scaffold for catalytic activity.

Emergence of Enterobacteriaceae harboring metallo‐β‐lactamases (MBL) has raised global threats due to their broad antibiotic resistance profiles and the lack of effective inhibitors against them. We have been studied one of the emerging environmental MBL, the L1 from Stenotrophomonas maltophilia K279a. We determined several crystal structures of L1 complexes with three different classes of β‐lactam antibiotics (penicillin G, moxalactam, meropenem, and imipenem), with the inhibitor captopril and different metal ions (Zn+2, Cd+2, and Cu+2). All hydrolyzed antibiotics and the inhibitor were found binding to two Zn+2 ions mainly through the opened lactam ring and some hydrophobic interactions with the binding pocket atoms. Without a metal ion, the active site is very similarly maintained as that of the native form with two Zn+2 ions, however, the protein does not bind the substrate moxalactam. When two Zn+2 ions were replaced with other metal ions, the same di‐metal scaffold was maintained and the added moxalactam was found hydrolyzed in the active site. Differential scanning fluorimetry and isothermal titration calorimetry were used to study thermodynamic properties of L1 MBL compared with New Deli Metallo‐β‐lactamase‐1 (NDM‐1). Both enzymes are significantly stabilized by Zn+2 and other divalent metals but showed different dependency. These studies also suggest that moxalactam and its hydrolyzed form may bind and dissociate with different kinetic modes with or without Zn+2 for each of L1 and NDM‐1. Our analysis implicates metal ions, in forming a distinct di‐metal scaffold, which is central to the enzyme stability, promiscuous substrate binding and versatile catalytic activity.StatementThe L1 metallo‐β‐lactamase from an environmental multidrug‐resistant opportunistic pathogen Stenotrophomonas maltophilia K279a has been studied by determining 3D structures of L1 enzyme in the complexes with several β‐lactam antibiotics and different divalent metals and characterizing its biochemical and ligand binding properties. We found that the two‐metal center in the active site is critical in the enzymatic process including antibiotics recognition and binding, which explains the enzyme's activity toward diverse antibiotic substrates. This study provides the critical information for understanding the ligand recognition and for advanced drug development.

Real-time activity assays of β-lactamases in living bacterial cells: application to the inhibition of antibiotic-resistant E. coli strains.

The emergence of antibiotic resistance caused by β-lactamases, including serine β-lactamases (SβLs) and metallo-β-lactamases (MβLs), is a global public health threat. L1, a B3 subclass MβL, hydrolyzes almost all of known β-lactam antibiotics. We report a simple and straightforward UV-Vis approach for real-time activity assays of β-lactamases inside living bacterial cells, and this method has been exemplified by choosing antibiotics, L1 enzyme, Escherichia coli expressing L1 (L1 E. coli), Escherichia coli expressing extended-spectrum β-lactamases (ESBL-E. coli), clinical bacterial strains, and reported MβL and SβL inhibitors. The cell-based studies demonstrated that cefazolin was hydrolyzed by L1 E. coli and clinical strains, and confirmed the hydrolysis to be inhibited by two known L1 inhibitors EDTA and azolylthioacetamide (ATAA), with an IC50 value of 1.6 and 18.9 μM, respectively. Also, it has been confirmed that the breakdown of cefazolin caused by ESBL-E. coli was inhibited by clavulanic acid, the first SβL inhibitor approved by FDA. The data gained through this approach are closely related to the biological function of the target enzyme in its physiological environment. The UV-Vis method proposed here can be applied to target-based whole-cell screening to search for potent β-lactamase inhibitors, and to assays of reactions in complex biological systems, for instance in medical assays.

Modified nitrocefin-EDTA method to differentially quantify the induced L1 and L2 β-lactamases inStenotrophomonas maltophilia

To estimate the ethylene diamine tetraacetic acid (EDTA) concentration at which the L1 enzyme activity in the cell extracts of Stenotrophomonas maltophilia can be mostly inhibited. The effective inhibition concentration of EDTA against the L1 enzyme in the cell extracts was firstly evaluated by using the L2 isogenic mutant of S. maltophilia KJ, KJDeltaL2, as the assayed strain. Approximately 92% L1 activity was inhibited by 10 mmol l(-1) EDTA, which is 100-fold higher than that from previously reported protocols (0.1 mmol l(-1)). Three phylogenetic clusters of L1 proteins were revealed from 11 clinical S. maltophilia isolates, with a L1 protein divergence of 0-11%. The EDTA concentration required to inhibit the L1 enzymes of different phylogenetic clusters was estimated to be 10 mmol l(-1). conclusion: The previous nitrocefin-EDTA protocol for differentially quantifying the L1 and L2 activity in the cell extracts has been modified by raising the added EDTA concentration to 10 mmol l(-1). A rapid and accurate method for determination of L1 and L2 activity will provide a convenient tool for enzyme characterization and induction mechanism study of S. maltophilia.

Antibiotic Binding to Dizinc β-Lactamase L1 from Stenotrophomonas maltophilia: SCC-DFTB/CHARMM and DFT Studies

A dizinc beta-lactamase (L1 from Stenotrophomonas maltophilia) complexed with an antibiotic compound (moxalactam) has been studied using a hybrid quantum mechanical/molecular mechanical (QM/MM) approach. The QM region is described by the self-consistent charge-density functional tight binding (SCC-DFTB) model while the MM by CHARMM. The Michaelis complex, which is constructed from a recent X-ray structure of the L1 enzyme with the hydrolyzed moxalactam, is simulated by molecular dynamics. The simulation yields valuable insights into substrate-enzyme interaction, whose implications in the enzyme catalysis are discussed. Finally, the QM/MM results are compared with a high-level density functional theory study of a truncated active-site model and the agreement provides strong support for the SCC-DFTB treatment of the QM region.

Homo-cysteinyl peptide inhibitors of the L1 metallo-β-lactamase, and SAR as determined by combinatorial library synthesis

Homo-cysteinyl peptides were found to be more active than cysteinyl peptides toward L1 metallo-β-lactamase as reversible competitive inhibitors. A combinatorial library of more than 90 homo-cysteinyl peptides was synthesized and screened for their inhibitory activity toward the L1 enzyme. A systematic structure–activity relationship analysis has revealed the preferred interaction groups for L1 conserved binding sites of β-lactam substrates. The most active compound 95b, had a K i of 2.1 nM.

Antibiotic Recognition by Binuclear Metallo-β-Lactamases Revealed by X-ray Crystallography

Metallo-beta-lactamases are zinc-dependent enzymes responsible for resistance to beta-lactam antibiotics in a variety of host bacteria, usually Gram-negative species that act as opportunist pathogens. They hydrolyze all classes of beta-lactam antibiotics, including carbapenems, and escape the action of available beta-lactamase inhibitors. Efforts to develop effective inhibitors have been hampered by the lack of structural information regarding how these enzymes recognize and turn over beta-lactam substrates. We report here the crystal structure of the Stenotrophomonas maltophilia L1 enzyme in complex with the hydrolysis product of the 7alpha-methoxyoxacephem, moxalactam. The on-enzyme complex is a 3'-exo-methylene species generated by elimination of the 1-methyltetrazolyl-5-thiolate anion from the 3'-methyl group. Moxalactam binding to L1 involves direct interaction of the two active site zinc ions with the beta-lactam amide and C4 carboxylate, groups that are common to all beta-lactam substrates. The 7beta-[(4-hydroxyphenyl)malonyl]-amino substituent makes limited hydrophobic and hydrogen bonding contacts with the active site groove. The mode of binding provides strong evidence that a water molecule situated between the two metal ions is the most likely nucleophile in the hydrolytic reaction. These data suggest a reaction mechanism for metallo-beta-lactamases in which both metal ions contribute to catalysis by activating the bridging water/hydroxide nucleophile, polarizing the substrate amide bond for attack and stabilizing anionic nitrogen intermediates. The structure illustrates how a binuclear zinc site confers upon metallo-beta-lactamases the ability both to recognize and efficiently hydrolyze a wide variety of beta-lactam substrates.

Three-dimensional Structure of FEZ-1, a Monomeric Subclass B3 Metallo-β-lactamase from Fluoribacter gormanii, in Native Form and in Complex with d-Captopril

The β-lactamases are involved in bacterial resistance to penicillin and related compounds. Members of the metallo-enzyme class are now found in many pathogenic bacteria and are thus becoming of major clinical importance. The structures of the Zn-β-lactamase from Fluoribacter gormanii (FEZ-1) in the native and in the complex form are reported here. FEZ-1 is a monomeric enzyme, which possesses two zinc-binding sites. These structures are discussed in comparison with those of the tetrameric L1 enzyme produced by Stenotrophomonas maltophilia. From this analysis, amino acids involved in the oligomerization of L1 are clearly identified. Despite the similarity in fold, the active site of FEZ-1 was found to be significantly different. Two residues, which were previously implicated in function, are not present in L1 or in FEZ-1. The broad-spectrum substrate profile of Zn-β-lactamases arises from the rather wide active-site cleft, where various β-lactam compounds can be accommodated.

Characterization of Monomeric L1 Metallo-β-lactamase and the Role of the N-terminal Extension in Negative Cooperativity and Antibiotic Hydrolysis

The L1 metallo-beta-lactamase from Stenotrophomonas maltophilia is unique among this class of enzymes because it is tetrameric. Previous work predicted that the two regions of important intersubunit interaction were the residue Met-140 and the N-terminal extensions of each subunit. The N-terminal extension was also implicated in beta-lactam binding. Mutation of methionine 140 to aspartic acid results in a monomeric L1 beta-lactamase with a greatly altered substrate specificity profile. A 20-amino acid N-terminal deletion mutant enzyme (N-Del) could be isolated in a tetrameric form but demonstrated greatly reduced rates of beta-lactam hydrolysis and different substrate profiles compared with that of the parent enzyme. Specific site-directed mutations of individual N terminus residues were made (Y11S, W17S, and a double mutant L5A/L8A). All N-terminal mutant enzymes were tetramers and all showed higher K(m) values for ampicillin and nitrocefin, hydrolyzed ceftazidime poorly, and hydrolyzed imipenem more efficiently than ampicillin in contrast to wild-type L1. Nitrocefin turnover was significantly increased, probably because of an increased rate of breakdown of the intermediate species due to a lack of stabilizing forces. K(m) values for monomeric L1 were greatly increased for all antibiotics tested. A model of a highly mobile N-terminal extension in the monomeric enzyme is proposed to explain these findings. Tetrameric L1 shows negative cooperativity, which is not present in either the monomer or N-terminal deletion enzymes, suggesting that the cooperative effect is mediated via N-terminal intersubunit interactions. These data indicate that while the N terminus of L1 is not essential for beta-lactam hydrolysis, it is clearly important to its activity and substrate specificity.

CAU-1, a subclass B3 metallo-beta-lactamase of low substrate affinity encoded by an ortholog present in the Caulobacter crescentus chromosome.

The sequenced chromosome of Caulobacter crescentus CB15 encodes a hypothetical protein that exhibits significant similarity (30 to 35% identical residues) to metallo-beta-lactamases of subclass B3. An allelic variant of this gene (divergent by 3% of its nucleotides) was cloned in Escherichia coli from C. crescentus type strain DSM4727. Expression studies confirmed the metallo-beta-lactamase activity of its product, CAU-1. The enzyme produced in E. coli was purified by two ion-exchange chromatography steps. CAU-1 contains a 29-kDa polypeptide with an alkaline isoelectric pH (> 9), and unlike the L1 enzyme of Stenotrophomonas maltophilia, the native form is monomeric. Kinetic analysis revealed a preferential activity toward penicillins, carbapenems, and narrow-spectrum cephalosporins, while oxyimino cephalosporins were poorly or not hydrolyzed. Affinities for the various beta-lactams were poor overall (K(m) values were always > 100 microM and often > 400 microM). The interaction with divalent ion chelators appeared to occur by a mechanism similar to that prevailing in other members of subclass B3. In C. crescentus, the CAU-1 enzyme is produced independently of beta-lactam exposure and, interestingly, the bla(CAU) determinant is bracketed by three other genes, including two genes encoding enzymes involved in methionine biosynthesis and a gene encoding a putative transcriptional regulator, in an operon-like structure. The CAU-1 enzyme is the first example of a metallo-beta-lactamase in a member of the alpha subdivision of the class Proteobacteria:

Novel Mechanism of Hydrolysis of Therapeutic β-Lactams byStenotrophomonas maltophilia L1 Metallo-β-lactamase

Stopped-flow tryptophan fluorescence under single turnover and pseudo-first-order conditions has been used to investigate the kinetic mechanism of beta-lactam hydrolysis by the Stenotrophomonas maltophilia L1 metallo-beta-lactamase. For the cephalosporin substrates nitrocefin and cefaclor and the carbapenem meropenem, a substantial quench of fluorescence is observed on association of substrate with enzyme. We have assigned this to a rearrangement event subsequent to formation of an initial collision complex. For the colorimetric compound nitrocefin, decay of this dark inter- mediate represents the overall rate-determining step for the reaction and is equivalent to decay of a previously observed state in which the beta-lactam amide bond has already been cleaved. For both cefaclor and meropenem, the rate-determining step for hydrolysis is loss of a second, less quenched state, in which, however, the beta-lactam amide bond remains intact. We suggest, therefore, that the mechanism of hydrolysis of nitrocefin by binuclear metallo-beta-lactamases may be atypical and that cleavage of the beta-lactam amide bond is the rate-determining step for breakdown of the majority of beta-lactam substrates by the L1 enzyme.

Metallo-beta-lactamase producers in environmental microbiota: new molecular class B enzyme in Janthinobacterium lividum.

Eleven environmental samples from different sources were screened for the presence of metallo-beta-lactamase-producing bacteria by using a selective enrichment medium containing a carbapenem antibiotic and subsequently testing each isolate for production of EDTA-inhibitable carbapenemase activity. A total of 15 metallo-beta-lactamase-producing isolates, including 10 Stenotrophomonas maltophilia isolates, 3 Chryseobacterium spp., one Aeromonas hydrophila isolate, and one Janthinobacterium lividum isolate (a species in which production of metallo-beta-lactamase activity was not previously reported), were obtained from 8 samples. In the J. lividum isolate, named JAC1, production of metallo-beta-lactamase activity was elicited upon exposure to beta-lactams. Screening of a JAC1 genomic library for clones showing a reduced imipenem susceptibility led to the isolation of a metallo-beta-lactamase determinant encoding a new member (named THIN-B) of the highly divergent subclass B3 lineage of metallo-beta-lactamases. THIN-B is most closely related (35.6% identical residues) to the L1 enzyme of S. maltophilia and more distantly related to the FEZ-1 enzyme of Legionella gormanii (27.8% identity) and to the GOB-1 enzyme of Chryseobacterium meningosepticum (24.2% identity). Sequences related to bla(THIN-B), and inducible production of metallo-beta-lactamase activity, were also detected in the J. lividum type strain DSM1522. Expression of the bla(THIN-B) gene in Escherichia coli resulted in decreased susceptibility to several beta-lactams, including penicillins, cephalosporins (including cephamycins and oxyimino cephalosporins), and carbapenems, revealing a broad substrate specificity of the enzyme. The results of this study indicated that metallo-beta-lactamase-producing bacteria are widespread in the environment and identified a new molecular class B enzyme in the environmental species J. lividum.

Plasmid location and molecular heterogeneity of the L1 and L2 beta-lactamase genes of Stenotrophomonas maltophilia.

An approximately 200-kb plasmid has been purified from clinical isolates of Stenotrophomonas maltophilia. This plasmid was found in all of the 10 isolates examined and contains both the L1 and the L2 beta-lactamase genes. The location of L1 and L2 on a plasmid makes it more likely that they could spread to other gram-negative bacteria, potentially causing clinical problems. Sequence analysis of the 10 L1 genes revealed three novel genes, L1c, L1d, and L1e, with 8, 12, and 20% divergence from the published strain IID 1275 L1 (L1a), respectively. The most unusual L1 enzyme (L1e) displayed markedly different kinetic properties, with respect to hydrolysis of nitrocefin and imipenem, compared to those of L1a (250- and 100-fold lower k(cat)/K(m) ratios respectively). L1c and L1d, in contrast, displayed levels of hydrolysis very similar to that of L1a. Several nonconservative amino acid differences with respect to L1a, L1b, L1c, and L1d were observed in the substrate binding-catalytic regions of L1e, and this could explain the kinetic differences. Three novel L2 genes (L2b, L2c, and L2d) were sequenced from the same isolates, and their sequences diverge from the published sequence of strain IID 1275 L2 (L2a) by 4, 9, and 25%, respectively. Differences in L1 and L2 gene sequences were not accompanied by similar divergences in 16S rRNA gene sequences, for which differences of <1% were found. It is therefore apparent that the L1 and L2 genes have evolved relatively quickly, perhaps because of their presence on a plasmid.

Purification and characterization of laccase-1 from Pleurotus florida.

Pleurotus florida (ITCC 3308) produces two laccase enzymes (L1 and L2) in potato-dextrose media containing 0.5% yeast extract. Concentrated culture filtrate was separated on DEAE-Sephadex (A-50) column into two enzyme peaks, subsequently named L1 and L2. The L1 enzyme has been purified to homogeneity by ion-exchange and gel-permeation chromatography. L1 is a monomeric glycoprotein with a molar mass of 77 and 82 kDa as determined by SDS-PAGE and gel-filtration chromatography, respectively. The pI value of L1 has been determined to be 4.1. The optimum reaction temperature of the enzyme is 50 degrees C. The Km and some other kinetic parameters of L1 have been determined. Cyanide and azide completely inhibit the enzyme activity. The enzyme was fully active in 1:1 (V/V) buffer-chloroform for at least 2 h. Spectroscopic analysis revealed that the enzyme has four copper atoms, a type 1 copper, a type 2 copper and a type 3 binuclear copper.

The Legionella (Fluoribacter) gormanii metallo-beta-lactamase: a new member of the highly divergent lineage of molecular-subclass B3 beta-lactamases.

A metallo-beta-lactamase determinant was cloned from a genomic library of Legionella (Fluoribacter) gormanii ATCC 33297(T) constructed in the plasmid vector pACYC184 and transformed into Escherichia coli DH5alpha, by screening for clones showing a reduced susceptibility to imipenem. The product of the cloned determinant, named FEZ-1, contains a 30-kDa polypeptide and exhibits an isoelectric pH of 7.6. Sequencing revealed that FEZ-1 is a molecular-class B beta-lactamase which shares the closest structural similarity (29.7% of identical residues) with the L1 enzyme of Stenotrophomonas maltophilia, being a new member of the highly divergent subclass B3 lineage. All the residues that in L1 are known to be directly or indirectly involved in coordination of the zinc ions were found to be conserved also in FEZ-1, suggesting that the geometry of zinc coordination in the active site of the latter enzyme is identical to that of L1. Unlike L1, however, FEZ-1 appeared to be monomeric in gel permeation chromatography experiments and exhibited a distinctive substrate specificity with a marked preference for cephalosporins and meropenem. The properties of FEZ-1 overall resembled those of a beta-lactamase previously purified from the same strain of L. gormanii (T. Fujii, K. Sato, K. Miyata, M. Inoue, and S. Mitsuhashi, Antimicrob. Agents Chemother. 29:925-926, 1986) and are as yet unique among class B enzymes, reinforcing the notion that considerable functional heterogeneity can be encountered among members of this class. A system for overexpression of the bla(FEZ-1) gene in E. coli, based on the T7 phage promoter, was also developed.

Interactions of beta-lactamases with sanfetrinem (GV 104326) compared to those with imipenem and with oral beta-lactams.

Sanfetrinem is a trinem beta-lactam which can be administered orally as a hexatil ester. We examined whether its beta-lactamase interactions resembled those of the available carbapenems, i.e., stable to AmpC and extended-spectrum beta-lactamases but labile to class B and functional group 2f enzymes. The comparator drugs were imipenem, oral cephalosporins, and amoxicillin. MICs were determined for beta-lactamase expression variants, and hydrolysis was examined directly with representative enzymes. Sanfetrinem was a weak inducer of AmpC beta-lactamases below the MIC and had slight lability, with a kcat of 0.00033 s(-1) for the Enterobacter cloacae enzyme. Its MICs for AmpC-derepressed E. cloacae and Citrobacter freundii were 4 to 8 microg/ml, compared with MICs of 0.12 to 2 microg/ml for AmpC-inducible and -basal strains; MICs for AmpC-derepressed Serratia marcescens and Morganella morganii were not raised. Cefixime and cefpodoxime were more labile than sanfetrinem to the E. cloacae AmpC enzyme, and AmpC-derepressed mutants showed much greater resistance; imipenem was more stable and retained full activity against derepressed mutants. Like imipenem, sanfetrinem was stable to TEM-1 and TEM-10 enzymes and retained full activity against isolates and transconjugants with various extended-spectrum TEM and SHV enzymes, whereas these organisms were resistant to cefixime and cefpodoxime. Sanfetrinem, like imipenem and cefixime but unlike cefpodoxime, also retained activity against Proteus vulgaris and Klebsiella oxytoca strains that hyperproduced potent chromosomal class A beta-lactamases. Functional group 2f enzymes, including Sme-1, NMC-A, and an unnamed enzyme from Acinetobacter spp., increased the sanfetrinem MICs by up to 64-fold. These enzymes also compromised the activities of imipenem and amoxicillin but not those of the cephalosporins. The hydrolysis of sanfetrinem was examined with a purified Sme-1 enzyme, and biphasic kinetics were found. Finally, zinc beta-lactamases, including IMP-1 and the L1 enzyme of Stenotrophomonas maltophilia, conferred resistance to sanfetrinem and all other beta-lactams tested, and hydrolysis was confirmed with the IMP-1 enzyme. We conclude that sanfetrinem has beta-lactamase interactions similar to those of the available carbapenems except that it is a weaker inducer of AmpC types, with some tendency to select derepressed mutants, unlike imipenem and meropenem.

Susceptibility to -lactam antibiotics of mutant strains of Xanthomonas maltophilia with high- and low-level constitutive expression of L1 and L2 -lactamases

Xanthomonas maltophilia produces two inducible beta-lactamases, L1 and L2, and resists the antimicrobial activity of beta-lactam antibiotics, including carbapenems. L1 is a zinc-metaloenzyme with carbapenemase activity; L2 is an unusual cephalosporinase. Mutant strains with high- and low-level constitutive expression of these enzymes were derived from three reference strains of X. maltophilia. With a single exception, the mutant strains had altered expression of both enzymes, indicating that these beta-lactamases share regulatory components. The exception was a mutant strain that had low-level constitutive (basal) expression of L1 enzyme but remained inducible for L2. A parent strain with low-level beta-lactamase inducibility was more susceptible to penicillins, cephalosporins and carbapenems than were those in which higher levels of enzyme activity were inducible. Mutations that caused high-level constitutive beta-lactamase expression increased resistance to penicillins and newer cephalosporins. beta-Lactamase basal mutant strains, including the one that remained inducible for L2 enzyme, were more susceptible than inducible strains to these drugs. Organisms with inducible or high-level constitutive beta-lactamase expression were equally resistant to meropenem and imipenem but basal mutant strains, including the one that remained inducible for L2 enzyme, were more susceptible to meropenem than imipenem. Minimal inhibitory concentrations of meropenem, penicillins and cephalosporins, but not imipenem, were greater on Mueller Hinton agar than on IsoSensitest or Diagnostic Sensitivity Test agars. This behaviour was independent of beta-lactamase inducibility, and may reflect permeability differences between cells grown on different media.