The global proliferation of metallo-carbapenemase-producing Enterobacteriaceae has created an unmet need for inhibitors of these enzymes. The rational design of metallo-carbapenemase inhibitors requires detailed knowledge of their catalytic mechanisms. Nine cephalosporins, structurally identical except for the systematic substitution of electron-donating and withdrawing groups in the para position of the styrylbenzene ring, were synthesized and utilized to probe the catalytic mechanism of New Delhi metallo-β-lactamase (NDM-1). Under steady-state conditions, Km values were all in the micromolar range (1.5–8.1 μM), whereas kcat values varied widely (17–220 s−1). There were large solvent deuterium isotope effects for all substrates under saturating conditions, suggesting a proton transfer is involved in the rate-limiting step. Pre-steady-state UV–visible scans demonstrated the formation of short-lived intermediates for all compounds. Hammett plots yielded reaction constants (ρ) of −0.34±0.02 and −1.15±0.08 for intermediate formation and breakdown, respectively. Temperature-dependence experiments yielded ΔG values that were consistent with the Hammett results. These results establish the commonality of the formation of an azanide intermediate in the NDM-1-catalysed hydrolysis of a range cephalosporins with differing electronic properties. This intermediate is a promising target for judiciously designed β-lactam antibiotics that are poor NDM-1 substrates and inhibitors with enhanced active-site residence times.

INTRODUCTION

β-Lactam-containing antibiotics, including the penicillins, cephalosporins, carbapenems and monobactams, are the safest and most effective drugs to combat bacterial infections [13]. These antibiotics block bacterial cell wall synthesis and maintenance by inactivating penicillin-binding proteins (PBPs), and the resulting bacterial cells are susceptible to osmotic lysis [4]. β-Lactam-containing antibiotics are the most widely used antimicrobial agents and make up more than 50% of antibiotic prescriptions to combat bacterial infections [5]. However, bacteria have evolved to resist β-lactam-containing antibiotics through three major mechanisms: changing drug targets, expressing efflux pumps, and producing β-lactamase(s) [6]. Production of β-lactamases is the most common way that Gram-negative bacteria become antibiotic-resistant [7]. β-Lactamases have been grouped into four classes: A, B, C and D [8]. Class A, C and D enzymes are serine-β-lactamases, which utilize an active-site serine residue as a nucleophile to cleave the β-lactam ring in these antibiotics. Class B enzymes, also called metallo-β-lactamases (MBLs), require one or two equivalents of Zn2+ per enzyme, and these enzymes utilize a Zn2+-OH for nucleophilic attack on the β-lactam bond. MBLs have been further subgrouped into B1, B2 and B3 subclasses according to amino acid sequence homologies and metal content [9]. For the last 60 years, serine-β-lactamases have been the most clinically important β-lactamases, and, fortunately, there are clinical inhibitors (sulbactam, clavulanate and tazobactam), which are given in combination with existing β-lactam-containing antibiotics to effectively treat infections caused by several class A- and D-producing pathogens [10,11]. However, there are no clinical inhibitors towards other serine β-lactamases or towards the MBLs. The recent emergence and rapid spread of several MBLs [Verona integron-encoded metallo-β-lactamase (VIM), New Delhi metallo-β-lactamase (NDM) and IMP] in clinically important pathogens [such as carbapenem-resistant Enterobacteriaceae (CRE)] has resulted in the urgent need for new inhibitors [1215].

NDM-1 belongs to the B1 subclass (or B1a) [16] and was first isolated from Klebsiella pneumoniae and Escherichia coli in a Swedish patient who had undergone elective surgery in India in 2008 [17]. The isolated Gram-negative bacteria were found to be resistant to almost all antibiotics, except colistin and tetracycline, mostly due to the presence of plasmids harbouring blaNDM-1 [18]. BlaNDM-1, which also harbours other antibiotic-resistance genes, is horizontally transferred, which has resulted in the rapid global dissemination of the gene among several pathogens [19].

MBLs, once thought to be clinically irrelevant but chemically interesting enzymes, have become increasing prevalent in pathogenic strains. The most clinically significant MBLs appear to be the NDM, IMP and VIM variants, and these enzymes have appeared in K. pneumoniae, Pseudomonas aeruginosa, Acinetobacter baumannii and other organisms [20,21]. A number of reversible and metal-chelating inhibitors of the MBLs have been reported; however, there are no clinically approved inhibitors [2226]. A mechanism-based inhibitor was reported in 2005; however, little development of this compound as a lead has been reported in the subsequent 9 years [27]. A previous article reviewed the status of current inhibitors of the MBLs, and the authors hypothesized that substrate analogues may offer the best hope for MBL inhibitors [28]. It is possible that substrate analogues could be mechanism-based or transition-state analogue inhibitors. Mechanism-based or transition-state analogue inhibitors would be preferred, since these inhibitors typically exhibit higher specificities, lower toxicities and more favourable inhibition potencies [2932]. Indeed, all of the clinically approved inhibitors for the serine β-lactamases are mechanism-based [7,3335]. To design mechanism-based or transition-state analogue inhibitors, scientists need information on the reaction mechanism of the targeted enzyme(s).

There is a great deal known about the reaction mechanisms of the MBLs. Crystal structures of MBLs from all three subclasses have been reported, and many of these structures are of enzyme–product/inhibitor complexes [3638]. Time-dependent spectroscopic studies have been reported on native Zn2+-containing and Co2+-substituted MBLs, and these studies have revealed the role of the metal ions in catalysis [3941]. Steady-state and pre-steady-state kinetic studies have been used to probe the hydrolysis reaction directly, and several minimal kinetic schemes have been proposed for the MBLs. With MBLs CcrA, L1, Bla2, GOB and NDM-1, a ring-opened nitrogen anionic intermediate is formed during the reaction when using nitrocefin or chromacef as substrate [4146]. Mechanistic studies demonstrated that the rate-limiting step of the hydrolysis reaction was protonation of the intermediate. Similar nitrogen anionic intermediates were suggested using clinical substrates imipenem and meropenem with Co2+-substituted BcII [47]. A crystal structure showed a plausible intermediate in the active metal site of mono-zinc CphA when using biapenem as the substrate [48]. It has been argued that mechanistic insight gleaned using the ‘activated’ substrates (i.e. nitrocefin and chromacef) may not reflect the reaction mechanisms of MBLs when using other β-lactam-containing compounds [49]. No nitrogen anionic intermediate was observed in studies with BcII and nitrocefin [50]. There is controversy regarding whether an anionic intermediate was observed in the hydrolysis of nitrocefin by IMP-1 [5153]. Nonetheless, we reasoned that compounds with a sufficiently stabilized intermediate may ‘tie up’ some of the MBLs, preventing the hydrolysis of additional β-lactam containing antibiotics. To further characterize the anionic nitrogen intermediate and the catalytic mechanism, chromacef analogues with a styrylbenzene substituent containing different electron-donating and -withdrawing groups were synthesized and used in kinetic studies. Temperature-dependent pre-steady-state kinetic studies were performed to evaluate the activation energies of transition state(s) involved in catalysis. Enthalpies and entropies of activation for each styrylbenzene-containing compound were calculated by fitting the data to the Eyring–Polanyi equation. The data suggest that cephalosporins containing substituted styrylbenzenes may serve as an excellent scaffold for the generation of new substrate analogue inhibitors.

EXPERIMENTAL

Materials and methods

NDM-1 was purified as described previously [44]. Nitrocefin was purchased from Becton Dickinson. The synthesis of S2 (chromacef) has been described by Yu et al. [54]. The synthetic pathway of compounds S3–S10 is shown in Figure 1. The syntheses of compounds S3–S10 were accomplished via a modification of the published procedure [54]. A 10 mmol amount of p-methoxybenzyl-3-chloromethyl-7-(2-phenylacetamido)-ceph-3-em-4-carboxylate (GCLE; Otsuka Chemical) was dissolved in 250 ml of a mixture of acetone/dichloromethane (4:1, v/v). One equivalent of reagent grade triphenylphosphine and 1.5 equivalents of potassium iodide were sequentially added to this solution. The reaction vessel was covered with aluminum foil and stirred overnight at ambient temperature. The reaction progress was monitored by using TLC via the disappearance of the GCLE spot (RF=0.2–0.25, silica gel 60 F254, EMD-TLC, toluene/ethyl acetate, 5:1, v/v). The precipitated salts in the solvent were removed by gravity filtration under vacuum followed by solvent removal under vacuum.

Synthetic route to substrates

Figure 1
Synthetic route to substrates

PMB is p-methoxybenzyl, and KOTMS is potassium trimethylsilanoate. MeCl2, methylene chloride; Ph3P, triphenylposphine; TFA, trifluoroacetic acid.

Figure 1
Synthetic route to substrates

PMB is p-methoxybenzyl, and KOTMS is potassium trimethylsilanoate. MeCl2, methylene chloride; Ph3P, triphenylposphine; TFA, trifluoroacetic acid.

The residue was then redissolved in 200 ml of tetrahydrofuran (THF)/dicholoromethane (3:1, v/v), and the solution was chilled to −10°C. One equivalent of potassium trimethylsilanoate (Sigma–Aldrich), dissolved in 5 ml of THF, was slowly added to the chilled mixture. After all of the trimethylsilanoate was added, the mixture was stirred for an additional 30 min. One equivalent of aldehyde dissolved in dichloromethane was then added to the mixture. The mixture was then allowed to warm and stirred overnight at room temperature with reactant aldehydes containing electron-withdrawing groups or under reflux with reactant aldehydes containing electron-donating groups. Progress of the reaction was monitored by the formation of a TLC spot (RF=0.25 in toluene/ethyl acetate, 5:1, v/v).

The mixture was vacuum-filtered through a thin pad of silica, and the solvent was removed under vacuum after the reaction was completed. The intermediate product was isolated by flash chromatography performed by using the Biotage SP1 Flash System with a SNAP KP-Sil Column and chloroform/ethyl acetate (4:1, v/v) as the mobile phase (RF=0.5). The desired fractions were pooled, and the solvent was removed under vacuum in a tared round-bottom flask. Deprotection was accomplished by dissolving the residue in a minimal amount (typically 10–15 ml) of chloroform followed by the addition of 0.1 ml of TFA (trifluoroacetic acid)/anisole (5:1, v/v) per mmol intermediate product at 0°C for 30 min. The mixture was then neutralized with a saturated NaHCO3 solution. The organic layer was separated, dried with MgSO4, and removed under vacuum. Solid compounds were obtained by triturating the residue with diethyl ether and petroleum ether and further purified either by recrystallization from acetonitrile or flash chromatography using a acetonitrile/water (5:1, v/v) mobile phase.

Kinetic studies

Steady-state kinetic studies with NDM-1 and cephalosporin scaffolds were performed on a Hewlett-Packard 5480A UV–visible spectrophotometer at 25°C. Substrate concentrations were varied from 1 to 100 μM. Initial velocities were recorded by monitoring the disappearance of substrate using the continuous UV detecting mode. Absorbance readings were converted into concentrations using the following effective molar absorption coefficients of substrate decay: Δε398 nm=−11500 M−1·cm−1 (S1 nitrocefin), Δε378 nm=−14290 M−1·cm−1 (S2 NO2), Δε335 nm=−14100 M−1·cm−1 (S3 CN), Δε317 nm=−3900 M−1·cm−1 (S5 H), Δε325 nm=−1200 M−1·cm−1 (S6 CH3), Δε321 nm=−3450 M−1·cm−1 (S7 Br), Δε318 nm=−5400 M−1·cm−1 (S8 F), Δε326 nm=−1100 M−1·cm−1 (S9 OCH3), Δε325 nm=−10000 M−1·cm−1 (S10 CF3). The effective molar absorption coefficients were measured by dissolving 2–4 mg of the compounds in 50 mM Hepes, pH 7.0, containing less than 1% DMSO, and diluting the compounds to concentrations between 10 and 100 μM. Measurements were performed in triplicate at three different concentrations to obtain the average molar absorption coefficients. Km and kcat values were determined by fitting velocity compared with substrate concentration data to the Michaelis–Menten equation with GraphPad Prism 5 software. The hydrolysis of S1–S10 by OH was monitored in 1 M KH2PO4 adjusted to pH 12 with KOH.

Stopped-flow kinetics

Rapid UV–visible scans were performed on an Applied Photophysics SX 20 stopped-flow instrument equipped with a photodiode array detector. The buffer used in these studies was 50 mM Hepes, pH 7.0, containing 10 μM ZnCl2. Single turnover reactions were performed by mixing 40 μM NDM-1 with 20 μM substrates at different temperatures (5°C, 10°C, 15°C and 20°C). Single intermediate kinetic traces of absorbance were fitted with a bi-exponential equation using the Pro-K software package to obtain the observed rate constants of intermediate formation kobs2 and intermediate decay kobs3 (Supplementary Table S2). Then substrate concentrations were varied between 5 to 40 μM, whereas the enzyme concentration was held at 40 μM. The kon rates were set to diffusion control 108 M−1·s−1, and koff rates were approxi-mated to be multiplying kon by Ki values determined by using reporter substrate cefaclor. All the intermediate kinetic traces were globally fitted with the mechanism in 1 with Dynafit software [55].

Kinetic mechanism of NDM-1

Scheme 1
Kinetic mechanism of NDM-1
Scheme 1
Kinetic mechanism of NDM-1

RESULTS

Synthesis of styryl-substituted analogues

Typically, overall yields ranged from approximately 20% for products with electron-donating groups (S6 and S9) up to 50% for products with electron-withdrawing groups (S2 and S3). Each compound's expected molecular ion peak was identified by its M+23 peak in mass spectra. Each compound's designed structure was consistent with its 400 MHz 1H NMR spectrum and integration. MS and NMR raw spectra are not shown but are available upon request from M.W.C.

Steady-state kinetic studies with NDM-1 and p-substituted styrylbenzyl analogues

Compounds S2, S3, S5, S6, S7, S8, S9 and S10 were tested as substrates for NDM-1, and steady-state kinetic constants, kcat and Km, are shown in Table 1. The Km values for all compounds with NDM-1 were similar, with values ranging between 1.5 and 8.1 μM. To verify binding, the styrylbenzene-substituted compounds were used as competitive inhibitors in reactions with cefaclor as the reporter substrate. Ki values ranged from 0.2 to 5.1 μM (Table 1), suggesting to a first approximation, that all of the compounds bind well to the enzyme [56]. However, there was wide variability in kcat values among the compounds, with the compounds containing the strongest electron-withdrawing groups exhibiting the lowest values for kcat and the compounds with more electron-donating groups exhibiting the largest values for kcat. For example, NDM-1 hydrolysed S2 (NO2) with a kcat of 17±3 s−1 and hydrolysed S9 (OCH3) with a kcat of 220±25 s−1.

Table 1
Steady-state kinetic studies on p-substituted styrylbenzene-substituted cephalosporins at 25°C
Substratekcat (s−1)Km (μM)kcat/Km (M−1·s−1)kcatD (s−1)KmD (μM)kcatH/kcatDKi (μM)*Second-order rate constant for hydrolysis by OH (M−1·s−1)
S1 (Nitrocefin) 9.1±0.5 4.0±0.8 2.3×106 5.1±0.3 4.3±0.5 1.8±0.2 0.2±0.1 2.0×10−1 
S2 (NO217±3 7.8±4.1 2.2×106 7.5±1.0 3.9±0.5 2.6±0.4 1.4±0.6 9.2×10−2 
S3 (CN) 33±1 1.7±0.1 1.9×107 21±2 11±3 1.6±0.2 0.8±0.4 9.1×10−2 
S5 (H) 106±6 3.8±0.6 2.8×107 20±2 2.9±1.3 5.3±0.9 3.0±1.4 3.4×10−2 
S6 (CH3150±13 1.4±0.5 1.1×108 39±1 5.2±0.5 3.8±0.5 5.1±2.7 1.9×10−2 
S7 (Br) 54±2 6.1±2.8 8.9×106 8.8±1.0 2.3±0.8 6.1±1.1 1.3±0.7 2.8×10−2 
S8 (F) 44±3 1.5±0.4 2.9×107 30±4 5.1±2.3 1.5±0.3 1.8±0.8 1.7×10−2 
S9 (OCH3220±25 8.1±1.4 2.7×107 63±6 7.7±3.2 3.9±0.4 2.4±1.2 1.4×10−2 
S10 (CF330±4 1.5±0.4 2.0×107 24±2 8.3±2.0 1.3±0.2 2.4±1.2 4.4×10−2 
Substratekcat (s−1)Km (μM)kcat/Km (M−1·s−1)kcatD (s−1)KmD (μM)kcatH/kcatDKi (μM)*Second-order rate constant for hydrolysis by OH (M−1·s−1)
S1 (Nitrocefin) 9.1±0.5 4.0±0.8 2.3×106 5.1±0.3 4.3±0.5 1.8±0.2 0.2±0.1 2.0×10−1 
S2 (NO217±3 7.8±4.1 2.2×106 7.5±1.0 3.9±0.5 2.6±0.4 1.4±0.6 9.2×10−2 
S3 (CN) 33±1 1.7±0.1 1.9×107 21±2 11±3 1.6±0.2 0.8±0.4 9.1×10−2 
S5 (H) 106±6 3.8±0.6 2.8×107 20±2 2.9±1.3 5.3±0.9 3.0±1.4 3.4×10−2 
S6 (CH3150±13 1.4±0.5 1.1×108 39±1 5.2±0.5 3.8±0.5 5.1±2.7 1.9×10−2 
S7 (Br) 54±2 6.1±2.8 8.9×106 8.8±1.0 2.3±0.8 6.1±1.1 1.3±0.7 2.8×10−2 
S8 (F) 44±3 1.5±0.4 2.9×107 30±4 5.1±2.3 1.5±0.3 1.8±0.8 1.7×10−2 
S9 (OCH3220±25 8.1±1.4 2.7×107 63±6 7.7±3.2 3.9±0.4 2.4±1.2 1.4×10−2 
S10 (CF330±4 1.5±0.4 2.0×107 24±2 8.3±2.0 1.3±0.2 2.4±1.2 4.4×10−2 
*

Ki values were determined by using cefaclor as a reporter substrate and assuming a competitive inhibition mode.

In order to explore the correlation between the steady-state kcat and the electron-donating/withdrawing characteristics of the compounds containing the p-substituted styrylbenzenes, a Hammett plot was constructed (Figure 2). A linear relationship was observed between ln(kcat/kH) (steady-state kcat, ○) and σ, and the slope of the line was ρ=−1.44±0.16. The negative value of ρ suggests that the rate-limiting step involves a negatively charged transition state during the reaction of NDM-1 and these compounds. Such a transition state is consistent with the anionic nitrogen intermediate that was proposed during the hydrolysis of chromacef by NDM-1 [44].

Hammett plot of p-substituted styrylbenzyl analogues under different reaction conditions

Figure 2
Hammett plot of p-substituted styrylbenzyl analogues under different reaction conditions

Plot of steady-state kcat (○) compared with σ yielded ρ=−1.44. Plot of pre-steady-state rate of intermediate formation at 5°C (□) compared with σ yielded ρ=−0.34. Plot of pre-steady-state rate of intermediate breakdown (●) compared with σ yielded ρ=−1.15.

Figure 2
Hammett plot of p-substituted styrylbenzyl analogues under different reaction conditions

Plot of steady-state kcat (○) compared with σ yielded ρ=−1.44. Plot of pre-steady-state rate of intermediate formation at 5°C (□) compared with σ yielded ρ=−0.34. Plot of pre-steady-state rate of intermediate breakdown (●) compared with σ yielded ρ=−1.15.

Solvent isotope studies were conducted to further probe the rate-limiting breakdown of the styrylbenzene-substituted compounds (Table 1). All compounds exhibited kcatH/kcatD ratios greater than 1, and the kcatH/kcatD ratios ranged between 1.3 and 6.1; this range is precedented with other enzymes that exhibit rate-limiting proton transfers [57]. Previous solvent isotope studies on MBLs (BcII, L1 and ImiS) demonstrated solvent isotope effects ranging from 1.5 to 5.6 [5862].

Catalytic efficiencies were calculated from the data in Table 1. The kcat/Km values for the styrylbenzene-substituted compounds ranged over two orders of magnitude, from 2.3±0.6×106 M−1·s−1 for the poorest cephalosporins with strongly electron-withdrawing groups to rates approaching diffusion control (1.1±0.5×108 M−1·s−1) for the best substrates containing electron-donating groups. Values of kcat near diffusion-control rates were previously reported for IMP-catalysed hydrolysis of some cephalosporins [63]. Similar catalytic efficiencies were observed for other cephalosporins, carbapenems and a penicillin (Supplementary Table S1). For comparison's sake, we tested the hydrolysis of the styrylbenzene-containing compounds by OH ions at pH 12, and second-order rate constants were 107—109-fold slower for the OH catalysed reactions (Table 1).

Pre-steady-state kinetic studies with NDM-1 and p-substituted styrylbenzene analogues

Representative spectra of rapidly-mixed S3 with NDM-1 are shown in Figure 3. The peak at 350 nm corresponds to substrate decay, and the peak at 375 nm corresponds to product formation. During the course of the reaction, a peak with maximum absorbance at 460 nm increased over the first 4 ms of the reaction and then decreased over the subsequent 500 ms. As with previous studies using nitrocefin and chromacef as substrate [44], we assign the peak at 460 nm to a ring-opened nitrogen anionic intermediate. Similar studies with the other substrates revealed that the peaks corresponding to the intermediates at 10 ms were found between 375 and 460 nm (Figure 4). The substrates containing the more electron-withdrawing substituents on the styrylbenzene ring exhibited peaks at longer wavelengths, and the substrates with more electron-donating groups on the styrylbenzene ring exhibited peaks at shorter wavelengths. The intensities of the peaks corresponding to the intermediates varied, with the compounds having the more electron-withdrawing substituents also having the most intense intermediate peaks.

Stop-flow absorbance spectra of the reaction of 20 μm S3 and 40 μM NDM-1 at pH 7 and 5°C

Figure 3
Stop-flow absorbance spectra of the reaction of 20 μm S3 and 40 μM NDM-1 at pH 7 and 5°C

Spectra were taken every 4 ms for 0.2 s. The peak at 350 nm corresponds to the spectrum of substrate, the peak at 375 nm corresponds to the spectrum of product, and the peak at 460 nm corresponds to the spectrum of intermediate. The inset represents the kinetic trace of intermediate increase and decay at 460 nm.

Figure 3
Stop-flow absorbance spectra of the reaction of 20 μm S3 and 40 μM NDM-1 at pH 7 and 5°C

Spectra were taken every 4 ms for 0.2 s. The peak at 350 nm corresponds to the spectrum of substrate, the peak at 375 nm corresponds to the spectrum of product, and the peak at 460 nm corresponds to the spectrum of intermediate. The inset represents the kinetic trace of intermediate increase and decay at 460 nm.

Stopped-flow absorbance spectra of intermediates at 10 ms

Figure 4
Stopped-flow absorbance spectra of intermediates at 10 ms
Figure 4
Stopped-flow absorbance spectra of intermediates at 10 ms

The rate constants for intermediate formation (k2), decay (k3) and the reverse rate for intermediate formation (k−2) were determined through global analysis of stopped-flow absorbance spectra of the reactions of NDM-1 with the p-substituted styrylbenzene analogues (Table 2 and Supplementary Figures S1–S9). At 20°C, the rates of intermediate formation (k2) varied between 240 and 1000 s−1. The rates of intermediate decay (k3) varied between 9 and 57 s−1. The determined values of k3 and k2 are lower than the actual rates because there is significant overlap of the absorbance peaks of substrate, product, and intermediate, particularly with the substrates with electron-donating substituents (Supplementary Figures S11–S19). The reverse rates of intermediate formation decreased as the electron-withdrawing strength of the p-substituted styrylbenzene analogues increased. For example, the di-nitro-substituted S1 compound exhibited the lowest intermediate decay rate of 9±1 s−1, whereas the most electron-donating methoxy group in S6 exhibited the highest intermediate decay rate of 57±1 s−1 (Table 1). The ring-opening rate of these compounds catalysed by NDM-1 generally decreased with increasing electron-withdrawing strength in contrast with the compounds’ susceptibility to base-catalysed hydrolysis (Table 1). For instance, nitrocefin was most unstable in pH 12 buffer. This result suggests that the catalytic pathway is different for enzyme catalysis and base catalysis.

Table 2
Obtained rate constants of intermediate formation and decay from Dynafit global analysis at different temperatures

The concentrations of enzyme were fixed at 40 μM, and substrate concentrations were varied between 5 and 40 μM for global analyses. k1 was set to the diffusion-control limit of 109 M−1·s−1, and k−1 was approximated by assuming k−1=Ki.k1. σ constants are from [73]. NA, no data reported.

SpeciesRate constants (s−1)5°C9°C15°C20°Cσ
S1 k2 170±10 190±10 200±10 240±10 NA 
 k−2 <1 <1 <1 <1  
 k3 2±1 3±1 6±1 9±1  
S2 k2 160±10 220±10 260±10 470±10 1.27 
 k−2 <1 <1 <1 5±1  
 k3 3±1 6±1 9±1 15±1  
S3 k2 260±10 400±10 560±10 590±10 1.00 
 k−2 <1 <1 <1 <1  
 k3 5±1 7±1 11±1 18±1  
S5 k2 240±10 350±10 500±10 600±10 
 k−2 36±11 100±10 60±2 100±10  
 k3 11±1 18±1 23±1 33±1  
S6 k2 260±20 570±50 980±90 1000±70 −0.17 
 k−2 90±10 65±10 70±20 94±20  
 k3 27±1 26±1 33±1 57±1  
S7 k2 290±10 520±10 670±20 840±20 0.23 
 k−2 <1 <1 <1 <1  
 k3 10±1 14±1 19±1 27±1  
S8 k2 240±10 430±10 480±10 700±10 0.06 
 k−2 <1 <1 <1 <1  
 k3 11±1 15±1 21±1 32±1  
S9 k2 300±10 440±20 640±50 920±70 −0.27 
 k−2 20±5 50±10 100±10 130±20  
 k3 16±1 23±1 31±1 46±1  
S10 k2 210±10 300±10 420±10 550±10 0.50 
 k−2 <1 <1 <1 <1  
 k3 8±1 12±1 18±1 25±1  
SpeciesRate constants (s−1)5°C9°C15°C20°Cσ
S1 k2 170±10 190±10 200±10 240±10 NA 
 k−2 <1 <1 <1 <1  
 k3 2±1 3±1 6±1 9±1  
S2 k2 160±10 220±10 260±10 470±10 1.27 
 k−2 <1 <1 <1 5±1  
 k3 3±1 6±1 9±1 15±1  
S3 k2 260±10 400±10 560±10 590±10 1.00 
 k−2 <1 <1 <1 <1  
 k3 5±1 7±1 11±1 18±1  
S5 k2 240±10 350±10 500±10 600±10 
 k−2 36±11 100±10 60±2 100±10  
 k3 11±1 18±1 23±1 33±1  
S6 k2 260±20 570±50 980±90 1000±70 −0.17 
 k−2 90±10 65±10 70±20 94±20  
 k3 27±1 26±1 33±1 57±1  
S7 k2 290±10 520±10 670±20 840±20 0.23 
 k−2 <1 <1 <1 <1  
 k3 10±1 14±1 19±1 27±1  
S8 k2 240±10 430±10 480±10 700±10 0.06 
 k−2 <1 <1 <1 <1  
 k3 11±1 15±1 21±1 32±1  
S9 k2 300±10 440±20 640±50 920±70 −0.27 
 k−2 20±5 50±10 100±10 130±20  
 k3 16±1 23±1 31±1 46±1  
S10 k2 210±10 300±10 420±10 550±10 0.50 
 k−2 <1 <1 <1 <1  
 k3 8±1 12±1 18±1 25±1  

To further explore the effect of the electron-withdrawing effect of the substituent on the rates of intermediate formation (□) and decay (□), a Hammett plot was constructed with observed rate constants at temperature 5°C (Figure 2). The plots of ln(kobs2/kH) and ln(kobs3/kH) compared with σ demonstrated negative slopes of the lines. The Hammett plot of intermediate formation rate as a function of σ yielded a slope (ρ) of −0.34±0.02, suggesting that the C–N bond cleavage rate is dependent on the electron-withdrawing ability of the p-substituted styrylbenzene group. The Hammett plot of intermediate decay rate as a function of σ yielded a slope (ρ) of −1.15±0.08, which suggests that the lifetime of the intermediate is also dependent on the electron-withdrawing ability of the styrylbenzene substituent.

Temperature-dependence of pre-steady-state kinetics

In order to probe the energy of activation of the transition states of the hydrolysis reactions, the temperature-dependence of rate constants was studied using the Eyring–Polanyi equation (

graphic
k: reaction rate, T: absolute temperature, ΔH: enthalphy of activation, ΔS: entropy of activation, kB: Boltzmann's constant, h: Planck's constant) [64]. The lnk2/T and lnk3/T data were plotted against 1/T to determine the enthalpies and entropies of activation. Excellent linear correlations for the plots containing kobs2 and kobs3 were observed for each substrate in Figures 5(A) and 5(B). The Gibbs energies of activation (ΔG) at 288.1 K were calculated for comparison using equation ΔGHTΔS (Table 3), and under these conditions, ΔG is approximately equivalent to the energy barrier from reactants to the transition state [65]. In general, the enthalpies of activation with kobs3 are larger than with kobs2, except for S6 and S9. However, after compensating for the entropy of activation, the Gibbs energy of activation for k3 is approximately 63 kJ·mol−1. This value is larger than ΔG with kobs2 for all of the tested compounds, indicating that, thermodynamically, intermediate breakdown is the rate-limiting step. The Gibbs energy of activation for the second transition state (kobs3) is reflected to be linear among S2–S10 compounds, which is consistent with the prediction of Hammett analyses.

Temperature effect on the intermediate formation and breakdown of S1–S10 compounds catalysed by NDM-1

Figure 5
Temperature effect on the intermediate formation and breakdown of S1–S10 compounds catalysed by NDM-1

(A) Temperature effect of intermediate formation rate; (B) temperature effect of intermediate decay. The lines were fitted by using the Eyring–Polanyi equation. ○: S1; □: S2; ◇: S3; x: S5; +:S6; △: S7; ●: S8; ■: S9; ◆: S10.

Figure 5
Temperature effect on the intermediate formation and breakdown of S1–S10 compounds catalysed by NDM-1

(A) Temperature effect of intermediate formation rate; (B) temperature effect of intermediate decay. The lines were fitted by using the Eyring–Polanyi equation. ○: S1; □: S2; ◇: S3; x: S5; +:S6; △: S7; ●: S8; ■: S9; ◆: S10.

Table 3
Free energy activation of intermediate formation and decay
Intermediate formationIntermediate decay
SpeciesΔH (kJ·mol−1)ΔS (J·mol−1·K−1)ΔG (kJ·mol−1)ΔH (kJ·mol−1)ΔS (J·mol−1·K−1)ΔG (kJ·mol−1)
S1 25±1 −112±1 56±2 76±1 29±1 68±1 
S2 19±1 −133±1 56±1 61±1 −18±1 66±1 
S3 36±1 −67±1 55±1 52±1 −46±1 65±1 
S5 29±1 −89±1 55±1 49±1 −47±1 63±1 
S6 50±1 −15±1 54±1 47±1 −54±1 62±1 
S7 30±1 −88±1 55±2 49±1 −55±1 64±1 
S8 16±1 −139±1 55±1 47±1 −56±1 62±1 
S9 47±1 −25±1 54±1 34±2 −97±1 62±4 
S10 13±1 −150±1 55±3 48±1 −53±1 63±1 
Intermediate formationIntermediate decay
SpeciesΔH (kJ·mol−1)ΔS (J·mol−1·K−1)ΔG (kJ·mol−1)ΔH (kJ·mol−1)ΔS (J·mol−1·K−1)ΔG (kJ·mol−1)
S1 25±1 −112±1 56±2 76±1 29±1 68±1 
S2 19±1 −133±1 56±1 61±1 −18±1 66±1 
S3 36±1 −67±1 55±1 52±1 −46±1 65±1 
S5 29±1 −89±1 55±1 49±1 −47±1 63±1 
S6 50±1 −15±1 54±1 47±1 −54±1 62±1 
S7 30±1 −88±1 55±2 49±1 −55±1 64±1 
S8 16±1 −139±1 55±1 47±1 −56±1 62±1 
S9 47±1 −25±1 54±1 34±2 −97±1 62±4 
S10 13±1 −150±1 55±3 48±1 −53±1 63±1 

DISCUSSION

The occurrence and global dissemination of NDM-positive strains especially among pathogens have caused alarm since β-lactam-containing antibiotics are the largest class of clinical antibiotics [5]. Furthermore, the plasmids harbouring blaNDM-1 may also contain resistance determinants for macrolides, aminoglycosides, rifampicin, sulfamethoxazole and monobactams [17], resulting in bacterial infections that are not treatable with current antibiotic therapies. Therefore, there is an urgent need to discover new clinical inhibitors of the MBLs, and these inhibitors could be co-administered with existing β-lactam-containing antibiotics. Many non-clinical inhibitors have been reported [2226], and several of these compounds target the metal centre(s) in the MBLs. Mechanism-based inhibitors originating from substrate analogues may offer the best hope for new and effective MBL inhibitors [66].

In order to shed light on the hydrolysis mechanism of p-substituted styrylbenzene-containing compounds, S2–S10 compounds were synthesized and characterized. All compounds tested were shown to be substrates for NDM-1, and, as hypothesized, the electron-donating or -withdrawing ability of the p-substituted styrylbenzene substituent greatly influenced the hydrolysis rates of the substrates. The hydrolysis of all of the compounds resulted in the appearance of short-lived intermediates, which exhibited maximal absorbances between 310 and 750 nm. As previously explained for studies with nitrocefin [42], the highly conjugated π system of the styrylbenzene group and the dihydrothiazine ring in all compounds tested facilitates electron delocalization and results in the red-shifted UV absorbance peaks of the intermediates [67]. The absorbance maximum, which is characteristic of the intermediate's energy, is dependent on the electron-donating/withdrawing abilities of the substituent group on the styrylbenzene moiety, with nitrocefin and chromacef exhibiting the lowest energy intermediates (Figure 4). For all compounds tested, intermediate decay rates are much smaller than the substrate decay/intermediate formation rates, indicating that the breakdown of intermediates is rate-limiting. Previous mechanistic studies with nitrocefin and L1 suggested that protonation of the intermediate was the rate-limiting chemical step [42,43] and the large solvent isotope effects exhibited by NDM-1 with all tested compounds suggest involvement of solvent, such as protonation of the intermediates by water, is also rate-limiting.

The variety of substituents on the compounds used in the present study allowed for Hammett analysis to probe further the chemical reaction with transition state theory [29]. Hammett plots are often applied in organic chemistry to predict unknown reactions and identify transition states involved in certain chemical transformations [68,69]. Under steady-state conditions with NDM-1, the ln(kcat/kH) against σ plot demonstrated a linear correlation between the electron-withdrawing/donating ability of the p-substituted styrylbenzene substituent and hydrolysis rate. In other words, compounds with strong electron-withdrawing substituents were hydrolysed slower than those substrates with electron-donating substituents. The negative slope in the Hammett plot is consistent with the proposed nitrogen anionic intermediate [42,43]. It is well known that catalysts can enhance reaction rates by many orders of magnitude by increasing the concentration and affinity of activated complex or transition state, which then decomposes into product [29,70]. The compounds with strong electron-withdrawing groups are thought to delocalize the negative charge on the intermediate and decrease the affinity of the nitrogen anion for a proton, resulting in a higher activation energy and a lower rate.

Pre-steady-state kinetic studies allowed for Hammett analyses of the microscopic rate constants for intermediate formation and decay. Previously, we reported that the minimal kinetic/reaction mechanism for the hydrolysis of chromacef by NDM-1 involved two steps. In step 1 after substrate binds, a Zn2+-OH attacks the activated β-lactam carbonyl forming a tetrahedral transition state that decomposes (C-N bond cleavage) to a ring-opened nitrogen anionic intermediate (EI). The second and rate-limiting step is protonation of the nitrogen anion and product release. Hammett plots of kobs2 (rate of intermediate formation) and kobs3 (rate of intermediate decay) support this proposed mechanism. The Hammett plot using kobs2 is linear with a slope of −0.34±0.02 (middle line in Figure 2), while the plot using kobs3 is linear with a slope of −1.15±0.08 (top line in Figure 2). The negative slopes suggest negatively charged transition states for each step, and the magnitude of the slopes suggest that the electron-withdrawing strength of the substituent affects intermediate decay more than it affects intermediate formation. This result is not surprising because the nitrogen anion in the intermediate is strongly conjugated to the styrylbenzene substituent. In contrast, the β-lactam carbonyl would not be as strongly (if at all) conjugated to the styrylbenzene group. The ring-opening rate of these compounds by simple OH ions was analysed by Hammett plot (Supplementary Figure S10) with a positive ρ of 1.23±0.06, which is strikingly different from the ring-opening step catalysed by NDM-1 enzyme. Based on this observation, we reasoned that the catalytic mechanism of β-lactam hydrolysis by NDM-1 is different from simple OH ions. Most likely, the enzymatic catalysis bisects the catalysis into two steps and shifts the rate-limiting step to protonation of an anionic nitrogen intermediate rather than the formation of tetrahedral oxyanion catalysed by simple OH ions [71,72]. It is also interesting that the slopes of Hammett plots for the two steps of enzymatic catalysis add to roughly equal to the slope (ρ of −1.44±0.16) of the Hammett plot using the steady-state kcat. This result highlights that intermediate formation (substrate attack by OH, tetrahedral transition state and C–N bond cleavage) and intermediate decay are the key mechanistic steps in the reactions catalysed by NDM-1. The large solvent isotope effect and relative rates (of intermediate formation compared with decay) suggest that the intermediate decay plays a larger role in the overall mechanism with the compounds tested.

Temperature-dependence studies support the relative large role that intermediate decay plays in the overall reaction. The ΔG values corresponding to kobs2 (intermediate formation rate) are roughly equal for all compounds tested, suggesting that the electron-withdrawing abilities of the p-substituted styrylbenzene groups play almost no role in the activation energy of this step (Table 3). This result is consistent with the Hammett analyses described above. Nonetheless, the ΔG values vary by 5 kJ/mol when examining the different compounds and following kobs3 (rate of intermediate decay). The smallest values for ΔG were found for the compounds with the most electron-donating substituents, and the largest values of ΔG were observed for the compounds with the most electron-withdrawing substituents (Table 3). These results argue that the transition state for the intermediate decay step is less stabilized by compounds that donate electron density, and these compounds have smaller activation barriers and exhibit higher hydrolysis rates.

All MBLs resist inhibition by mechanism-based inhibitors of serine-β-lactamases [11] that utilize a transient covalent intermediate to elicit electronic rearrangements that lead to irreversible inhibition. Consequently, most efforts to prepare new inhibitors towards the MBLs involve finding small molecules that bind to the active sites of the enzymes [2226]. The results of the present study suggest that a different strategy to inhibit MBLs may be possible. The chemical behaviour of the substituents can greatly affect antibiotic hydrolysis rates. It may be possible to design mechanism-based inhibitors (suicide substrates) with long residence times in MBLs that would enhance inhibition. This concept is depicted in qualitative free energy diagrams modelling the rate-determining step for the MBL-catalysed hydrolysis of β-lactams in Figure 6.

Qualitative free energy diagram for the breakdown of the azanide intermediate of NDM-1-catalysed hydrolysis of styrylbenzene-substituted cephalosporins

Figure 6
Qualitative free energy diagram for the breakdown of the azanide intermediate of NDM-1-catalysed hydrolysis of styrylbenzene-substituted cephalosporins

(A) Breakdown of the intermediate of a good substrate. (B) Breakdown of the intermediate of a poor substrate. Open arrows represent energies of activation. I and t represent the azanide intermediate and transition state structures, respectively.

Figure 6
Qualitative free energy diagram for the breakdown of the azanide intermediate of NDM-1-catalysed hydrolysis of styrylbenzene-substituted cephalosporins

(A) Breakdown of the intermediate of a good substrate. (B) Breakdown of the intermediate of a poor substrate. Open arrows represent energies of activation. I and t represent the azanide intermediate and transition state structures, respectively.

In Figure 6, A represents the reaction energy diagram for the breakdown of a good substrate such as one that contains an electron-donating group which forms a strongly anionic intermediate (I). B, in contrast, represents that for breakdown of a poor substrate such as one with a strongly electron-withdrawing group that dissipates the anionic charge on the intermediate. Note that the energies of activation (open arrows) for the breakdown of the intermediate are much smaller for A compared with that B. A judiciously designed suicide substrate that stabilizes this azanide intermediate would be trapped in this deeper energy well, increasing its residence time in the enzyme's active site giving more time for inhibition chemistry to occur before diffusing away. Efforts at preparing such compounds are currently underway.

Abbreviations

     
  • CRE

    carbapenem-resistant Enterobacteriaceae

  •  
  • GCLE

    p-methoxybenzyl-3-chloromethyl-7-(2-phenylacetamido)-ceph-3-em-carboxylate

  •  
  • MBL

    metallo-β-lactamase

  •  
  • NDM

    New Delhi metallo-β-lactamase

  •  
  • PBP

    penicillin-binding protein

  •  
  • THF

    tetrahydrofuran

  •  
  • VIM

    Verona integron-encoded metallo-β-lactamase

AUTHOR CONTRIBUTION

Hao Yang carried out steady-state, pre-steady-state kinetic experiments, and wrote the paper. Heather Young participated in purifying NDM-1 proteins. Sophia Yu synthesized compounds S2–S10. Larry Sutton and Michael Crowder supervised the project and wrote the paper.

FUNDING

This work was supported by the National Science Foundation [grant number CHE-1151658 (to N.W.C.)].

References

References
1
Kardos
 
N.
Demain
 
A. L.
 
Penicillin: the medicine with the greatest impact on therapeutic outcomes
Appl. Microbiol. Biotechnol.
2011
, vol. 
92
 (pg. 
677
-
687
)
[PubMed]
2
El-Gamal
 
M. I.
Oh
 
C. H.
 
Current status of carbapenem antibiotics
Curr. Top. Med. Chem.
2010
, vol. 
10
 (pg. 
1882
-
1897
)
[PubMed]
3
Demain
 
A. L.
Elander
 
R. P.
 
The β-lactam antibiotics: past, present, and future
Antonie Van Leeuwenhoek
1999
, vol. 
75
 (pg. 
5
-
19
)
[PubMed]
4
Koch
 
A. L.
 
Penicillin binding proteins, β-lactams, and lactamases: offensives, attacks, and defensive countermeasures
Crit. Rev. Microbiol.
2000
, vol. 
26
 (pg. 
205
-
220
)
[PubMed]
5
Essack
 
S. Y.
 
The development of β-lactam antibiotics in response to the evolution of β-lactamases
Pharm. Res.
2001
, vol. 
18
 (pg. 
1391
-
1399
)
[PubMed]
6
Poole
 
K.
 
Resistance to β-lactam antibiotics
Cell. Mol. Life Sci.
2004
, vol. 
61
 (pg. 
2200
-
2223
)
[PubMed]
7
Sandanayaka
 
V. P.
Prashad
 
A. S.
 
Resistance to β-lactam antibiotics: structure and mechanism based design of β-lactamase inhibitors
Curr. Med. Chem.
2002
, vol. 
9
 (pg. 
1145
-
1165
)
[PubMed]
8
Bush
 
K.
Jacoby
 
G. A.
 
Updated functional classification of β-lactamases
Antimicrob. Agents Chemother.
2010
, vol. 
54
 (pg. 
969
-
976
)
[PubMed]
9
Garau
 
G.
Garcia-Saez
 
I.
Bebrone
 
C.
Anne
 
C.
Mercuri
 
P.
Galleni
 
M.
Frère
 
J.-M.
Dideberg
 
O.
 
Update of the standard numbering scheme for class B β-lactamases
Antimicrob. Agents Chemother.
2004
, vol. 
48
 (pg. 
2347
-
2349
)
[PubMed]
10
Cullmann
 
W.
Binder
 
S.
Stieglitz
 
M.
 
Sulbactam and clavulanic acid: enzyme kinetics and synergy with ampicillin and mezlocillin
Int. J. Antimicrob. Agents
1996
, vol. 
6
 (pg. 
S1
-
S7
)
[PubMed]
11
Drawz
 
S. M.
Bonomo
 
R. A.
 
Three decades of β-lactamase inhibitors
Clin. Microbiol. Rev.
2010
, vol. 
23
 (pg. 
160
-
201
)
[PubMed]
12
Boucher
 
H. W.
Talbot
 
G. H.
Benjamin
 
D. K.
Bradley
 
J.
Guidos
 
R. J.
Jones
 
R. N.
Murray
 
B. E.
Bonomo
 
R. A.
Gilbert
 
D.
Infect Dis Soc
 
A.
 
10 x ‘20 Progress-development of new drugs active against Gram-negative bacilli: an update from the Infectious Diseases Society of America
Clin. Infect. Dis.
2013
, vol. 
56
 (pg. 
1685
-
1694
)
[PubMed]
13
Buynak
 
J. D.
 
β-Lactamase inhibitors: a review of the patent literature (2010–2013)
Expert Opin. Ther. Pat.
2013
, vol. 
23
 (pg. 
1469
-
1481
)
[PubMed]
14
Fast
 
W.
Sutton
 
L. D.
 
Metallo-β-lactamase: inhibitors and reporter substrates
Biochim. Biophys. Acta
2013
, vol. 
1834
 (pg. 
1648
-
1659
)
[PubMed]
15
King
 
D. T.
Strynadka
 
N. C.
 
Targeting metallo-β-lactamase enzymes in antibiotic resistance
Fut. Med. Chem.
2013
, vol. 
5
 (pg. 
1243
-
1263
)
16
Bush
 
K.
 
The ABCD's of β-lactamase nomenclature
J. Infect. Chemother.
2013
, vol. 
19
 (pg. 
549
-
559
)
[PubMed]
17
Yong
 
D.
Toleman
 
M. A.
Giske
 
C. G.
Cho
 
H. S.
Sundman
 
K.
Lee
 
K.
Walsh
 
T. R.
 
Characterization of a new metallo-β-lactamase gene, bla(NDM-1), and a novel erythromycin esterase gene carried on a unique genetic structure in Klebsiella pneumoniae sequence Type 14 from India
Antimicrob. Agents Chemother.
2009
, vol. 
53
 (pg. 
5046
-
5054
)
[PubMed]
18
Kumarasamy
 
K. K.
Toleman
 
M. A.
Walsh
 
T. R.
Bagaria
 
J.
Butt
 
F.
Balakrishnan
 
R.
Chaudhary
 
U.
Doumith
 
M.
Giske
 
C. G.
Irfan
 
S.
, et al 
Emergence of a new antibiotic resistance mechanism in India, Pakistan, and the UK: a molecular, biological, and epidemiological study
Lancet Infect. Dis.
2010
, vol. 
10
 (pg. 
597
-
602
)
[PubMed]
19
Nordmann
 
P.
Poirel
 
L.
Walsh
 
T. R.
Livermore
 
D. M.
 
The emerging NDM carbapenemases
Trends Microbiol.
2011
, vol. 
19
 (pg. 
588
-
595
)
[PubMed]
20
Gupta
 
V.
 
Metallo β lactamases in Pseudomonas aeruginosa and Acinetobacter species
Expert Opin. Investig. Drugs.
2008
, vol. 
17
 (pg. 
131
-
143
)
[PubMed]
21
Walsh
 
T. R.
Weeks
 
J.
Livermore
 
D. M.
Toleman
 
M. A.
 
Dissemination of NDM-1 positive bacteria in the New Delhi environment and its implications for human health: an environmental point prevalence study
Lancet Infect. Dis.
2011
, vol. 
11
 (pg. 
355
-
362
)
[PubMed]
22
Mollard
 
C.
Moali
 
C.
Papamicael
 
C.
Damblon
 
C.
Vessilier
 
S.
Amicosante
 
G.
Schofield
 
C. J.
Galleni
 
M.
Frère
 
J.-M.
Roberts
 
G. C.
 
Thiomandelic acid, a broad spectrum inhibitor of zinc β-lactamases: kinetic and spectroscopic studies
J. Biol. Chem.
2001
, vol. 
276
 (pg. 
45015
-
45023
)
[PubMed]
23
Siemann
 
S.
Evanoff
 
D. P.
Marrone
 
L.
Clarke
 
A. J.
Viswanatha
 
T.
Dmitrienko
 
G. I.
 
N-arylsulfonyl hydrazones as inhibitors of IMP-1 metallo-β -lactamase
Antimicrob. Agents Chemother.
2002
, vol. 
46
 (pg. 
2450
-
2457
)
[PubMed]
24
Faridoon, Hussein
 
W. M.
Vella
 
P.
Islam
 
N. U.
Ollis
 
D. L.
Schenk
 
G.
McGeary
 
R. P.
 
3-Mercapto-1,2,4-triazoles and N-acylated thiosemicarbazides as metallo-β-lactamase inhibitors
Bioorg. Med. Chem. Lett.
2012
, vol. 
22
 (pg. 
380
-
386
)
[PubMed]
25
Chen
 
P.
Horton
 
L. B.
Mikulski
 
R. L.
Deng
 
L.
Sundriyal
 
S.
Palzkill
 
T.
Song
 
Y.
 
2-Substituted 4,5-dihydrothiazole-4-carboxylic acids are novel inhibitors of metallo-β-lactamases
Bioorg. Med. Chem. Lett.
2012
, vol. 
22
 (pg. 
6229
-
6232
)
[PubMed]
26
Schlesinger
 
S. R.
Bruner
 
B.
Farmer
 
P. J.
Kim
 
S. K.
 
Kinetic characterization of a slow-binding inhibitor of Bla2: thiomaltol
J. Enzyme. Inhib. Med. Chem.
2013
, vol. 
28
 (pg. 
137
-
142
)
[PubMed]
27
Kurosaki
 
H.
Yamaguchi
 
Y.
Higashi
 
T.
Soga
 
K.
Matsueda
 
S.
Yumoto
 
H.
Misumi
 
S.
Yamagata
 
Y.
Arakawa
 
Y.
Goto
 
M.
 
Irreversible inhibition of metallo-β-lactamase (IMP-1) by 3-(3-mercaptopropionylsulfanyl)propionic acid pentafluorophenyl ester
Angew. Chem. Int. Ed. Engl.
2005
, vol. 
44
 (pg. 
3861
-
3864
)
[PubMed]
28
Biondi
 
S.
Long
 
S.
Panunzio
 
M.
Qin
 
W. L.
 
Current trends in β-lactam based β-lactamases inhibitors
Curr. Med. Chem.
2011
, vol. 
18
 (pg. 
4223
-
4236
)
[PubMed]
29
Wolfenden
 
R.
 
Transition-state analog inhibitors and enzyme catalysis
Annu. Rev. Biophys. Bioeng.
1976
, vol. 
5
 (pg. 
271
-
306
)
[PubMed]
30
Pratt
 
R. F.
 
On the definition and classification of mechanism-based enzyme-inhibitors
Bioorg. Med. Chem. Lett.
1992
, vol. 
2
 (pg. 
1323
-
1326
)
31
Lienhard
 
G. E.
 
Enzymatic catalysis and transition-state theory
Science
1973
, vol. 
180
 (pg. 
149
-
154
)
[PubMed]
32
Schramm
 
V. L.
 
Enzymatic transition state theory and transition state analogue design
J. Biol. Chem.
2007
, vol. 
282
 (pg. 
28297
-
28300
)
[PubMed]
33
Yang
 
Y.
Janota
 
K.
Tabei
 
K.
Huang
 
N.
Siegel
 
M. M.
Lin
 
Y. I.
Rasmussen
 
B. A.
Shlaes
 
D. M.
 
Mechanism of inhibition of the class A β-lactamases PC1 and TEM-1 by tazobactam—observation of reaction products by electrospray ionization mass spectrometry
J. Biol. Chem.
2000
, vol. 
275
 (pg. 
26674
-
26682
)
[PubMed]
34
Bush
 
K.
Macalintal
 
C.
Rasmussen
 
B. A.
Lee
 
V. J.
Yang
 
Y. J.
 
Kinetic interactions of tazobactam with β-lactamases from all major structural classes
Antimicrob. Agents Chemother.
1993
, vol. 
37
 (pg. 
851
-
858
)
[PubMed]
35
Charnas
 
R. L.
Fisher
 
J.
Knowles
 
J. R.
 
Chemical studies on the inactivation of Escherichia coli RTEM β-lactamase by clavulanic acid
Biochemistry
1978
, vol. 
17
 (pg. 
2185
-
2189
)
[PubMed]
36
King
 
D. T.
Worrall
 
L. J.
Gruninger
 
R.
Strynadka
 
N. C. J.
 
New Delhi metallo-β-lactamase: structural insights into β-lactam recognition and inhibition
J. Am. Chem. Soc.
2012
, vol. 
134
 (pg. 
11362
-
11365
)
[PubMed]
37
Spencer
 
J.
Read
 
J.
Sessions
 
R. B.
Howell
 
S.
Blackburn
 
G. M.
Gamblin
 
S. J.
 
Antibiotic recognition by binuclear metallo-β-lactamases revealed by X-ray crystallography
J. Am. Chem. Soc.
2005
, vol. 
127
 (pg. 
14439
-
14444
)
[PubMed]
38
Horsfall
 
L. E.
Garau
 
G.
Lienard
 
B. M. R.
Dideberg
 
O.
Schofield
 
C. J.
Frère
 
J.-M.
Galleni
 
M.
 
Competitive inhibitors of the CphA metallo-β-lactamase from Aeromonas hydrophila
Antimicrob. Agents Chemother.
2007
, vol. 
51
 (pg. 
2136
-
2142
)
[PubMed]
39
Hu
 
Z. X.
Periyannan
 
G.
Bennett
 
B.
Crowder
 
M. W.
 
Role of the Zn1 and Zn2 sites in metallo-β-lactamase L1
J. Am. Chem. Soc.
2008
, vol. 
130
 (pg. 
14207
-
14216
)
[PubMed]
40
Garrity
 
J. D.
Bennett
 
B.
Crowder
 
M. W.
 
Direct evidence that the reaction intermediate of metallo-β-lactamase L1 is metal bound
Biochemistry
2005
, vol. 
44
 (pg. 
1078
-
1087
)
[PubMed]
41
Yang
 
H.
Aitha
 
M.
Marts
 
A. R.
Hetrick
 
A. M.
Bennett
 
B.
Crowder
 
M. W.
Tierney
 
D. L.
 
Spectroscopic and mechanistic studies of heterodimetallic forms of metallo-β-lactamase NDM-1
J. Am. Chem. Soc.
2014
, vol. 
136
 (pg. 
7273
-
7285
)
[PubMed]
42
Wang
 
Z. G.
Fast
 
W.
Benkovic
 
S. J.
 
Direct observation of an enzyme-bound intermediate in the catalytic cycle of the metallo-β-lactamase from Bacteroides fragilis
J. Am. Chem. Soc.
1998
, vol. 
120
 (pg. 
10788
-
10789
)
43
McManus-Munoz
 
S.
Crowder
 
M. W.
 
Kinetic mechanism of metallo-β-lactamase L1 from Stenotrophomonas maltophilia
Biochemistry
1999
, vol. 
38
 (pg. 
1547
-
1553
)
[PubMed]
44
Yang
 
H.
Aitha
 
M.
Hetrick
 
A. M.
Richmond
 
T. K.
Tierney
 
D. L.
Crowder
 
M. W.
 
Mechanistic and spectroscopic studies of metallo-β-lactamase NDM-1
Biochemistry
2012
, vol. 
51
 (pg. 
3839
-
3847
)
[PubMed]
45
Lisa
 
M. N.
Hemmingsen
 
L.
Vila
 
A. J.
 
Catalytic role of the metal ion in the metallo-β-lactamase GOB
J. Biol. Chem.
2010
, vol. 
285
 (pg. 
4570
-
4577
)
[PubMed]
46
Hawk
 
M. J.
Breece
 
R. M.
Hajdin
 
C. E.
Bender
 
K. M.
Hu
 
Z.
Costello
 
A. L.
Bennett
 
B.
Tierney
 
D. L.
Crowder
 
M. W.
 
Differential binding of Co (II) and Zn (II) to metallo-β-lactamase Bla2 from Bacillus anthracis
J. Am. Chem. Soc.
2009
, vol. 
131
 (pg. 
10753
-
10762
)
[PubMed]
47
Tioni
 
M. F.
Llarrull
 
L. I.
Poeylaut-Palena
 
A. A.
Marti
 
M. A.
Saggu
 
M.
Periyannan
 
G. R.
Mata
 
E. G.
Bennett
 
B.
Murgida
 
D. H.
Vila
 
A. J.
 
Trapping and characterization of a reaction intermediate in carbapenem hydrolysis by B. cereus metallo-β-lactamase
J. Am. Chem. Soc.
2008
, vol. 
130
 (pg. 
15852
-
15863
)
[PubMed]
48
Garau
 
G.
Bebrone
 
C.
Anne
 
C.
Galleni
 
M.
Frère
 
J.-M.
Dideberg
 
O.
 
A metallo-β-lactamase enzyme in action: crystal structures of the monozinc carbapenemase CphA and its complex with biapenem
J. Mol. Biol.
2005
, vol. 
345
 (pg. 
785
-
795
)
[PubMed]
49
Spencer
 
J.
Clarke
 
A. R.
Walsh
 
T. R.
 
Novel mechanism of hydrolysis of therapeutic β-lactams by Stenotrophomonas maltophilia L1 metallo-β-lactamase
J. Biol. Chem.
2001
, vol. 
276
 (pg. 
33638
-
33644
)
[PubMed]
50
Rasia
 
R. M.
Vila
 
A. J.
 
Mechanistic study of the hydrolysis of nitrocefin mediated by B. cereus metallo-β-lactamase
Arkivoc
2003
, vol. 
3
 (pg. 
507
-
516
)
51
Moali
 
C.
Anne
 
C.
Lamotte-Brasseur
 
J.
Groslambert
 
S.
Devreese
 
B.
Van Beeumen
 
J.
Galleni
 
M.
Frère
 
J.-M.
 
Analysis of the importance of the metallo-β-lactamase active site loop in substrate binding and catalysis
Chem. Biol.
2003
, vol. 
10
 (pg. 
319
-
329
)
[PubMed]
52
Griffin
 
D. H.
Richmond
 
T. K.
Sanchez
 
C.
Moller
 
A. J.
Breece
 
R. M.
Tierney
 
D. L.
Bennett
 
B.
Crowder
 
M. W.
 
Structural and kinetic studies on metallo-β-lactamase IMP-1
Biochemistry
2011
, vol. 
50
 (pg. 
9125
-
9134
)
[PubMed]
53
Yamaguchi
 
Y.
Kuroki
 
T.
Yasuzawa
 
H.
Higashi
 
T.
Jin
 
W.
Kawanami
 
A.
Yamagata
 
Y.
Arakawa
 
Y.
Goto
 
M.
Kurosaki
 
H.
 
Probing the role of Asp-120(81) of metallo-β-lactamase (IMP-1) by site-directed mutagenesis, kinetic studies, and X-ray crystallography
J. Biol. Chem.
2005
, vol. 
280
 (pg. 
20824
-
20832
)
[PubMed]
54
Yu
 
S.
Vosbeek
 
A.
Corbella
 
K.
Severson
 
J.
Schesser
 
J.
Sutton
 
L. D.
 
A chromogenic cephalosporin for β-lactamase inhibitor screening assays
Anal. Biochem.
2012
, vol. 
428
 (pg. 
96
-
98
)
[PubMed]
55
Kuzmič
 
P.
 
DynaFit: a software package for enzymology
Methods Enzymol.
2009
, vol. 
467
 (pg. 
247
-
280
)
[PubMed]
56
Fersht
 
A.
 
Enzyme Structure and Mechanism
1985
New York
W.H. Freeman
57
Schowen
 
K. J.
 
Gandour
 
R.
Schowen
 
R.
 
Solvent hydrogen isotope effects
Transition States of Biochemical Processes
1978
New York
Springer
(pg. 
225
-
283
)
58
Bounaga
 
S.
Laws
 
A.
Galleni
 
M.
Page
 
M.
 
The mechanism of catalysis and the inhibition of the Bacillus cereus zinc-dependent β-lactamase
Biochem. J.
1998
, vol. 
331
 (pg. 
703
-
711
)
[PubMed]
59
Badarau
 
A.
Page
 
M. I.
 
Enzyme deactivation due to metal-ion dissociation during turnover of the cobalt-β-lactamase catalyzed hydrolysis of β-lactams
Biochemistry
2006
, vol. 
45
 (pg. 
11012
-
11020
)
[PubMed]
60
Garrity
 
J. D.
Carenbauer
 
A. L.
Herron
 
L. R.
Crowder
 
M. W.
 
Metal binding Asp-120 in metallo-β-lactamase L1 from Stenotrophomonas maltophilia plays a crucial role in catalysis
J. Biol. Chem.
2004
, vol. 
279
 (pg. 
920
-
927
)
[PubMed]
61
Llarrull
 
L. I.
Fabiane
 
S. M.
Kowalski
 
J. M.
Bennett
 
B.
Sutton
 
B. J.
Vila
 
A. J.
 
Asp-120 locates Zn2 for optimal metallo-β-lactamase activity
J. Biol. Chem.
2007
, vol. 
282
 (pg. 
18276
-
18285
)
[PubMed]
62
Sharma
 
N. P.
Hajdin
 
C.
Chandrasekar
 
S.
Bennett
 
B.
Yang
 
K.-W.
Crowder
 
M. W.
 
Mechanistic studies on the mononuclear ZnII-containing metallo-β-lactamase ImiS from Aeromonas sobria
Biochemistry
2006
, vol. 
45
 (pg. 
10729
-
10738
)
[PubMed]
63
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
)
[PubMed]
64
Eyring
 
H.
 
The activated complex in chemical reactions
J. Chem. Phys.
1935
, vol. 
3
 pg. 
107
 
65
Laidler
 
K. J.
King
 
M. C.
 
Development of transition-state theory
J. Phys. Chem.
1983
, vol. 
87
 (pg. 
2657
-
2664
)
66
Spencer
 
J.
Walsh
 
T. R.
 
A new approach to the inhibition of metallo-β-lactamases
Angew. Chem. Int. Ed. Engl.
2006
, vol. 
45
 (pg. 
1022
-
1026
)
[PubMed]
67
Kaminskaia
 
N. V.
Spingler
 
B.
Lippard
 
S. J.
 
Intermediate in β-lactam hydrolysis catalyzed by a dinuclear zinc(II) complex: relevance to the mechanism of metallo-β-lactamase
J. Am. Chem. Soc.
2001
, vol. 
123
 (pg. 
6555
-
6563
)
[PubMed]
68
Desai
 
L. V.
Stowers
 
K. J.
Sanford
 
M. S.
 
Insights into directing group ability in palladium-catalyzed C-H bond functionalization
J. Am. Chem. Soc.
2008
, vol. 
130
 (pg. 
13285
-
13293
)
[PubMed]
69
Wang
 
L.
Moss
 
R. A.
Krogh-Jespersen
 
K.
 
Hammett analyses of halocarbene-halocarbanion equilibria
Org. Lett.
2013
, vol. 
15
 (pg. 
2014
-
2017
)
[PubMed]
70
Schramm
 
V. L.
 
Enzymatic transition states and transition state analog design
Annu. Rev. Biochem.
1998
, vol. 
67
 (pg. 
693
-
720
)
[PubMed]
71
Page
 
M. I.
 
The mechanisms of reactions of β-lactam antibiotics
Acc. Chem. Res.
1984
, vol. 
17
 (pg. 
144
-
151
)
72
Proctor
 
P.
Gensmantel
 
N. P.
Page
 
M. I.
 
The chemical reactivity of penicillins and other β-lactam antibiotics
J. Chem. Soc. Perkin Trans.
1982
, vol. 
2
 (pg. 
1185
-
1192
)
73
Hammett
 
L. P.
 
The effect of structure upon the reactions of organic compounds: benzene derivatives
J. Am. Chem. Soc.
1937
, vol. 
59
 (pg. 
96
-
103
)