Substrate Inhibition of Beta-Lactamases, a Method for Predicting Enzymatic Stability of Cephalosporins

Transcription

1 ANTIMICROBIAL AGZNTS AND CHIMOTHERAPY, Sept. 1976, p Copyright C 1976 American Society for Microbiology Vol. 10, No. 3 Printed in U.S.A. Substrate Inhibition of Beta-Lactamases, a Method for Predicting Enzymatic Stability of Cephalosporins DAVID F. MAHONEY, G. A. KOPPEL, AND J. R. TURNER* Lilly Research Laboratories, Eli Lilly & Co., Indianapolis, Indiana Received for publication 28 April 1976 Selected cephalosporins, including cefamandole, cephaloridine, cephaloglycin, and cefoxitin, were examined for their ability to inhibit the enzymatic activity of and act as substrates for beta-lactamases produced by Enterobacter cloacae and Staphylococcus aureus. Enzyme inhibition was determined by Michaelis-Menten kinetic measurements and by a spot plate assay using a chromogenic substrate (Glaxo compound 87/312). These two methods provide comparable estimates of kinetic parameters. Inhibition of beta-lactamase, as measured by these two methods, was generally found to correlate with resistance to hydrolysis and is proposed as a preliminary method of assessing susceptibility of cephalosporins to beta-lactamase hydrolysis. Four 7-aOCH3, 7-aH cephalosporin analogue pairs were also examined. The presence of the 7-aOCH3 substituent invariably resulted in reduced susceptibility to enzymatic hydrolysis, regardless of the other C7 substituent. The 7-aOCH3 compounds were also better inhibitors than were their 7-aH analogues, with the exception that 7-aOCH3 compounds having C7 adipic acid substituents were less inhibitory to the S. aureus enzyme than were the corresponding 7-aH analogues. Response of these two enzymes to 7-aOCH3 and 7-aH cephalosporins suggests that beta-lactamase hydrolysis of these compounds involves attack at the alpha side of the betalactam ring. 470 Two naturally occurring 7-a-methoxycephalosporins, 7-a-methoxycephalosporin C and deacetyl-3-o-carbamoyl-7-a-methoxycephalosporin C, were originally isolated from two species of Streptomyces by Nagarajan et al. (9). Both antibiotics exhibited greater activity against gram-negative organisms than did cephalosporin C or deacetyl-3-o-carbamoylcephalosporin C, their 7-hydrogen analogues, and the carbamoyloxymethyl derivative was reported (18) to be more active against gramnegative than gram-positive bacteria. Cefoxitin, a thienylacetyl derivative of deacetyl-3-0- carbamoyl-7-a-methoxycephalosporin C, has been reported to have activity against grampositive and gram-negative organisms (21). This 7-methoxylated cephalosporin generally showed resistance to the action of beta-lactamases from both gram-positive and gram-negative bacteria (14), and it was suggested that this characteristic contributed to the broad antimicrobial spectrum observed. There is considerable disagreement as to the relative importance of beta-lactamases in bacterial resistance to beta-lactam antibiotics (20). Because of the lack of uniform correlation between in vitro minimal inhibitory concentration values and lability of these compounds to enzymatic hydrolysis (1, 10, 11), this lability, by itself, is not sufficient evidence to predict bacterial susceptibility. Despite this, it is probable that the action of these enzymes does contribute to resistance although not always as the limiting factor. Given this assumption, the affinity of a betalactamase for a given cephalosporin should play a role in overall resistance of a beta-lactamaseproducing organism. The affinity (Ki) can be measured directly by using the compound as substrate or alternatively can be estimated from the Ki as determined from inhibition studies. Estimation of affinity by the latter method is particularly useful with beta-lactamases because many interesting cephalosporins are hydrolyzed at such low rates that accurate measurement ofkm values is not possible. The discovery of a chromogenic substrate for beta-lactamase (12), compound 87/312, has made this type of evaluation relatively straightforward since hydrolysis of 87/312 can be followed spectrophotometrically and is a good substrate for essentially all beta-lactamases. This report describes a spot plate assay for beta-lactamase inhibition and shows a comparison of the results to standard kinetic inhibitor studies. Data are also presented that show

2 VOL. 10, 1976 some effects of the 7-a-OCH3 substituent in cephalosporins on their stability and capacity to inhibit beta-lactamases. (This report was presented in part at the 15th Interscience Conference on Antimicrobial Agents and Chemotherapy, Washington, D.C., September 1975.) MATERIALS AND METHODS Organisms. Enterobacter cloacae 265A was obtained from the culture collection of Lilly Research Laboratories. The beta-lactamase-producing, methicillin-resistant Staphylococcus aureus was provided by M. Richmond and is a type A variant (16). Both organisms were maintained on Trypticase soy agar (BBL). Cultures for preparation of betalactamase were grown at 37 C in Trypticase soy broth (BBL). Antibiotics. Cephaloridine, cephaloglycin, cefamandole, D-O-formyl cefamandole, L-O-formyl cefamandole, and compounds II (3-carbamoyloxymethyl-7-[2-(thienyl)acetamido]-3-cephem-4-carboxylic acid), III (7-[2-carboxy-2-phenylacetamido]-7-amethoxycephalosporanic acid), and IV (7-[2-carboxy-2-phenylacetamido]cephalosporanic acid) were supplied by Lilly Research Laboratories. Cefoxitin (compound I) was obtained from Merck Sharp & Dohme, West Point, Pa. Cephalosporin C (compound VI) and compounds V (7-a-methoxycephalosporin C), VIII (deacetyl-3-0-carbamoylcephalosporin C), and VII were derived from or are fermentation products isolated at Lilly Research Laboratories. The chromogenic cephalosporin substrate compound 87/312 was obtained from Glaxo Research Ltd., Greenford, Middlesex, England. Beta-lactamase purification. The beta-lactamase (Richmond tiype I) frome. cloacae 265A was purified in a manner similar to that described for E. cloacae P99 (17), except that chromatography was performed on diethylaminoethyl-sephadex (Pharmacia Fine Chemicals, Inc., Piscataway, N.J.), ph 7.5, instead of QAE-Sephadex. The S. aureus betalactamase was purified according to a modification of a previous method (15). The modification consisted of the induction of midlog-phase cells with 1 mg of penicillin G (Eli Lilly & Co., Indianapolis, Ind.) per ml for 2.5 h at 37 C, adsorption of the enzyme on cellulose phosphate (grade P-li, Whatman), and subsequent elution with 2 M tris(hydroxymethyl)aminomethane-hydrochloride (Sigma Chemical Co., St. Louis, Mo.) buffer, ph 7.5. Fractions with detectable beta-lactamase activity were pooled, and bovine serum albumin (Calbiochem, Los Angeles, Calif.) was added to 0.07 mg of protein per ml. The enzyme was immediately dialyzed overnight (with one change) against 0.1 M sodium acetate buffer, ph 5.9. Portions (1 ml) were frozen and stored under liquid nitrogen. Protein determination. Protein was determined by the method of Lowry et al. (8). Enzymatic hydrolysis of cephalosporins. Hydrolysis rates of cephalosporins for both enzymes were determined by monitoring the optical density at the SUBSTRATE INHIBITION OF BETA-LACTAMASES 471 wavelength associated with maximum absorption of the beta-lactam ring (13). The maximum absorption varied slightly among the compounds examined, but all were within the range from 260 to 270 nm. The relative hydrolysis rate (RHR) is the rate as a percentage of the cephaloridine rate. The reaction mixture for determination of the hydrolysis rate consisted of 0.4 ml of bovine serum albumin (0.5 mg of protein per ml of solution in 0.05 M NaPO4, ph 7.0), 0.15 mg of substrate (dissolved in 0.05 M NaPO4, ph 7.0), an appropriate amount of enzyme to give linear rates, and sufficient 0.05 M NaPO4 buffer, ph 7.0, to give a total volume of 3.0 ml. The reaction mixture was placed in a 3-ml (1-cm light path) cuvette, and the optical density was monitored in a Gilford recording spectrophotometer equipped with thermal spacers to maintain temperature at 37 C. Hydrolysis rates were determined from the de*ease in optical density that occurred during the first 5 min of the reaction. Kinetic measurements. Determination of K{ for the various cephalosporins was made by assessing the ability of these antibiotics to inhibit the hydrolysis of compound 87/312. Reaction mixtures identical to that described above were used, except that 87/312 was used as the substrate and the optical density was measured at 482 nm. Inhibition constants (K1) were determined from either Lineweaver-Burk (7) or Dixon plots (2). Spot plate assay for determination of inhibition. Utilizing 87/312 as substrate, a spot plate assay was developed for the identification of cephalosporins that inhibit beta-lactamases. The assay is based on visual observation of the extent of hydrolysis of the chromogenic substrate 87/312 in the presence of a decreasing concentration of the putative inhibitor. Spot tests with 87/312 have previously been employed solely for the detection and quantitation of beta-lactamase activity (6, 12). Cephalosporin solutions were prepared at concentrations of 2 mg/ml (or greater) in 0.05 M NaPO4, ph 7.0, buffer. A portion (100,uI) ofthis solution was placed in a well of a spot plate (model 96U-WS; Linbro Chemical Co., Inc., New Haven, Conn.), and the cephalosporin concentration was serially diluted 1:5 by the sequential transfer, with an Eppendorf pipette, of 20 1.l of solution to wells containing 80,ul of 0.05 M NaPO4 buffer, ph 7.0. Enzyme (0.005 U) was then added to each well and mixed, and the mixture was incubated for 10 min at room temperature. One unit is the amount of enzyme required to gmol of 87/312 in 1 min at 37 C and ph hydrolyze A 20-,ul amount of the chromogenic substrate 87/ 312 (0.516 mg/ml in 0.05 M NaPO4, ph 7.0) was then added to each well and incubation was continued at room temperature. The final volume in each well was 120,ul. The end point for inhibition was read at 5 min after the addition of 87/312 and is defined as the highest concentration of the inhibitor that had no observable effect on the hydrolysis of 87/312. These end points are quite stable after a 5-min incubation period. To aid in end point determination, control wells that did not contain enzyme or inhibitor were included, and these were compared with test wells.

3 472 MAHONEY, KOPPEL, AND TURNER RESULTS Substrate and inhibitor specificity. Since 87/312 plays an important role in the determination of K{, as well as end point, the kinetic constants (Km and Vmax) for this compound are given in Table 1. The beta-lactamase from S. aureus that normally does not hydrolyze cephalosporins very rapidly can hydrolyze 87/312 at a relatively rapid rate. This substrate appears to be hydrolyzed at rates in excess of most cephalosporins that we have examined. The beta-lactamases from E. cloacae 265A and S. aureus were characterized by their substrate specificity and inhibition, by assay against a selected group of cephalosporins (Table 2). End points obtained from the spot plate assay were compared with Ki v4ues determined by kinetic measurements. The spot plate end points and Ki values are expressed as 10-6 M for each enzyme. End point and Ki values for the E. cloacae 265A and S. aureus enzymes were either very close or within fourfold of each other for a given compound. Since serial 1:5 dilutions of the inhibitors were used, this correlation is well within the resolution of the spot plate assay and shows good agreement between the spot plate and kinetic measurements. Significant quantitative differences in inhibition between the two enzymes were observed in the presence of some compounds. Cephaloridine and cefamandole are potent inhibitors of the enzyme as judged by low end point and K{ values, although cephaloridine is a relatively poor inhibitor of the 265A enzyme, with an end point of 6.37 x 10-4 M and a Ki value of 9.68 x 10-4 M. The D and L isomers of O-formyl cefa- TABLE 1. lactamases from strains 265A and Source of 3-lactanase Km (AM) Vmaxa E. cloacae 265A S. aureus Kinetics of hydrolysis of 87/312 by,& a Micromoles per minute per milligram of protein. ANTIMICROB. AGENTS CHEMOTHER. mandole did not show large differences in ability to inhibit the enzyme although a fivefold difference in end point appears for the 265A enzyme, indicating that a change in configuration from L to D has a slightly greater effect on the 265A enzyme. Cephaloglycin, which is quite resistant to enzymatic hydrolysis, shows the lowest affinity for the enzyme while having a similar affinity for the 265A enzyme as do three other compounds in this group. The inhibition of these two enzymes is also qualitatively different. The enzyme was competitively inhibited by all the cephalosporins in this group. The 265A enzyme, however, showed competitive inhibition only by cephaloridine, whereas the other four compounds acted as noncompetitive inhibitors. Since all of these compounds show some degree of affinity for beta-lactamases, one might be able to predict beta-lactamase stability of beta-lactams by measuring their ability to inhibit these enzymes. In Table 2, the RHRs are given for these cephalosporins. For comparison, the hydrolysis rates were normalized to the hydrolysis rate ofcephaloridine. There is a good correlation between the ability of a compound to inhibit the 265A enzyme and the lability of that compound to hydrolysis. That is, those compounds that are good substrates, such as cephaloridine and D-O-formyl cefamandole, tend to be poor inhibitors as judged by high Ki values and high spot plate end points, whereas cephaloglycin, cefamandole, and L-O-formyl cefamandole are poor substrates but good inhibitors. This coyrelation does not appear to hold in all instances for the enzyme. For example, the data show approximately equal RHRs for cephaloridine and D--formyl cefamandole, although the Ki values are 100-fold different. We do not have sufficient data as yet to know whether this is a general characteristic of the enzyme. Stability studies of 7-aOCH3 cephalosporins. The usefulness of determining hydrolysis rates and kinetic parameters in the study of the interaction between beta-lactamases and beta- TABLE 2. Comparison of spot plate end points, K*, and RHR E. cloacae 265A S. aureus Compound End point K, (PM) RHR End point K, (PM) RHR (itm) KIM) RR(PM) K,M H Cephaloridine a a Cephaloglycin Cefamandole D-O-formyl cefamandole L-O-formyl cefamandole a Micromoles per minute per milligram of protein relative to the rate obtained with cephaloridine.

4 VOL. 10, 1976 SUBSTRATE INHIBITION OF BETA-LACTAMASES 473 lactam antibiotics is shown in Table 3. The structure of each cephalosporin is given in Fig. 1. The results (Table 3) show the effect of the 7- aoch3 group on hydrolysis by and inhibition of the beta-lactamases from strains 265A and Each 7-aOCH3 cephalosporin is paired with its 7-aH analogue. As seen before, spot plate end points correlated well with K, values obtained for the 265A enzyme and reasonably well with the '13136 enzyme. The 7-aOCH3 cephalosporins, compounds V and VII, are weak inhibitors of the enzyme, whereas the 265A enzyme was inhibited by both compounds, with Ki values of 1.54 x 10-4 and 1.21 x 10-4 M, respectively. The 7-aH analogues, compounds VI (cephalosporin C) and VIII, are essentially ineffective inhibitors of the enzyme (Ki values could not be determined), with end point values of 1.61 x 10-2 M as opposed to end point values of about 6 x 10-4 M for their 7- aoch3 analogues, compounds V and VII. Following the pattern set by the other compounds, Ki values for compounds VI (cephalosporin C) and VIII were expected to be up to fivefold greater than their end point values, but could not be determined due to the insensitivity of the enzyme to these compounds. A Ki value for compound III could not be determined due to the limited solubility of the compound. Substitution of -aoch3 for -ah in this group of compounds results in an increased affinity for the 265A enzyme. The magnitude of this change varies from 100-fold for compound I (cefoxitin) and its 7-aH analogue (compound II) to TABLE 3. Comparison of spot plate end points, Ki, and RHR, of 7-methoxycephalosporins and their 7- hydrogen analogues E. cloacae 265A S. aureus Compound Enpon K1(M )End pomt Ki (M) RHR End point (,um) K (,um) RHR I. 7-aOCH a a II. 7-aH < III. 7-aOCH3 < IDb 0 IV. 7-aH V. 7-aOCH ,489 0 VI. 7-aH ,103 NDC 79 VII. 7-aOCH ,192 0 VIII. 7-aH ,064 ND 92 a Micromoles per minute per milligram of protein relative to the rate obtained with cephaloridine (100). bid, Indeterminate (see text). c ND, Not determined. 0 OCH3 s>ch2-c-nh4 1 N 0 CH2-0-C-NH2 (CEFOXITIN) 0 Cj-CH2-C-NH S O o,, N CH2-0-C-NH2 11 o OCH HOOC-CH-(CH2)3-C-NH S IO NH2 N CH2-O-C-CH3 V 0 0j OCH3 CH-C-NH 0 N - CH2-0-C-CH3 111 a CH-C-NH 0 > N, CH2-O-C-CH3 Hooc-c-cH1> IV 0 OCH 3 0-H4 HOOC-CH-(CH2)3-C-NH 11-~ NH2 N - CH2-O-C-NH2 VIl 0 HOOC - CH -(CH2)3-C- N H 'T-CN -r-- HOOC-CH-(CH2)3-C-NH 1 NH2 I/ N-N CH2-O-C-CH2 NH2 TN' - CH2-0-C-NH2 VI (CEPHALOSPORIN C) Vlill FIG. 1. Structure of the 7-a-methoxycephalosporins and their 7-a-hydrogen analogues.

5 474 MAHONEY, KOPPEL, AND TURNER about 2-fold for the compounds containing an adipic acid side chain. A similar increased affinity for the enzyme is seen with 7- caoch3-substituted compounds, providing that the adipic acid chain is also present. Affinity for the enzyme is reduced by the -aoch3 group when substituted on compounds having side chains other than adipic acid, that is, compounds I (cefoxitin) and III. As opposed to the mixed response toward affinity caused by the substitution of 7-aH with 7-aOCH3, the effect of this change on hydrolysis appears to be straightforward. In all cases, the compound containing the 7-aOCH3 group is more resistant to hydrolysis than is its 7-aH analogue. The type of inhibition found for both enzymes differs for this group of compounds (Table 4). For the first two analogue pairs, the 7-aH analogues (compounds II and IV) showed noncompetitive inhibition of the 265A enzyme and competitive inhibition of the enzyme. Compound I (cefoxitin) showed competitive inhibition with both enzymes. Unfortunately, the 7- aoch3 analogue of the second pair (compound III) could not be classified because of a nonlinear reaction rate in the presence of this compound. With the weaker inhibitors, i.e., the third and fourth analogue pairs, both 7-aOCH3 and 7-aH compounds (compounds V through VIII) were competitive inhibitors of the 265A enzyme, whereas the 7-aOCH3 analogues (compounds V and VII) showed noncompetitive inhibition of the enzyme. Inhibition of the enzyme by the 7-aH analogues of these two pairs (compounds VI and VIII) could not be classified because of the low affinity of these two compounds for the enzyme, as shown by the spot plate end points in Table 3. DISCUSSION As a preliminary step in the analysis ofa new group of cephalosporins, the spot plate assay appears useful for predicting beta-lactamase stability since those compounds that were good inhibitors generally were stable to enzymatic hydrolysis. Exceptions to this were the responses of both enzymes to compound V and compound VII, where the Ki or end point (when Ki could not be determined) indicated weak inhibition (_10-4 M) while the relative hydrolysis indicated good stability (<50% of cephaloridine). Another exception was found with both enzymes to compound II, where the Ki indicated good inhibition, but RHRs were high. Two other compounds, cephaloridine and D-0- formyl cefamandole, showed the same response to the enzyme, that is, good inhibition but ANTimicRoB. AGENTS CHZMMHZR. TBnLz 4. Inhibition of265a and 13136,& lactamases by 7-aOCH3 cephalosporins and their 7- ah analogues Type of inhibition Compound E. cloacae 265A S. aureus I. 7-aOCH3 Competitive Competitive II. 7-aH Noncompetitive Competitive III. 7-aOCH3 Non-Michaelis- IDa Menten IV. 7-aH Noncompetitive Competitive V. 7-aOCH3 Competitive Noncompetitive VI. 7-aH Competitive NDb VII. 7-aOCH3 Competitive Noncompetitive VIII. 7-aH Competitive ND a ID, Indeterminate (see text). b ND, Not determined. relatively high hydrolysis rates. Overall, the hydrolysis rates for 10 of the 13 compounds correlated with affinity for the 265A enzyme, whereas hydrolysis rates for 8 of 13 compounds correlated with affinity for the enzyme. There is, of course, no a priori reason for a slow reaction velocity to be correlated with high affinity or vice versa. However, within the context of this study and generally with respect to the examination of beta-lactamases in which the numbers of potential substrates and enzymes are great, the observation that high affinity is generally associated with resistance to hydrolysis allows one to use a qualitative assay for inhibition to predict hydrolysis. The reaction velocities and affinities found for the Enterobacter enzyme are in good agreement with preliminary studies of other gramnegative enzymes (D. Mahoney, unpublished data). The generally low activity of S. aureus enzymes against cephalosporins (12) probably is responsible for the slightly poorer correlation between affinity and reaction velocities found with this enzyme. Substitution of 7-aH with 7-aOCH3 in the cephalosporins examined here has a direct and unequivocal effect on enzymatic hydrolysis in that the 7-aOCH3 analogue was always more resistant to enzymatic degradation. However, studies on the chemical hydrolysis of these compounds show that addition of a 7-aOCH3 group has no effect on the stability of cephalosporins to nucleophiles because the beta-face of the beta-lactam ring remains open to attack (5). In contrast, the addition of a 6-aOCH3 group to a penicillin results in chemical stability since both faces of the ring are then blocked. Since the rate of enzymatic hydrolysis is severely reduced by the addition of 7-aOCH3 to cephalosporins, it seems reasonable that enzymatic at-

6 VOL. 10, 1976 tack may originate from the a-face rather than the beta-face of the beta-lactam ring and further that the beta-face may be inaccessible when the compound is bound to the enzyme. Other related reports (3, 4) show that 7- aoch3-substituted cephalosporins are 10-fold better inhibitors ofescherichia coli transpeptidase than are their 7-aH analogues. The finding in this study that 7-aOCH3-substituted cephalosporins invariably have greater affinity for the Enterobacter beta-lactamase and usually have greater affinity for the S. aureus beta-lactamase suggests again (19) the possibility that active centers on transpeptidase and beta-lactamase enzymes may share similar configurations. Therefore, a study of the mechanism by which compounds inhibit beta-lactamases may provide important information on the capability of these compounds to inhibit the activity of transpeptidases. ACKNOWLEDGMENTS We wish to thank Joseph M. Indelicato for his helpful discussions during preparation of this manuscript. LITERATURE CITED 1. Cornelis, G Distribution of p8-lactamases A and B in some groups of Yersinia enterocolitica and their role in resistance. J. Gen. Microbiol. 91: Dixon, M The determination of enzyme inhibitor constants. Biochem. J. 55: Ho, P. P. K., R. D. Towner, J. M. Indelicato, W. A. Spitzer, and G. A. Koppel Biochemical and microbiological studies on 6-substituted penicillins. J. Antibiot. 25: Ho, P. P. K., R. D. Towner, J. M. Indelicato, W. L. Wilham, W. A. Spitzer, and G. A. Koppel Biochemical and microbiological studies on 7-methoxycephalosporins. J. Antibiot. 26: Indelicato, J. M., and W. L. Willm Effect of 6-a substitution in penicillins and 7-a substitution in cephalosporins upon,b-lactam reactivity. J. Med. Chem. 17: Kammer, R. B., D. A. Preston, J. R. Turner, and L. C. Hawley Rapid detection of ampicillin-resistant Haemophilus influenzae and their susceptibility to sixteen antibiotics. Antimicrob. Agents Chemother. 8: Lineweaver, H., and D. Burk The determination SUBSTRATE INHIBITION OF BETA-LACTAMASES 475 of enzyme dissociation constants. J. Am. Chem. Soc. 56: Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: Nagarajan, R., L. D. Boeck, M. Gorman, R. L. Hamill, C. E. Higgens, M. M. Hoehn, W. M. Stark, and J. G. Whitney ,B-Lactam antibiotics from Streptomyces. J. Am. Chem. Soc. 93: Neu, H. C Cefoxitin, a semisynthetic cephamycin antibiotic: antibacterial spectrum and resistance to hydrolysis by gram-negative beta-lactamases. Antimicrob. Agents Chemother. 6: Neu, H. C Cefamandole, a cephalosporin antibiotic with an unusually wide spectrum of activity. Antimicrob. Agents Chemother. 6: O'Callaghan, C. H., A. Morris, S. M. Kirby, and A. H. Shingler Novel method for detection of(-lactamases by using a chromogenic cephalosporin substrate. Antimicrob. Agents Chemother. 1: O'Callaghan, C. H., P. W. Muggleton, and G. W. Ross Effects of,b-lactamase from gram-negative organisms on cephalosporins and penicillins, p Antimicrob. Agents Chemother Onishi, H. R., D. R. Daonst, S. B. Zimmerman, D. Hendlin, and E. 0. Stapley Cefoxitin, a semisynthetic cephamycin antibiotic: resistance to betalactamase inactivation. Antimicrob. Agents Chemother. 5: Richmond, M. H Purification and properties of the exopenicillinase from Staphylococcus aureus. Biochem. J. 88: Richmond, M. H Wild-type variants of exopenicillinase from Staphylococcus aureus. Biochem. J. 94: Ross, G. W., and M. G. Boulton Purification of /3- lactamases on QAE-Sephadex. Biochim. Biophys. Acta 309: Stapley, E. O., M. Jackson, S. Hernandez, S. B. Zimmerman, S. A. Currie, S. Mochales, J. M. Mata, H. B. Woodruff, and D. Hendlin Cephamycins, a new family of /3-lactam antibiotics. I. Production by Actinomyces, including Streptomyces lactamdurans sp. n. Antimicrob. Agents Chemother. 2: Tipper, D. J., and J. L. Strominger 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. 54: Tsang, J. C., G. A. Sansing, and M. A. Miller Relation of beta-lactamase activity to antimicrobial susceptibility in Serratia marcescens. Antimicrob. Agents Chemother. 8: Wallick, H., and D. Hendlin Cefoxitin, a semisynthetic cephamycin antibiotic: susceptibility studies. Antimicrob. Agents Chemother. 5:25-32.