Escherichia coli Resistant to Cephalosporins and Quinolones Is Still

Transcription

1 ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, May 1991, p /91/ $02.00/0 Copyright 1991, American Society for Microbiology Vol. 35, No. 5 Escherichia coli Resistant to Cephalosporins and Quinolones Is Still Susceptible to the Cephalosporin-Quinolone Ester Ro J. PACE, A. BERTASSO, AND N. H. GEORGOPAPADAKOU* Roche Research Center, Hoffmann-La Roche, Inc., Nutley, New Jersey Received 23 August 1990/Accepted 5 March 1991 Ro is a broad-spectrum antibacterial agent consisting of a cephalosporin (desacetylcefotaxime) linked through an ester bond to a fluoroquinolone (fleroxacin). Its activity against mutants of Escherichia coli TE18 resistant to both antibacterial components was examined. E. coli TE18 overproduces the AmpC,I-lactamase and is resistant to several cephalosporins, including desacetylcefotaxime (MIC, 50,ug/ml), although it is still susceptible to Ro (MIC, 0.2,ug/ml). Thirty-five spontaneous, two-step mutants of E. coli TE18 which were resistant to fleroxacin (MIC, 50,ig/ml) were isolated. In the mutants, replicative DNA biosynthesis (permeabilized cells) was resistant to fleroxacin, and some mutants had porin abnormalities. However, all remained susceptible to Ro (MIC, 0.5,ig/ml). Examination of,b-lactamase activity in the parent strain revealed that it hydrolyzes desacetylcefotaxime more rapidly than it does Ro Thus, Ro may be less susceptible to the gram-negative, chromosomal,-lactamases that hydrolyze several broad-spectrum cephalosporins and may be effective in cases in which neither of its two components is active. Ro is a synthetic, broad-spectrum antibacterial agent consisting of a cephalosporin (desacetylcefotaxime) linked through an ester bond to a fluoroquinolone (fleroxacin). Ro combines the antibacterial spectrum and potency of its two components (2, 3, 10, 14). It binds to essential penicillin-binding proteins (PBPs) of Escherichia coli and Staphylococcus aureus (10). It also inhibits replicative DNA biosynthesis and produces filamentation in E. coli (10). However, it is unclear whether the intact molecule is responsible for the antibacterial activity. It has been suggested that Ro initially acts as a cephalosporin but that upon decomposition free fleroxacin is released and quinolone activity appears (10). Considering that a low level of fleroxacin is present in the sera of animals receiving Ro , it is plausible that, in vivo, the agent acts as a quinolone prodrug (14). However, Ro may be more resistant to animal esterases than are esters of other cephalosporins and P-lactams (11, 15, 26), as it is excreted predominantly intact in the urine of primates (14). This study was undertaken to determine whether intact Ro can be responsible for the antibacterial effect in E. coli. To that end, fleroxacin-resistant mutants were selected from E. coli TE18. This strain is already resistant to the cephalosporin component, desacetylcefotaxime, because of AmpC P-lactamase overproduction (20) (Table 1). Therefore, the mutants selected would be resistant to both decomposition products of Ro , and any antibacterial activity should arise from the intact molecule. The parent strain, E. coli TE18, and the mutants were subsequently examined with respect to porn expression, permeability, P-lactamase activity, PBP profile, replicative DNA biosynthesis, and antibiotic susceptibility profile. MATERIALS AND METHODS Bacterial strains and bacteriophages. E. coli TE18, a K-12 strain containing the ampc gene and overproducing chromosomal 1-lactamase (20), was kindly provided by S. Normark of Washington University (St. Louis, Mo.). E. coli ATCC * Corresponding author , a reference strain for MIC determinations, was purchased from the American Type Culture Collection (Rockville, Md.). E. coli JF568, JF701, and JF703, used as reference strains for porin expression (23), were kindly provided by J. Foulds of the National Institutes of Health (Bethesda, Md.). E. coli DCO, a reference strain for PBP analysis, was a gift of D. Clark of Southern Illinois University (Carbondale). Bacteriophages SS4 and K20, used for porin identification (24), were kindly provided by C. Schnaitman of Arizona State University (Tempe). Antibiotics. Ro , desacetylcefotaxime (Ro ), and fleroxacin (Ro ) were obtained from Roche Laboratories (Nutley, N.J.). Cefoxitin and norfloxacin were from Merck Sharp & Dohme Research Laboratories (Rahway, N.J.), ofloxacin was from Ortho Pharmaceutical Corp. (Raritan, N.J.), pefloxacin was from Rhone- Poulenc Pharmaceuticals (Monmouth Junction, N.J.), ciprofloxacin was from Miles Inc., Pharmaceuticals Division (West Haven, Conn.), cefotaxime was from Hoechst-Roussel Pharmaceuticals Inc. (Somerville, N.J.), cefazolin was from SmithKline Beecham (Philadelphia, Pa.), cephalothin, cephalexin, and cefaclor were from Eli Lilly & Co. (Indianapolis, Ind.), amifloxacin was from Sterling-Winthrop Institute (Rensselaer, N.Y.), and cephaloridine was from Glaxo Group Research, Ltd. (Greenford, United Kingdom). Chloramphenicol and tetracycline were from Sigma Chemical Co. (St. Louis, Mo.). Isolation of fleroxacin-resistant mutants. First-step fleroxacin-resistant mutants were isolated from E. coli TE18 after spread-plating log-phase (A6. = 0.6) cells (1.3 x 108 CFU per plate) on Antibiotic Medium 3 agar (AB3; Difco Laboratories, Detroit, Mich.) containing 1 jig of fleroxacin per ml and then incubating for 24 h at 37 C. Second-step mutants were selected in a similar manner from the most resistant first-step mutant (designated JP28) on plates containing 20 jig of fleroxacin per ml. Mutants were characterized by using Enterotube II (Roche Diagnostic Systems). Antibiotic susceptibility. Susceptibilities of the strains to the compounds were determined by the broth microdilution method (200 RI [104 CFU/ml] per well) with AB3. The MIC

2 VOL. 35, 1991 E. COLI SUSCEPTIBLE TO Ro TABLE 1. Susceptibility of E. coli to Ro and related antibiotics MIC (>Lg/ml) Strain Ro Desacetyl cefotaxime Cefotaxime Fleroxacin ATCC ' JF ' TE JP JP JP JP JP JP JP JP JP JP JP JP JP JP JP JP JP JP JP Strain TABLE 2. was the concentration which inhibited visible growth after 18 h of incubation at 37 C (10). Bacteriophage susceptibility. Susceptibility to OmpF- or OmpC-specific bacteriophages was determined, as an indicator of porin expression in E. coli TE18 and the fieroxacinresistant mutants. Mid-logarithmic phase cells (200,ul), grown in modified Luria broth (MLB; Luria broth lacking NaCl but containing 0.25% glucose and 2.5 mm KPO4, ph 7.5) to enhance OmpF production, were mixed with 2 ml of molten soft agar (0.75% agar) and poured onto the surface of an MLB agar plate (1.5% agar). Bacteriophage stocks were diluted with SM buffer (0.1 M NaCl, 17 mm MgSO4, 5 mm Tris-HCl [ph 7.5], 0.01% gelatin) (17), and 5-,ul aliquots were spotted onto the seeded soft agar overlays. Plates were incubated at 37 C, and the PFU were counted after 24 h. Plaquing efficiency was calculated by dividing the number of PFU/ml obtained with the mutant strain by the number of PFU/ml obtained with E. coli JF568. Biochemical assays. Outer membrane proteins were prepared by Sarkosyl extraction of bacterial cells and separated in 8 M urea-sodium dodecyl sulfate (SDS)-polyacrylamide gels (16). Permeability and,-lactamase activities were determined as described by Zimmermann and Rosselet (30) and by Nikaido and Normark (20). Quinolone uptake was monitored fluorimetrically, as the amount of cell-associated drug, by the method of Chapman and Georgopapadakou (6). PBP binding was measured in Triton X-100-solubilized membranes as inhibition of [14C]penicillin G binding by the test compound (10); proteins were separated by SDS-polyacrylamide gel electrophoresis, and PBPs were detected by fluorography. Replicative DNA biosynthesis, an indicator of gyrase activity, was measured as ATP-dependent incorporation of [3H]thymidine into trichloroacetic acid-insoluble material by toluene-treated cells (18). RESULTS Isolation and characterization of fleroxacin-resistant strains. Spontaneous fleroxacin-resistant mutants were selected from E. coli TE18 at a frequency of 1V7 mutants per Susceptibility of E. coli to quinolones, tetracycline, and chloramphenicol MIC (,ug/ml) Pefloxacin Norfloxacin Ciprofloxacin Amifloxacin Ofloxacin Tetracycline Chloramphenicol ATCC '0.1 so.1 so.1 s JF s0.1 so.1 so JE <O.1 s0.1 s JP JP JP JP JP JP JP JP JP JP JP JP JP JP JP JP JP JP JP

3 912 PACE ET AL. TABLE 3.,B-Lactamase activity and permeability of E. coli TE18' V.. Km P (cmls) Compound (nmollminlmg)b (qm)b (1o05) Ro ± ± ± Desacetyl ± ± ± 0.00 cefotaxime Cefotaxime 0.40 ± ± ± 0.00 a All values are expressed as mean ± standard deviation. b Sonicated cells (dry weight). c P, Permeation. CFU in the presence of 1,i.g of fleroxacin per ml (first-step mutants). Mutants were further selected for higher levels of fleroxacin resistance at a frequency of 4.0 x 10' mutants per CFU in the presence of 20,ug of fleroxacin per ml (second-step mutants). Antibiotic susceptibilities of the fleroxacin-resistant mutants are shown in Tables 1 and 2. Predictably, the greatest increase in MICs was that of fleroxacin, where 50- and 500-fold increases were found for the first-step (JP28 to JP225) and second-step (JP28-1 to JP28-13) mutants, respectively. The first-step mutants were cross-resistant to other quinolones (Table 2). The second-step mutants showed specific resistance to fleroxacin and ofloxacin; increases in MICs of other quinolones were about twofold (Table 2). With cephalosporins other than cefotaxime, MICs for the mutants did not increase greatly relative to those for E. coli TE18. For E. coli TE18, cephalosporin MICs ranged from 50 to >200,ug/ml, while for strains ATCC and JF568, MICs were approximately 0.1,ug/ml (data not shown). MICs of chloramphenicol and tetracycline were unchanged (Table 2). P-Lactamase activity and outer membrane permeability. 13-Lactamase in sonicated cells of E. coli TE18 had an apparent Km for Ro and desacetylcefotaxime substantially lower than that for cefotaxime (Table 3), which was in turn lower than the Km for cephaloridine or cephalothin (Table 4). Significantly, the hydrolysis rate of desacetylcefotaxime was more than 30 times greater than that of Ito (Table 3). Cefotaxime was hydrolyzed at a rate approximately one-third that of Ro , while cephaloridine and cephalothin were hydrolyzed at much higher rates. Km values for cephaloridine and cephalothin remained unchanged in the first-step (JP28) and second-step (JP28-1) mutants (Table 4). Outer membrane permeations of cephaloridine and cephalothin were also similar in E. coli TE18, JP28, and JP28-1 (Table 4). The ability of Ro to permeate E. coli TE18 was substantially lower than the abilities of the related drugs cefotaxime and desacetylcefotaxime to do so (Table 3). Porin expression. Outer membrane protein profiles and ANTIMICROB. AGENTS CHEMOTHER. susceptibilities to porin-specific bacteriophages indicated that the parent strain, TE18, had reduced levels of the OmpF porin relative to the levels in the reference strain, E. coli JF568 (Fig. 1 and Table 5). Still, most of the fleroxacinresistant mutants expressed less than 7% of the OmpF protein found in TE18 (Table 5). Quinolone uptake. There was a slight reduction in fleroxacin uptake in the first-step mutant JP28, and there was no reduction in uptake in the second-step mutant JP28-1 (data not shown). PBP profiles. PBPs of E. coli TE18 and the fleroxacinresistant mutants were similar to those of E. coli DCO (Fig. 2). Interestingly, fluorograms of polyacrylamide gels of TE18 as well as of the mutants showed an extra PBP with a molecular weight of approximately 37,000. Replicative DNA biosynthesis. Replicative DNA biosynthesis was resistant to fleroxacin, pefloxacin, and Ro in strains JP28, JP28-1, and JP28-2, compared with E. coli TE18 (Table 6). The 50% inhibitory concentrations of these compounds were 10- to 100-fold higher for the fleroxacinresistant mutants. The increase was most pronounced in the first-step mutants and only slight in the second-step mutants. DISCUSSION The cephalosporin 3'-quinolone ester Ro is a broad-spectrum antibacterial agent active both in vitro and in vivo (2, 3, 14). The compound is bactericidal, free of inoculum effects, and active against cephalosporin- and quinolone-resistant organisms. Previous in vitro studies suggested that Ro acts both as a cephalosporin and as a quinolone prodrug (10). However, these studies were complicated by the fact that Ro hydrolyzes to yield desacetylcefotaxime and fleroxacin, both of which possess antibacterial activity of their own. In fact, in many cases it is not entirely clear whether the intact molecule or its degradation products are responsible for the antibacterial activity (2, 14). The present study sought to address this issue by examining E. coli mutants resistant to both the cephalosporin and quinolone components in order to examine antibacterial activity due solely to the intact Ro The parent strain used in the study, E. coli TE18, is a K-12 strain previously reported to overproduce a chromosomally encoded AmpC,3-lactamase and to have reduced susceptibility or outright resistance to a number of cephalosporins (20). We found it to be highly resistant to desacetylcefotaxime, though not to Ro or cefotaxime (Table 1). In addition to the P-lactamase overproduction, the strain produced reduced amounts of the OmpF porin relative to the amount in E. coli JF568 (Table 5 and Figure 1), which might contribute to its cephalosporin resistance. Its PBP profile, though typical for an E. coli strain, included an extra 37,000-Da PBP which is probably the AmpC enzyme (4). Fleroxacin-resistant mutants were isolated from E. coli TABLE 4.,-Lactamase activity and permeability of E. coli TE18, JP28, and JP28-1a Cephaloridine Cephalothin Strain V.. Km P (cm/s) Vma Km P (cm/s) (,umol/min/mg) (>LM) (10-5) (,umol/min/mg) (AM) (10-5) TE ± ± ± ± ± 0.26 JP ± ± ± ± ± ± 0.03 JP ± ± ± ± ± ± 0.25 a All values are expressed as mean ± standard deviation. P, Permeation.

4 VOL. 35, 1991 E. COLI SUSCEPTIBLE TO Ro OmpA* FIG. 1. Outer membrane protein profiles of E. coli strains. (A) Lanes a to h, strains JF568, JF703, TE18, JP57, JP63, JP155, JP204, and JP225, respectively. (B) Lanes a to g, strains TE18, JP28, JP28-1, JP28-2, JP28-3, JP28-4, and JP28-5, respectively. TE18 at a frequency higher than that previously reported for fleroxacin in E. coli JF568 (5) or for other quinolones (12, 13). Predictably, the first-step mutants were cross-resistant to other quinolones, but surprisingly, the second-step mutants were cross-resistant only to the tricyclic quinolone ofloxacin. Resistance to amifloxacin was slightly reduced in the second-step mutants. The nature of this selectivity is at present unclear. TABLE 5. Strain Plaquing efficiency of OmpF- and OmpC-specific bacteriophages on E. coli strains SS4 a Plaquing efficiency of: K20b JF JF701 <1.0 X JF x 10-5 TE X 10-5 JP x 10-5 JP x 10-5 JP x 10-5 JP <1.0 X 10-5 JP <1.0 x i0-5 JP x l0-5 JP x 10-5 JP <1.0 x io-5 JP <1.0 x 10i- JP X 10-5 JP x 10-5 JP x 10-5 JP x 10-5 JP x 10-5 JP X lo-5 JP x 10-5 JP <1.0 x io-5 JP x i0-5 JP x lo-5 a OmpC-specific bacteriophage. b OmpF-specific bacteriophage. FIG. 2. Binding of Ro (A) and desacetylcefotaxime (B) to PBPs of E. coli TE18 and the fieroxacin-resistant mutants JP28 and JP28-1. E. coli DCO is also included for comparison. Antibiotic concentrations (in micrograms per milliliter) are indicated at the top. Arrowheads mark the position of the extra 37,000-Da PBP. In all fleroxacin-resistant mutants examined, the amounts of OmpF porin were reduced relative to the amount in TE18, and replicative DNA biosynthesis, an index of DNA gyrase activity, was resistant to quinolones. Similar effects have been shown in other quinolone-resistant gram-negative bacteria (1, 4, 7-9, 27, 28). The clinical implications of OmpF porin reduction are unclear, as OmpF expression, and thus OmpF-associated permeability, is normally repressed by the high osmolarity and temperature which prevail in infections (19). Ro had reduced outer membrane permeation in E. coli TE18 compared with other P-lactam antibiotics (Tables 3 and 4). This could be due to its large size (molecular TABLE 6. Inhibition of replicative DNA biosynthesis in E. coli TE18 and fleroxacin-resistant mutants by Ro , fleroxacin, and pefloxacin IC50 (gml)a Strain Ro Fleroxacin Pefloxacin TE18 20 (22) 0.8 (2.2) 1.2 (3.6) JP28 >200 (>220) 60 (165) 26 (79) JP28-1 >200 (>220) 125 (339) 52 (155) JP28-2 >200 (>220) 47 (129) 40 (120) a IC50, 50% inhibitory concentration. Numbers in parentheses are the micromolar concentrations.

5 914 PACE ET AL. weight, 732) and to the steric limitations posed by the aminothiazole-methoxyimino side chain (29). Still, Ro must be able to pass through the outer membrane, as both the parent strain and the mutants were highly susceptible to it (Table 1). The agent most likely enters through the OmpF porin channel (10). Entry through the narrower OmpC porin channel might also be possible if Ro can assume a cylindrical conformation; such a conformation would permit a solute to enter through a channel with a radius smaller than the Stokes' radius of the solute (21). Ro probably does not pass through the outer membrane via the self-promoted pathway proposed for quinolones (6), as there was no antagonism by magnesium (data not shown). Also, the lack of potentiation by transferrin under conditions in which MICs of catechol cephalosporins are reduced 10-fold (data not shown) precludes entry of Ro via the iron uptake pathway proposed for catechol-containing,3-lactams (22). The P-lactamase stability of Ro (Table 3) would permit it to diffuse intact through the periplasmic space and reach its target PBPs in E. coli TE18. The activity of Ro against desacetylcefotaxime-resistant, fleroxacinresistant E. coli is thus most likely due to its reduced hydrolysis by P-lactamase and its affinity for PBPs, which make it effective despite its poor ability to permeate the outer membrane (10). The interplay of PBP affinity, outer membrane permeability, and,-lactamase stability in producing the overall antibacterial effect has been recently studied quantitatively in E. coli for a number of 1-lactam antibiotics (20). The relative stability of Ro against AmpC which 1-lactamase may translate to other enterobacteria in cephalosporins are hydrolyzed by chromosomal P-lactamases (25). In conclusion, the susceptibility of E. coli strains to Ro but not to its degradation products, desacetylcefotaxime and fleroxacin, has allowed us to examine the activity of the intact molecule in a well-characterized organism. Ro acts in this case as a 1-lactamase-resistant cephalosporin. The present study could have clinical implications if the findings in preliminary animal studies suggesting that Ro is eliminated from the body largely intact (6a) are confirmed in humans. ACKNOWLEDGMENTS We thank D. Clark, J. Foulds, S. Normark, and C. Schnaitman for their generous gifts of E. coli strains and bacteriophages and D. Pruess for helpful discussions. REFERENCES 1. Bedard, J., S. Chamberland, S. Wong, T. Schallaardt, and L. E. Bryan Contribution of permeability and sensitivity to inhibition of DNA synthesis in determining susceptibilities of Escherichia coli, Pseudomonas aeruginosa, and Alcaligenes faecalis to ciprofloxacin. Antimicrob. Agents Chemother. 33: Beskid, G., V. Fallat, E. R. Lipschitz, D. H. McGarry, R. Cleeland, K. Chan, D. D. Keith, and J. Unowsky In vitro activities of a dual-action antibacterial agent, Ro , and comparative agents. Antimicrob. Agents Chemother. 33: Beskid, G., J. Siebelist, C. M. McGarry, R. Cleeland, K. Chan, and D. D. Keith In vivo evaluation of a dual-action antibacterial, Ro , compared to cefotaxime and fleroxacin. Chemotherapy 36: Bush, K Classification of P-lactamases: groups 1, 2a, 2b, and 2b'. Antimicrob. Agents Chemother. 33: Chapman, J. S., A. Bertasso, and N. H. Georgopapadakou Fleroxacin resistance in Escherichia coli. Antimicrob. Agents ANTIMICROB. AGENTS CHEMOTHER. Chemother. 33: Chapman, J. S., and N. H. Georgopapadakou Routes of quinolone permeation in Escherichia coli. Antimicrob. Agents Chemother. 32: a.Christenson, J. G., K. K. Chan, H. H. Farrish, I. H. Patel, and A. Specian Program Abstr. 28th Intersci. Conf. Antimicrob. Agents Chemother., abstr Cohen, S. P., D. C. Hooper, J. S. Wolfson, K. S. Souza, L. M. McMurry, and S. B. Levy Endogenous active efflux of norfloxacin in susceptible Escherichia coli. Antimicrob. Agents Chemother. 32: Cohen, S. P., L. M. McMurry, D. C. Hooper, J. S. Wolfson, and S. B. Levy Cross-resistance to fluoroquinolones in multiple-antibiotic-resistant (Mar) Escherichia coli selected by tetracycline or chloramphenicol: decreased drug accumulation associated with membrane changes in addition to OmpF reduction. Antimicrob. Agents Chemother. 33: Cohen, S. P., L. M. McMurry, and S. B. Levy mara locus causes decreased expression of OmpF porin in multiple-antibiotic-resistant (Mar) mutants of Escherichia coli. J. Bacteriol. 170: Georgopapadakou, N. H., A. Bertasso, K. K. Chan, J. S. Chapman, R. Cleeland, L. M. Cummings, B. A. Dix, and D. D. Keith Mode of action of the dual-action cephalosporin Ro Antimicrob. Agents Chemother. 33: Harding, S. M., P. E. 0. Williams, and J. Ayrton Pharmacology of cefuroxime as the 1-acetoxyethyl ester in volunteers. Antimicrob. Agents Chemother. 25: Hirai, K., H. Aoyama, S. Suzue, T. Irikura, S. lyobe, and S. Mitsuhashi Isolation and characterization of norfloxacinresistant mutants of Escherichia coli K-12. Antimicrob. Agents Chemother. 30: Hooper, D. C., J. S. Wolfson, K. S. Souza, C. Tung, G. L. McHugh, and M. N. Swartz Genetic and biochemical characterization of norfloxacin resistance in Escherichia coli. Antimicrob. Agents Chemother. 29: Jones, R. N., A. L. Barry, and C. Thornsberry Antimicrobial activity of Ro , a novel ester-linked codrug of fleroxacin and desacetylcefotaxime. Antimicrob. Agents Chemother. 33: Lund, F., K. Roholt, L. Tybring, and W. 0. Godtfredsen Mecillinam and pivmecillinam-new P-lactam antibiotics with high activity against gram-negative bacilli, p In J. D. Williams and A. M. Geddes (ed.), Chemotherapy, vol. 5. Plenum Publishing Corp., New York. 16. Lundrigan, M. D., and C. F. Earhart Gene envy of Escherichia coli K-12 affects thermoregulation of major porin expression. J. Bacteriol. 157: Maniatis, T., E. F. Fritsch, and J. Sambrook Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 18. Moses, R. E., and C. C. Richardson Replication and repair of DNA in cells of Escherichia coli treated with toluene. Proc. Natl. Acad. Sci. USA 67: Nikaido, H Role of permeability barriers in resistance to P-lactam antibiotics. Pharmacol. Ther. 27: Nikaido, H., and S. Normark Sensitivity of Escherichia coli to various,b-lactams is determined by the interplay of outer membrane permeability and degradation by periplasmic P-lactamases: a quantitative predictive treatment. Mol. Microbiol. 1: Nikaido, H., and E. Y. Rosenberg Effect of solute size on diffusion rates through the transmembrane pores of the outer membrane of Escherichia coli. J. Gen. Physiol. 77: Nikaido, H., and E. Y. Rosenberg Cir and Fiu proteins in the outer membrane of Escherichia coli catalyze transport of monomeric catechols: study with,-lactam antibiotics containing catechol and analogous groups. J. Bacteriol. 172: Nikaido, H., E. Y. Rosenberg, and J. Foulds Porin channels in Escherichia coli: studies with f-lactams in intact cells. J. Bacteriol. 153: Schnaitman, C. A., and G. A. McDonald Regulation of outer membrane protein synthesis in Escherichia coli K-12:

6 VOL. 35, 1991 E. COLI SUSCEPTIBLE TO Ro deletion of ompc affects expression of the OmpF protein. J. Bacteriol. 159: Seeberg, A. H., R. M. Tolxdorff-Neutzling, and B. Wiedemann Chromosomal f3-lactamases of Enterobacter cloacae are responsible for resistance to third-generation cephalosporins. Antimicrob. Agents Chemother. 23: Wise, R After prodrugs-mutual prodrugs. J. Antimicrob. Chemother. 7: Wolfson, J. S., and D. C. Hooper Bacterial resistance to quinolones: mechanisms and clinical importance. Rev. Infect. Dis. ll:s960-s Wolfson, J. S., D. C. Hooper, D. J. Shih, G. L. McHugh, and M. N. Swartz Isolation and characterization of an Escherichia coli strain exhibiting partial tolerance to quinolones. Antimicrob. Agents Chemother. 33: Yoshimura, F., and H. Nikaido Diffusion of P-lactam antibiotics through porin channels of Escherichia coli K-12. Antimicrob. Agents Chemother. 27: Zimmermann, W., and A. Rosselet Function of the outer membrane of Escherichia coli as a permeability barrier to beta-lactam antibiotics. Antimicrob. Agents Chemother. 12: