Contribution of beta-lactamase hydrolysis and outer membrane permeability to ceftriaxone resistance in Enterobacter cloacae.

Mechanisms of ceftriaxone resistance were examined in Enterobacter cloacae. Clones were selected from four strains: susceptible (S), resistant (R1), selected by plating on ceftriaxone-containing agar, and highly resistant (R2), selected in ceftriaxone-treated mice infected with S clones. According to 14C-labeled beta-lactam binding assays, ceftriaxone resistance was not associated with altered target proteins. R1 and R2 clones stably produced 50 to 1,500 times more beta-lactamase than S clones; this production increased after cefoxitin induction in all S and some R1 clones. Experiments conducted with strain 218 suggested that ceftriaxone resistance involved beta-lactamase hydrolysis. Half-lives for the beta-lactamase-beta-lactam complexes at 37 degrees C were 0.4 and 2.2 min for ceftriaxone and Sch 34343, a drug not affected by the resistance, respectively; in chromatography experiments, 218 intact R1 cells (2 x 10(9) to 3 x 10(9) CFU) suspended in ceftriaxone-containing buffer (2 micrograms/ml) hydrolyzed 80% of the antibiotic in 30 min. Three observations also suggested decreased permeability in some clones, (i) Most of the R1 and R2 clones showed decreased expression of outer membrane proteins of 37,000 to 38,000 molecular weight (37K to 38K proteins) by electrophoresis, often associated with increased amounts of 42K protein. (ii) [14C]Sch 34343 labeling of intact cells proceeded more slowly in 218 R2 (with altered 37K to 38K proteins) than in 218 R1 (without this alteration), a difference persisting after competition with unlabeled cloxacillin. Delays in binding were not caused by different enzymatic activities, since 218 R1 and 218 R2 produce, in similar amounts, beta-lactamases undistinguishable in isoelectric point and Km of cephaloridine. (iii) Intact cells from 218 R2 hydrolyzed ceftriaxone more slowly (20% in 30 min) than did those from 218 R1. In 218 R1, beta-lactamase overproduction was responsible for a 15- to 200-fold increase in the MIC's of ceftriaxone, ceftazidime, carbenicillin, piperacillin, moxalactam, aztreonam, carumonam, and BMY 28142. Imipenem and Sch 34343 were not affected; an additional three- to fivefold increase in the MIC's of these antibiotics (with the exception of piperacillin, imipenem, Sch 34343), seen with 218 R2, was associated with decreased permeability.