Abstract
Citrobacter freundii, Enterobacter cloacae, and Serratia marcescens are members of the Enterobacteriaceae family and are significant nosocomial pathogens. These organisms encode inducible, chromosomal AmpC ß-lactamases which can cause, if overproduced, resistance to all ß-lactams except the fourth generation cephalosporins and the carbapenems. The interaction of ß-lactams with penicillin binding proteins (PBPs) disrupts cell wall synthesis and in the case of some ß-lactams, initiates the ampC induction pathway. Several factors are involved in the induction of ampC in all genera including the interaction of a ß-lactam inducer with the penicillin binding proteins, the byproducts of the bacterial cell wall synthesis, and the products of ampG, ampD, and ampR. It was hypothesized that inducible, chromosomal ampC ß-lactamase was regulated differently and at multiple levels in different genera of Enterobacteriaceae. The 5'UTR of the S. marcescens ampC transcript forms a stem-loop structure and is involved in regulating the stability of the transcript while nothing is known about the role of the 5'UTR of the ampC transcripts of C. freundii and E. cloacae plays in transcript stability. Preliminary studies suggested that the regulation of S. marcescens ampC is growth rate dependent while the growth phase regulation of ampC from C. freundii and E. cloacae is unknown. Furthermore, the potential of ampC induction for different ß-lactams varies among genera of Enterobacteriaceae and among drugs when tested on the same organism. In order to test the hypothesis that variations occur in the regulation of ampC expression, the regulation of ampC ß-lactamase in C. freundii, E. cloacae, and S. marcescens was examined and compared at the transcript half-life level, during bacterial growth phase, and upon induction with ß-lactams.|Growth-phase experiments evaluating ampC RNA expression by Real-Time RT- PCR analysis revealed that the regulation of ampC expression was growth phase dependent. The half-lives of ampC transcripts were compared among the three genera using rifampicin by Semi-Quantitative RT-PCR. The half-life of ampC transcript was 6 minutes for S. marcescens and 8 minutes for E. cloacae and C. freundii. The S. marcescens ampC transcript decay pattern was sharp after rifampicin treatment. In addition, it started to decline during late log phase to early stationary phase. The ampC transcript decay patterns for E. cloacae and C. freundii were gradual after rifampicin treatment. Furthermore, the decay patterns of these two transcripts were biphasic in which ampC expression was reduced by half during mid and late stationary phase. The difference in the kinetics of transcript degradation among these genera suggested that different mechanisms may govern the regulation of ampC expression.|To determine the role of the genetic background on the transcript half-life, the half-life study was performed on E. coli isogenic clones, pCf pEc, and pSm expressing inducible ampC of C. freundii, E. cloacae, and S. marcescens, respectively. The half-life of ampC transcript was similar in all clones (approximately 8 minutes) with a biphasic decay pattern. The half-life of the ampQ. transcript for C. freundii and E. cloacae did not change in E. coli background compared to the clinical strains. In addition, the 5'UTR of the S. marcescens ampC transcript is involved in the transcript stability. These data suggested that the half-lives of ampC transcripts in these genera were intrinsic to the|IV transcript. However, the change in the transcript decay patterns of the transcripts in the clinical strains compared to the E. coli clones suggested that the transcript half-life was influenced by the genetic background.|ampC induction was performed in the clinical strains and pCf pEc, and pSin clones. The induction of ampC ß-lactamase was determined at the transcriptional level using Real-Time RT-PCR and 1 /4th the MIC of six ß-lactams; cefoxitin, ampicillin, imipenem, cefamandole, aztreonam, and ceftazidime. The induction of ampC was correlated with the binding of these drugs to penicillin binding proteins (PBPs). Penicillin binding protein assays were performed using the fluorescently labeled penicillin V, bocillin 650/665 as the reporter drug. The basal level of ampC expression was the highest in S. marcescens followed by E. cloacae and then C. freundii ampC clinical strains. This trend was the same in the clones. Among the tested ß-lactams, the induction potential of cefoxitin was the highest in C. freundii, S. marcescens, and the three clones and correlated with binding to PBPs-4-6. Ampicillin was a good ampC inducer in all clinical strains. The highest induction of E. cloacae ampC was by ampicillin and correlated with binding of PBPs-2, -4, and -5/6. Imipenem induced ampC of all strains and correlated with its specific binding to PBP-2. Only the ampC of S. marcescens was induced by cefamandole and correlated with its binding to PBPs-2-4. These data indicated that the potential of ampC induction varied among the clones and among the tested ß-lactams, and suggested factors other than the interaction of a drug with the PBPs may be involved in ampC induction. These factors could be involved in the expression of ampC at the promoter level or post-transcriptional level.|This is the first study to 1) correlate ampC induction at the transcriptional level with the interaction between ß-lactams and PBPs, 2) measure the half-lives of ampC transcripts in C. freundii and E. cloacae, and 3) identify growth rate regulation of ampC. The data presented in this dissertation reveal that subtle differences occur in the regulation of ampC among genera even though the general mechanism is similar.