Epidemiological and Molecular Investigations on Antibiotic Resistance in Group B Streptococcus and Escherichia coli.

Abstract

Antibiotic resistance is currently one of the greatest challenges to clinicians and to public health. The detection of resistance mechanisms to antibiotics has become more and more epidemiologically and clinically important. This dissertation investigates rates, mechanisms of antibiotic resistance and adaptation in Group B Streptococcus (GBS) and Escherichia coli by the combined power of epidemiologic principles and molecular methods. In chapter 3 and 4, I describe the rates and mechanisms of resistance to fluoroquinolones in GBS. My colleagues and I report a strikingly high frequency of resistance to different fluoroquinolones, especially to norfloxacin (~93%) among a collection of 1075 GBS strains isolated from South Korea between 2006 and 2008. Ciprofloxacin and levofloxacin resistance was higher in invasive than in colonizing GBS isolates (10.6% versus 2.5% (p < 0.001) in ciprofloxacin resistance, 9.8% versus 2.1% (p < 0.001) in levofloxacin resistance). Mutations in gyrase and topoisomerase were found to be the resistance mechanism to ciprofloxacin, levofloxacin and moxifloxacin while efflux pumps appear to be the predominant resistance mechanism among GBS strains resistant to only norfloxacin. Strain serotypes were not associated with susceptibility to fluoroquinolones. In chapter 5, I turn my attention to Escherichia coli adaptation measured by differences in antibiotic resistance and in the distribution of CRISPRs (Clustered Regularly Interspaced Short Palindromic Repeats) between commensal E. coli and uropathogenic E. coli (UPEC) isolates. Recently, CRISPRs were identified as an immune system protecting numerous bacteria against invasion by phages, plasmids or other forms of foreign DNA. In my study on 81 matched pairs of commensal E. coli and UPEC strains isolated from UTI women, commensal E. coli isolates were found to have more repeats (p = 0.009) and more unique spacers (p < 0.0001) than UPEC isolates. Additionally, UPEC isolates were more likely to be resistant to the antibiotics tested (cefazolin (p < 0.0001), trimethoprim/sulfamethoxazole (p = 0.05)) than commensal E. coli isolates. Association between CRISPRs and antibiotic resistance was not well identified in this study. These findings support the hypothesis of better adaptability of UPEC and are suggestive of the positive role of E. coli CRISPRs as a defense system.