Incidence, emergence, persistence and mechanisms of antimicrobial resistance : In clinical isolates and normal microbiota

Abstract: The increasing resistance rate to antimicrobial agents is of global concern and needs to be readily addressed. Options for successful treatments of infectious diseases are becoming limited due to the spread of resistant strains and resistance genes in the clinic and in the community. The normal human microbiota, and especially the intestinal flora, is an important reservoir for the presence and transmission of resistance determinants. The Bacteroides fragilis group and Propionibacterium acnes are anaerobic opportunistic pathogens and members of the normal microbiota. Clinical isolates of these species are commonly resistant to a variety of antibiotics. Knowledge of the mechanisms behind the resistance is important for our understanding of its emergence, spread and persistence. It has also implications for the choice of antimicrobial therapy and the design of new drugs. In the present thesis, data on different aspects of antimicrobial resistance are presented. Bacteroides species are resistant to a number of antimicrobial agents. The resistance mechanisms against metronidazole, an important drug for the treatment of anaerobic infections, are still not sufficiently elucidated. We could demonstrate that nim genes are present in a sub-population of European clinical Bacteroides spp. strains and are significantly associated with reduced susceptibility to metronidazole. These strains could be induced to form a highly resistant sub-population. We detected a new variant of the nim gene as well as a variety of upstream regulatory elements. Patients with acne vulgaris are often heavily treated with antimicrobial agents creating a force for resistance development, which may also complicate the treatment of other infections. In the present study a number of known mechanisms behind erythromycin, clindamycin and tetracycline resistance in P. acnes isolates were characterised, but possibly newly evolved mechanisms were also involved. Different PFGE genotypes were shown to be distributed throughout Europe. Antimicrobial agents do not only affect the pathogens targeted but have also an impact on the normal microbiota. A 7-day administration of clindamycin was shown to cause long-term disturbance in the faecal bacterial flora of healthy volunteers. This perturbation was recorded by analysing total faecal DNA and Bacteroides isolates from consecutively collected faecal samples. Molecular analysis of the faecal DNA revealed a shift in the Bacteroides community and enrichment of erm genes for up to two years following clindamycin exposure. Comparing these shifts with those in an unexposed control group where only minor changes in the bacterial community were recorded further strengthens the dramatic impact of the antimicrobial administration. The negative influence the clindamycin exposure had on the Bacteroides community was also demonstrated among the cultured isolates. An immediate decrease in species and clonal diversity was noted and there was a strong selection of resistant strains and resistance determinants, especially the clindamycin specific erm genes, lasting for up to two years. The long-term stabilisation of in vivo acquired erm genes was shown to be due to a restoration of bacterial fitness, confirmed by an in vitro competition assay of isogenic isolates collected 21 days and 18 months after initiated administration. The initial biological cost of acquiring an erm-gene was further confirmed in an in vitro acquired transconjugant. In conclusion, short-term antibiotic administration may cause decreased clonal diversity and persistent enrichment of resistant clones in the normal intestinal microbiota, creating a reservoir of resistance, further adding to the global burden of resistance.