Consortium leader: Professor CHARLOTTA EDLUND,
Division of Clinical Bacteriology F82, Huddinge University Hospital, Karolinska Institutet SE-141 86 Stockholm, Sweden. +46-8-585 811 39, Charlotta.Edlund@labmed.ki.se
Partners of the consortium:
Pentti Huovinen, Research professor, KTL, National Public Health Institute, Department of Human Microbial Ecology and Inflammation, Kiinamyllynkatu 13, 20520 Turku, Finland,
+358-2-3316601, pentti.huovinen@kl.fi
Janet Jansson, professor Department of Microbiology, Swedish University of Agricultural Sciences, Box 7025, 750 07 Uppsala, Sweden, +46-(0) 18 67 3201, janet.jansson@mikrob.slu.se
Dan Andersson, Professor, Swedish Institute for Infectious Disease Control, Nobels väg 18,
S-171 82 Stockholm, +46 8 4572432, dan.andersson@smi.ki.se
Doctoral students of the consortium:
Sonja Löfmark, sonja.lofmark@labmed.ki.se
Cecilia Jernberg, cecilia.jernberg@sh.se
Sofia Nyberg, sofia.nyberg@ktl.fi
Marianne Lindgren, marianne.lindgren@ktl.fi
Johan Dicksved, johan.dicksved@mikrob.slu.se
Ann-Chatrine Palmgren, Ann-Chatrin.Palmgren@labmed.ki.se
Key words: Normal microflora, antimicrobial resistance, resistance genes, and fitness
Results
The aim of the investigation was to study potential long-term ecological consequences of one week antibiotic administration, regarding diversity, emergence of resistance and resistance genes in the intestinal microflora.
Eight healthy volunteers divided into two groups were included in the study. None of them had received any antibiotics at least one year prior to the study. Four subjects received 150 mg clindamycin capsules perorally qid for 7 days, and four who did not receive any antimicrobial agents during the study period composed a control group. Faecal samples were collected prior to clindamycin administration (day 0), and at day 7, day 21 and 3, 6, 9, 12 and 24 months post-exposure. For the control group, faecal sampling was performed at corresponding intervals. A total of 64 samples were cultured on agar media and from each sample and 20 colonies each of Bacteroides spp., Enterococcus spp. and Enterobacteriaceae was isolated for further analyses. The isolates were identified to species level by biochemical tests and further genotyped by pulsed-field gel electrophoresis and rep-PCR. Antimicrobial susceptibility testing was performed by the agar dilution method and presence of specific resistance genes were monitored by conventional and real-time PCR, confirmed by sequencing when needed.
The results showed that administration of clindamycin for one week had a dramatic long-term impact on the composition of the intestinal microflora regarding both clonal stability and levels of antibiotic resistant bacteria and resistance genes, most prominently shown among Bacteroides species, which compose about 25% of the intestinal community. The high diversity in the Bacteroides community consistently seen in the control group and in pre-treatment samples of the clindamycin group, were significantly reduced after exposure. A wide variety of susceptible clones were replaced by few resistant clones, dominated by Bacteroides thetaiotaomicron, which in two of the four exposed subjects could be isolated up to 24 months after clindamycin administration. In contrast, for enterococci and Enterobacteriaceae, the diversity increased after the administration since dominating susceptible clones of Enterococcus faecalis, Enterococcus faecium and Escherichia coli were partly replaced by less susceptible species.
After exposure there was a strong selection for resistant strains, most prominent for clindamycin resistance in Bacteroides species, which unexpectedly was not normalised even two years after the administration. Despite the fact that Enterobacteriaceae are intrinsically resistant to clindamycin, the administration caused major alterations in the susceptibility pattern to several other unrelated antimicrobial agents, e.g. ampicilin. The impact of clindamycin administration was not as large among enterococci and enterobacteria as in the Bacteriodes community, even though the level of resistance in E. coli isolates was elevated after the treatment and persisted up to three months post-treatment. Accordingly, there was a treatment induced enrichment of resistance genes in the intestinal microflora, which could be monitored both directly from faecal DNA and by analysing individual isolates, such as ermB in enterococci and ermF in Bacteroides spp, both encoding ribosomal methylases, and blaTEM in Enterobacteriaceae, encoding TEM-1 betalactamases.
Generally, resistance levels and presence of resistance genes were low throughout the study period in the control group and at baseline in the clindamycin group. Acquisition of erm-genes was detected in two bacteroides clones shortly after the administration period, indicating the importance of the intestinal microflora as a gene-pool of resistance genes.
The present results indicate that even a short administration of the antibiotic clindamycin, an agent commonly used for treatment of out-patients and hospitalised patients, causes long-term deleterious consequences in the dynamics of the normal intestinal microflora of individual subjects. Increased knowledge of the role of the intestinal tract as a reservoir of resistance genes in the total global burden of increasing resistance is of great importance.
Selected publication:
Jernberg C, Sullivan Å, Edlund C, Jansson J. K. (2005) Monitoring antibiotic induced alterations in the human intestinal microflora and detection of probiotic strains by the use of T-RFLP. Applied and Environm Microbiol 71 (1): 501-506.
ESPAR RESEARCH GROUP
An abstract of the reserch plan (January 2003)