Gut microbiome in rats: Effects of diet on community structure and host-microbiome interactions

Abstract

Host-microbe interactions are now considered essential for maintaining host health. It is known that short and long term dietary interventions influences the structure and activity of gut bacterial communities. However, our understanding of the forces shaping the gut microbiota is still limited and controversial, and most of the studies of the gut microbiota use the microbiota from faeces as a proxy for the intestinal tract populations. As such, the overarching aim of this thesis is to contribute to the understanding of host-microbiome interactions using an animal model.

In this thesis I describe the effect of diet changes on microbial community structure and host-microbiome interactions following 14 weeks on one of the three experimental diets. The diets consisted of a basal diet low in fibre (LF); the basal diet together with 26 % cellulose; a difficult to ferment fibre (HF); and the basal diet together with 50% dried cooked red kidney beans (B); a diet relatively high in easily fermentable fibre. These diets were fed to 45, 21 day old female Wistar rats originating from 6 litters for 14 weeks.

Diet had little effect on rat growth rates or adult body mass. However, diet had profound effects on gastro-intestinal morphology and dynamics. Caecum size was smallest in animals fed the LF diet, and caecums were about 2x as large in animals fed the B diet, while animals on the HF diet had intermediate-sized caecums. Food transit times were slowest in animals on the B and LF diets and fastest in animals on the HF diets. At the end of the diet experiment, colon and caecum contents were collected when the animals were killed and short chain fatty acids, nitrogen, carbon, as well fibre concentrations were determined. These data showed that the ‘chemical’ environment of the hindgut varied substantially among animals fed the different diets.

E. coli diversity and dynamics were described by characterizing more than three thousand isolates. E. coli diversity was low, and more than 97% of the isolates were represent by three strains: one phylogroup B2 strain and two phylogroup B1 strains. A decline of the frequency of the B2 strain in the animals fed on the bean diet was observed.

The faecal microbiota was characterized when the animals were 21 days old, while faecal, caecal and rectal microbial communities characterized at the end of the experiment. 16S amplicon sequencing of the V4 region on the Ion Torrent platform was the approach used to characterize the microbiota.

Members of 23 microbial families were detected in communities of the animals before and after 14 weeks on the experimental diets. At the start of the experiment there were significant litter membership effects on the structure of the faecal microbial communities. After 14 weeks on the experimental diets, both litter and diet explained a significant amount of the variation in microbial community structure. There were substantial differences in the microbial communities of the caecum and rectum and the extent of these differences depended on diet and on the time taken for material to move through the hindgut.

The outcomes of the present study make a contribution to our understanding of the factors that shape gut microbial communities. Microbial characterization of faecal samples is frequently used as proxy of gut microbiota. However, stool samples are probably most likely representative of the microbial communities in the rectum than other parts of the gastrointestinal tract. Indeed, the findings also throw doubt on the value of faecal community characterization as a means to understand community structure and function in the gastro-intestinal tract. Further, the results of these experiments suggest that efforts attempting to achieve positive health outcomes through diet manipulation may have limited success in general due to among individual differences in microbial community composition, and in how these different communities respond to dietary manipulation.