Taxonomic and metagenomic approaches to diet, health, and breeds of ruminants

Abstract

Ruminant mammals maintain a symbiotic relationship with their gut microbiome, the organisms of which break complex plant matter into metabolites that are absorbed by the host animal for its nutrition. Building on prior work describing the taxonomic composition of this ecosystem, we sought to explore its biochemical operation with a systems biology approach. We constructed a merged host/microbial metabolic network by using shotgun metagenomic data matched to a reference enzyme database. We used this approach to explore the complexities of the rumen ecosystem and to assess the effects of feed additives in diet, feed efficiency, and breed-differences on this ecosystem. We linked the metabolic network of the microbial community to that of the host animal by defining a set of interface metabolites shared between them. The value of our approach is illustrated with our comparison of the rumen microbiomes of animals of low and high feed efficiencies. Strikingly, there was no individual enzyme that was statistically different in its abundance between these two groups. However, by analyzing the network as a whole, we observed that the high efficiency animals had a microbial network that was closer in network space to the host's reactions. In our antibiotics study, we observed that there were changes in the taxonomic and functional structure when sheep were fed a diet supplemented with antibiotics. Sheep fed antibiotics had a set of microbial enzymes that were closer in in the network space to the host metabolism than were the enzymes of the microbes of sheep not fed antibiotics. There were 546 individual microbial reactions that were significantly different between the two groups. Finally in the third study described, we found that breed differences in these community metabolic networks between Angus and Charolais cattle were only detected by analyzing the global network structure: there were no individual microbial taxa or enzymes that differed significantly between the breeds. These studies have shown the importance of using a metabolic network approach to better characterize the interactions between gut microbiota and host organisms as compared to solely studying the individual taxonomic and metabolic reactions.