Gut Microbial Ecology of Five Species of Sympatric Desert Rodents in Relation to Herbivorous and Insectivorous Feeding Strategies

The gut microbial communities of mammals provide numerous benefits to their hosts. However, given the recent development of the microbiome field, we still lack a thorough understanding of the variety of ecological and evolutionary factors that structure these communities across species. Metabarcoding is a powerful technique that allows for multiple microbial ecology questions to be investigated simultaneously. Here, we employed DNA metabarcoding techniques, predictive metagenomics, and culture-dependent techniques to inventory the gut microbial communities of several species of rodent collected from the same environment that employ different natural feeding strategies [granivorous pocket mice (Chaetodipus penicillatus); granivorous kangaroo rats (Dipodomys merriami); herbivorous woodrats (Neotoma albigula); omnivorous cactus mice (Peromyscus eremicus); and insectivorous grasshopper mice (Onychomys torridus)]. Of particular interest were shifts in gut microbial communities in rodent species with herbivorous and insectivorous diets, given the high amounts of indigestible fibers and chitinous exoskeleton in these diets, respectively. We found that herbivorous woodrats harbored the greatest microbial diversity. Granivorous pocket mice and kangaroo rats had the highest abundances of the genus Ruminococcus and highest predicted abundances of genes related to the digestion of fiber, representing potential adaptations in these species to the fiber content of seeds and the limitations to digestion given their small body size. Insectivorous grasshopper mice exhibited the greatest inter-individual variation in the membership of their microbiomes, and also exhibited the highest predicted abundances of chitin-degrading genes. Culture-based approaches identified 178 microbial isolates (primarily Bacillus and Enterococcus), with some capable of degrading cellulose and chitin. We observed several instances of strain-level diversity in these metabolic capabilities across isolates, somewhat highlighting the limitations and hidden diversity underlying DNA metabarcoding techniques. However, these methods offer power in allowing the investigation of several questions concurrently, thus enhancing our understanding of gut microbial ecology.