|Abstract:||The study of the primate microbiome (genetic and environmental content of the microbiota) has served as a critical tool for understanding the impact of microbial communities on host health and disease. An in-depth understanding of the healthy primate microbial community composition and function is needed in order to identify changes in the microbiome that cause and/or result in unhealthy states in the host. In this dissertation project, I focused my efforts on better understanding the factors that influence or shape a healthy non-human primate (NHP) microbiome using high throughput sequencing (16S rRNA and metagenomics), bioinformatics and various statistical tools.
In the first study, investigating the confounding effects of sample storage preservatives in the form of technical variations in the microbiome, I was able to determine whether technical variations due to sample storage preservatives (no preservative, 96% ethanol and RNAlater) outweighed alterations due to biological factors of interest. The results showed that while these technical variations in the microbiome are not large enough to obscure the differences due to host species, body site and habitat differences, they may promote misleading interpretations if they have large enough effect sizes compared to the biological factors such as host population.
While most microbiome studies focus on one body site from one host species at a time, I leveraged the advantages of multi-body site and multi host species dataset to determine the extent to which the various factors influence the different types of microbiome in different host species. In the second study, I analyzed the microbial community composition of eight distinct sites (rectal, fecal, oral, nasal, otic/ear, vaginal, penile and skin) from 17 non-human primate species, the majority of which were wild. The results revealed a possibly conserved NHP oral microbiome with relatively low alpha and beta diversities as seen by the distinct clustering of the oral samples away from samples from all other body sites. Without the oral samples however, host species differences shaped the microbiome with samples clustered by body site within each host species due to niche specialization. The results also showed a stronger captivity effect on the gut microbiome in comparison to the vaginal microbiome possibly due to dietary differences between wild and captive individuals which in turn affects the microbiome more than the vaginal microbiome. These results help to elucidate the wide variation present in the microbiomes from various body sites both within and between NHPs.
To better understand the possibly conserved NHP oral microbiome observed in the second study, I analyzed the oral samples of 12 predominantly wild non-human primates (NHPs) to better understand the extent to which factors including host species, host phylogeny, diet and habitat influence the NHP oral microbiome in this study. While both alpha and beta diversity results showed statistically significant differences in the NHP oral microbiome due to all four factors, LEfSe and MaAsLin2 results revealed significant associations between the microbiome and host species. These results are also indicative of a distinct, possibly conserved NHP oral microbiome with certain host specific differences.
Finally, I also analyzed and compared the gut microbiomes of five closely related wild lemurs (Propithecus verreauxi, Propithecus diadema, Propithecus edwardsi, Varecia variegata, and Lemur catta) occupying different dietary niches. The results revealed host-specific gut microbiota clustered within broad clusters of host phylogenetic and dietary groups. While the lowest diversity was observed in the sifakas (folivores), the omnivorous L. catta showed the highest alpha diversity possibly due to their generalist feeding strategy with wide range of food sources. With the clear confoundment between host phylogeny and diet, I included three old world monkeys (Procolobus rufomitratus tephrosceles, Cercopithecus ascanius schmidti and Papio Anubis) in the analysis and obtained 22 discriminative biomarkers due host phylogeny compared to 5 due diet suggestive of a stronger host phylogenetic group effect on the NHP gut microbiome compared to diet.