PS 7-95 - Using a novel sample type to measure correlates of temporal variation in wild lemur (Eulemur rubriventer) microbiomes

Abstract: The number and types of microbes that make up the gut microbiome profoundly affect host health, but we don’t understand what drives changes in an individual animal’s microbiome over their lifetime. There is mounting evidence that short-term variation in gut microbiome composition is linked to an animal’s stress response, as measured by hormone levels. However, paired hormone-microbiome data sets are unavailable in most systems, such that very little work has been done to see if these relationships are borne out in wild populations experiencing natural stressors. As such, we don’t know if there are long-term relationships between the gut microbiome and the stress response, or if these relationships are consistent across different ecological contexts.

Wild primate systems provide excellent models for ‘natural experiments’ on hormone-microbiome relationships, as many long-term primate studies collect individualized data on life history and environmental variables, with corresponding hormone samples. However, fecal samples collected for hormone analysis during fieldwork are often preserved via air-drying (“desiccation”), an approach not used for preserving gut microbiome samples. Here, for the first time, we demonstrate that primate fecal samples collected and preserved for hormone analysis via desiccation can also yield gut microbiome community data, allowing for paired hormone-microbiome analyses. Specifically, the goals of our study were (1) establish if desiccated fecal samples yield viable gut microbial data; and (2) leverage longitudinally collected desiccated samples to test if individual host effects on gut microbial composition were stronger than ecological effects.

To do so, we used desiccated fecal samples collected over six months from two ecologically distinct populations of wild, red-bellied lemurs (Eulemur rubriventer) that are part of a long-term evolutionary endocrinology study, as well as lemur fecal samples stored in ethanol. We found that, with minor DNA extraction methods modifications, desiccated samples yielded sufficient high-quality 16S rRNA V4 sequence reads with low rates of contamination. We also found that desiccated samples were compositionally similar to ethanol samples, with microbial taxa relative abundances that were similar to past lemur microbiome studies. Further, microbiome diversity in desiccated samples was predicted by variables shown to affect lemur microbiomes in past studies, including collection month, sex, and local ecology. Taken together, these results suggests that desiccated samples are a viable approach for studying gut microbiota. Our work opens the door for other study systems to utilize historic biological sample data sets to answer novel hormone-microbiome questions in an ecological context.