2020 ESA Annual Meeting (August 3 - 6)

COS 22 Abstract - Environmentally-dependent heritability of the gut microbiome across 16,234 wild baboon (Papio cynocephalus) samples

Laura Grieneisen1, Mauna Dasari2, Johannes Bjork2, Trevor Gould1, Jean-Christophe Grenier3, Vania Yotova3, David Jansen2, Neil R. Gottel4, Jacob B Gordon5, Niki H Learn6, Laurence R Gesquiere5, Tim L Wango7,8, Raphael Mututua7, Kinyua Warutere7, Long'ida Siodi7, Jack A. Gilbert4, Luis Barreiro3, Susan C. Alberts9, Jenny Tung9, Elizabeth A. Archie2 and Ran Blekhman1,10, (1)Department of Genetics, Cell Biology, and Development, University of Minnesota, Minneapolis, MN, (2)Department of Biological Sciences, University of Notre Dame, Notre Dame, IN, (3)CHU Sainte Justine Research Center, Université de Montréal, Montréal, QC, Canada, (4)Pediatrics and Scripps Institute of Oceanography, University of California San Diego, La Jolla, CA, (5)Biology, Duke University, Durham, NC, (6)Department of Ecology and Evolutionary Biology, Princeton University, Princeton, NJ, (7)Amboseli Baboon Research Project, Amboseli National Park, Kenya, (8)Department of Veterinary Anatomy and Animal Physiology, University of Nairobi, Nairobi, Kenya, (9)Department of Biology, Duke University, Durham, NC, (10)Department of Ecology, Evolution, and Behavior, University of Minnesota, Minneapolis, MN
Background/Question/Methods

Variation in gut microbiome composition is linked to key health outcomes across vertebrates, and is driven by both environmental and host genetic effects. However, it is often difficult to disentangle genetic from environmental effects because relatives often use the same resources and have similar diets. Moreover, it is difficult to account for temporal dynamism in microbiome composition since longitudinal profiles are rarely available. Here, we quantified the heritability of the gut microbiome using an unprecedentedly large dataset of 16,234 wild baboon microbiome profiles (Papio cynocephalus). These samples encompass 14 years of longitudinal sampling, complemented by rich data on host traits, genetic relatedness, and environmental variation (e.g., diet, social group membership, age, rainfall, and seasonality). We identified heritable microbes while controlling for non-genetic effects by including granular metadata on host traits, diet, social group, and weather.

Results/Conclusions

We show that host genetic effects on the microbiome are near-universal but environmentally contingent. Controlling for diet, age, and socioecological variation, >95% of the analyzed microbiome phenotypes (96/100) are significantly heritable, although heritability estimates are typically modest (mean=0.065). Host genetics accounts for 11-21% of the variation in the abundances of the 15 most heritable microbial phenotypes, including Christensenellaceae, a bacterial family that is also highly heritable in humans. This suggests that similar microbiome traits are under host genetic control across primate species. Further, we find that the relative abundance of heritable taxa changes over time within individuals, leading to temporal variation in estimates of heritability. Specifically, heritability estimates vary between hydrological years and by season, probably reflecting fluctuations in diet diversity. We also observe that microbiome heritability tends to increase with baboon age. Due to this temporal dynamism, we find that sample size had profound effects on both heritability estimates and our ability to detect heritable phenotypes. Our findings highlight the temporally dynamic and environmentally-contingent nature of heritability, and show that microbiome heritability is structured by season, host age, and host diet. These results emphasize that genetic effects themselves depend on environmental context, but that large, longitudinal datasets will often be necessary to characterize this dependence. Further, our results indicate scope for selection on microbiome characteristics as a host phenotype.