2020 ESA Annual Meeting (August 3 - 6)

COS 181 Abstract - Leveraging the gut microbiome as a noninvasive biomarker of aging in wild primates

Mauna Dasari1, Susan C. Alberts2,3, Jeanne Altmann4, Luis Barreiro5, Jack A. Gilbert6, Ran Blekhman7,8, Jenny Tung2,3 and Elizabeth A. Archie1, (1)Department of Biological Sciences, University of Notre Dame, Notre Dame, IN, (2)Department of Biology, Duke University, Durham, NC, (3)Department of Evolutionary Anthropology, Duke University, Durham, NC, (4)Department of Ecology and Evolutionary Biology, Princeton University, Princeton, NJ, (5)CHU Sainte Justine Research Center, Université de Montréal, Montréal, QC, Canada, (6)Pediatrics and Scripps Institute of Oceanography, University of California San Diego, La Jolla, CA, (7)Department of Ecology, Evolution, and Behavior, University of Minnesota, Minneapolis, MN, (8)Department of Genetics, Cell Biology, and Development, University of Minnesota, Minneapolis, MN
Background/Question/Methods: Mammalian gut microbiomes are highly individualized, dynamic communities. These dynamics are driven by, and in turn may drive variation in host metabolism, immune function, behavior, and environmental microbial exposures. These factors change as hosts age, raising the possibility that the microbiome can serve as a marker of host development and senescence. In support, cross-sectional sampling in humans indicates changes in gut microbial composition in early life and old age. However, research linking aging to gut microbiome composition in wild, non-human animals is rare. To fill this gap, we evaluated whether the gut microbiome could be used as a noninvasive biomarker of aging by creating a “microbiome aging clock” based on a machine learning model. We trained our microbiome aging clock using a unique longitudinal data set spanning 13,563 fecal samples from 479 known-age individual baboons (Papio cynocephalus) in the Amboseli ecosystem in Kenya over 14 years. From these samples, we profiled the 16S rRNA gene region and generated 9,575 microbial features, ranging from diversity metrics to taxonomic abundances. These samples are complemented by concurrent data on demography and survival, allowing us to test whether patterns of microbiome aging predict the timing of host maturation or mortality risk.

Results/Conclusions: We built a microbiome aging clock, which predicted host age with an R2 of 0.42 and median error of 2.2 years (median error is the median absolute difference between chronological age and predicted microbiome age in the data set). 14% of microbial features predicted chronological age. Metrics of ecosystem diversity, like alpha and beta diversity, did not predict age, but the abundances of specific microbial taxa did. We then tested whether baboons who were “microbially old for age” (predicted microbiome age > chronological age) had lower survival or achieved maturational milestones sooner than baboons who were “microbially young for age” (predicted microbiome age < chronological age). Microbial age did not predict maturational milestones or survival in females. This result suggests that links between microbiome function and major life history components like survival and maturation may be relatively weak. Together, our results characterize how gut microbial communities change with age in a wild primate population. By understanding how the microbiome ages with the host, this research highlights the weakness of relying on taxonomic composition and the need to focus on functional aspects of the microbiome (e.g. gene content) to predict host traits.