Combining DNA-metabarcoding with stable isotope analyses of bat guano provides a more robust picture of bat diet

Rachel Reid, Geosciences, Virginia Tech, Blacksburg, VA and Christine Edwards, Center for Conservation and Sustainable Development, Missouri Botanical Garden, St. Louis, MO

Background/Question/Methods
Recent advances in high-throughput DNA sequencing have made it possible to quickly and accurately gain dietary information from fecal samples. Nonetheless, while DNA metabarcoding can characterize dietary diversity at high resolution, whether it provides accurate estimates of the relative abundances of consumed dietary resources is debatable. Stable carbon and nitrogen isotope analyses of fecal samples and dietary resources can be used to model the proportional contribution of different dietary resources to an organism, but the resolution of this approach is improved by having an accurate list of resources consumed by a species. Here we combine these complementary methodological approaches (DNA metabarcoding and stable isotope analysis) to study the diets of gray bats (Myotis grisescens), a federally endangered species, and evaluate the resolution of this multimethod approach to dietary analysis. We collected fresh guano samples from Mary Lawson Cave, Missouri, a gray bat maternity colony in June 2019. We also opportunistically collected arthropods in the vicinity of the cave. We used DNA metabarcoding of 16s and COI gene regions to catalog the dietary diversity of gray bats at the site and used carbon and nitrogen isotope mixing models to estimate the proportional contributions of different insect orders to gray bat diet
Results/Conclusions
We collected 44 gray bat fecal samples, which we split for genetic and isotopic analysis, as well as 20 local arthropod samples representing four arthropod orders (Lepidoptera, Diptera, Coleoptera, Hymenoptera). The markers employed for metabarcoding varied in taxonomic resolution, but both showed a large proportion of sequences assigned only to class Insecta, likely indicating that DNA sequence databases lack reference sequences for many diet items; additional sequencing of potential diet items is needed to improve species identification. Most of the remaining sequences were assigned to Diptera and Ephemeroptera. The fecal samples have δ13C values ranging from -34.4 to -24.2 ‰ (mean ± 1 SD = -26.7 ± 2.0 ‰) and δ15N values ranging from 0.0 to 6.9 ‰ (4.7 ± 1.7 ‰). δ13C values of local arthropods are quite similar to the bat fecal samples (-27.7 ± 1.9 ‰) while their δ15N values average a trophic position lower (2.6 ± 2.6 ‰). To quantify the proportional contributions of various arthropod orders to the bat diets, we used a Bayesian stable isotope mixing model (MixSIAR) and included additional arthropod isotope data from the literature. This preliminary analysis suggests that Lepidoptera and Diptera make up the greatest proportions of the bats’ diets.