Symbiotic microbial communities (microbiomes) are ubiquitous inhabitants of multicellular organisms and are increasingly recognized to play roles in host health and development. The skin-associated microbiome of amphibians may affect resistance to the potentially lethal disease, chytridiomycosis, which is caused by the fungal pathogen Batrachochytrium dendrobatidis (Bd). The Sierra Nevada yellow-legged frog (Rana sierrae ) is threatened with extinction, and Bd is a major driver of declines in these frogs. However, some R. sierrae populations are able to co-exist with the pathogen, persisting despite Bd infection. Understanding why some R. sierrae populations persist with Bd may provide clues for how to minimize the impact of Bd. However, while preliminary studies highlight the possibility that symbiotic bacteria affect disease resistance, very little is known about natural diversity, stability, or function of the amphibian skin microbiome.

I tested the hypothesis that differences in skin-associated bacterial communities (referred to simply as the microbiome for brevity) can account for differences in the outcome of R. sierrae infection, i.e. population extinction versus persistence. I surveyed microbiomes from multiple R. sierrae populations and showed that populations that persisted or declined to extinction due to Bd indeed harbored different skin bacterial communities, consistent with a protective effect of bacteria in persistent populations. At the same time I found evidence that Bd infection itself may drive variation in the microbiome, both at the local scale (among frogs within a given population), and landscape scale (among populations).

Correlations between microbiome composition and Bd infection severity among wild frogs may be the result of either microbiome-mediated resistance to disease, in which bacterial assemblages control the degree of Bd infection, or, alternatively could result from Bd-induced disturbance causing changes in the microbiome. Using a laboratory experiment, I demonstrated that Bd infection alters the R. sierrae microbiome, while I found no direct evidence that the microbiome limits the severity of Bd infection, at least within the range of microbiome variability represented in this study. In addition, some of the bacterial taxa that were sensitive to Bd infection might have been predicted to inhibit the growth of Bd based on results in other study systems. These results highlight the importance of considering microbiome stability when assessing the potential for the microbiome to limit pathogen growth.

Understanding why microbial communities vary among individuals or populations is an important step toward understanding the diversity and function of the microbiome. I present an experiment that tested the effects of natural variation in the aquatic environment and genetic variation among R. sierrae individuals in shaping the skin microbiome.

By integrating controlled experiments with field surveys at multiple spatial scales, these studies reveal new insights into amphibian skin microbiome assembly, function, and sensitivity in the face of infection by an important fungal pathogen, and show the importance of these characteristics of the microbiome in wild amphibian populations.