Geophysiology of Wood Frogs: Landscape Patterns of Prevalence of Disease and Circulating Hormone Concentrations across the Eastern Range

Abstract

One of the major challenges for conservation physiologists is to determine how current or future environmental conditions relate to the health of animals at the population level. In this study, we measured prevalence of disease, mean condition of the body, and mean resting levels of corticosterone and testosterone in a total of 28 populations across the years 2011 and 2012, and correlated these measures of health to climatic suitability of habitat, using estimates from a model of the ecological niche of the wood frog's geographic range. Using the core-periphery hypothesis as a theoretical framework, we predicted a higher prevalence and intensity of infection of Batrachochytrium dendrobatidis (Bd) and ranaviruses, two major amphibian pathogens causing disease, and higher resting levels of circulating corticosterone, an indicator of allostatic load incurred from living in marginal habitats. We found that Bd infections were rare (2% of individuals tested), while infections with ranavirus were much more common: ranavirus-infected individuals were found in 92% of ponds tested over the 2 years. Contrary to our predictions, rates of infection with ranaviruses were positively correlated with quality of the habitat with the highest prevalence at the core of the range, and plasma corticosterone concentrations measured when frogs were at rest were not correlated with quality of the habitat, the prevalence of ranavirus, or the intensity of infection. Prevalence and mean viral titers of ranavirus infection were higher in 2012 than in 2011, which coincided with lower levels of circulating corticosterone and testosterone and an extremely early time of breeding due to relatively higher temperatures during the winter. In addition, the odds of having a ranavirus infection increased with decreased body condition, and if animals had an infection, viral titers were positively correlated to levels of circulating testosterone concentration. By resolving these patterns, experiments can be designed to test hypotheses about the mechanisms that produce them, such as whether transmission of the ranavirus and tolerance of the host are greater or whether virulence is lower in populations within core habitats. While there is debate about which metrics serve as the best bioindicators of population health, the findings of this study demonstrate the importance of long-term monitoring of multiple physiological parameters to better understand the dynamic relationship between the environment and the health of wildlife populations over space and time.