Understanding host and environmental factors in the immunology and epidemiology of chytridiomycosis in anuran populations in Australia

Emerging infectious diseases (EIDs) affecting biodiversity (hereafter 'biodiversity diseases') have tremendous and increasing social, environmental, economic and political impacts worldwide. The devastating amphibian skin disease, chytridiomycosis, caused by the fungus, Batrachochytrium dendrobatidis (hereafter Bd), is an example of such a disease, and has been an important driver of species declines and extinctions since its recent emergence. Bd is now considered endemic throughout most of its climatically suitable range. However this does not imply that the disease is now benign.\ud \ud The aims of my research were to 1) investigate and characterize population- and individual-level impacts of endemic chytridiomycosis, 2) investigate the amphibian host immune response to Bd infection to determine the practical utility of immunization, and investigate the potential for evolution of resistance, and 3) evaluate strategies to mitigate endemic chytridiomycosis and minimize the impact of future emerging biodiversity diseases.\ud \ud I performed Cormack-Jolly-Seber and Pradel analysis of an intensive two-year mark-recapture data set from two populations of the common mistfrog (Litoria rheocola) in the lowland wet tropics of Queensland, Australia. I found that endemic chytridiomycosis continues to have substantial seasonally fluctuating population-level effects on amphibian survival which necessitates increased recruitment for population persistence. Populations at both sites exhibited very low annual survival probabilities but high recruitment. My results suggest that similarly endemically infected amphibian populations may thus be under continued threat from chytridiomycosis, which may render them vulnerable to other threatening processes, particularly those affecting recruitment success.\ud \ud Multistate Mark Recapture analysis of L. rheocola from the Tully population, and examination of abundance, distribution and transmission of Bd between hosts, allowed me to identify and characterize pathogen aggregation as a key feature of endemic chytridiomycosis, and evaluate its implications for study, modeling and management of the disease. Examination of transition probabilities revealed that more infections occurred in cooler months, that recoveries were frequent throughout the year, and that survival probabilities were dependent on infection intensity. In order to account for the effects of over-dispersed pathogen distributions in future studies, I recommend the quantification of individual infection burdens as well as prevalence where possible.\ud \ud I performed a controlled exposure experiment of individually-housed captive-bred Bd-naïve Booroolong's frogs (Litoria booroolongensis) involving prior exposure and treatment with the antifungal itraconazole as a form of immunization. I compared survival and infection intensities between immunized and infection-naïve frogs, and was unable to demonstrate differences indicative of clinically protective host adaptive immunity. My results are consistent with other studies suggesting Bd-induced suppression of the host adaptive immune system, and indicate that immunization for reintroduced frogs is unlikely to be an effective management strategy in the near future.\ud \ud I performed a second controlled exposure experiment instead with alpine tree frogs (Litoria verreauxii alpina) from multiple populations with differing long-term Bd-exposure histories. I demonstrated population, clutch and individual-level differences in susceptibility to chytridiomycosis (measured as survival and infection intensities). I found that frogs from one long-exposed population survived significantly longer when compared with frogs from two other long-exposed populations and the naïve population. My results demonstrate differences in survival and infection dynamics between populations and clutches associated with infection exposure history that are consistent with selection for disease resistance. Features of the natural history of this species, such as lack of Bd-exposure and high survival until after breeding at two years of age, may limit opportunities for natural selection of disease resistance. My finding of a more resistant population, however, holds promise for the future management of species threatened by chytridiomycosis.\ud \ud I analysed skin, liver and spleen tissue samples from a subset of the frogs from this latter experiment using a systems biology approach in order to examine underlying mechanisms contributing to observed differences in population susceptibility to chytridiomycosis. Via total RNA extraction, next-generation sequencing, de novo transcriptome assembly, functional annotation and differential gene expression analysis, I found marked evidence for activation of gene pathways associated with immune responses in Bd-infected frogs, that differed among populations and times of sampling post exposure. I demonstrated a link between a chytridiomycosis-resistant phenotype from a long-exposed population, and evidence for a more robust early immune response at the level of gene expression compared with other populations. These differences in gene expression may putatively explain a large component of population-level differences observed in survival in the larger experiment.\ud \ud In addition to gene expression analysis, via metabolite extraction, gas chromatography-mass spectrometry, and both univariate and multivariate analyses, I identified a number of key metabolites in the skin and liver tissues that predominantly differentiate moribund frogs with chytridiomycosis from both unexposed control and subclinical frogs. I also identified metabolites related to differences in population of origin, which may be associated with variation in phenotypic resistance between populations. These findings help build our understanding of the key mechanisms and pathways involved in pathogenesis.\ud \ud In a review of current and upcoming techniques, I contributed substantially to developing a conceptual framework for management of endemic chytridiomycosis. This framework consisted of interventions to guide experimental management and applied research, and involved two main approaches, 1) reducing Bd in the environment or on amphibians, and 2) increasing the capacity of populations to persist despite increased mortality from disease. In this review I recommended trialling several promising management actions including habitat manipulation, antifungal treatments, animal translocation, bioaugmentation, head starting and selection for resistance.\ud \ud Finally, to help improve timely mitigation of future emerging biodiversity diseases, I reviewed and evaluated the efficacy of current surveillance approaches for these diseases. Barriers to effective surveillance include a relative lack of social and political will, and the inherent complexity and cost of implementing surveillance for multiple and diverse free-ranging populations. I synthesized recommendations to address these challenges by 1) extending global animal disease surveillance systems to emphasize diseases that could predominantly affect biodiversity, and 2) utilizing a systematic, population-based and self-evaluative approach to improve timely disease recognition and management, with the aim of reducing species loss.\ud \ud In summary, I found that endemic chytridiomycosis impacts amphibian population dynamics in Australia and is characterized by high mortality rate and turnover. I found that immunization is currently an ineffective strategy. However, from clinical and systems biology data, the evolution of innate immunity is possible and hence assisted selection may be a viable management strategy among other approaches