Virulence and pathogenesis of chytridiomycosis: a lethal disease of amphibians

Few fungi are highly virulent to terrestrial vertebrates. Yet the disease chytridiomycosis, caused by the fungal pathogen Batrachochytrium dendrobatidis (Bd), is one of the causes of global amphibian declines. How a superficial skin fungus can cause catastrophic extirpations is perplexing. To date most investigations have focused on ecological aspects of the host-pathogen dynamic: understanding the seasonal dynamics of the disease, mapping the distribution of the pathogen and determining its impact on amphibian populations. Relatively few studies have considered the importance of differential virulence, and evolution of virulence, of Bd. Additionally, the mechanisms of pathogenesis in chytridiomycosis remain largely unresolved. I examined the growth and developmental response of Bd to different biotic and abiotic conditions over multiple generations with an underlying objective of understanding Bd virulence. I also used pathophysiological techniques to determine the cause of mortality in frogs with severe chytridiomycosis.\ud \ud In some susceptible amphibian species severe disease is closely associated with high burdens of Bd. Therefore, rate of zoospore production is likely to be an important determinant of Bd virulence. I quantified zoospore densities in multiple isolates and examined growth and development of Bd in different nutrient and temperature conditions over multiple generations. In short term experiments Bd responds to different temperature and nutrient conditions by adjusting its life history. I found that, after multiple passages, Bd is phenotypically plastic in its response to low nutrient concentrations, but may have an adaptive response to long-term maintenance in low temperatures. Cultures that were originally derived from a single cryo-archived isolate and passaged in 0.2% tryptone TGhL (tryptone/gelatin hydrolysate/lactose media) for 24 passages had higher zoospore densities when inoculated into 1.6% tryptone TGhL, suggesting that Bd is phenotypically plastic in its response to nutrient conditions after 24 passages. In a reciprocol transplant experiment, cultures maintained in 4ºC for 20 passages released zoospores earlier and had a longer period of high zoospore densities than cultures of the same isolate and passage history, but that were maintained at 23ºC. This pattern of early zoospore release was consistent for cultures maintained in low temperatures at 4ºC and at 23ºC, suggesting an adaptive response to lower temperatures.\ud \ud The effects of serial passage on growth of Bd cultures were also examined. Two cultures that were originally derived from the same cryo-archived isolate, but had higher and lower passage histories, had different zoospore densities in in vitro experiments; after 50 passages cultures had significantly higher zoospore densities than cultures with a passage history of 10. These patterns of zoospore densities in vitro corresponded with differences in prevalence and intensities of infection in experimentally exposed Litoria caerulea. However, the differences in these response variables (prevalence and intensities of infection) were not significant and no mortality occurred in any experimental group. These results suggest that variation can exist within a single Bd isolate and that certain environmental conditions may exert selective pressures on Bd, which could influence the host-pathogen dynamic in important ways. For example, adaptive adjustments to low temperatures could enhance transmission and substantially alter the impact of the disease in amphibian populations. The practical applications of these results are that Bd may be evolving in particular ways due to long-term culturing practices, which should be a consideration for laboratory experiments aimed at understanding chytridiomycosis.\ud \ud The pathophysiological changes associated with chytridiomycosis were investigated by tracking Bd infection in experimentally exposed L. caerulea and measuring a wide range of biochemical and physiological parameters. Infected L. caerulea that developed clinical signs of severe chytridiomycosis had the highest burdens of Bd. Ussing chamber tests which measure transepithelial current and resistance demonstrated that skin samples from experimentally infected L. caerulea had inhibited electrolyte (sodium and chloride) transport across the skin surface. Plasma electrolyte concentrations, including potassium, sodium, magnesium and chloride, were reduced in the terminal stages of disease. Surgically implanted biotransmitters that were continuously recording cardiac electrograms revealed that asystolic cardiac arrest (which can be triggered by shifts in electrolytes) was the terminal event in L. caerulea with severe chytridiomycosis.\ud \ud Diseased frogs that received an electrolyte supplement became more active and lived longer than diseased frogs that received no treatment. Because I found no significant changes in haematocrit, albumin, total protein or body mass, it appears that the reductions in electrolyte concentrations were due to depletion from circulation rather than water uptake. It is the disproportionate loss of electrolytes compared with water that signifies an imbalance in osmotic homeostasis. Loss of electrolytes could occur via the skin or the kidney. Histological analysis of the kidney samples was inconclusive but the skin was severely damaged when assessed with histology and electrolyte transport (Ussing chamber) tests, suggesting that the skin is the primary organ involved in the extensive electrolyte shifts that lead to mortality. Amphibians can tolerate greater electrolyte fluctuations than other terrestrial vertebrates, but my results support the epidermal dysfunction hypothesis, which suggests that the disruption to cutaneous functioning, and the extent of electrolyte imbalance that occurs in severe chytridiomycosis, produce a life compromising\ud pathophysiology.\ud \ud The unique importance of the skin in maintaining amphibian homeostasis and the ability of Bd to disrupt epidermal functioning are two key factors that help explain how\ud mortality can occur in a wide range of amphibian species. Additionally, the ability of Bd to respond to a wide range of environmental conditions (temperatures and nutrient conditions) in ways that potentially alter the virulence and impact of chytridiomycosis, makes Bd a formidable pathogen. These disease characteristics, combined with the ability to spread rapidly and persist at low host densities, create a lethal suite of concomitant variables that, taken separately, might not be so devastating, but together are threatening amphibians worldwide