Chytridiomycosis causes catastrophic organism-wide metabolic dysregulation including profound failure of cellular energy pathways.

Chytridiomycosis is among several recently emerged fungal diseases of wildlife that have caused decline or extinction of naïve populations. Despite recent advances in understanding pathogenesis, host response to infection remains poorly understood. Here we modelled a total of 162 metabolites across skin and liver tissues of 61 frogs from four populations (three long-exposed and one naïve to the fungus) of the Australian alpine tree frog (Litoria verreauxii alpina) throughout a longitudinal exposure experiment involving both infected and negative control individuals. We found that chytridiomycosis dramatically altered the organism-wide metabolism of clinically diseased frogs. Chytridiomycosis caused catastrophic failure of normal homeostatic mechanisms (interruption of biosynthetic and degradation metabolic pathways), and pronounced dysregulation of cellular energy metabolism. Key intermediates of the tricarboxylic acid cycle were markedly depleted, including in particular α-ketoglutarate and glutamate that together constitute a key nutrient pathway for immune processes. This study was the first to apply a non-targeted metabolomics approach to a fungal wildlife disease and specifically to dissect the host-pathogen interface of Bd-infected frogs. The patterns of metabolite accumulation we have identified reveal whole-body metabolic dysfunction induced by a fungal skin infection, and these findings have broad relevance for other fungal diseases.

Survival, gene and metabolite responses of Litoria verreauxii alpina frogs to fungal disease chytridiomycosis.

The fungal skin disease chytridiomycosis has caused the devastating decline and extinction of hundreds of amphibian species globally, yet the potential for evolving resistance, and the underlying pathophysiological mechanisms remain poorly understood. We exposed 406 naïve, captive-raised alpine tree frogs (Litoria verreauxii alpina) from multiple populations (one evolutionarily naïve to chytridiomycosis) to the aetiological agent Batrachochytrium dendrobatidis in two concurrent and controlled infection experiments. We investigated (A) survival outcomes and clinical pathogen burdens between populations and clutches, and (B) individual host tissue responses to chytridiomycosis. Here we present multiple interrelated datasets associated with these exposure experiments, including animal signalment, survival and pathogen burden of 355 animals from Experiment A, and the following datasets related to 61 animals from Experiment B: animal signalment and pathogen burden; raw RNA-Seq reads from skin, liver and spleen tissues; de novo assembled transcriptomes for each tissue type; raw gene expression data; annotation data for each gene; and raw metabolite expression data from skin and liver tissues. These data provide an extensive baseline for future analyses.

Evolution of resistance to chytridiomycosis is associated with a robust early immune response.

Potentiating the evolution of immunity is a promising strategy for addressing biodiversity diseases. Assisted selection for infection resistance may enable the recovery and persistence of amphibians threatened by chytridiomycosis, a devastating fungal skin disease threatening hundreds of species globally. However, knowledge of the mechanisms involved in the natural evolution of immunity to chytridiomycosis is limited. Understanding the mechanisms of such resistance may help speed-assisted selection. Using a transcriptomics approach, we examined gene expression responses of endangered alpine tree frogs (Litoria verreauxii alpina) to subclinical infection, comparing two long-exposed populations with a naïve population. We performed a blinded, randomized and controlled exposure experiment, collecting skin, liver and spleen tissues at 4, 8 and 14 days postexposure from 51 wild-caught captively reared infection-naïve adult frogs for transcriptome assembly and differential gene expression analyses. We analysed our results in conjunction with infection intensity data, and the results of a large clinical survival experiment run concurrently with individuals from the same clutches. Here, we show that frogs from an evolutionarily long-exposed and phenotypically more resistant population of the highly susceptible alpine tree frog demonstrate a more robust innate and adaptive immune response at the critical early subclinical stage of infection when compared with two more susceptible populations. These results are consistent with the occurrence of evolution of resistance against chytridiomycosis, help to explain underlying resistance mechanisms, and provide genes of potential interest and sequence data for future research. We recommend further investigation of cell-mediated immunity pathways, the role of interferons and mechanisms of lymphocyte suppression.

Endemicity of chytridiomycosis features pathogen overdispersion.

Pathogens can be critical drivers of the abundance and distribution of wild animal populations. The presence of an overdispersed pathogen load distribution between hosts (where few hosts harbour heavy parasite burdens and light infections are common) can have an important stabilizing effect on host-pathogen dynamics where infection intensity determines pathogenicity. This may potentially lead to endemicity of an introduced pathogen rather than extirpation of the host and/or pathogen. Overdispersed pathogen load distributions have rarely been considered in wild animal populations as an important component of the infection dynamics of microparasites such as bacteria, viruses, protozoa and fungi. Here we examined the abundance, distribution and transmission of the model fungal pathogen Batrachochytrium dendrobatidis (Bd, cause of amphibian chytridiomycosis) between wild-caught Litoria rheocola (common mist frogs) to investigate the effects of an overdispersed pathogen load distribution on the host population in the wild. We quantified host survival, infection incidence and recovery probabilities relative to infectious burden, and compared the results of models where pathogen overdispersion either was or was not considered an important feature of host-pathogen dynamics. We found the distribution of Bd load between hosts to be highly overdispersed. We found that host survival was related to infection burden and that accounting for pathogen overdispersion allowed us to better understand infection dynamics and their implications for disease control. In addition, we found that the pattern of host infections and recoveries varied markedly with season whereby (i) infections established more in winter, consistent with temperature-dependent effects on fungal growth, and (ii) recoveries (loss of infection) occurred frequently in the field throughout the year but were less likely in winter. Our results suggest that pathogen overdispersion is an important feature of endemic chytridiomycosis and that intensity of infection determines disease impact. These findings have important implications for our understanding of chytridiomycosis dynamics and the application of management strategies for disease mitigation. We recommend quantifying individual infectious burdens rather than infection state where possible in microparasitic diseases.

Susceptibility of amphibians to chytridiomycosis is associated with MHC class II conformation.

The pathogenic chytrid fungus Batrachochytrium dendrobatidis (Bd) can cause precipitous population declines in its amphibian hosts. Responses of individuals to infection vary greatly with the capacity of their immune system to respond to the pathogen. We used a combination of comparative and experimental approaches to identify major histocompatibility complex class II (MHC-II) alleles encoding molecules that foster the survival of Bd-infected amphibians. We found that Bd-resistant amphibians across four continents share common amino acids in three binding pockets of the MHC-II antigen-binding groove. Moreover, strong signals of selection acting on these specific sites were evident among all species co-existing with the pathogen. In the laboratory, we experimentally inoculated Australian tree frogs with Bd to test how each binding pocket conformation influences disease resistance. Only the conformation of MHC-II pocket 9 of surviving subjects matched those of Bd-resistant species. This MHC-II conformation thus may determine amphibian resistance to Bd, although other MHC-II binding pockets also may contribute to resistance. Rescuing amphibian biodiversity will depend on our understanding of amphibian immune defence mechanisms against Bd. The identification of adaptive genetic markers for Bd resistance represents an important step forward towards that goal.

Experimental evolution alters the rate and temporal pattern of population growth in Batrachochytrium dendrobatidis, a lethal fungal pathogen of amphibians.

Virulence of infectious pathogens can be unstable and evolve rapidly depending on the evolutionary dynamics of the organism. Experimental evolution can be used to characterize pathogen evolution, often with the underlying objective of understanding evolution of virulence. We used experimental evolution techniques (serial transfer experiments) to investigate differential growth and virulence of Batrachochytrium dendrobatidis (Bd), a fungal pathogen that causes amphibian chytridiomycosis. We tested two lineages of Bd that were derived from a single cryo-archived isolate; one lineage (P10) was passaged 10 times, whereas the second lineage (P50) was passaged 50 times. We quantified time to zoospore release, maximum zoospore densities, and timing of zoospore activity and then modeled population growth rates. We also conducted exposure experiments with a susceptible amphibian species, the common green tree frog (Litoria caerulea) to test the differential pathogenicity. We found that the P50 lineage had shorter time to zoospore production (T min ), faster rate of sporangia death (d s ), and an overall greater intrinsic population growth rate (λ). These patterns of population growth in vitro corresponded with higher prevalence and intensities of infection in exposed Litoria caerulea, although the differences were not significant. Our results corroborate studies that suggest that Bd may be able to evolve relatively rapidly. Our findings also challenge the general assumption that pathogens will always attenuate in culture because shifts in Bd virulence may depend on laboratory culturing practices. These findings have practical implications for the laboratory maintenance of Bd isolates and underscore the importance of understanding the evolution of virulence in amphibian chytridiomycosis.

Chytridiomycosis and seasonal mortality of tropical stream-associated frogs 15 years after introduction of Batrachochytrium dendrobatidis.

Assessing the effects of diseases on wildlife populations can be difficult in the absence of observed mortalities, but it is crucial for threat assessment and conservation. We performed an intensive capture-mark-recapture study across seasons and years to investigate the effect of chytridiomycosis on demographics in 2 populations of the threatened common mist frog (Litoria rheocola) in the lowland wet tropics of Queensland, Australia. Infection prevalence was the best predictor for apparent survival probability in adult males and varied widely with season (0-65%). Infection prevalence was highest in winter months when monthly survival probabilities were low (approximately 70%). Populations at both sites exhibited very low annual survival probabilities (12-15%) but high recruitment (71-91%), which resulted in population growth rates that fluctuated seasonally. Our results suggest that even in the absence of observed mortalities and continued declines, and despite host-pathogen co-existence for multiple host generations over almost 2 decades, chytridiomycosis continues to have substantial seasonally fluctuating population-level effects on amphibian survival, which necessitates increased recruitment for population persistence. Similarly infected populations may thus be under continued threat from chytridiomycosis which may render them vulnerable to other threatening processes, particularly those affecting recruitment success.

Prior infection does not improve survival against the amphibian disease Chytridiomycosis.

Many amphibians have declined globally due to introduction of the pathogenic fungus Batrachochytrium dendrobatidis (Bd). Hundreds of species, many in well-protected habitats, remain as small populations at risk of extinction. Currently the only proven conservation strategy is to maintain species in captivity to be reintroduced at a later date. However, methods to abate the disease in the wild are urgently needed so that reintroduced and wild animals can survive in the presence of Bd. Vaccination has been widely suggested as a potential strategy to improve survival. We used captive-bred offspring of critically endangered booroolong frogs (Litoria booroolongensis) to test if vaccination in the form of prior infection improves survival following re exposure. We infected frogs with a local Bd isolate, cleared infection after 30 days (d) using itraconazole just prior to the onset of clinical signs, and then re-exposed animals to Bd at 110 d. We found prior exposure had no effect on survival or infection intensities, clearly showing that real infections do not stimulate a protective adaptive immune response in this species. This result supports recent studies suggesting Bd may evade or suppress host immune functions. Our results suggest vaccination is unlikely to be useful in mitigating chytridiomycosis. However, survival of some individuals from all experimental groups indicates existence of protective innate immunity. Understanding and promoting this innate resistance holds potential for enabling species recovery.

Temperature alters reproductive life history patterns in Batrachochytrium dendrobatidis, a lethal pathogen associated with the global loss of amphibians.

Understanding how pathogens respond to changing environmental conditions is a central challenge in disease ecology. The environmentally sensitive fungal pathogen Batrachochytrium dendrobatidis (Bd), which causes the amphibian disease chytridiomycosis, has spread globally causing amphibian extirpations in a wide variety of climatic regions. To gain an in-depth understanding of Bd's responses to temperature, we used an integrative approach, combining empirical laboratory experiments with mathematical modeling. First, we selected a single Bd isolate and serially propagated two lineages of the isolate for multiple generations in two stable thermal conditions: 4°C (cold-adapted lineage) and 23°C (warm-adapted lineage). We quantified the production of infectious zoospores (fecundity), the timing of zoospore release, and zoospore activity in reciprocal temperature transplant experiments in which both Bd lineages were grown in either high or low temperature conditions. We then developed population growth models for the Bd lineages under each set of temperature conditions. We found that Bd had lower population growth rates, but longer periods of zoospore activity in the low temperature treatment (4°C) compared to the high temperature treatment (23°C). This effect was more pronounced in Bd lineages that were propagated in the low temperature treatment (4°C), suggesting a shift in Bd's response to low temperature conditions. Our results provide novel insights into the mechanisms by which Bd can thrive in a wide variety of temperature conditions, potentially altering the dynamics of chytridiomycosis and thus, the propensity for Bd to cause amphibian population collapse. We also suggest that the adaptive responses of Bd to thermal conditions warrant further investigation, especially in the face of global climate change.

Environmental refuge from disease-driven amphibian extinction.

Species that are tolerant of broad environmental gradients may be less vulnerable to epizootic outbreaks of disease. Chytridriomycosis, caused by the fungus Batrachochytrium dendrobatidis, has been linked to extirpations and extinctions of amphibian species in many regions. The pathogen thrives in cool, moist environments, and high amphibian mortality rates have commonly occurred during chytridiomycosis outbreaks in amphibian populations in high-elevation tropical rainforests. In Australia several high-elevation species, including the armored mist frog (Litoria lorica), which is designated as critically endangered by the International Union for the Conservation of Nature (IUCN), were believed to have gone extinct during chytridiomycosis outbreaks in the 1980s and early 1990s. Species with greater elevational ranges disappeared from higher elevations, but remained common in the lowlands. In June 2008, we surveyed a stream in a high-elevation dry sclerophyll forest and discovered a previously unknown population of L. lorica and a population of the waterfall frog (Litoria nannotis). We conducted 6 additional surveys in June 2008, September 2008, March 2009, and August 2009. Prevalences of B. dendrobatidis infection (number infected per total sampled) were consistently high in frogs (mean 82.5%, minimum 69%) of both species and in tadpoles (100%) during both winter (starting July) and summer (starting February). However, no individuals of either species showed clinical signs of disease, and they remained abundant (3.25 - 8.75 individuals of L. lorica and 6.5-12.5 individuals of L. nannotis found/person/100 m over 13 months). The high-elevation dry sclerophyll site had little canopy cover, low annual precipitation, and a more defined dry season than a nearby rainforest site, where L. nannotis was more negatively affected by chytridiomycosis. We hypothesize this lack of canopy cover allowed the rocks on which frogs perched to warm up, thereby slowing growth and reproduction of the pathogen on the hosts. In addition, we suggest surveys for apparently extinct or rare species should not be limited to core environments.

Mitigating amphibian disease: strategies to maintain wild populations and control chytridiomycosis.

BACKGROUND: Rescuing amphibian diversity is an achievable conservation challenge. Disease mitigation is one essential component of population management. Here we assess existing disease mitigation strategies, some in early experimental stages, which focus on the globally emerging chytrid fungus Batrachochytrium dendrobatidis. We discuss the precedent for each strategy in systems ranging from agriculture to human medicine, and the outlook for each strategy in terms of research needs and long-term potential. RESULTS: We find that the effects of exposure to Batrachochytrium dendrobatidis occur on a spectrum from transient commensal to lethal pathogen. Management priorities are divided between (1) halting pathogen spread and developing survival assurance colonies, and (2) prophylactic or remedial disease treatment. Epidemiological models of chytridiomycosis suggest that mitigation strategies can control disease without eliminating the pathogen. Ecological ethics guide wildlife disease research, but several ethical questions remain for managing disease in the field. CONCLUSIONS: Because sustainable conservation of amphibians in nature is dependent on long-term population persistence and co-evolution with potentially lethal pathogens, we suggest that disease mitigation not focus exclusively on the elimination or containment of the pathogen, or on the captive breeding of amphibian hosts. Rather, successful disease mitigation must be context specific with epidemiologically informed strategies to manage already infected populations by decreasing pathogenicity and host susceptibility. We propose population level treatments based on three steps: first, identify mechanisms of disease suppression; second, parameterize epizootiological models of disease and population dynamics for testing under semi-natural conditions; and third, begin a process of adaptive management in field trials with natural populations.

Batrachochytrium dendrobatidis: requirement for further isolate collection and archiving.

The fungal pathogen Batrachochytrium dendrobatidis (Bd) causes the disease chytridiomycosis, which is lethal to many species of amphibians worldwide. Many studies have investigated the epidemiology of chytridiomycosis in amphibian populations, but few have considered possible host-pathogen coevolution. More specifically, investigations focused on the evolution of Bd, and the link with Bd virulence, are needed. Such studies, which may be important for conservation management of amphibians, depend on access to Bd isolates. Here we provide a summary of known Bd isolates that have been collected and archived in various locations around the world. Of 257 Bd isolates, we found that 53% originate from ranids in the United States. In many cases, detailed information on isolate origin is unavailable, and it is unknown how many isolates are cryo-archived. We suggest the creation of a centralized database of isolate information, and we urge researchers and managers to isolate and archive Bd to facilitate future research on chytridiomycosis.

Application of the survey protocol for chytridiomycosis to Queensland, Australia.

Spread of the amphibian chytrid fungus Batrachochytrium dendrobatidis (Bd), which causes chytridiomycosis, has resulted in the extinction of frogs, but the distribution of Bd is incompletely known. We trialled the survey protocol for Bd by attempting to systematically map its distribution in Queensland, Australia. Bd was easily detected in known infected areas, such as the Wet Tropics and South East Queensland. It was not detected in bioregions adjacent to, but inland from or to the north of, infected regions: Einasleigh Uplands and Cape York adjacent to the infected Wet Tropics; and Brigalow Belt South adjacent to the infected South East Queensland bioregion. These regions where Bd was not detected have bordered infected regions for between 15 yr (in northern Queensland) and 30 yr (in southern Queensland), and so they define the geographical limits of Bd with regard to the long-term environmental conditions in Queensland. The Gulf Plains, a bioregion distant from infected bioregions, was also negative. Bd was confined to rainforest and bordering habitats, such as wet eucalypt forests. Infections were largely confined to permanent water-associated species, consistent with this being an important cause of this group having the greatest declines. Our data supports biogeographic climatic models that show much of inland and northern Australia to be too hot and dry to support Bd. As there is limited opportunity for Bd to spread further in Queensland, the priority for management is reducing the impact of Bd in affected populations and assisting frogs to disperse into their former distributions. Given that the survey protocol has been applied successfully in Australia it may be useful for mapping the distribution of Bd in other parts of the world.

Sodium hypochlorite denatures the DNA of the amphibian chytrid fungus Batrachochytrium dendrobatidis.

Batrachochytrium dendrobatidis, an aquatic amphibian fungus, has been implicated in many amphibian declines and extinctions. A real-time polymerase chain reaction (PCR) TaqMan assay is now used to detect and quantify B. dendrobatidis on amphibians and other substrates via tissue samples, swabbing and filtration. The extreme sensitivity of this diagnostic test makes it necessary to rigorously avoid cross-contamination of samples, which can produce false positives. One technique used to eliminate contamination is to destroy the contaminating DNA by chemical means. We tested 3 concentrations of sodium hypochlorite (NaOCl) (1, 6 and 12%) over 4 time periods (1, 6, 15 and 24 h) to determine if NaOCl denatures B. dendrobatidis DNA sufficiently to prevent its recognition and amplification in PCR tests for the fungus. Soaking in 12% NaOCl denatured 100% of DNA within 1 h. Six percent NaOCl was on average 99.999% effective across all exposure periods, with only very low numbers of zoospores detected following treatment. One percent NaOCl was ineffective across all treatment periods. Under ideal, clean conditions treatment with 6% NaOCl may be sufficient to destroy DNA and prevent cross-contamination of samples; however, we recommend treatment with 12% NaOCl for 1 h to be confident all B. dendrobatidis DNA is destroyed.