A possible reservoir of Batrachochytrium dendrobatidis in Australia

Epidemiological determinants regulate the frequency and pattern of disease events: for example, the persistent presence of a reservoir of infection should increase both the probability of a susceptible host becoming infected and the severity of the outcome of the infection process. Identification of a reservoir, therefore, is not only important from an academic perspective, but can inform disease mitigation strategies like culling, environmental disinfection and immunization. Given the importance of amphibian chytridiomycosis as a threat to biodiversity, it should be no surprise that significant effort has been expended identifying environmental and biological reservoirs of fungal infection and characterizing the role they play in pathogen maintenance and disease dynamics (e.g. Mitchell et al., 2008; Briggs, Knapp & Vredenburg, 2010; McMahon et al., 2012; Adams et al., 2017; Wilber et al., 2017). In the Featured Paper, Brannelly et al. (2017) report what they believe is an important reservoir for Batrachochytrium dendrobatidis (Bd) in Australia, Crinia signifera, a species commonly infected with the fungus but for which disease-driven declines have not been reported. Australia has been particularly bedevilled by chytridiomycosis, which played a lead role in the declines of several Australian frog species and has been postulated as the primary cause of several frog species extinctions (Australian Government Department of the Environment and Heritage, 2004, 2006). Although some Australian anurans exhibit high prevalence of infection with no overt signs of disease, population declines and mass mortality, recent work has described how infection may be causing cryptic declines or interact with other threatening processes to drive down abundance of species previously presumed to be safe from the deleterious effects of chytridiomycosis (Scheele et al., 2016). Since the threat of decline due to chytridiomycosis is ongoing in Australia (Scheele et al., 2017a), it makes good sense to determine which hosts are responsible for pathogen proliferation as well as those that are at risk of decline due to chytridiomycosis. The results reported by Brannelly et al. (2017) are strongly suggestive that C. signifera has the capacity to act as a biological reservoir of infection. In particular the combination of patchy but long-term field data combined with short-term laboratory data supports the conclusion that C. signifera is persistently infected and has been experiencing infection for at least a decade. Consistent availability of C. signifera at the sampling sites, the distribution of estimated ages in the skeletochronology sample and the survival rates of laboratory animals also suggest that any conservation threat posed by infection with Bd in C. signifera is weaker than those described for Australian anurans directly threatened by chytridiomycosis. Caution is warranted, though, as the age structure sample is small and without the benefit of comparisons to age structures of populations that lack infection, it is impossible to know if disease is truncating C. signifera populations, as has been described in other Australian species (Scheele et al., 2016). But is this strong evidence that C. signifera actually fulfils the role of reservoir in their study system? What is missing from this story is any quantification of inter- versus intrahost transmission dynamics, or even fundamental data on the ability of and opportunity for C. signifera to transmit to any of the three other species investigated in this study. Close geographic proximity and relatively shared habitat does not predicate interhost transmission opportunity, or ensure that these types of transmission events occur at a significant rate. Examples of weak interspecific host transmission rates in syntopic host species and disease dynamics driven predominantly by intraspecific host transmission dynamics exist in the Bd literature (Briggs et al., 2010; Bielby et al., 2013; Wilber et al., 2017). A previous study by the authors addresses this issue to some degree (Scheele et al., 2017b), but also lacks estimates of transmission probabilities among or within any of the host species studied by Brannelly et al. (2017). Methods exist to estimate the relative contributions of different hosts to pathogen maintenance in a multihost community (e.g. Fenton et al., 2015) and their application could go some ways towards estimating the reservoir potential of any of the host species included in this study. No one study can fully describe the infection dynamics of a multihost system, however, and this is another in a long series of research outputs from the One Health Research Group of James Cook University, the primary research team producing the research that predominantly informs Australian amphibian conservation efforts to combat chytridiomycosis. The group has a history of producing links in an evidentiary chain that has retrospectively pinned the epidemiological tail on the chytridiomycosis donkey that has plagued Australia's amphibians, and identified the extant species that are at risk of decline due to disease. These more recent efforts to investigate the multihost species infection dynamics are beginning to construct the picture of where the risk lies, this time at the community level. They are showing how amphibian species not only are at risk, but may contribute to it, which is equally important if we are ever to mitigate chytridiomycosis.