Our results suggest that the Bd dose administered is capable of inducing a subclinical infection in the palmate newt within one to two weeks after inoculation. We found no evidence to suggest that Bd caused signs of chytridiomycosis  or death, and indeed, virtually all exposed newts appeared to have cleared any infection by one month post-inoculation. These results complement circumstantial field evidence documenting that in Bd areas, palmate newts appear not to suffer mass mortality (e.g., [37, 38]). Nevertheless, our evidence also suggest that such apparent resistance to Bd comes at a cost of increased mass loss during the aquatic phase and a more rapid transition to terrestrial t-phase compared to non-exposed controls. Within exposed newts, both the amount of mass lost and probability of entering t-phase increased as a function of increasing pathogen load clearance. By contrast, the rate of loss of secondary sexual characteristics were generally not influenced by Bd infection, with the exception of hind foot webbing that remained longer in exposed newts than controls. While the devastating impacts of Bd on amphibians are well publicised [7–9, 47], much less is known about the extent, form and underlying causes of more subtle symptoms in apparently resistant amphibian species. Our results suggest that caution should be exercised before concluding that Bd has negligible consequences for apparently resistant species.
Bd is known to invade the host epidermis; feeding on various nutrients (e.g., keratin), causing pathological abnormalities and impairing critical cutaneous functions, such as the maintenance of osmotic balance (reviewed in ). Although Bd infection can have devastating consequences (see Introduction), accumulating evidence suggests that some amphibians only exhibit subclinical symptoms and might be able to effectively clear the infection through mechanisms such as antimicrobial peptides [13, 48], Bd killing microbial flora on their skin [49–51], anti-Bd immunoglobulin’s [52–54], increasing body temperature during the infection  and improvements to dietary condition  (reviewed in ). In such circumstances, individuals might still suffer costs: (i) because pathogens impair body functioning; (ii) because mounting an immune response or repairing damaged tissues requires energy; (iii) because pathogens actually consume host energy resources; and/or (iv) because immune-associated illness-induced anorexia reduces energy intake [23, 58–60]. For instance, in wild frog populations, Bd infection has found to be associated with smaller body size [29, 61], although the mechanism(s) causing the reduction in body size in these frog studies was unclear. The evidence for our study suggests that increased mass loss might be mediated by a cost of immunity , but verification of this as a specific mechanism needs elucidating through more targeted experimentation in our and other studies. For example, we are not able to rule out a role of adaptive anorexia, but we suggest that such an effect is unlikely to explain our results fully, since newts were not fed adlib and we noticed no obvious surplus of food in experimental tanks. Indeed, that a recent study has shown experimentally that mounting an innate immune response (skin peptides) against Bd comes at cost to host body condition , provides some tentative support for our conclusions.
We found little evidence to suggest that the regression of secondary sexual characteristics were hastened by exposure to Bd, but we found some support for the possibility that breeding season duration might be curtailed. Both the regression of sexually selected characteristics and transition into t-phase are thought to be largely under hormonal control [63–65]. Although, we were not able to measure neuroendocrine changes of the exposed and infected newts, our results fit well with the current knowledge of amphibians’ stress responses and its impacts on reproduction. In amphibians, exposure to pathogens can cause a rapid release of anti-microbial peptides [66, 67] through activation of hypothalamic-pituitary-adrenal (HPA) axis (=stress axis in mammals) [67–69]. Stimulation of this axis can result in inhibition of production and release of stress hormone (i.e. corticosterone) [70, 71] which, in urodeles, inhibits the courtship behaviour, development/maintenance of male secondary sexual traits and triggers the migration toward the terrestrial habitat [63, 64, 72–74]. At the behavioural level, a decrease of prolactin triggers the termination of aquatic phase and migration toward terrestrial habitat while at a morphological level the decrease of prolactin induces the decline of tail crest [63, 64, 73, 74]. However, the decline of hind feet webbing is mediated by a different mechanism which involves a synergetic effect of several hormones [73, 75]. Therefore, the slowed reduction of hind feet webbings, in comparison to tail crest and tail filament, might be due to the difference of the hormonal bases which control these traits. In order to elucidate the potential mechanisms of slowed regression of foot webbing (in comparison with other secondary sexual traits, or vice versa) as well as early entry into t-phase, more studies are required. In addition, in order to understand the potential consequences, the exact role and relative importance of each sexually selected trait in female choice is required, as is the consequences of the size of each trait for survival on transition to the t-phase.
Although, for obvious ethical and conservation reasons, we were unable to measure the consequences of experimental infection for individual fitness in the wild, we suggest the consequences of Bd that we observed are likely to be significant [26, 29, 30, 76]. In palmate newts, mating success is likely to be influenced by the duration of their aquatic phase, and is known to be condition-dependent: female fecundity and male display rate are both highly demanding energetically [77–79]. It is also highly probable that the success of terrestrial migrations are at least partly associated with having sufficient energy reserves as is the ability to survive winter hibernation, since the annual rate of survival of newts is fairly low (i.e. ≤ 50%, see [80, 81]) and newts consume almost all their resources during the winter . Our ability to project the population consequences of sub-lethal infections requires an understanding of whether or not individuals can acquire adaptive immunity to Bd or whether individuals with primed immune system remain susceptible to Bd. Where the former is the case, we would expect Bd to have little impact on palmate newt’s populations once resistance spreads in the populations (e.g. see [62, 82]). On the other hand, if the latter is true, the sub-lethal consequences observed in this study are likely to have more significant population consequences, with possible impairment of female fecundity, juvenile recruitment and adult survival. Currently, it is unclear whether amphibian species that suffer subclinical effects of Bd are declining, as one might expect from our results. We urge that future studies are careful to monitor population sizes of all amphibian species in a given area, and attempt to determine whether Bd can also have population consequences, even for apparently resistant species. Further, in the advent of such declines being apparent, it is important to determine whether such declines are generated through biased effects on each sex or age class.