BackgroundGlucocorticoid hormones are known to play a key role in mediating a cascade of physiological responses to social and ecological stressors and can therefore influence animals’ behaviour and ultimately fitness. Yet, how glucocorticoid levels are associated with reproductive success or survival in a natural setting has received little empirical attention so far. Here, we examined links between survival and levels of glucocorticoid in a small, short-lived primate, the grey mouse lemur (Microcebus murinus), using for the first time an indicator of long-term stress load (hair cortisol concentration). Using a capture-mark-recapture modelling approach, we assessed the effect of stress on survival in a broad context (semi-annual rates), but also under a specific period of high energetic demands during the reproductive season. We further assessed the power of other commonly used health indicators (body condition and parasitism) in predicting survival outcomes relative to the effect of long-term stress.ResultsWe found that high levels of hair cortisol were associated with reduced survival probabilities both at the semi-annual scale and over the reproductive season. Additionally, very good body condition (measured as scaled mass index) was related to increased survival at the semi-annual scale, but not during the breeding season. In contrast, variation in parasitism failed to predict survival.ConclusionAltogether, our results indicate that long-term increased glucocorticoid levels can be related to survival and hence population dynamics, and suggest differential strength of selection acting on glucocorticoids, body condition, and parasite infection.Electronic supplementary materialThe online version of this article (doi:10.1186/s12898-017-0140-1) contains supplementary material, which is available to authorized users.

Understanding how animals react to human-induced changes in their environment is a key question in conservation biology. Owing to their potential correlation with fitness, several physiological parameters are commonly used to assess the effect of habitat disturbance on animals’ general health status. Here, we studied how two lemur species, the fat-tailed dwarf lemur (Cheirogaleus medius) and the grey mouse lemur (Microcebus murinus), respond to changing environmental conditions by comparing their stress levels (measured as hair cortisol concentration), parasitism and general body condition across four habitats ordered along a gradient of human disturbance at Kirindy Forest, Western Madagascar. These two species previously revealed contrasting responses to human disturbance; whereas M. murinus is known as a resilient species, C. medius is rarely encountered in highly disturbed habitats. However, neither hair cortisol concentrations nor parasitism patterns (prevalence, parasite species richness and rate of multiple infections) and body condition varied across the gradient of anthropogenic disturbance. Our results indicate that the effect of anthropogenic activities at Kirindy Forest is not reflected in the general health status of both species, which may have developed a range of behavioural adaptations to deal with suboptimal conditions. Nonetheless, a difference in relative density among sites suggests that the carrying capacity of disturbed habitat is lower, and both species respond differently to environmental changes, with C. medius being more negatively affected. Thus, even for behaviourally flexible species, extended habitat deterioration could hamper long-term viability of populations.

Genes of the major histocompatibility complex (MHC) play a central role in adaptive immune responses of vertebrates. They exhibit remarkable polymorphism, often crossing species boundaries with similar alleles or allelic motifs shared across species. This pattern may reflect parallel parasite-mediated selective pressures, either favouring the long maintenance of ancestral MHC allelic lineages across successive speciation events by balancing selection ("trans-species polymorphism"), or alternatively favouring the independent emergence of functionally similar alleles post-speciation via convergent evolution. Here, we investigate the origins of MHC similarity across several species of dwarf and mouse lemurs (Cheirogaleidae). We examined MHC class II variation in two highly polymorphic loci (DRB, DQB) and evaluated the overlap of gut-parasite communities in four sympatric lemurs. We tested for parasite-MHC associations across species to determine whether similar parasite pressures may select for similar MHC alleles in different species. Next, we integrated our MHC data with those previously obtained from other Cheirogaleidae to investigate the relative contribution of convergent evolution and co-ancestry to shared MHC polymorphism by contrasting patterns of codon usage at functional vs. neutral sites. Our results indicate that parasites shared across species may select for functionally similar MHC alleles, implying that the dynamics of MHC-parasite co-evolution should be envisaged at the community level. We further show that balancing selection maintaining trans-species polymorphism, rather than convergent evolution, is the primary mechanism explaining shared MHC sequence motifs between species that diverged up to 30 million years ago.