Climate variation influences host specificity in avian malaria parasites

Abstract

Parasites with low host specificity (e.g. infecting a large diversity of host species) are of special interest in disease ecology, as they are likely more capable of circumventing ecological or evolutionary barriers to infect new hosts than are specialist parasites. Yet for many parasites, host specificity is not fixed and can vary in response to environmental conditions. Using data on host associations for avian malaria parasites (Apicomplexa: Haemosporida), we develop a hierarchical model that quantifies this environmental dependency by partitioning host specificity variation into region‐ and parasite‐level effects. Parasites were generally phylogenetic host specialists, infecting phylogenetically clustered subsets of available avian hosts. However, the magnitude of this specialisation varied biogeographically, with parasites exhibiting higher host specificity in regions with more pronounced rainfall seasonality and wetter dry seasons. Recognising the environmental dependency of parasite specialisation can provide useful leverage for improving predictions of infection risk in response to global climate change.

An inverse latitudinal gradient in infection probability and phylogenetic diversity for Leucocytozoon blood parasites in New World birds

“…Whereas these host traits are often phylogenetically conserved, they may also be influenced by environmental conditions. Therefore, rather than host ecology and evolution, regional climate determined primarily by temperature seems to have an important impact on Leucocytozoon host range expansion as demonstrated recently for its two related genera (Fecchio, Wells, et al, 2019).…”

“…For each of the 69 sampling sites in our dataset, we extracted altitude as well as several long‐term climate variables that could relate to Leucocytozoon infection prevalence by influencing the distributions and densities of vectors (see Appendix S1). Previous studies have found evidence that temperature and precipitation variables, particularly those that capture variation in summer and winter climates, may be linked to important aspects of haemosporidian community assembly and transmission (Clark et al, ; Fecchio, Wells, et al, ; Oakgrove et al, ; Sehgal et al, ). We included maximum temperature of the warmest month, maximum rainfall of the wettest month, minimum temperature of the coldest month and minimum rainfall of the driest month as predictors in our analysis.…”

“…Priors for species‐level non‐phylogenetic intercepts ( α ) were sampled from normal distributions with mean = 0 and SD = 10 (see Appendices S1 and S2 for model specifications in jags language). To reduce the model's parameter space and identify the most important predictors, we implemented a form of Gibbs variable selection (GVS) when estimating β coefficients that ensured effects were shrunk towards zero when evidence for their inclusion in the model was limited (Fecchio, Wells, et al, ; O'Hara & Sillanpää, ). Our GVS method implemented a spike and slab prior of the form P ( β | ind ) = N (0, SD = 10)(1 − ind ) + N (0, SD = 0.01)( ind ), where variable selection indicator ( ind ) was a Bernoulli draw with a low prior belief of inclusion ( p ~ .2).…”

Generalists versus specialists in fluctuating environments: a bet-hedging perspective

“…It has been suggested that environmental variation can affect the host specificity of parasites in other systems as well, including avian malaria parasites Plasmodium and Haemoproteus spp. (Fecchio et al 2019). The short life cycles of vector-borne pathogens will lead them to experience environmental variation (host availability) mostly at the among-generation level.…”

Generalists versus specialists in fluctuating environments: a bet‐hedging perspective

“…It has been suggested that environmental variation can affect the host specificity of parasites in other systems as well, including avian malaria parasites Plasmodium and Haemoproteus spp. (Fecchio et al ). The short life cycles of vector‐borne pathogens will lead them to experience environmental variation (host availability) mostly at the among‐generation level.…”