Host migration and environmental temperature influence avian haemosporidians prevalence : a molecular survey in a Brazilian Atlantic rainforest.

Abstract

Avian haemosporidians are parasites with great capacity to spread to new environments and new hosts, being considered a good model to host-parasite interactions studies. Here, we examine avian haemosporidian parasites in a protected area covered by Restinga vegetation in northeastern Brazil, to test the hypothesis that haemosporidian prevalence is related to individual-level traits (age and breeding season), species-specific traits (diet, foraging strata, period of activity, species body weight, migratory status, and nest shape), and climate factors (temperature and rainfall). We screened DNA from 1,466 birds of 70 species captured monthly from April 2013 to March 2015. We detected an overall prevalence (Plasmodium/ Haemoproteus infection) of 22% (44 host species) and parasite’s lineages were identified by mitochondrial cyt b gene. Our results showed that migration can be an important factor predicting the prevalence of Haemoproteus (Parahaemoproteus), but not Plasmodium, in hosts. Besides, the temperature, but not rainfall, seems to predict the prevalence of Plasmodium in this bird community. Neither individual-level traits analyzed nor the other species-specific traits tested were related to the probability of a bird becoming infected by haemosporidians. Our results point the importance of conducting local studies in particular environments to understand the degree of generality of factors impacting parasite prevalence in bird communities. Despite our attempts to find patterns of infection in this bird community, we should be aware that an avian haemosporidian community organization is highly complex and this complexity can be attributed to an intricate net of factors, some of which were not observed in this study and should be evaluated in future studies. We evidence the importance of looking to host-parasite relationships in a more close scale, to assure that some effects may not be obfuscated by differences in host life-history.