No abstract

The microbiota has a broad range of impacts on host physiology and behaviour, pointing out the need to improve our comprehension of the drivers of host-microbiota composition. Of particular interest is whether the microbiota is acquired passively, or whether and to what extent hosts themselves shape the acquisition and maintenance of their microbiota. In birds, the uropygial gland produces oily secretions used to coat feathers that have been suggested to act as an antimicrobial defence mechanism regulating body feather microbiota. However, our comprehension of this process is still limited. In this study, we for the first time coupled high-throughput sequencing of the microbiota of both body feathers and the direct environment (i.e., the nest) in great tits with chemical analyses of the composition of uropygial gland secretions to examine whether host chemicals have either specific effects on some bacteria or nonspecific broad-spectrum effects on the body feather microbiota. Using a network approach investigating the patterns of co-occurrence or co-exclusions between chemicals and bacteria within the body feather microbiota, we found no evidence for specific promicrobial or antimicrobial effects of uropygial gland chemicals. However, we found that one group of chemicals was negatively correlated to bacterial richness on body feathers, and a higher production of these chemicals was associated with a poorer body feather bacterial richness compared to the nest microbiota. Our study provides evidence that chemicals produced by the host might function as a nonspecific broad-spectrum antimicrobial defence mechanism limiting colonization and/or maintenance of bacteria on body feathers, providing new insight about the drivers of the host's microbiota composition in wild organisms.

Understanding the ecology and evolution of parasites is contingent on identifying the selection pressures they face across their infection landscape. Such a task is made challenging by the fact that these pressures will likely vary across time and space, as a result of seasonal and geographical differences in host susceptibility or transmission opportunities. Avian haemosporidian blood parasites are capable of infecting multiple co‐occurring hosts within their ranges, yet whether their distribution across time and space varies similarly in their different host species remains unclear. Here, we applied a new PCR method to detect avian haemosporidia (genera Haemoproteus, Leucocytozoon, and Plasmodium) and to determine parasite prevalence in two closely related and co‐occurring host species, blue tits (Cyanistes caeruleus, N = 529) and great tits (Parus major, N = 443). Our samples were collected between autumn and spring, along an elevational gradient in the French Pyrenees and over a three‐year period. Most parasites were found to infect both host species, and while these generalist parasites displayed similar elevational patterns of prevalence in the two host species, this was not always the case for seasonal prevalence patterns. For example, Leucocytozoon group A parasites showed inverse seasonal prevalence when comparing between the two host species, being highest in winter and spring in blue tits but higher in autumn in great tits. While Plasmodium relictum prevalence was overall lower in spring relative to winter or autumn in both species, spring prevalence was also lower in blue tits than in great tits. Together, these results reveal how generalist parasites can exhibit host‐specific epidemiology, which is likely to complicate predictions of host–parasite co‐evolution.