Evolution of immune genes in island birds: reduction in population sizes can explain island syndrome

Shared ecological conditions encountered by species that colonize islands often lead to the evolution of convergent phenotypes, commonly referred to as “island syndrome”. Reduced immune functions have been previously proposed to be part of this syndrome, as a consequence of the reduced diversity of pathogens on island ecosystems. According to this hypothesis, immune genes are expected to exhibit genomic signatures of relaxed selection pressure in island species. In this study, we used comparative genomic methods to study immune genes in island species (N = 20) and their mainland relatives (N = 14). We gathered public data as well as generated new data on innate (TLR: Toll-Like Receptors, BD: Beta Defensins) and acquired immune genes (MHC: Major Histocompatibility Complex classes I and II), but also on hundreds of genes with various immune functions. As a control, we used a set of 97 genes, not known to be involved in immune functions based on the literature, to account for the increased drift effects of the lower effective population sizes in island species. We used synonymous and non-synonymous variants to estimate the selection pressure acting on immune genes. We found that BDs and TLRs have higher ratios of non-synonymous over synonymous polymorphisms (Pn/Ps) than randomly selected control genes, suggesting that they evolve under a different selection regime. However, simulations show that this is unlikely to be explained by ongoing positive selection or balancing selection. For the MHC genes, which evolve under balancing selection, we used simulations to estimate the impact of population size variation. We found a significant effect of drift on immune genes of island species leading to a reduction in genetic diversity and efficacy of selection. However, the intensity of relaxed selection was not significantly different from control genes, except for MHC class II genes. These genes exhibit a significantly higher level of non-synonymous loss of polymorphism than expected assuming only drift and evolution under frequency dependent selection, possibly due to a reduction of extracellular parasite communities on islands. Overall, our results showed that demographic effects lead to a decrease in the immune functions of island species, but the relaxed selection that is expected to be caused by a reduced parasite pressure may only occur in some categories of immune genes.