Resolving the fine-scale population structure of Nesospiza buntings using a genetic multi-marker system

Abstract

Elucidating the factors driving population diversification and speciation has been a much debated and investigated topic since the time of Darwin. Genetic and environmental factors that underlie adaptive evolution are of particular interest in the study of evolutionary biology. Oceanic islands are ideal study systems because they offer ecological opportunities linked to their depauperate communities, new food sources, and low levels of competition, that frequently drive rapid adaptive radiations. The Nesospiza buntings that inhabit the South Atlantic Tristan da Cunha archipelago provide an excellent system for studying a simple adaptive radiation that has occurred in parallel on two of the islands. Speciation has occurred to completion on Nightingale Island, where a large-billed habitat specialist and a small-billed habitat generalist occur with no known interbreeding. On the larger Inaccessible Island, where habitat diversity is greater, the population structure is more complex and three sub-species have been identified based on morphology. An upland and a lowland morph of the small-billed habitat generalist, and a large-billed habitat specialist occur at different locations on the island according to the seed size distribution. There is extensive hybridisation between all three subspecies across the ecotone where a wide range of seed sizes are available. The purpose of this study is to resolve the fine-scale population structure on Inaccessible Island by incorporating morphological and geographic information with molecular data. The incomplete speciation with ongoing hybridisation on Inaccessible Island also provides an ideal system with which to investigate some of the factors underlying avian speciation, such as genes and traits under selection. I make use of neutral microsatellite and nuclear SNP loci in addition to bill-size and geographic sampling data to examine the fine-scale population structure on Inaccessible Island. Results show a weak association of bill-size with genetic variation, but no pattern of isolation by distance. Extensive gene flow between all sampling localities is likely to counteract divergent selection by homogenising genetic variation. I made use of the candidate gene approach to identify and investigate six genes that are likely to be under divergent selection in Nesospiza. No sequence variation was found within or between Nesospiza populations. This lack of variation may be due to functional constraint acting on the genes while phenotypic variation may be attributed to differences in gene expression or due to cis- and/or trans-acting elements, rather than DNA sequence variation in the coding regions of the genes. To further investigate selection on Inaccessible Island, MHC class IIâ genes were characterised in Nesospiza buntings. Results suggest that there is little divergent selection on geographically separated populations. A low parasite load or the absence of strong mate preference may result in a lack of local population structure. This study finds evidence of fine-scale population structure of Nesospiza buntings at the Tristan da Cunha archipelago. On Inaccessible Island, the maintenance of genetic and morphological differentiation despite ongoing gene flow is likely due to divergent selection on morphological traits, such as bill-size, for optimal exploitation of the local seed size availability. Copyright