Phylogeography of a successful aerial disperser: the golden orb spider Nephila on Indian Ocean islands

BACKGROUND: The origin and diversification patterns of lineages across the Indian Ocean islands are varied due to the interplay of the complex geographic and geologic island histories, the varying dispersal abilities of biotas, and the proximity to major continental landmasses. Our aim was to reconstruct phylogeographic history of the giant orbweaving spider (Nephila) on western Indian Ocean islands (Madagascar, Mayotte, Réunion, Mauritius, Rodrigues), to test its origin and route of dispersal, and to examine the consequences of good dispersal abilities for colonization and diversification, in comparison with related spiders (Nephilengys) inhabiting the same islands, and with other organisms known for over water dispersal. We used mitochondrial (COI) and nuclear (ITS2) markers to examine phylogenetic and population genetic patterns in Nephila populations and species. We employed Bayesian and parsimony methods to reconstruct phylogenies and haplotype networks, respectively, and calculated genetic distances, fixation indices, and estimated clade ages under a relaxed clock model. RESULTS: Our results suggest an African origin of Madagascar Nephila inaurata populations via Cenozoic dispersal, and the colonization of the Mascarene islands from Madagascar. We find evidence of gene flow across Madagascar and Comoros. The Mascarene islands share a common 'ancestral' COI haplotype closely related to those found on Madagascar, but itself absent, or as yet unsampled, from Madagascar. Each island has one or more unique haplotypes related to the ancestral Mascarene haplotype. The Indian Ocean N. inaurata are genetically distinct from the African populations. CONCLUSIONS: Nephila spiders colonized Madagascar from Africa about 2.5 (0.6-5.3) Ma. Our results are consistent with subsequent, recent and rapid, colonization of all three Mascarene islands. On each island, however, we detected unique haplotypes, consistent with a limited gene flow among the islands subsequent to colonization, a scenario that might be referred to as speciation in progress. However, due to relatively small sample sizes, we cannot rule out that we simply failed to collect Mascarene haplotypes on Madagascar, a scenario that might imply human mediated dispersal. Nonetheless, the former interpretation better fits the available data and results in a pattern similar to the related Nephilengys. Nephilengys, however, shows higher genetic divergences with diversification on more remote islands. That the better disperser of the two lineages, Nephila, has colonized more islands but failed to diversify, demonstrates how dispersal ability can shape both the patterns of colonization and formation of species across archipelagos.