Biome specificity of distinct genetic lineages within the four-striped mouse Rhabdomys pumilio (Rodentia: Muridae) from southern Africa with implications for taxonomy.

Within southern Africa, a link between past climatic changes and faunal diversification has been hypothesized for a diversity of taxa. To test the hypothesis that evolutionary divergences may be correlated to vegetation changes (induced by changes in climate), we selected the widely distributed four-striped mouse, Rhabdomys, as a model. Two species are currently recognized, the mesic-adapted R. dilectus and arid-adapted R. pumilio. However, the morphology-based taxonomy and the distribution boundaries of previously described subspecies remain poorly defined. The current study, which spans seven biomes, focuses on the spatial genetic structure of the arid-adapted R. pumilio (521 specimens from 31 localities), but also includes limited sampling of the mesic-adapted R. dilectus (33 specimens from 10 localities) to act as a reference for interspecific variation within the genus. The mitochondrial COI gene and four nuclear introns (Eef1a1, MGF, SPTBN1, Bfib7) were used for the construction of gene trees. Mitochondrial DNA analyses indicate that Rhabdomys consists of four reciprocally monophyletic, geographically structured clades, with three distinct lineages present within the arid-adapted R. pumilio. These monophyletic lineages differ by at least 7.9% (±0.3) and these results are partly confirmed by a multilocus network of the combined nuclear intron dataset. Ecological niche modeling in MaxEnt supports a strong correlation between regional biomes and the distribution of distinct evolutionary lineages of Rhabdomys. A Bayesian relaxed molecular clock suggests that the geographic clades diverged between 3.09 and 4.30Ma, supporting the hypothesis that the radiation within the genus coincides with paleoclimatic changes (and the establishment of the biomes) characterizing the Miocene-Pliocene boundary. Marked genetic divergence at the mitochondrial DNA level, coupled with strong nuclear and mtDNA signals of non-monophyly of R. pumilio, support the notion that a taxonomic revision of the genus is needed.