Phylogeography of Muller's termite frog suggests the vicariant role of the Central Brazilian Plateau

Aim To test three different scenarios to account for the geographic distribution of genetic variation in Dermatonotus muelleri, a fossorial frog endemic of the South American diagonal of open formations (DOF) formed by Chaco, Cerrado, and Caatinga biomes: (a) The pan-DOF hypothesis, where these biomes behave as a single biogeographical unit and DOF species lack population structure; (b) the southwest-northeast (SW-NE) hypothesis, where structured populations along the DOF have either asymmetric or unidirectional gene flow in the northeast direction; (c) the vicariance hypothesis, where a pattern of population structure with no or reduced gene flow is observed. Location Methods Caatinga, Cerrado, and Chaco in South America. We sampled 179 individuals across the species distribution and sequenced one mitochondrial and two nuclear markers. We evaluated population structure, the presence, direction, and asymmetry of gene flow, and demographic changes to test three different phylogeographic hypotheses for D. muelleri using a model-based simulation approach. Results Main conclusions We found two geographically structured lineages with no gene flow, refuting pan-DOF hypothesis. The Southwest lineage occurs in Chaco and western Cerrado and the Northeast lineage in eastern Cerrado and Caatinga. Our results support the model with demographic change for the Northeast lineage and constant size for Southwest lineage. The vicariance model with divergence time estimated around 4 Ma was preferred. Vicariance was the best explanation for the geographic distribution of genetic variation in Dermatonotus. Divergence time supports the role of Neogene orogeny in central Brazil for the splitting of Dermatonotus lineages. The influence of Pleistocene climatic fluctuations seems restricted to changes in one lineage's population size. Therefore, both ancient and recent events shaped the phylogeography of Dermatonotus. Genetic divergence, lack of gene flow, and the presence of a geographic barrier between lineages indicate the existence of a cryptic, undescribed species in northeastern Brazil.