Differentiation underground: Range-wide multilocus genetic structure of the silvery mole-rat does not support current taxonomy based on mitochondrial sequences

The silvery mole-rat (Heliophobius argenteocinereus) is a solitary subterranean rodent with its distribution centred mainly in miombo woodlands of eastern Africa. This part of the continent was significantly influenced by the formation of the East African Rift System (EARS) during the last 25 Mya and by pronounced climatic changes in the Plio-Pleistocene that have caused genetic differentiation leading even to speciation events in many organisms. Recently, based on analysis of mitochondrial (mt) DNA sequences, it was suggested that H. argenteocinereus is a complex of six to eight species that diverged from the early to middle Miocene. In the present study, we significantly extended the sampling, re-analysed mtDNA datasets and analysed nuclear markers with the aim to assess the evolutionary history of Heliophobius. If we do not consider the old museum samples from south-eastern Democratic Republic of Congo (very divergent short mtDNA sequences obtained from ancient DNA, requiring further study from fresh material), the genus Heliophobius is composed of three major mtDNA lineages with parapatric distribution. The Rukwa Rift (+Mbeya triple junction) at the Zambia-Tanzania border, Lake Malawi, and the Eastern Arc Mountains form the biogeographical divides among these clades. The relatively shallow differences among the mitochondrial clades, divergence dating based on the use of the fossilised birth-death ratio model and a multi-locus dataset, and a very similar pattern of genetic structure to other rodents inhabiting the same area and habitat, suggest that the evolutionary history of the extant silvery mole-rat was predominantly influenced by the climatic fluctuations in the Plio-Pleistocene. Awaiting further studies employing genomic, ecological, morphological or behavioural data, we advocate for using the single name H. argenteocinereus for all evolutionary lineages within this taxon, because (1) comparison of the genetic structure observed in mtDNA and nuclear markers suggest hybridization between at least some mtDNA lineages in the contact zones; and (2) new samples close to the type localities suggest incorrect use of previous names.