What does mitogenomics tell us about the evolutionary history of the Drosophila buzzatii cluster (repleta group)

Abstract

The Drosophila repleta group is an array of more than 100 cactophilic species endemic to the \"New World\". The acquisition of the ability to utilize decaying cactus tissues as breeding and feeding sites is a key aspect that allowed the successful diversification of the repleta group in the American deserts. Within this group, the Drosophila buzzatii cluster is a South American clade of seven cactophilic closely related species in different stages of divergence, a feature that makes it a valuable model system for evolutionary research. However, even though substantial effort has been devoted to elucidating the phylogenetic relationships among members of the D. buzzatii cluster, the issue is still controversial. In effect, molecular phylogenetic studies performed to date generated ambiguous results since tree topologies depend on the kind of molecular marker employed. Curiously, even though mitochondrial DNA has become a popular marker in evolutionary biology and population genetics, none of the more than twenty Drosophila mitogenomes assembled so far belongs to this cluster. In this work we report the assembly of six complete mitogenomes of five species: D. antonietae, D. borborema, D. buzzatii, D. seriema and two strains of D. koepferae, with the aim to revisit the phylogenetic relationships and divergence times by means of a mitogenomic approach. The recovered topology using complete mitogenomes gives support to the hypothesis of the monophyly of that the D. buzzatii cluster and shows two main clades, one including D. buzzatii and D. koepferae (both strains) and the other the remaining species. These results are in agreement with previous reports based on a few mitochondrial and/or nuclear genes but in conflict with the results of a recent large-scale nuclear phylogeny, suggesting that nuclear and mitochondrial genomes depict different evolutionary histories.