Phylogenetic relationships of the Chinese Labeoninae (Teleostei, Cypriniformes) derived from two nuclear and three mitochondrial genes

The majority of genera within the Labeoninae occur in the South of China and the phylogeny of the subfamily Labeoninae has been a controversial topic over the years. The early and more recent phylogenetic results based on morphology are not in agreement, and some of the molecular analyses contradict those based on morphology. However, none of the previous studies has included an extensive sampling of Labeoninae genera. In this study, partial sequences of two nuclear (exon 3 of recombination activating protein 1 and rhodopsin) and three mitochondrial genes (cytochrome b, cytochrome oxidase subunit I and 16S ribosomal RNA) from 39 ingroup taxa and 11 outgroup taxa were used analysed to provide a hypothesis of phylogenetic relationships within the Chinese Labeoninae. The results supported the monophyly of Labeoninae and refuted the subdivision within Labeoninae based on the oromandibular morphology. It also further confirmed that the presence of a disc on lower lip arose through convergent evolution. Labeo was the basal clade and Osteochilus was closely related to Cirrhinus, forming the second basal clade. The monophyly of Garra and Bangana was refuted in this study, and they were both subdivided into two lineages. One lineage of Garra had closer relationships with Crossocheilus and Akrokolioplax, and the other lineage had closer relationships with Placocheilus. One lineage of Bangana represented the species with broadly interrupted postlabial groove; these species represented the true Bangana. The other lineage of Bangana included the species with continuous postlabial groove, and these species should be assigned to a new genus. Discogobio, as currently conceived, was paraphyletic, but it rendered monophyletic with the inclusion of Discocheilus. These results indicated that Discocheilus should be synonymised with Discogobio. Hongshuia and Sinocrossocheilus, respectively, formed independent lineages related to a lineage consisting of Pseudocrossocheilus plus seven other nominal genera. The relationships received weak support, but the repeatability of the clades could form the basis of recognising only the three genera mentioned above.