Species boundaries and phylogenetic relationships in the critically endangered Asian box turtle genus Cuora

Turtles are currently the most endangered major clade of vertebrates on earth, and Asian box turtles (Cuora) are in catastrophic decline. Effective management of this diverse turtle clade has been hampered by human-mediated, and perhaps natural hybridization, resulting in discordance between mitochondrial and nuclear markers and confusion regarding species boundaries and phylogenetic relationships among hypothesized species of Cuora. Here, we present analyses of mitochondrial and nuclear DNA data for all 12 currently hypothesized species to resolve both species boundaries and phylogenetic relationships. Our 15-gene, 40-individual nuclear data set was frequently in conflict with our mitochondrial data set; based on its general concordance with published morphological analyses and the strength of 15 independent estimates of evolutionary history, we interpret the nuclear data as representing the most reliable estimate of species boundaries and phylogeny of Cuora. Our results strongly reiterate the necessity of using multiple nuclear markers for phylogeny and species delimitation in these animals, including any form of DNA “barcoding”, and point to Cuora as an important case study where reliance on mitochondrial DNA can lead to incorrect species identification.

Figure optionsDownload as PowerPoint slideHighlights
► We generate phylogenies and networks from one mitochondrial and 15 nuclear loci.
► We estimate the species tree for all species of Cuora.
► We find introgression is prevalent among the Cuora trifasciata species complex.
► Due to introgression, mitochondrial barcoding would fail to correctly identify species for this group of turtles.
► The multilocus nuclear gene tree is a more accurate estimate of phylogeny than the mitochondrial gene tree.