Comparative phylogeography of the Yellow River schizothoracine fishes (Cyprinidae): vicariance, expansion, and recent coalescence in response to the Quaternary environmental upheaval in the Tibetan Plateau.
Ziyuan Duan,Kai Zhao,Zuogang Peng,Junbing Li,Rui Diogo,Xinquan Zhao,Shunping He
DOI: https://doi.org/10.1016/j.ympev.2009.03.034
IF: 5.019
2009-01-01
Molecular Phylogenetics and Evolution
Abstract:Highly specialized grade (HSG; genera Gymnocypris, Oxygymnocypris, Schizopygopsis, Platypharodon and Chuanchia) of the Schizothoracinae (Cypriniformes: Cyprinidae) are endemic to the Qinghai-Tibet Plateau (QTP). Previously, two distinct ecomorphs were recognized according to trophic traits. One was a limnetic omnivore with normal lower jaw morphology, terminal mouth, and moderate or dense gill rakers, mostly inhabiting in open water of lakes, including Gymnocypris and Oxygymnocypris. Another was a benthic feeder with inferior mouth, sparse gill rakers and sharp horny sheath on the lower jaw for scraping of attached prey off hard substrates, including Schizopygopsis, Platypharodon and Chuanchia. However, traditional taxonomy of HSG based on these trophic traits presented extensive conflicts with the molecular studies in recent years. The possible cause could be convergent evolution in morphology, retention of ancestral polymorphisms or mitochondrial introgression, but these hypotheses could not be assessed due to incomplete taxon sampling and only mitochondrial data employed in previous works. Here, we conducted the most comprehensive molecular analysis on HSG fishes to date, using four mitochondrial loci and 152,464 genome-wide SNPs, and including 21 of 24 putative species and one undescribed Schizopygopsis species. Both SNP and mtDNA trees confirmed extensive paraphyly of genera Gymnocypris and Schizopygopsis, where species often were clustered together by watershed instead of by genus. Basal split into the north clade B and the south clade C (ca. 3.03 Ma) approximately by the Tanggula-Tanitawen Mountains in SLAF tree coincided with a violent uplift of the QTP during the phase A of ‘Qingzang movement’ (ca. 3.6 Ma). Ancestral state reconstruction of the trophic ecomorph showed that the limnetic omnivore ecomorph had evolved repeatedly in clade B and C. Furthermore, we presented a striking case of convergent evolution between two ‘subspecies’ Gymnocypris chui chui and G. chui longimandibularis, which had diverged as early as two million years ago (ca. 2.42 Ma). Ecological analyses revealed that similar food utilization, particularly in zooplankton, was the main underlying driving force. This work showed an example of taxonomy with the most extensive errors at the genus/species levels due to convergent evolution and suggested that trophic traits could be misleading in fish taxonomy. Therefore, we propose a major generic revision for HSG species.