Species-level monitoring of rare and invasive fishes using eDNA metabarcoding in the middle and upper Yarlung Zangbo River, Tibet

Abstract

Fish diversity of the Yarlung Zangbo River is very sensitive and vulnerable to biological invasion, anthropogenic activities and climate change, especially in the upper and middle reaches where several endemic fishes have become endangered and nearly ten invasive fishes have been established. Here, we used environmental DNA (eDNA) metabarcoding to monitor rare and invasive fishes, and to assess diversity in 25 sites from two wetlands (Lalu and Chabalang) and the main channel (YT), within the upper and middle reaches. To obtain a species-level resolution, we evaluated species discrimination potentials of three mitochondrial markers and found Cytb had the highest average genetic distance at each taxonomic level followed by COI and 12S. The 12S was unqualified for species assignment, as two species shared identical haplotypes. The newly designed Cytb primers used for metabarcoding showed an average mismatch of 0.28 and amplified well across species. In total, 8942 operational taxonomic units (OTUs) were obtained based on a 100% identity threshold, among which 98.1% were assigned to 24 fishes based on our custom-made database and the remaining were assigned to six fishes based on the NCBI nt database. Almost all captured fishes were detected by the eDNA method except for two species. However, 12 fishes detected by the eDNA method were not listed in catch data for several sites, including one endangered species (Oxygymnocypris stewartii), four previously recorded non-native species and two unrecorded non-native species (Monopterus albus and Siniperca chuatsi). The alpha diversities estimated by eDNA and capture-based methods were correlated for sites at Lalu. Both methods revealed significant differences in community composition between YT and the wetlands. Our results provide both basic information for conservation and management of rare and invasive fishes in the Yarlung Zangbo River and a framework of fish eDNA metabarcoding with a species-level resolution.