Development of an environmental DNA metabarcoding assay for aquatic vascular plant communities

Abstract

AbstractEnvironmental DNA (eDNA) metabarcodes allow for the simultaneous detection of multiple taxa if the barcode regions meet several key requirements including conserved primer‐binding sites, interspecific variability that exceeds intraspecific variability, and relatively short amplicons. Currently, there are no established metabarcoding assays for aquatic vascular plants, which could limit biodiversity assessments and the early detection of alien species. We used a combination of novel and pre‐existing primers to generate eDNA metabarcodes from three gene regions that are commonly used for plant barcoding: two regions of chloroplast DNA (rbcL and matK) plus a segment of an internal transcribed spacer (ITS2). We optimized the assay on a mock community of 25 known species and then applied it to wild samples collected from two waterbodies in southern Ontario, Canada (Black River and Seymour Lake). Our novel rbcL primers, which amplify a fragment of ~220 bp, provided the most comprehensive description of the mock community, identifying all but one of the taxa to species or genus. In addition, after pooling data from five sites within each sampled waterbody, metabarcodes generated by this same primer pair identified more taxa than all other primer pairs; more specifically, 20 and 26 taxa were identified from Black River and Seymour Lake, respectively, to species or genus. Across the two sites, nine of the identified taxa are alien invasive aquatic plants. Five of these invasive species have no previous reports from our sites, and in some cases have no known established Ontario populations; our data therefore suggest an urgent need to increase surveillance for these aliens. Overall, our study showed that eDNA metabarcoding with a novel rbcL primer pair provides a cost‐effective method for simultaneously detecting multiple aquatic vascular plant taxa and is a valuable tool for the early detection of invasive species.