Limited genetic connectivity of aquatic invertebrate communities in Grand Canyon revealed using environmental DNA metabarcoding

Authors: Jared Freedman, Theodore Kennedy, Molly Burke, Dave Lytle

Year: 2023

Abstract

The ability to disperse between habitat patches is a key life history trait that influences how aquatic invertebrates spread their genes, escape predation pressure, and seek out optimal habitats, with impacts on both the community and population scale. While the dispersal ability of every species is different, aquatic invertebrates with complex life histories face unique dispersal barriers such as stream network configuration, flow regime variation, and inhospitable terrestrial landscapes. Furthermore, anthropogenic alterations such as dam regulation and climate change have the potential to influence community-wide patterns of dispersal and genetic connectivity. In Grand Canyon, anthropogenetic habitat alterations have greatly reduced species diversity in aquatic invertebrate communities in the mainstem Colorado River, though little is known about the ability of invertebrates to disperse among tributary habitats. While dispersal is difficult to observe directly, gene flow can reveal dispersal patterns since individuals must move between populations for gene flow to occur. Following the isolation-by-distance (IBD) framework of Hutchison and Templeton (1999), the relationship between genetic and geographic distance can be used to classify gene flow across the study area, and therefore estimate the dispersal ability of each community member. Environmental DNA (eDNA) metabarcoding presents a promising method for this type of community-wide study of genetic connectivity and dispersal by enabling the simultaneous quantification of invertebrate community composition and the genetic diversity of the constituent species. To investigate gene flow and dispersal of invertebrates in Grand Canyon tributaries, we collected eDNA water samples from 22 perennial tributaries, as well as adjacent upstream mainstem Colorado River habitats, and performed metabarcoding sequencing on a short region of the aquatic invertebrate COI mitochondrial gene fragment. Applying strict denoising and filtering algorithms, we extracted exact sequence variants (ESVs; analogous to haplotypes) before clustering into operational taxonomic units (OTUs; analogous to species). For each OTU with sufficient diversity for population genetic analysis, we calculated pairwise Fst and pairwise river network distance and compared them using linear regression. We then classified population genetic structure of each OTU as low gene flow (low dispersal), medium gene flow (medium dispersal), or high gene flow (high dispersal). We identified 1,201 ESVs across 448 OTUs, with 148 OTUs comprised of multiple haplotypes. Of these 148 OTUs, 73 contained sufficient genetic variability to perform population genetic analyses, with 58 OTUs exhibiting low gene flow, 10 OTUs exhibiting medium gene flow, and 5 OTUs exhibiting high gene flow. This result suggests a widespread barrier to aquatic invertebrate dispersal that limits gene flow between tributary communities. The genetic isolation of many Grand Canyon invertebrates highlights the difficulties these invertebrates face in dispersing between viable habitat patches when separated by the dam-impacted Colorado River. Aquatic invertebrates form the base of the aquatic food web, so reduced gene flow creates fragmented and potentially vulnerable populations that can have detrimental upstream effects on fish and other consumers.