Skip to main content
Dryad logo

Micro-endemic species of snails and amphipods show population genetic structure across very small geographic ranges


Walters, Ashley (2022), Micro-endemic species of snails and amphipods show population genetic structure across very small geographic ranges, Dryad, Dataset,


Understanding variation in population genetic structure, even across small distances and for species with extremely limited ranges, is critical for conservation planning and the development of effective management strategies for imperiled species. Organisms that occupy the same geographic extent can maintain different population structures, ranging from highly diverged to panmictic. Such differences can result from differences in biological characteristics such as dispersal ability or demographic history. We used microsatellite loci to evaluate population genetic structure and variation of four desert spring invertebrates having high to low dispersal ability: the lung snail Physa acuta, two species of gilled snails (Juturnia kosteri and Pyrgulopsis roswellensis; family Hydrobiidae) and the amphipod Gammarus desperatus. The study location represents entire species ranges for the micro-endemic hydrobiids and G. desperatus, while P. acuta is ubiquitous throughout much of North America. We found little evidence of significant population genetic structure for P. acuta and J. kosteri, but much more for P. roswellensis and G. desperatus. Our results demonstrate differences in habitat preference and/or dispersal ability between the species. While significant isolation-by-distance was detected in the two hydrobiids, dispersal is likely more limited in P. roswellensis than J. kosteri. This information provides insight into how gene flow shapes varying population genetic structure between species across small spatial scales (<100 km2). Most importantly, our results suggest that conservation agencies should not consider these microendemic species to be composed of single populations, but rather, that management plans for such species should account for population genetic variation across the species’ ranges.


From 2014-2016, we sampled 16 sites throughout the refuge for all four species, attempting to obtain 24 individuals of each species per site; however, presence and abundance varied among species and across sites. The sampled area includes the entire ranges of the hydrobiid snails and amphipods (Fig. 1). Each species was sampled throughout the refuge within a single sampling effort (i.e. all Gammarus and Physa specimens were collected in 2015, all hydrobiids in 2014) with the exception of Unit 5 sites (U5A, U5B, U5C) which were sampled in 2016. We sorted individuals while in the field; P. roswellensis and J. kosteri, were distinguished by operculum color using a hand lens or dissecting microscope. All specimens were stored in 95% ethanol until DNA was extracted using DNeasy kits (Qiagen, Inc.).  We used published microsatellite primers and methods to genotype 11 loci for Physa (Ansah, Inoue, Lang & Berg, 2014), 16 loci for J. kosteri and 20 loci for P. roswellensis (both described in Holste, Inoue, Lang & Berg, 2016) to assess genetic diversity of these species.

For G. desperatus, we developed novel microsatellite primers using the Plant-Microbe Genomics Facility at Ohio State University to perform de novo pyrosequencing in 1/4th of a picotiterplate using a Roche 454 FLX Titanium Genome Sequencer which produced 55,692 reads. From sequencing outputs, we used MSATCOMMANDER v1.0.8 (Faircloth, 2008) to identify putative microsatellite loci and PRIMER3 (Rozen & Skaletsky, 2000) to design primers on flanking regions. Tail sequences of M13R or CAG were added to the 5’-end of either the forward or reverse primer. We identified 202 putative microsatellite markers and we were able to design primers for 120 microsatellite loci. Trinucleotides were the most abundant repeat type (46.7% of all loci), followed by dinucleotides (42.5%), and tetranucleotides (10.8%). Twenty loci (Table S1, Supplementary Information) were consistently amplified and used for population-level analyses. These loci were amplified in 10 µL polymerase chain reactions (PCR) using Gotaq Master Mix (Promega Corporation), 0.2 µM of universal fluorescently labeled primer (M13R or CAG) including the non-tailed primer, 0.04 µM of tailed primer, and 10 ng of DNA. For each locus, we identified optimum annealing temperature using the following PCR conditions: initial denaturing at 95°C for 2 min; followed by 35 cycles at 95°C for 30 s, annealing at 48-62°C (or 55-65°C as a higher temperature gradient, if needed) for 45s, extension at 72°C for 1 min; and final extension at 72°C for 30 min.

For all species examined, fragment analyses were performed on an ABI Genetic Analyzer with LIZ600 size standard (Applied Biosystems, Inc.). We used PEAKSCANNER v1.0 (Applied Biosystems, Inc.) to score alleles and TANDEM v1.07 (Matschiner & Salzburger, 2009) to assign integers to DNA fragment sizes.