COS 174-6 - Investigating the phylogeography of the North American medicinal leech, Macrobdella decora.

Background/Question/Methods

The data regarding population structure and biogeographic affinities of leeches are scarce despite their important roles in many ecosystems, where they function as prey for fish, birds and reptiles, and are vectors of blood-borne diseases. To begin to remedy this knowledge gap, my research focuses on performing a phylogeographic analysis of the North American medicinal leech, Macrobdella decora. The species is widespread throughout eastern North America, and although the species is thought to disperse with its hosts while feeding, there is a conspicuous paucity of data relating to their population structure. A total of 226 M. decora specimens were collected from 36 localities, including from the type locality in Lake Vermillion, Minnesota. We sequenced 4 loci for each specimen: mitochondrial COI and ND1, and nuclear 28S and 18S rDNA. Using these sequences, I identified boundaries between sampled populations using a pluralistic approach to phylogenetic tree-building and haplotype networks. I ran an AMOVA test to quantitatively determine whether there is significant genetic structure between populations, and a Mantel test to determine whether or not there is a correlation between genetic distance and geographical distance within our samples.

Results/Conclusions

When examining the phylogeny, there is an overall lack of geographic structure among the different populations of M. decora. Similarly, the haplotype network shows that identical haplotypes are found across distant localities, and there is little fidelity to watersheds. Despite this apparent lack of structure, the AMOVA found that there are significant differences between the genetic variation within populations and between populations (FST = 0.65, P = 0.00), although specimens compared between watersheds were not genetically distinct (FST = 0, P = 0.48). Finally, the results from the Mantel test also showed a significant correlation between geographic distance and genetic distance (r = 0.18, P = 0.02). This subtle structure may be due to periodic glaciation, which historically covered the sampled area. As glaciers advanced and retreated several times during the Late Pleistocene and Early Holocene, M. decora may have also periodically contracted its range into refuge populations and then expanded again as the ice retreated, which would result in periodic bottlenecks to gene flow in a region. While some genetic structure was preserved as seen with the AMOVA and Mantel test, these range contractions during the ice age may have muddied the structure in the phylogeny and haplotype dispersal.