Ecological and genetic divergence in Gunnison’s prairie dogs (Cynomys gunnisoni)

Background/Question/Methods

One of the principal goals of ecology and evolutionary biology is describing the mechanisms that lead to the diversity of life on earth. Our understanding of ecological divergence has improved greatly in the last several years; however, the ecological speciation literature is primarily focused on elucidating responses to selection at the genomic level. An important prerequisite to describing genomic divergence is understanding how ecology itself contributes to divergence. Thus, research should attempt to determine which types of divergent ecological selection pressures facilitate speciation. This study evaluates the evolutionary divergence of Gunnison’s prairie dogs (Cynomys gunnisoni), whose populations are found in starkly different habitats, presumably exerting distinct selection pressures. As a result, some have advocated the recognition of two distinct subspecies. Geographical differences in alarm calls and morphology are linked to habitat structure; both vocalizations and morphological crypsis are of paramount importance to predator avoidance. However, it remains unknown whether these differences translate into evolutionary divergence. This study integrates ecology and evolution by examining the extent to which selective differences in habitat structure contribute to divergence of behavior (vocalizations) that translates into the early stages of evolutionary divergence.

Results/Conclusions

Habitat parameters were quantified at 48 sites across the range of Gunnison’s prairie dogs using ArcMap for each site, and alarm calls were analyzed from 18 colonies.  We genotyped 838 individuals at two mitochondrial genes and 15 microsatellite loci to assess evolutionary divergence.  We found significant differences in vocalizations that were linked to habitat; genetic differences were due to geographic location (i.e., isolation by distance) and habitat type.  Phylogenetic reconstruction in BEAST resolved three, rather than two, distinct clades.  Fifteen populations contained haplotypes from multiple clades or were admixed with respect to microsatellite genotypes (three were considered admixed for both types of markers).  Gene flow was high among populations both within and among clades, but we also observed significant differentiation among almost all pairs of populations (and clades).  We conclude that there is significant genetic structure among the three clades that partially parallels the observed behavioral and ecological differences. Gunnison’s prairie dogs may be in the early stages of evolutionary divergence.  Future work should address whether mate preference is linked to differences in behavior such as vocalizations, and whether fitness differences exist between prairie dog subspecies in distinct habitats.