Host plant association and spatial autocorrelation as drivers of genetic differentiation among populations of a regionally host-specific insect herbivore
MetadataShow full metadata
Disentangling the processes responsible for structuring patterns of biodiversity at all spatial scales challenges biologists as such patterns represent evolutionary and ecological processes coupled with spatial autocorrelation among sample units. The phytophagous insect, Belonocnema treatae (Hymenoptera: Cynipidae) exhibits regional specialization on three species of live oaks throughout its geographic range across the southern USA. Here I ask whether populations of B. treatae affiliated with each host plant species exhibit genetic differentiation that parallels host plant phylogeography while controlling for spatial autocorrelation among sampling locations. I used genotyping-by-sequencing of 1,219 B. treatae collected from 58 sites distributed across the geographic ranges of the three host plants to identify 6,987 common single nucleotide variants. Population genomic structure was then investigated using a hierarchical Bayesian model to assign individuals to genetic clusters and estimate admixture proportions. To control for spatial autocorrelation when investigating the role of host plant affiliation in determining patterns of among-population genetic differentiation, Distance-based Moran’s eigenvector mapping was used to construct regression variables summarizing spatial structure inherent in the sampling design. Redundancy analysis (RDA) incorporating these spatial variables was then used to simultaneously examine the roles of host plant affiliation and spatial autocorrelation in determining patterns of among-population genetic differentiation. Patterns of genomic variation indicate a distinct geographic division east and west of Mississippi, coupled with discrete host associated lineages in the eastern portion of the species' range and clinal host-associated lineages in the west. RDA confirmed host association as a significant predictor of genomic variation, but longitudinal spatial autocorrelation explained a larger proportion of B. treatae's genomic variation. These results suggest B. treatae and the host plants share a common evolutionary history that links their patterns of genomic differentiation.