An important step in the integration of ecology and genomics is the progression from molecular studies of relatively simple model systems to complex field systems. The recent availability of sequenced genomes from key plants is leading to a new understanding of the molecular drivers of community composition and ecosystem processes. As genome sequences accumulate for species that form intimate associations in nature, a detailed view may emerge as to how these associations cause changes among species at the nucleotide level. This advance could dramatically alter views about the structure and evolution of communities and ecosystems.
To test the hypothesis that genes have extended phenotypes on the community, we quantified how genetic differences among cottonwoods affect the diversity, abundance, and composition of the dependent arthropod community. Over two years, five major patterns were observed in both field and common-garden studies that focused on two species of cottonwoods and their naturally occurring F 1 and backcross hybrids (collectively referred to as four different cross types). We did not find overall significant differences in arthropod species richness or abundance among cottonwood cross types. We found significant differences in arthropod community composition among all cross types except backcross and narrowleaf cottonwoods. Thus, even though we found similar richness among cross types, the species that composed the community were significantly different. Using vector analysis, we found that the shift in arthropod community composition was correlated with percent Fremont alleles in the host plant, which suggests that the arthropod community responds to the underlying genetic differences among trees. We found 13 arthropod species representing different trophic levels that were significant indicators of the four different cross types. Even though arthropod communities changed in species composition from one year to the next, the overall patterns of community differences remained remarkably stable, suggesting that the genetic differences among cross types exert a strong organizing influence on the arthropod community. Together, these results support the extended phenotype concept. Few studies have observationally and experimentally shown that entire arthropod communities can be structured by genetic differences in their host plants. These findings contribute to the developing field of community genetics and suggest a strategy for conserving arthropod diversity by promoting genetic diversity in their host plants.
To examine how genetic variation in a plant population affects arthropod community richness and composition, we quantified the arthropod communities on a synthetic population of Eucalyptus amygdalina, E. risdonii, and their F1 and advanced-generation hybrids. Five major patterns emerged. First, the pure species and hybrid populations supported significantly different communities. Second, species richness was significantly greatest on hybrids (F1 > F2 > E. amygdalina > E. risdonii). These results are similar to those from a wild population of the same species and represent the first case in which both synthetic and wild population studies confirm a genetic component to community structure. Hybrids also acted as centers of biodiversity by accumulating both the common and specialist taxa of both parental species (100% in the wild and 80% in the synthetic population). Third, species richness was significantly greater on F1s than the single F2 family, suggesting that the increased insect abundance on hybrids may not be caused by the breakup of coadapted gene complexes. Fourth, specialist arthropod taxa were most likely to show a dominance response to F1 hybrids, whereas generalist taxa exhibited a susceptible response. Fifth, in an analysis of 31 leaf terpenoids that are thought to play a role in plant defense, hybrids were generally intermediate to the parental chemotypes. Within the single F2 family, we found significant associations between the communities of individual trees and five individual oil components, including oil yield, demonstrating that there is a genetic effect on plant defensive chemistry that, in turn, may affect community structure. These studies argue that hybridization has important community-level consequences and that the genetic variation present in hybrid zones can be used to explore the genetic-based mechanisms that structure communities.