I am an evolutionary ecologist interested in how species interactions affect the generation and maintenance of biological diversity, with a focus on evolution in plant-pollinator mutualisms. My research covers a range of ecological scenarios: plants without pollinators, ones relying on a single pollinator species, and plants visited by many insects.
How does floral evolution change when pollinators are lost?
My dissertation focused on the evolution of stamens (male reproductive organs) in the mustard plant family (Brassicaceae). Most mustards have both long and short stamens within a flower. In insect-pollinated (“outcrossing”) species, short stamens seem to increase offspring fathered by a plant, but the mechanism is unclear. In species that pollinate themselves (“selfing”), they have no known function. I studied the evolution of short stamens in both an outcrossing species and a selfer to understand how ecology and genetics contribute to the maintenance of this trait.
Wild radish (Raphanus raphanistrum) is an outcrossing species pollinated by a wide variety of insects. Evidence supports short stamens increasing fitness in this species. Using experimental stamen removal in field arrays and paternity analysis of the resulting seeds, I showed that the contribution of short stamens to fitness varies with the frequency of visitation. In particular, short stamens seem to be more advantageous at higher pollinator visitation rates.
By contrast, the model plant species Arabidopsis thaliana (also a mustard) is almost exclusively selfing. I showed experimentally that short stamens do not increase seed set in the absence of pollinators. By growing plants from across Europe, I found that partial loss of the useless short stamens is common – especially in the south, where genetic diversity is higher. Using quantitative trait locus mapping, I discovered short stamen loss is controlled by at least three separate areas of the genome, which interact to reduce their total effect. While the ecological shift to selfing should favor elimination of short stamens, two aspects of genetics (low diversity and genetic interactions) are likely slowing adaptive evolution.
Can tight interactions between species drive speciation?
The idea that biotic interactions foster diversification is supported by substantial circumstantial evidence, but is challenging to test. In my postdoc, I am using cutting-edge genomics approaches to ask whether putative coevolution in a plant-pollinator mutualism is contributing to speciation.
Two subspecies of Joshua tree (Yucca brevifolia brevifolia and Y. b. jaegeriana) have reciprocally obligate relationships with their pollinators, two sister species of yucca moth: both trees and moths depend solely on each other for reproduction. These plant-pollinator pairs exhibit trait matching suggestive of coevolution in key characters involved in the interaction – style length in the trees, ovipositor length in the moths. Although speciation in the moths is complete, the trees hybridize in a narrow contact zone in southern Nevada. We hypothesize that coevolution is contributing to reproductive isolation in the trees. If this is true, we expect to find disruptive selection on style length – in which trees with long and short styles produce many seeds, but intermediate-style length results in low seed set – in the hybrid zone.
I tested this prediction using RAD sequencing to identify 9516 genetic markers across the Joshua tree genome. I performed a genome wide association study (GWAS) to find loci associated with variation in tree style length. Fst-outlier tests identified loci that differed significantly between subspecies, suggesting that selection is driving them apart. Finally, I used Bayesian Genomic Clines analysis to test for disruptive selection in the hybrid zone. I determined that style-length loci are often under selection, and that selection is often disruptive. These results support the plant-pollinator interaction as a source of reproductive isolation, potentially contributing to speciation in Joshua trees.
How does a specialist evolve in response to a generalist partner?
Mutually obligate plant-pollinator relationships are relatively rare in nature, e.g, a plant pollinated by a specialist is usually visited by many other insects. In the Midwest, the common wildflower Claytonia virginica (Spring Beauty) exemplifies this. The solitary bee Andrena erigeniae specializes on C. virginica, but many other insects pollinate it. Data I collected as a PhD student show that C. virginica populations vary widely in floral color, shape, and selection on these traits. I expect strong selection on the specialist bee from the plants, particularly through chemicals responsible for flower color, but measuring selection on A. erigeniae was not possible. I look forward to applying population genomics to this system in the future.