Friday, August 6, 2010

PS 107-135: Insect herbivory and flower color variation in the Arctic: Floral pigmentation is associated with foliar herbivory and induction of chemical defenses in Parrya nudicaulis (Brassicaceae)

Matthew L. Carlson1, Justen B. Whittall2, Justin R. Fulkerson1, Cindy Dick2, Savannah L. Hamilton1, Mimi Portilla2, Laura C. Schneller1, Heidi Cossentine2, and Daniel J. Kliebenstein3. (1) University of Alaska Anchorage, (2) Santa Clara University, (3) University of California, Davis

Background/Question/Methods

Increasingly, biotic forces are recognized as structuring Arctic ecosystems and influencing selection regimes. Little emphasis has been directed to insect herbivores in the Arctic, which are considered to be uncommon. At lower latitudes invertebrate herbivory is an important driver in plant chemical defenses and can affect evolution of seemingly unrelated traits, such as flower color. In radish (Raphanus sativus), floral anthocyanin pigments are associated with lower herbivory and higher induced chemical defense levels (Irwin et al. 2003). As part of a cross-university undergraduate Arctic Biology Ecology and Genomics course, we characterized levels of herbivory in Alaskan populations of the arctic mustard, Parrya nudicaulis and have investigated the metabolomic and molecular basis of defense. Specifically, we tested if 1) if individuals with greater floral pigmentation have reduced herbivory 2), if floral anthocyanin pigmentation is positively correlated with constitutive and induced glucosinolate defensive compound concentrations, and 3) if genetic polymorphisms in glucosinolate genes are related to flower color.

Results/Conclusions

Across populations, 37.5% of individuals showed signs of herbivory. In one population, more than 80% of individuals were damaged, in many cases with greater than 50% lost biomass. Plants with evidence of herbivory had lighter colored flowers than plants without evidence of herbivory in one population. Plants with intermediate flower color had lower constitutive levels of 3 and 4 carbon glucosinolates than the lighter and darker colored phenotypes. Plants of all flower colors had significant induction of 5 and 6 carbon glucosinolates, with the dark and intermediate phenotypes tending to induce greater concentrations of this class of compound following simulated herbivory. At the molecular level, transcriptome analysis on the Illumina platform identified homologues of all eight glucosinolate biosynthetic genes, covering an average of 47% of their protein coding regions (18.8% - 95% coverage). In a more detailed examination of one of these genes, the MAM locus that is involved in carbon chain elongation in glucosinolates, we have identified at least 13 single nucleotide polymorphisms that correlate with flower color out of 1762 base pairs surveyed. Our results indicate that insect herbivory can be common in this arctic system. Lower herbivory in dark and intermediate plants may be due to a greater ability to induce 5 and 6 carbon glucosinolates. Last, defense and flower color genes appear to be associated through selection or linkage.