Understanding the Mechanisms for Aphid-derived Toxicity Toward Ladybeetles

The standing paradigm for tri-trophic chemical interactions between plants, herbivores, and natural enemies is that herbivorous insects use chemicals from their host plants to defend themselves against predators. Consequently, when an insect species is found to have to have different defensive qualities when feeding on different host plants, it is generally assumed that differential toxicity is a property of underlying differences in plant chemistry, rather than variation in the herbivore. In the cowpea aphid, Aphis craccivora, this assumption is faulty. It has been documented for decades that A. craccivora from some host plants (e.g. black locust) are toxic to the Asian ladybeetle, Harmonia axyridis. Our preliminary data shows that this toxic effect is not purely a property of the plant, but is also a function of aphid lineage. Furthermore, we have found that a different ladybeetle species, Coleomegilla maculata, was not killed by the aphids, although development rate and adult size were negatively affected by locust-origin A. craccivora lineages. Thus we have an ideal system for comparatively investigating the mechanism of a novel herbivore-derived toxicity. I propose to use whole transcriptome shotgun sequencing (RNA-seq) to compare gene expression of "toxic" and "nontoxic" A. craccivora lineages, and also to compare the transcriptomes of susceptible (H. axyridis) and resistant (C. maculata) ladybeetles feeding on the two aphid lineages. This approach will allow me to specifically test the hypothesis that differential metabolism of the non-protein amino acid canavanine (which has previously been shown to be present in locust) contributes to aphid toxicity. I will also be able to test whether a facultative bacterial endosymbiont that is present in the locust-origin aphid lineages might play a role in this process.