Resistance assays of Drosophila suzukii populations from seven locations in Michigan and one location in Oregon

Host-parasitoid interactions are tied in coevolutionary arms races where parasitoids continuously have to evolve increased virulence as hosts evolve increased resistance. Over time geographic structure in virulence and resistance can arise because of spatial and temporal differences in parasitoid communities, in the strength of reciprocal selection pressures, in genetic variation in local populations, and as trade-offs are balanced between defense and fitness traits. It is crucial to understand how invasion changes the coevolutionary dynamics of host-parasitoid interactions to successfully implement biological control programs against invasive insect hosts. We investigated spatial and temporal variation in resistance of the invasive Drosophila suzukii in seven geographically distinct populations in Michigan and of one population from Oregon against a newly approved biocontrol agent, the larval parasitoid Ganaspis brasiliensis. We found regional and temporal variation in resistance (encapsulation rates of parasitoid eggs) of D. suzukii populations that ranged from 11– 48%. The northernmost site with the lowest encapsulation rate had the highest rate of parasitism suggesting that parasitoids may be able to detect the defensive capacities of their hosts and adjust attack rates accordingly. The lowest encapsulation rates at the northernmost and thus coldest site suggest a negative effect of temperature on resistance. Large regional differences in resistance of D. suzukii populations can render the ensuing biocontrol program more variable and less predictable, and release strategies may need to be altered at sites where flies have high resistance to accelerate biocontrol.  

Methods
Drosophila suzukii was collected from seven locations in Michigan and one location in Oregon. Field collected flies were reared in the laboratory for one generation and then assayed for resistance to parasitism by Ganaspis brasiliensis.