Gravediggers: Parasitoid flies manipulate bumblebee host behavior and may select for host body size

Background/Question/Methods

Many parasites and parasitoids manipulate the behavior of their hosts to maximize their own survival and/or reproductive fitness. Parasitoids are also known to selectively infect their hosts based on certain physical characteristics that optimize the fitness of the parasite or its offspring. Conopid flies (Conopidae) are parasitoids of bumblebees (Bombus spp.) in Europe and North America. In one European bumblebee species (B. terrestris), it has been shown that individuals harboring endoparasitic conopid larvae bury themselves in soil just before death, a behavior that may confer fitness benefits on the conopid pupae overwintering in the bodies of their hosts.  Some evidence suggests that adult conopid females may also selectively target bees based on body size, a host characteristic that often correlates positively with increased survival and fitness of endoparasitoids. This study examined three locally common bumblebees (B. bimaculatus, B. griseocollis, and B. impatiens) at the University of Virginia’s Blandy Experimental Farm (Boyce, VA) in the summer of 2012 to determine whether self-burial behavior was apparent in parasitized hosts, whether this behavior was more or less apparent in particular species, and whether host body size influenced the mass of conopid pupae. Bees were kept in tanks with soil and dissected upon expiration.

Results/Conclusions

Self-burial behavior was found in parasitized bees of all three bumblebee species. Of all parasitized bees (n=140 of 498), a majority buried themselves (61%). Of the bees that were buried, 85% were parasitized. The frequency of self-burial was not significantly different among parasitized bees across species. Collectively, this indicates that late-instar conopid larvae consistently induce this behavior in bumblebee hosts irrespective of species identity.  However, species may not be equally competent hosts. Unexpectedly, pupae were found significantly less frequently in B. griseocollis (13%) than in B. bimaculatus (46%) and B. impatiens (35%). In 2011, larvae were found with similar frequencies in these same three species. Further research is needed to explain this discrepancy; one possibility is that the survival rate of conopid larvae may be lower in B. griseocollis than in the other two species. In accordance with previous research, this study found that the likelihood of conopid parasitism increased with host body size, and that larger pupae developed in larger-bodied bees. The findings of this study are novel: we demonstrate a parasite-induced modification of bumblebee behavior previously undocumented in North American bumblebees, and, for the first time, show this behavior in multiple Bombus species.