COS 200-3 - A phylloplane fungus influences cutting of leaves by leafcutter bees (Megachile sp.)

Background/Question/Methods

The outcomes of species interactions are often affected by biotic context. Such biotic contexts include host-associated microbiomes, which can affect host traits relevant to behavior, health, and fitness of interacting partners. We investigated interactions between foliar microbiomes and leafcutter bees (Megachile sp.), which line their nests with discs cut from leaves. Leaf discs carry microbes that could impact Megachile eggs and larvae detrimentally (e.g., causing disease, spoiling provisions, or overgrowing nests) or beneficially (e.g., defending against natural enemies). We previously showed that leaves of cultivated rose plants (Rosa x hybrida) that were selected for cutting by leafcutter bees hosted a distinctive microbiome relative to those not cut. In particular, the fungus Aspergillus was more abundant on cut leaves, whereas the fungus Alternaria and an unidentified bacterium were especially prevalent on uncut leaves. To establish a causal role of these microbes in influencing cutting behavior, we conducted an experiment in which we randomly assigned rose leaves to one of five treatments: inoculated with one of these three microbes, sham inoculated, or unmanipulated. We then measured the accumulation of leafcutter bee cuts over a 96-hour period. We predicted that Aspergillus would increase cutting, whereas Alternaria and the unidentified bacterium would decrease cutting.

Results/Conclusions

Inoculation with Alternaria and the unidentified bacterium did not influence the accumulation of cuts by leafcutter bees. However, inoculation with Aspergillus increased cutting relative to sham-inoculated leaves. Thus, it appears that Aspergillus acts as a positive cue for bees to cut leaves. Because some Aspergillus strains inhibit the growth of other plant-associated fungi, and/or influence their production of secondary metabolites, it is plausible that this fungus is beneficial for bee nests. Alternatively, its presence may be a reliable cue to bees that a given leaf has other desirable traits, or it may play an attractant role independent of its impact on the nest. Experiments to distinguish among these possibilities will clarify whether bees gain information from detecting Aspergillus, how they detect it, and whether it relates to leaf traits alone, nest health, or other factors. More broadly, our study highlights the capacity of a common element of the foliar microbiome to influence insect behavior, and provides the basis for hypothesis testing regarding mechanisms that drive this interaction.