Surviving in the absence of flowers: do nectar yeasts rely on overwintering bumblebee queens to complete their annual life cycle?

Floral nectar represents an ephemeral habitat that is restricted in time and space to zoophilous flowering vegetation. To survive in these habitats, nectar-inhabiting microorganisms rely on animal vectors to disperse from one flower to the next. However, it remains unclear how nectar yeasts persist when flowers and nectar cease to be present. Here, we tested the hypothesis that hibernating bumblebee queens function as a reservoir for nectar yeasts in the absence of plants or pollinators during winter. Our results show that the nectar yeast, Metschnikowia reukaufii, was present in the gastrointestinal tract of wild bumblebee queens that emerged from hibernation and that it could persist inside the gut of hibernating queens under experimental conditions. However, no evidence for such persistence was found in the case of the second most frequent nectar yeast, M. gruessii. Furthermore, a phylloplane yeast that occasionally inhabits nectar, Rhodotorula mucilaginosa, was able to colonize the gut under experimental conditions. Two bumblebee-associated yeasts, Candida bombi and C. bombiphila, were successfully passed down generations after administration in commercial lab-reared bumblebees. Overall, these results demonstrate that bumblebees could act as a reservoir for nectar yeasts during winter when floral nectar is absent.

Conserved queen pheromones in bumblebees: a reply to Amsalem et al

In a recent study, Amsalem, Orlova & Grozinger (2015) performed experiments with Bombus impatiens bumblebees to test the hypothesis that saturated cuticular hydrocarbons are evolutionarily conserved signals used to regulate reproductive division of labor in many Hymenopteran social insects. They concluded that the cuticular hydrocarbon pentacosane (C₂₅), previously identified as a queen pheromone in a congeneric bumblebee, does not affect worker reproduction in B. impatiens. Here we discuss some shortcomings of Amsalem et al.’s study that make its conclusions unreliable. In particular, several confounding effects may have affected the results of both experimental manipulations in the study. Additionally, the study’s low sample sizes (mean n per treatment = 13.6, range: 4–23) give it low power, not 96–99% power as claimed, such that its conclusions may be false negatives. Inappropriate statistical tests were also used, and our reanalysis found that C₂₅ substantially reduced and delayed worker egg laying in B. impatiens. We review the evidence that cuticular hydrocarbons act as queen pheromones, and offer some recommendations for future queen pheromone experiments.