Effects of a Resident Yeast from the Honeybee Gut on Immunity, Microbiota, and Nosema Disease

Abstract

The western honeybee (Apis mellifera) has a core bacterial microbiota that is well described and important for health. Honeybees also host a yeast community that is poorly understood with respect to host nutrition and immunity, and also the symbiotic bacterial microbiota. In this work, we present two studies focusing on the consequences of dysbiosis when honeybees were control-fed a yeast that was isolated from a honeybee midgut, Wickerhamomyces anomalus. Yeast augmentation for bees with developed microbiota appeared immunomodulatory (lowered immunity and hormone-related gene expression) and affected the microbial community, while yeast augmentation for newly emerged bees without an established bacterial background did not lead to decreased immunity- and hormone-related gene expression. In newly emerged bees that had a naturally occurring baseline level of W. anomalus, we observed that the addition of N. ceranae led to a decrease in yeast levels. Overall, we show that yeasts can affect the microbiome, immunity, and physiology.

Keywords: health; honeybee; microbes; microbiota; yeasts.

Figure 1

A heatmap amended with a dendrogram (assembled with heatmaply) for study A that shows median normalized, relative gene expression levels (as Δ Cq) of frame-captured bees that were either constantly fed the W. anomalus–sucrose solution or only the sucrose solution. Data were pooled from three independent cage feeding trials. The asterisk indicates a statistically significant change in Δ Cq between the treatments of at least p < 0.5 by the Wilcoxon pairwise test. Overall, we observed that many immune and hormone-like genes had lower expression after the introduction of W. anomalus, which cluster together at the top of the heatmap.

Figure 2

Boxplots for study B showing the relative gene expression values (as Δ Cq) of newly emerged bees that were conditioned with the yeast, followed by feeding N. ceranae and then continuous feeding of the yeast (“both”), and respective feeding controls (i.e., N. ceranae only, W. anomalus only or sugar only). The asterisks indicate statistically significant differences between the treatments of at least p < 0.05 by the Wilcoxon pairwise test. The statistical chance for W. anomalus-specific targeted qPCR between both-fed and sugar-fed was p = 0.0962. Note that when NEWs were not fed N. ceranae, there was no amplification of N. ceranae, for which a Δ Cq was nonetheless provided (Δ Cq = Cq _RPS5a−50). Overall, we observed that NEWs fed N. ceranae had detectable Nosema titers which were absent in the NEWs that were not fed N. ceranae, and that the naturally occurring W. anomalus population decreased after the introduction of N. ceranae (sugar vs. N. ceranae groups for W. anomalus-specific amplification).