Host Species and Geography Differentiate Honeybee Gut Bacterial Communities by Changing the Relative Contribution of Community Assembly Processes

Abstract

Honeybee gut microbiota modulates the health and fitness of honeybees, the ecologically and economically important pollinators and honey producers. However, which processes drive the assembly and shift of honeybee gut microbiota remains unknown. To explore the patterns of honeybee gut bacterial communities across host species and geographical sites and the relative contribution of different processes (i.e., homogeneous selection, variable selection, homogeneous dispersal, dispersal limitation, and an undominated process) in driving the patterns, two honeybee species (Apis cerana and Apis mellifera) were sampled from five geographically distant sites along a latitudinal gradient, followed by gut bacterial 16S rRNA gene sequencing. The gut bacterial communities differed significantly between A. cerana and A. mellifera, which was driven by the interhost dispersal limitation associated with the long-term coevolution between hosts and their prokaryotic symbionts. A. mellifera harbored more diverse but less varied gut bacterial communities than A. cerana due to the dominant role of homogeneous selection in converging A. mellifera intestinal communities. For each honeybee species, the gut bacterial communities differed across geographical sites, with individuals from lower latitudes harboring higher diversity; also, there was significant decay of gut community similarity against geographic distance. The geographical variation of honeybee gut bacterial communities was mainly driven by an undominated process (e.g., stochastic drift) rather than variable selection or dispersal limitation. This study elucidates that variations in host and geography alter the relative contribution of different processes in assembling honeybee gut microbiota and, thus, provides insights into the mechanisms underlying honeybee gut microbial shifts across evolutionary time. IMPORTANCE Honeybees provide crucial pollination services and valuable apiarian products. The symbiotic intestinal communities facilitate honeybee health and fitness by promoting nutrient assimilation, detoxifying toxins, and resisting pathogens. Thus, understanding the processes that govern honeybee gut bacterial communities is imperative for better managing gut microbiota to improve honeybee health. However, little is known about the processes driving the assembly and shift of honeybee gut bacterial communities. This study quantitatively deciphers the relative importance of selection, dispersal, and undominated processes in governing the assembly of honeybee gut bacterial communities and explores how their relative importance varies across biological and spatial scales. Our study provides new insights into the mechanisms underlying the maintenance and shift of honeybee gut microbiota.

Keywords: bacterial community; ecological processes; geography; gut microbiota; honeybee; host; neutrality; selection; stochasticity.

FIG 1

Dynamics of community patterns under the force of different processes. (a) Homogeneous selection and homogeneous dispersal cause communities to converge. (b) Variable selection and dispersal limitation cause communities to diverge. (c) Stochastic drift (i.e., undominated process) disperses communities.

FIG 2

Distinct gut bacterial communities between A. cerana and A. mellifera. The Chao1 richness (a) and phylogenetic diversity (b) of A. mellifera gut bacterial communities were higher than those of A. cerana. (c) Nonmetric multidimensional scaling (NMDS) profiling illustrated distinct gut bacterial communities between A. cerana and A. mellifera. (d) Comparison of gut bacterial similarities within A. cerana, within A. mellifera, and between host species. (e) Phylogenetic tree of the 81 operational taxonomic units (OTUs) with a relative abundance of >0.1% in at least one sample, where the core OTUs within A. cerana and A. mellifera guts were labeled with red and blue circles, respectively. Bars showed the mean relative abundance of the OTUs. (f) The bootstrap consensus tree of the COI-COII haplotypes observed in this study and the correspondence to 100 honeybee samples. Branches with less than 50% bootstrap replicates were collapsed. Numbers on the edges of the tree represent the bootstrap values.

FIG 3

Effects of geography on A. cerana and A. mellifera gut bacterial communities. (a) The sampling sites. (b) The latitudinal patterns of honeybee gut bacterial diversity. (c) Nonmetric multidimensional scaling (NMDS) profiling illustrated geographical site effects on honeybee gut bacterial communities. P values were adjusted with the BH method. (d) The distance-decay relationship between geographical distance and community similarity of either A. cerana or A. mellifera gut microbiota. (e) The community similarity between A. cerana and A. mellifera gut bacterial communities negatively correlated with latitude.

FIG 4

Relative contribution of host species and geographical site to honeybee gut bacterial communities. (a) Variance partitioning analysis (VPA) revealed that host species had higher effects on gut bacterial communities than geographical site. (b) The variations of host species and geographical site decreased the similarities of honeybee gut bacterial communities, while the decrease induced by host species was significantly larger than by geographical site. Error bars were 95% confidence intervals. ***, P < 0.001.

FIG 5

Ecological processes driving the assembly and shift of honeybee gut bacterial communities. (a) Ecological processes between paired samples. (b) The relative importance of ecological processes in governing the gut bacterial communities of A. cerana and A. mellifera without considering the effects of host species or geographical site. (c) The relative importance of ecological processes in governing honeybee gut bacterial communities of the same host at the same site (S.host and S.site), the same host at different sites (S.host and D.site), different hosts at the same site (D.host and S.site), and different hosts at different sites (D.host and S.site). (d) Host- and site-induced change of ecological processes’ relative importance in governing the shift of honeybee gut bacterial communities. Error bars were 95% confidence intervals. *, P < 0.05; ns (not significant), P > 0.05.

FIG 6

Historical stochasticity engines the coevolution of honeybee gut microbiota.