Divergence and convergence of gut microbiomes of wild insect pollinators.

Pollination services provided by wild insect pollinators are critical to natural ecosystems and crops around the world. There is an increasing appreciation that the gut microbiota of these insects influences their health and consequently their services. However, pollinator gut microbiota studies have focused on well-described social bees, but rarely include other, more phylogenetically divergent insect pollinators. To expand our understanding, we explored the insect pollinator microbiomes across three insect orders through two DNA sequencing approaches. First, in an exploratory 16S amplicon sequencing analysis of taxonomic community assemblages, we found lineage-specific divergences of dominant microbial genera and microbiota community composition across divergent insect pollinator genera. However, we found no evidence for a strong broad-scale phylogenetic signal, which we see for community relatedness at finer scales. Subsequently, we utilized metagenomic shotgun sequencing to obtain metagenome-assembled genomes and assess the functionality of the microbiota from pollinating flies and social wasps. We uncover a novel gut microbe from pollinating flies in the family Orbaceae that is closely related to Gilliamella spp. from social bees but with divergent functions. We propose this novel species be named Candidatus Gilliamella eristali. Further metagenomes of dominant fly and wasp microbiome members suggest that they are largely not host-insect adapted and instead may be environmentally derived. Overall, this study suggests selective processes involving ecology or physiology, or neutral processes determining microbe colonization may predominate in the turnover of lineages in insect pollinators broadly, while evolution with hosts may occur only under certain circumstances and on smaller phylogenetic scales. IMPORTANCE Wild insect pollinators provide many key ecosystem services, and the microbes associated with these insect pollinators may influence their health. Therefore, understanding the diversity in microbiota structure and function, along with the potential mechanisms shaping the microbiota across diverse insect pollinators, is critical. Our study expands beyond existing knowledge of well-studied social bees, like honey bees, including members from other bee, wasp, butterfly, and fly pollinators. We infer ecological and evolutionary factors that may influence microbiome structure across diverse insect pollinator hosts and the functions that microbiota members may play. We highlight significant differentiation of microbiomes among diverse pollinators. Closer analysis suggests that dominant members may show varying levels of host association and functions, even in a comparison of closely related microbes found in bees and flies. This work suggests varied importance of ecological, physiological, and non-evolutionary filters in determining structure and function across largely divergent wild insect pollinator microbiomes.

Non-linear relationships between density and demographic traits in three Aedes species.

Understanding the relationship of population dynamics to density is central to many ecological investigations. Despite the importance of density-dependence in determining population growth, the empirical relationship between density and per capita growth remains understudied in most systems and is often assumed to be linear. In experimental studies of interspecific competition, investigators often evaluate the predicted outcomes by assuming such linear relationships, fitting linear functions, and estimating parameters of competition models. In this paper, we experimentally describe the shape of the relationship between estimated population rate of change and initial density using laboratory-reared populations of three mosquito species. We estimated per capita growth rate for these experimental populations over a 30-fold range of larval densities at a standard resource abundance. We then compared fits of linear models and several different nonlinear models for the relationship of estimated rate of change and density. We find that that the relationship between density and per capita growth is strongly non-linear in Aedes aegypti (Linnaeus), Aedes albopictus (Skuse), and Aedes triseriatus (Say) mosquitoes. Components of population growth (survivorship, development time, adult size) are also nonlinearly related to initial density. The causes and consequences of this nonlinearity are likely to be important issues for population and community ecology.

Sex-specific effects of hatching order on nestling baseline corticosterone in a wild songbird.

Variation in nestling growth and survival is often influenced by hatching order, with first-hatched offspring having an advantage over later-hatched younger siblings. In house wrens (Troglodytes aedon), this effect of hatching order is especially evident in asynchronously hatched broods and can lead to sex-specific differences in the size and condition of nestlings. Females appear to allocate the sex of their offspring across the laying order to capitalize on these differences. We hypothesized that levels of circulating corticosterone, the primary metabolic hormone in birds, mediates these sex-specific effects in nestlings. We predicted that: i) baseline levels of corticosterone in nestlings should vary along the hatching order, ii) effects of hatching order on baseline corticosterone should be sex specific, and iii) any sex-specificity of hatching order on baseline corticosterone could be contingent on the degree of hatching synchrony. We tested these predictions in a study in which we measured baseline corticosterone in first- and last-hatched nestlings in synchronously and asynchronously hatching broods. To assess whether any differences in nestling baseline corticosterone levels could be attributed to pre-natal maternal effects, the post-natal environment, or both, we conducted two additional studies in which we measured i) yolk corticosterone in first- and last-laid eggs and ii) baseline corticosterone in nestlings that were cross-fostered to create simulated 'asynchronously' hatched broods. There was a significant interaction between sex and relative hatching order in their effects on nestling baseline corticosterone, but no effect of hatching synchrony. Corticosterone levels remained relatively constant across the hatching order in males but decreased in females. There was a significant effect of laying order on yolk corticosterone, with first-laid eggs containing significantly higher levels of yolk corticosterone than last-laid eggs. Cross-fostering of nestlings at different points of development had no significant effect on nestling corticosterone levels. These results indicate that sex-dependent differences in corticosterone levels across the hatching order may arise, at least in part, from embryonic exposure to maternally derived corticosterone, whereas the post-natal rearing environment plays, at best, a minimal role in determining nestling baseline corticosterone levels.

An interaction between host and microbe genotypes determines colonization success of a key bumble bee gut microbiota member.

There has been a proliferation of studies demonstrating an organism's health is influenced by its microbiota. However, factors influencing beneficial microbe colonization and the evolution of these relationships remain understudied relative to host-pathogen interactions. Vertically transmitted beneficial microbes are predicted to show high levels of specificity in colonization, including genotype matching, which may transpire through coevolution. We investigate how host and bacterial genotypes influence colonization of a core coevolved microbiota member in bumble bees. The hindgut colonizing Snodgrassella alvi confers direct benefits, but, as an early colonizer, also facilitates the further development of a healthy microbiota. Due to predominantly vertical transmission promoting tight evolution between colonization factors of bacteria and host lineages, we predict that genotype-by-genotype interactions will determine successful colonization. Germ-free adult bees from seven bumble bee colonies (host genotypic units) were inoculated with one of six genetically distinct strains of S. alvi. Subsequent colonization within host and microbe genotypes combinations ranged from 0 to 100%, and an interaction between host and microbe genotypes determined colonization success. This novel finding of a genotype-by-genotype interaction determining colonization in an animal host-beneficial microbe system has implications for the ecological and evolutionary dynamics of host and microbe, including associated host-fitness benefits.