Rounding up the Honey Bees (1)

by Christoph
 

This is a longread again comprising 3,600 words – 27.5 min reading – and although that's still a factor of 10 less words than bees in a typical hive it's split in two posts. This is the first part, the second part comes next week. I borrowed the title from TWiM episode #178 for the irresistible pun, just adding a drop of honey.
 

Why would I deviate here from STC's habit to rather not  feature the newest results of trendy re­search topics? The reason comes hidden in a short notice in Science by Warren Cornwall: "...The new research is certain to make a controversial herbicide even more of a flashpoint." He referred, of course, to a recent paper published in PNAS  by the lab of Nancy Moran at the University of Texas, Austin TX, as part of their ongoing Bee Microbiome Project and titled: "Glyphosate perturbs the gut microbiota of honey bees". You are certainly well informed about the "controversial her­bi­cide" glyphosate – the Wikipedia entry gives a comprehensive account – and aware of CCD, co­lony collapse disorder in honey bees, but you may also know that many biologists are deeply worried by the dramatic decline in insect diversity – including honey bees and all the other important yet less well-known pollinators – so I will not dwell on these points but consider what Erick da Silva Motta, Kasie Raymann and Nancy A. Moran found out about the small things, the honey bee's gut microbiota *).
 

The gut microbiota of the honey bee, Apis mellifera is, when compared to the human gut microbi­o­ta, truly "minimalistic" as it comprises just eight bacterial "core species" and few "others" at low cell numbers in adult "workers" or "foraging adults" (see here). Cell numbers for the eight core species range between 1×10⁷ and 9×10⁷ total in the guts of workers (see here). In relative abun­dance, these core spe­cies differ from each other but their composition is largely invariant across all worker bees of a colony (but varies between workers and the "reproductive castes" (queens, drones)) and is also largely invariant across different colonies. Given their habitat in the bee's gut, the core species are all facultative anaerobes or micro-aerophiles, and only rarely found in other places within a hive. Three core species, the Betaproteobacterium Snodgrassella alvi, and the Gam­maproteobacteria Gilliamella apicola and Frischella perrara were first isolated from bee guts (see here their unspectacular EM "portraits"). Taxonomically, the latter two belong to the Orbales, a novel sister clade of the Pasteurellales (think Haemophilus influenzae ). My characterization of the gut microbiota of bees as "minimalistic" is, admittedly, an oversimplification: all eight bacterial spe­cies were present in individual workers, and consequently in an entire colony, as several distinct lineages (or: strains), which have, according to genomic analyses, similar "core genomes" but high­ly variable "accessory genomes". The consequences of this intra-species variabilty for homeostasis and adaptability of the entire microbiota are not well understood yet.
 

(Click to enlarge )
Figure 1. Changes in gut microbiota composi­tion follow­ing glyphosate exposure of honey bees with established gut communities. Box­plots of bacterial 16S rDNA copies for control (C) and glyphosate-treated (G-5 and G-10) bees at posttreatment days 0 and 3 (n = 15 for each group and time point). Box-and-whisker plots show high, low, and median values, with lower and upper edges of each box denoting first and third quartiles, respectively. *P < 0.05 and **P < 0.01, Wilcoxon rank sum test followed by Bon­ferroni correction. Source. Frontpage: Apis mel­lifera. CC BY-NC 2.0 Vlad Proklov

Since it has been long known that the herbicide glypho­sate (Roundup) has a growth-inhibitory effect on various bacteria and a recent study by Herbert et al. (2014) pre­sented evidence for a slightly aberrant behavior in gly­pho­sate-exposed honey bees, Motta et al. sought to stu­dy whether glyphosate would affect the honey bee gut microbiota. For the experiment, they collected >100 adult worker bees from a single hive, treated them with either 5 mg/l glyphosate (G-5), 10 mg/l glyphosate (G-10) or, as control, with sterile sucrose syrup (C) for 5 days, and then returned them to their hive. For the treatment, the bees were temporarily immobilized by chilling, and paint-mar­ked on the thorax to make them distinguishable later in the hive. The glyphosate concentrations were chosen to mimic environmental levels, which typically range be­tween 1.4 and 7.6 mg/l, and may be encountered by bees foraging at flowering weeds but would not negatively af­fect their foraging-related behavior. To determine the ef­fects of glyphosate on the size and composition of the gut microbiome, they sampled 15 bees from each group before reintroduction to the hive (day 0) and post-reintroduction (day 3), and assessed the relative and absolute abundances of gut bacteria by deep sequencing (=quantification) of the V4 region of the bacterial 16S rRNA gene and by quantitative PCR (qPCR). Note that the time frame of the experiment was well within the average life span of adult workers (~20 d after the transition from hive/nurse bees to foraging adults/workers).
 

At day 0, glyphosate exposure had little effect on the bee gut microbiome size, but the absolute and relative abundances of one core species, S. alvi, were significantly lower in the G-10 group (Fi­gure 1). The effects on the bee gut microbiome were more prominent at day 3. The total number of gut bacteria decreased for both treatment groups, relative to control, but this drop was signi­fi­cant only for the G-5 group, which also exhibited more severe compositional shifts. The absolute abundances of four core gut bacteria, S. alvi, Bifidobacterium, Lactobacillus  Firm-4 and Firm-5 were decreased, and the relative abundance of G. apicola increased in the G-5 group. Surprisingly, only Lactobacillus Firm-5 decreased in absolute abundance in the G-10 group. They repeated this experiment using bees from a different hive and at a different season, and observed similar trends (see here). As in the first experiment, they found significant reduction in abundance for S. alvi  in glyphosate-exposed bees.
 

Motta et al. clearly showed that glyphosate "perturbs" the honey bee gut microbiota. The most prominent "victim" was the core species Snodgrassella alvi. The seemingly erratic response to gly­phosate of several of the core species may have two not necessarily mutually exclusive reasons: 1. the sensitivity of the cellular target(s) to glyphosate may vary among the species (see last section), 2. all core species except S. alvi  are deficient in the biosynthesis of two or more amino acids (=auxo­trophic) and depend therefore on cross-feeding by other members of the microbiota. The experi­mental design of re-introducing the glyphosate-exposed bees to the hive had the advantage of al­lowing to monitor them under close-to-natural conditions, that is, not interrupting their foraging behavior and social interactions. However, this design came at the price of not being able to moni­tor those bees that had left the hive without returning. Since they recovered less than 20% of the bees that were re-introduced to the hive (day 3) they consider it unlikely that these represent the total effect of glyphosate on the treatment groups.
 

Microbiota acquisition

Newly hatched bees – phonetically newbies, Motta et al. call them NEWs – are almost "germ free" and acquire their gut microbiota through feeding by their slightly elder sisters, the "nurse bees", which do not leave the hive for foraging as do the "workers", the adult bees (a whiff of bee biology: the gut microbiota of larvae is almost completely lost during pupation, see here). This "handover" from nurse bees to newbies works so reliably that the specific "core" gut microbiota of Apis melli­fe­ra and other social bees worldwide can be traced back to a shared origin ~80 million years ago. In an experiment where NEWs were exposed to glyphosate during feeding for the first days after hatching, one could monitor its effect on the crucial step of bacterial colonization of the gut. This is not a purely academic matter, but closely resembles the field situation. Here, glyphosate was found in hives and honey samples, indicating that foraging workers can transport residues of this herbi­ci­de back to the colony and contaminate other bees, including NEWs, and food resources. Motta et al. fed NEWs for 5 days with bee gut homogenate as "inoculum", then for another 2 days with sugar syrup containing 1 mM glyphosate (G) or syrup alone (C). 15 bees of each group were samp­led 2 days after the last feeding, and DNA and RNA extracted from their dissected guts. Quantify­ing both nucleic acids was important here as captive bees do not normally defecate, and dead bac­terial cells and their released DNA accumulate in the gut. By contrast, RNA is more easily degraded and thus a measure of living cells. Any significant difference between RNA, quantified by sequenc­ing, and DNA, quantified by qPCR or sequencing, would thus point to the presence of many dead cells in the sample. This was, however, not the case in this experiment, indicating that most bacte­ria were alive at the time point of sampling.
 

Figure 2. Changes in gut microbiota composi­ti­on follow­ing glyphosate exposure of young ho­ney bees. Boxplots of total bacterial 16S rDNA copies and of absolute and relative abundances of two gut bacterial species for control (n = 14) and glyphosate-treated (n = 11) bees. **P < 0.01, and ***P < 0.001,Wilcoxon rank sum test followed by Bonferroni correction. Source

They found all eight core species in both control and gly­phosate-treated samples, which indicated that low doses of glyphosate do not prevent colonization of the bee gut by any core member (note that they used 1 mM here, not 5 or 10 mM as in the experiment discussed above). The re­lative abundance of the 8 core species was similar in con­trol bees not exposed to glyphosate, but their absolute abundance varied by a factor 10, showing that coloni­za­tion was asynchronous in individual bees. The results were comparable in the glyphosate-treated bees, albeit with a somewhat higher variation in relative abundance. Snodgrassella alvi  was most strongly affected by glypho­sa­te and showed a marked decrease in both absolute and relative abundance, while Lactobacillus Firm-4 increased in relative abundance (Figure 2).
 

From microbiota to immune response

Honey bees have an innate immune system that normally protects them well against bacterial pa­thogens. There's a growing consensus among biologists that the immune systems of individual animals are dynamically shaped by interactions with their gut microbiota – although many (mole­cu­lar) mechanistical details are still unknown. Serratia marcescens  is an opportunistic pathogen of many plants and animals, including humans, and a virulent opportunistic pathogen of honey bees. In a survey of several locations in the USA, S. marcescens  was found to colonize the guts of ~60% of all bees at low cell numbers, but in lab experiments to kill the bees at high frequency when left unchecked. This suggested to Motta et al. that they determine the survival rate of glyphosate-exposed bees upon infection with S. marcescens  strain kz19, which was isolated from a bee gut. Since glyphosate does not prevent colonization of the bee gut by the core species of the bee mi­cro­biota (see previous section), this experiment became actually feasible because all necessary controls could be included. (I will go deeper into their experimental procedure to "expose" their carefully chosen controls.)
 

Figure 3. Glyphosate exposure of young honey bees and susceptibility to Serratia infection. The percent survival of age-controlled bees after Ser­ratia kz19 exposure, shown as a Kaplan-Meier survival curve.***P < 0.001, coxph model imple­mented in the "survival" package in R. GH, gut ho­mogenate-exposed bees; Gly, glyphosate treat­ment; MF, microbiota-free bees; Ser, Serratia challenge. Source

For the experiment, "hundreds" of late-stage pupae (they don't give the exact number) were removed from brood frames and allowed to emerge under sterile conditions in the lab. The NEWs were hand fed bee gut homogenate (GH) – thought to closely mimic their normal feeding by the hive bees, and providing a "starter" for their micobio­ta – or sterile sucrose syrup (MF). Each group was divided into two subgroups and treated for 5 d with 0.1 mM gly­pho­sate in sucrose syrup (Gly), resulting in a dose of ∼1.7 μg per bee, or sterile sucrose syrup. After 5 d, one half of the subgroups (likely >25 bees per subgroup) was chall­en­ged with S. marcescens kz19, the other half serving as control. They monitored the survival rate on a daily basis from day 0 through day 8 (Figure 3). The survival rate was close to 100% on days 1 and 2, then gradually tapered down to ~80% at day 8 for bees with "normal" gut micro­bio­ta (GH) and microbiota-free bees (MF). For these two groups, exposure to glyphosate had no significant effect on survival (GH+Gly, MF+Gly). The survival rate dropped to 50% for bees with "normal" gut micro­biota challenged with Serratia (GH+Ser). They had apparently hit the LD50 for Serratia pretty ex­actly (they don't mention it, but I assume they had "titrated" their Serratia inoculum before). How­ever, the survival rate dropped sharply to just ~10% for bees with "normal" gut microbiota expos­ed to glyphosate and challenged with Serratia (GH+Gly+Ser). Since the survival rate also dropped sharply to ~10% for microbiota-free bees challenged with Serratia (MF+Ser), they could safely con­clude that unperturbed microbiota are key for partial protection of bees against infection with Ser­ratia. The survival rates of microbiota-free bees challenged with Serratia were equally low whether or not they were exposed to glyphosate (MF+Ser, MF+Gly+Ser). This again indicates that glypho­sa­te at the tested concentration had no significant influence on the bees' health. Note that in bees exposed to glyphosate but not challenged with Serratia, the survival rates were in all cases higher than in the Serratia-challenged groups. Therefore, a direct effect of glyphosate on bees is clearly not the basis of the high mortality of glyphosate-exposed, pathogen-challenged bees.
 

Reference
Motta EVS, Raymann K, Moran NA. (2018). Glyphosate perturbs the gut microbiota of honey bees. Proc Natl Acad Sci U S A, 115 (41), 10305 – 10310, PMID 30249635 (Open Access PDF)

The Motta et al. paper was also subject in episode #187 of the podcast "This Week in Microbio­lo­gy" (TWiM), from minute 21:45 on.
 

*) The terms "Microbiome" and "Microbiota" are sometimes used as synonyms. Here, I understand "micro­bio­ta" as the community of all those microbes you actually meet "in person" at a certain location. The "mico­biome" would then be rather a census of these microbes based on their genomes, their ID cards so to say.