The molecular basis of socially induced egg-size plasticity in honey bees

  1. Bin Han
  2. Qiaohong Wei
  3. Esmaeil Amiri
  4. Han Hu
  5. Lifeng Meng
  6. Micheline K Strand
  7. David R Tarpy
  8. Shufa Xu
  9. Jianke Li
  10. Olav Rueppell

  • Curated by eLife

    eLife logo

    eLife assessment

    This study investigates factors that control egg size plasticity in a social insect, the honey bee Apis mellifera. It finds that honey bee queens vary egg size in response to size of their colony, and that the gene Rho1 is involved in egg-size determination. These findings inform our understanding of maternal control over egg size, a key form of maternal investment. The work is relevant to colleagues studying reproduction and social insects.

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

Reproduction involves the investment of resources into offspring. Although variation in reproductive effort often affects the number of offspring, adjustments of propagule size are also found in numerous species, including the Western honey bee, Apis mellifera . However, the proximate causes of these adjustments are insufficiently understood, especially in oviparous species with complex social organization in which adaptive evolution is shaped by kin selection. Here, we show in a series of experiments that queens predictably and reversibly increase egg size in small colonies and decrease egg size in large colonies, while their ovary size changes in the opposite direction. Additional results suggest that these effects cannot be solely explained by egg-laying rate and are due to the queens’ perception of colony size. Egg-size plasticity is associated with quantitative changes of 290 ovarian proteins, most of which relate to energy metabolism, protein transport, and cytoskeleton. Based on functional and network analyses, we further study the small GTPase Rho1 as a candidate regulator of egg size. Spatio-temporal expression analysis via RNAscope and qPCR supports an important role of Rho1 in egg-size determination, and subsequent RNAi-mediated gene knockdown confirmed that Rho1 has a major effect on egg size in honey bees. These results elucidate how the social environment of the honey bee colony may be translated into a specific cellular process to adjust maternal investment into eggs. It remains to be studied how widespread this mechanism is and whether it has consequences for population dynamics and epigenetic influences on offspring phenotype in honey bees and other species.

Article activity feed

  1. Version published to 10.7554/elife.80499 on eLife
  2. eLife

    Author Response

    Reviewer #1 (Public Review):

    This study presents a series of experiments that investigate maternal control over egg size in honey bees (Apis mellifera). Honey bees are social insects in which a single reproductive female (the queen) lays all the eggs in the colony. The first set of experiments presented here explore how queens change their egg size in response to changes in colony size. Specifically, they show that queens have relatively larger eggs in smaller colonies, and that egg size changes when queens are transplanted into colonies of a different size (i.e. confirming that egg size is a plastic trait in honey bee queens). The second set of experiments investigates candidate genes involved in egg size determination. Specifically, it shows that Rho1 plays a role in determining egg size in honey bee queens.

    In principle, we agree with this summary, although we find the experimental demonstration that perceived colony size affects egg size (first set of experiments) and the overall proteomic comparison of ovaries that produce small and large eggs (second set of experiments that indicate the upregulation of metabolism, protein transport, cytoskeleton organization, and a few other processes in large egg-producing ovaries) also important.

    A strength of the study is that it combines both manipulative field (apiary) experiments and molecular studies, and therefore attempts to consider broadly the mechanisms of plasticity in egg size. The link between these two types of dataset in the manuscript, however, is not strong. While the two parts are related, the molecular experiments do not follow from the conclusions of the field experiments but rather run in parallel (both using the same initial treatments of queens from large v small colonies).

    We would welcome suggestions on how to further strengthen the integration between the field experiments and our molecular studies. We sought to explore the molecular basis of the observed plasticity in reproductive behavior and thus focused on samples from the first set of experiments for our proteome comparisons, realizing that every additional field experiment could have entail a similar molecular follow-up. We attempted to bring molecular studies and field experiments back together with the RNAi-mediated knock-down of Rho1 in queens that produce eggs in differently-sized colonies under realistic apicultural conditions. There may be better, additional opportunities for a closer integration of molecular and field experiments, but we could not conceive of them.

    Another strength of the study is the focus on social cues for egg size control in a social insect. Particularly interesting is data showing that queens suddenly exposed to the cues of a larger colony (even where egg-laying opportunities did not actually increase) will decrease their egg size, in the same way as queens genuinely transplanted to larger colonies. That honey bee queens can control their egg sizes in response to cues in the colony is not unexpected, given that queens are known to vary egg size based on the cell type they are laying into (queen, drone or worker cell). Nevertheless, it is interesting to show that worker egg sizes over time are also mediated by social cues.

    We thank the reviewer for this positive assessment and want to highlight that this experiment not only controls for egg laying opportunities, but also for potentially greater resource availability in larger colonies. These results are therefore important for the key argument that egg size is actively regulated by honey bee queens.

    A weakness of the study is that the consequence of egg size on egg development and survival in honey bees is not made clear. The assumption is that larger egg size compensates for smaller colonies in some way. Do smaller eggs (i.e. those laid in large colonies) fare worse in smaller colonies than they do in large colonies? Showing that the variation in egg size is biologically relevant to fitness is an important piece of the puzzle.

    We agree that the consequences of egg size variation are important to address beyond our previously published data set and the benefits demonstrated in other contexts by other authors. However, to comprehensively resolve the consequences requires considerable additional experiments that exceed the scope of our current study, which is primarily focused on the causes of the queens’ reproductive plasticity.

    Also, the relationship between egg number and egg size in honey bees remains rather murky. Does egg size depend at least in part on daily egg laying rate (which is sure to be greater in larger colonies)? The study makes an effort to explore this by preventing queens from laying for two weeks and then comparing their egg size when they resume to those that did not have a pause in laying. Although egg size did not vary between the groups in this case, it is unclear whether the same effect would be seen if queens had simply been restricted from laying at such high rates (e.g. if available empty brood cells had been reduced rather than removed entirely).

    We agree that the relation between egg number and egg size is complicated. We have added more data that show that egg laying rates can be higher in larger colonies than in smaller colonies. We also report now that the egg size is negatively correlated to egg number, although not in all instances, which partially supports (and partially contradicts) our previous findings (Amiri et al. 2020). We have modified the discussion of our results to account for the additional results and point out the limitation of the experiment with caged queens. It is important to realize though that the queens were caged on comb and not restricted in typical, small queen cages that are used for queen transport. It is not clear whether this treatment resulted in a downregulation of the reproductive efforts and/or the resorption of eggs.

    Overall this study makes new contributions to our understanding of maternal control over egg size in honey bees. It provides stepping stones for further investigation of the molecular basis for egg size plasticity in insects.

    We agree that we could not resolve everything in this study and that more investigations are needed.

    Reviewer #2 (Public Review):

    This paper builds on recent work showing that honeybee queens can change the size of the eggs they lay over the course of their life. Here the authors identified an environmental condition that reversibly causes queens to change their egg sizes: namely, being in a relatively small or large colony context. Recently published work demonstrated the existence of this egg size plasticity, but it was completely unknown what signaled to the queen. In a series of simple and elegant experiments they confirmed the existence of this egg size plasticity, and narrowed down the set of environmental inputs to the queen that could be responsible for signaling the change in the environment. They also began the work of identifying genes and proteins that might be involved in controlling egg size. They did a comparative proteomic analysis between small-egg-laying ovaries and large-egg-laying ovaries, and then selected one candidate gene (Rho1). They showed that it is expressed during oogenesis, and that when it is knocked down, eggs get smaller.

    This is a good summary, although we think that it is fair to add that the expression of Rho1 is specific to the egg growth stage, and that we found an almost perfect correlation of Rho1 mRNA levels and egg size in two separate experiments (in addition the difference between large and small egg-producing ovaries at the protein level).

    The experiments on honeybee colonies are well-designed, and they provide fairly strong evidence that the queens are reversibly changing egg size and that it is (at least some component of) their perception of colony size that is the signal. One minor but unavoidable weakness is that experiments on honeybees are necessarily done with small sample sizes. The authors were clear about this, however, and it was very effective that they showed all individual data points. Alongside the previous work on which this paper builds, I found their core results to be rather convincing and important.

    We thank the reviewer for this positive evaluation.

    I found the parts of the paper on oogenesis to be useful, but overall less informative in answering the questions that the authors set out for those sections. On balance, I think the best way to interpret the oogenesis results is as "suggestive and exploratory". For instance, the experiment aimed at understanding the relationship between egg-laying rate and egg size does not include a direct measurement of egg-laying rate, but instead puts queens in a place with no suitable oviposition sites. The proteomic analysis was fine, but since they were using whole ovaries, with tissue pooled across all stages of oogenesis including mature oocytes, I would be cautious in interpreting the results to mean that they had identified proteins involved in making larger eggs. These proteins might just as easily be the proteins that are put into larger eggs. In fact, for the one candidate gene that is examined, its transcripts seem as though they are predominantly in the oocyte cell itself rather than in the supporting cells that actually control the egg size (although it is hard to tell from the micrographs without a label for cell interfaces).

    We have added data on the number of eggs produced in the first experiment, which actually show a negative correlation between egg size and egg number. In addition, we have cautioned our wording about the conclusions that can be drawn from the oviposition restriction experiment. Concerning the expression and role of Rho1, we apologize for the lack of a cell membrane marker. However, we share the reviewer’s interpretation that the mRNA is located in the oocyte. While we also agree that egg loading from the nurse cells is important, transport of vitellogenin from the follicle cells may also be quite significant for egg size (Wu et al. 2021 – doi:10.3389/fcell.2020.593613 and Fleig 1995 - doi:10.1016/0020-7322(95)98841-Z), a process that could be controlled by Rho1 in the documented location. We have added to the discussion to clarify this point.

    On that note, with the caveat that the sample sizes are quite small, I agree that there is some evidence that Rho1 is involved in honeybee oogenesis. If this was the only gene they knocked down, and given that it results in a small size change with such a small sample size, it strikes me as a bit of a stretch to say that these results are evidence that Rho1 plays an important role in egg size determination. It is essential to know if this is a generic result of inhibiting cytoskeletal function or a specific function of Rho1. That is beyond the scope of this study, but until those experiments are done, it is hard to know how to interpret these results. For context, in Drosophila, there are lots and lots of genes such that if you knock them down, you get a smaller or differently shaped egg, including genes involved in planar polarity, cytoskeleton, basement membrane, protrusion/motility, septate junctions, intercellular signaling and their signal transduction components, muscle functions, insect hormones, vitellogenesis, etc. This is helpful, perhaps, for thinking about how to interpret the knockdown of just one gene.

    We thank the reviewer for this perspective and have consequently cautioned our wording. The role of Rho1 in regulating the cytoskeletal function has been established in other organisms, but we do not have the tools to study the corresponding pathways and establish causality in honey bees. We have added to the discussion to alert the reader to the point that additional experiments are necessary.

    Overall, I found the results to be technically sound, and there are several clever manipulations on honeybee colonies that will doubtless be repeated and elaborated in the future to great effect. The core result-that queens can change the size of their eggs quickly and reversibly, in response to some perceived signal-was honestly pretty astonishing to me, and it reveals that there are non-nutritive plastic mechanisms in insect oogenesis that we had no idea existed. I look forward to follow-up studies with interest.

    We thank the reviewer for the overall evaluation and encouragement to continue our research.

  3. eLife

    eLife assessment

    This study investigates factors that control egg size plasticity in a social insect, the honey bee Apis mellifera. It finds that honey bee queens vary egg size in response to size of their colony, and that the gene Rho1 is involved in egg-size determination. These findings inform our understanding of maternal control over egg size, a key form of maternal investment. The work is relevant to colleagues studying reproduction and social insects.

  4. eLife

    Reviewer #1 (Public Review):

    This study presents a series of experiments that investigate maternal control over egg size in honey bees (Apis mellifera). Honey bees are social insects in which a single reproductive female (the queen) lays all the eggs in the colony. The first set of experiments presented here explore how queens change their egg size in response to changes in colony size. Specifically, they show that queens have relatively larger eggs in smaller colonies, and that egg size changes when queens are transplanted into colonies of a different size (i.e. confirming that egg size is a plastic trait in honey bee queens). The second set of experiments investigates candidate genes involved in egg size determination. Specifically, it shows that Rho1 plays a role in determining egg size in honey bee queens.

    A strength of the study is that it combines both manipulative field (apiary) experiments and molecular studies, and therefore attempts to consider broadly the mechanisms of plasticity in egg size. The link between these two types of dataset in the manuscript, however, is not strong. While the two parts are related, the molecular experiments do not follow from the conclusions of the field experiments but rather run in parallel (both using the same initial treatments of queens from large v small colonies).

    Another strength of the study is the focus on social cues for egg size control in a social insect. Particularly interesting is data showing that queens suddenly exposed to the cues of a larger colony (even where egg-laying opportunities did not actually increase) will decrease their egg size, in the same way as queens genuinely transplanted to larger colonies. That honey bee queens can control their egg sizes in response to cues in the colony is not unexpected, given that queens are known to vary egg size based on the cell type they are laying into (queen, drone or worker cell). Nevertheless, it is interesting to show that worker egg sizes over time are also mediated by social cues.

    A weakness of the study is that the consequence of egg size on egg development and survival in honey bees is not made clear. The assumption is that larger egg size compensates for smaller colonies in some way. Do smaller eggs (i.e. those laid in large colonies) fare worse in smaller colonies than they do in large colonies? Showing that the variation in egg size is biologically relevant to fitness is an important piece of the puzzle.

    Also, the relationship between egg number and egg size in honey bees remains rather murky. Does egg size depend at least in part on daily egg laying rate (which is sure to be greater in larger colonies)? The study makes an effort to explore this by preventing queens from laying for two weeks and then comparing their egg size when they resume to those that did not have a pause in laying. Although egg size did not vary between the groups in this case, it is unclear whether the same effect would be seen if queens had simply been restricted from laying at such high rates (e.g. if available empty brood cells had been reduced rather than removed entirely).

    Overall this study makes new contributions to our understanding of maternal control over egg size in honey bees. It provides stepping stones for further investigation of the molecular basis for egg size plasticity in insects.

  5. eLife

    Reviewer #2 (Public Review):

    This paper builds on recent work showing that honeybee queens can change the size of the eggs they lay over the course of their life. Here the authors identified an environmental condition that reversibly causes queens to change their egg sizes: namely, being in a relatively small or large colony context. Recently published work demonstrated the existence of this egg size plasticity, but it was completely unknown what signaled to the queen. In a series of simple and elegant experiments they confirmed the existence of this egg size plasticity, and narrowed down the set of environmental inputs to the queen that could be responsible for signaling the change in the environment. They also began the work of identifying genes and proteins that might be involved in controlling egg size. They did a comparative proteomic analysis between small-egg-laying ovaries and large-egg-laying ovaries, and then selected one candidate gene (Rho1). They showed that it is expressed during oogenesis, and that when it is knocked down, eggs get smaller.

    The experiments on honeybee colonies are well-designed, and they provide fairly strong evidence that the queens are reversibly changing egg size and that it is (at least some component of) their perception of colony size that is the signal. One minor but unavoidable weakness is that experiments on honeybees are necessarily done with small sample sizes. The authors were clear about this, however, and it was very effective that they showed all individual data points. Alongside the previous work on which this paper builds, I found their core results to be rather convincing and important.

    I found the parts of the paper on oogenesis to be useful, but overall less informative in answering the questions that the authors set out for those sections. On balance, I think the best way to interpret the oogenesis results is as "suggestive and exploratory". For instance, the experiment aimed at understanding the relationship between egg-laying rate and egg size does not include a direct measurement of egg-laying rate, but instead puts queens in a place with no suitable oviposition sites. The proteomic analysis was fine, but since they were using whole ovaries, with tissue pooled across all stages of oogenesis including mature oocytes, I would be cautious in interpreting the results to mean that they had identified proteins involved in making larger eggs. These proteins might just as easily be the proteins that are put into larger eggs. In fact, for the one candidate gene that is examined, its transcripts seem as though they are predominantly in the oocyte cell itself rather than in the supporting cells that actually control the egg size (although it is hard to tell from the micrographs without a label for cell interfaces).

    On that note, with the caveat that the sample sizes are quite small, I agree that there is some evidence that Rho1 is involved in honeybee oogenesis. If this was the only gene they knocked down, and given that it results in a small size change with such a small sample size, it strikes me as a bit of a stretch to say that these results are evidence that Rho1 plays an important role in egg size determination. It is essential to know if this is a generic result of inhibiting cytoskeletal function or a specific function of Rho1. That is beyond the scope of this study, but until those experiments are done, it is hard to know how to interpret these results. For context, in Drosophila, there are lots and lots of genes such that if you knock them down, you get a smaller or differently shaped egg, including genes involved in planar polarity, cytoskeleton, basement membrane, protrusion/motility, septate junctions, intercellular signaling and their signal transduction components, muscle functions, insect hormones, vitellogenesis, etc. This is helpful, perhaps, for thinking about how to interpret the knockdown of just one gene.

    Overall, I found the results to be technically sound, and there are several clever manipulations on honeybee colonies that will doubtless be repeated and elaborated in the future to great effect. The core result-that queens can change the size of their eggs quickly and reversibly, in response to some perceived signal-was honestly pretty astonishing to me, and it reveals that there are non-nutritive plastic mechanisms in insect oogenesis that we had no idea existed. I look forward to follow-up studies with interest.

  6. Version published to 10.1101/2022.05.22.492980v1 on bioRxiv