Periodic Reporting for period 4 - CASTECON (SHARING A GENOME: CASTE ANTAGONISM AND COADAPTATION IN SOCIAL INSECTS)
Berichtszeitraum: 2021-07-01 bis 2023-06-30
SHARING A GENOME: CASTE ANTAGONISM AND COADAPTATION IN SOCIAL INSECTS
As social organisms ourselves, humans have an interest in understanding animal societies. Eusociality is the most extreme form of social behaviour, where some individuals (‘workers’) sacrifice their own reproduction to rear the offspring of others (‘queens’). Queens are typically long-lived and specialize on egg-laying, while workers are short-lived and specialize on foraging. In order to maximize their performance, the two castes therefore need very different characteristics (dimorphism). Yet remarkably, both castes are produced from the same genetic material. This suggests an important constraint, ‘caste antagonism’: a gene that increases performance in one caste may decrease performance in the other. For example, a gene that increases fertility may benefit queens, but it may cause workers to use up resources unnecessarily on ovary development at the expense of reduced foraging efficiency. Antagonism may mean that queens and workers are prevented from reaching their optimal characteristics, so that the fitness of the population or social group is reduced (Figure 1). Antagonism could be resolved through caste-specific gene expression (Figure 1), but genetic barriers may prevent this.
The objectives of the CASTECON project were to use wild populations of sweat bees and paper wasps to:
(1) Determine, by analysing gene expression, whether antagonism really does constrain colony performance;
(2) Measure changes in queen-worker dimorphism across a latitudinal gradient;
(3) Use social manipulations and transplants to investigate how antagonism could be resolved, and the correlation between social phenotype and brain structure;
(4) Test whether queens manipulate worker body size in their own interests.
Major conclusions so far are as follows:
A. Our gene expression data provide good evidence that queen-worker dimorphism is constrained by caste antagonism. Furthermore, while dimorphism often involves queens and workers expressing different genes, expressing different versions of the same genes may be an alternative route to evolve dimorphism.
B. Dimorphism varies between populations, but is not correlated with latitude in a simple way.
C. Sweat bee workers show remarkable plasticity: when we removed the original queen, a worker took over her role and produced just as many offspring. We also found plasticity when the environment changed. For example, bees we transplanted from non-social populations to environments where the natives are social produced offspring that had more of the sensilla involved in olfactory perception (e.g. for nest-mate recognition). Micro-CT scanning revealed brain region differences between bees from social and non-social populations.
D. Cross-fostering created mismatches between queens and workers – for example, colonies with workers that were unusually large for the cross-fostered queen. Our results suggest that queens may limit worker size strategically to optimize foraging, rather than to reduce queen-worker conflict.
As social organisms ourselves, humans have an interest in understanding animal societies. Eusociality is the most extreme form of social behaviour, where some individuals (‘workers’) sacrifice their own reproduction to rear the offspring of others (‘queens’). Queens are typically long-lived and specialize on egg-laying, while workers are short-lived and specialize on foraging. In order to maximize their performance, the two castes therefore need very different characteristics (dimorphism). Yet remarkably, both castes are produced from the same genetic material. This suggests an important constraint, ‘caste antagonism’: a gene that increases performance in one caste may decrease performance in the other. For example, a gene that increases fertility may benefit queens, but it may cause workers to use up resources unnecessarily on ovary development at the expense of reduced foraging efficiency. Antagonism may mean that queens and workers are prevented from reaching their optimal characteristics, so that the fitness of the population or social group is reduced (Figure 1). Antagonism could be resolved through caste-specific gene expression (Figure 1), but genetic barriers may prevent this.
The objectives of the CASTECON project were to use wild populations of sweat bees and paper wasps to:
(1) Determine, by analysing gene expression, whether antagonism really does constrain colony performance;
(2) Measure changes in queen-worker dimorphism across a latitudinal gradient;
(3) Use social manipulations and transplants to investigate how antagonism could be resolved, and the correlation between social phenotype and brain structure;
(4) Test whether queens manipulate worker body size in their own interests.
Major conclusions so far are as follows:
A. Our gene expression data provide good evidence that queen-worker dimorphism is constrained by caste antagonism. Furthermore, while dimorphism often involves queens and workers expressing different genes, expressing different versions of the same genes may be an alternative route to evolve dimorphism.
B. Dimorphism varies between populations, but is not correlated with latitude in a simple way.
C. Sweat bee workers show remarkable plasticity: when we removed the original queen, a worker took over her role and produced just as many offspring. We also found plasticity when the environment changed. For example, bees we transplanted from non-social populations to environments where the natives are social produced offspring that had more of the sensilla involved in olfactory perception (e.g. for nest-mate recognition). Micro-CT scanning revealed brain region differences between bees from social and non-social populations.
D. Cross-fostering created mismatches between queens and workers – for example, colonies with workers that were unusually large for the cross-fostered queen. Our results suggest that queens may limit worker size strategically to optimize foraging, rather than to reduce queen-worker conflict.
Overview: Six of our seven planned experiments were successfully conducted. Results have been disseminated through 8 publications and at conferences, with a similar number of publications expected in future.
Main results so far:
1. Parsons et al. (2017) reported a new set of 24 genetic markers that we developed for the sweat bee L. malachurum so that we could determine whether the queen, or one of her workers, has produced each of the offspring.
2. Pennell et al. (2018) is an important review where we hoped to stimulate research focussed on caste antagonism. The review characterised antagonism, discussed how it might be resolved during evolution, suggested approaches to study it and reviewed the empirical evidence.
3. Couchoux & Field (2019) relates to objective (4) above. In social animals, parents might manipulate characteristics of their offspring in their own interests. In paper wasps, the first offspring produced are smaller than the queen and become workers. We investigated whether paper wasp queens benefit by producing small daughter workers. We used cross-fostering to create size mismatches. Before cross-fostering, there was a positive correlation between queen and worker body sizes: larger queens had larger workers. After cross-fostering, the correlation was absent. We then recorded foraging, reproductive effort and aggression. Queens were less likely to attack relatively larger workers, but larger workers did not forage less, did not invest more in ovarian development, and were not more aggressive themselves. But because small workers were no less successful foragers, producing a larger number of smaller workers may optimize colony work effort. Thus, queens may limit worker size strategically to optimize foraging, with limited evidence of size-based queen-worker conflict.
4. Parsons et al. (2017) analyse choice of social partner in paper wasps. Results suggest that kin groups form via a rule-of-thumb where wasps simply select nearby partners, rather than via kin recognition.
5. Pennell et al. (2021) showed that an unexpectedly large proportion of L. malachurum nests were orphaned: the original queen had died or been usurped. Colonies with original queens invested more in female offspring than did orphaned colonies (Pennell & Field 2021). Our data also suggested queen coercion – nests with larger queens produced sex ratios closer to the queen’s optimum.
6. Boulton & Field (2022) showed that social populations of the polymorphic H. rubicundus have denser antennal sensilla involved in olfactory perception than solitary populations. Sensilla density appears to be developmentally plastic: bees transplanted from north to south produced offspring (which developed in the south) with more olfactory sensilla.
7. In small colony social insects, individuals may switch from non-reproductive to reproductive roles (e.g. after orphaning). To investigate plasticity, Price & Field (2022) promoted L. malachurum workers experimentally to reproductive roles. Remarkably, we found worker-queens produced the same number of offspring as original foundress-queens, and worker-queens monopolised reproduction to the same extent. However, non-reproductives foraged less with worker-queens, producing less offspring biomass.
8. Gruber & Field (2022) tested an important assumption underlying a model concerning how eusociality evolves (Seger 1983 Nature). The model requires first brood males to survive and mate with females of the second brood. We obtained high estimates of male survival in H. rubicundus, supporting the model.
Main results so far:
1. Parsons et al. (2017) reported a new set of 24 genetic markers that we developed for the sweat bee L. malachurum so that we could determine whether the queen, or one of her workers, has produced each of the offspring.
2. Pennell et al. (2018) is an important review where we hoped to stimulate research focussed on caste antagonism. The review characterised antagonism, discussed how it might be resolved during evolution, suggested approaches to study it and reviewed the empirical evidence.
3. Couchoux & Field (2019) relates to objective (4) above. In social animals, parents might manipulate characteristics of their offspring in their own interests. In paper wasps, the first offspring produced are smaller than the queen and become workers. We investigated whether paper wasp queens benefit by producing small daughter workers. We used cross-fostering to create size mismatches. Before cross-fostering, there was a positive correlation between queen and worker body sizes: larger queens had larger workers. After cross-fostering, the correlation was absent. We then recorded foraging, reproductive effort and aggression. Queens were less likely to attack relatively larger workers, but larger workers did not forage less, did not invest more in ovarian development, and were not more aggressive themselves. But because small workers were no less successful foragers, producing a larger number of smaller workers may optimize colony work effort. Thus, queens may limit worker size strategically to optimize foraging, with limited evidence of size-based queen-worker conflict.
4. Parsons et al. (2017) analyse choice of social partner in paper wasps. Results suggest that kin groups form via a rule-of-thumb where wasps simply select nearby partners, rather than via kin recognition.
5. Pennell et al. (2021) showed that an unexpectedly large proportion of L. malachurum nests were orphaned: the original queen had died or been usurped. Colonies with original queens invested more in female offspring than did orphaned colonies (Pennell & Field 2021). Our data also suggested queen coercion – nests with larger queens produced sex ratios closer to the queen’s optimum.
6. Boulton & Field (2022) showed that social populations of the polymorphic H. rubicundus have denser antennal sensilla involved in olfactory perception than solitary populations. Sensilla density appears to be developmentally plastic: bees transplanted from north to south produced offspring (which developed in the south) with more olfactory sensilla.
7. In small colony social insects, individuals may switch from non-reproductive to reproductive roles (e.g. after orphaning). To investigate plasticity, Price & Field (2022) promoted L. malachurum workers experimentally to reproductive roles. Remarkably, we found worker-queens produced the same number of offspring as original foundress-queens, and worker-queens monopolised reproduction to the same extent. However, non-reproductives foraged less with worker-queens, producing less offspring biomass.
8. Gruber & Field (2022) tested an important assumption underlying a model concerning how eusociality evolves (Seger 1983 Nature). The model requires first brood males to survive and mate with females of the second brood. We obtained high estimates of male survival in H. rubicundus, supporting the model.
Our research programme was broadly novel and went beyond the state of the art by being the first in-depth investigation of genetic constraints on caste evolution. How dimorphism can evolve, the possibility that its evolution could be limited by genetic constraints, and the processes that could resolve those constraints, are topics that had hardly been considered before. Empirically, our work is unusual because our data were collected in the field, where natural selection operates, unlike almost all previous studies of sexual and caste antagonism. We are also the first to use cross-fostering to investigate the possibility that queens manipulate worker characteristics. The overall result is a new and exciting perspective on queen-worker coevolution.