Proximate and ultimate factors regulating reproduction and division of labour in insect societies

A major outstanding question in evolutionary biology is to explain why previously independently replicating subunits teamed up to form higher levels of organization, such as eukaryotic cells, multicellular organisms or animal societies. The evolution of eusociality in insects, marked by the evolution of a reproductive division of labour between queens and workers and complex forms of division of labour among the workers themselves, stands as one of the most spectacular examples of one of these “major transitions in evolution”. The Darwinian paradox presented by the reproductive altruism of the workers, who sacrifice their own reproduction for the benefit of the queen, led to the development of Hamilton’s inclusive fitness theory and fruitful research into the ultimate factors favouring sociality. More recently, more mechanistic factors have also been increasingly considered in treatments on the evolution of social behaviour and division of labour, focusing e.g., on the evo-devo of sociality. Here, the goal is to decipher how social traits in social species were constructed from specific physiological or gene regulatory building blocks in their pre-social ancestors. 

My doctoral research aimed to combine these proximate and ultimate frameworks to shed light on the evolution of division of labour in insect societies. In the first section of this thesis, I investigated the hormonal processes that regulate the division of labour of social insects, and determined whether the same hormonal pathways regulate different aspects of their division of labour, such as the reproductive division of labour between egg laying and non egg-laying females and reproductive signalling. Through experimental studies conducted on neotropical primitively eusocial wasps, I show that both processes are under shared juvenile hormone (JH) control, and argue that such a situation would have pre-adapted fertility signals in primitively social species to evolve into queen pheromones in advanced eusocial ones. Building on the insights into role of JH in primitively eusocial species, I then examined whether this same hormone might also regulate the age-related division of labour among the workers in highly eusocial wasps. While previous research predominantly centred on the extensively studied honey bee, little information existed about the role of JH in regulating division of labour in advanced eusocial wasps, which independently evolved sociality. The results of this study show that JH indeed strongly affects the workers' behavioural maturation and age-related division of labour, and acquired this function multiple times across different taxa and several independent origins of sociality.

Although the evolution of eusociality in social Hymenoptera can be well explained by kin selection, the pronounced inequality in reproductive potential between queens and workers also provides scope for conflict. This is because at an individual level, females would often benefit from developing as queens instead of workers. From an ultimate angle, my research focused on the expression and resolution of such conflicts over caste fate in advanced eusocial species with morphologically distinct queen and worker castes. I developed a new inclusive fitness model of caste fate conflict in bee societies, for the first time implementing dynamic aspects of colony growth and reproduction, and validated the model predictions based on an extensive meta-analysis of data from 42 stingless bee species and the honeybee. The model shows why in species where caste is self-determined, such as in stingless bees of the genus Melipona, queens are greatly overproduced, while in species where caste is nutritionally determined, levels of queen production are ca. 2 orders of magnitude lower, and queens end up being produced in line with colony needs. In Melipona, we also show that at the observed level of queen overproduction, where ca. 10% of all female larvae selfishly choose to develop as queens, an evolutionary equilibrium is reached, where the individual benefit of developing as a queen balances with the collective cost of queen overproduction. In a last chapter, I review the occurrence of caste fate conflict in relation to observed patterns of caste determination across different lineages of social insects, including in termites, where the evidence for caste fate conflict had not been detailed before. I argue that caste fate conflict illustrates a model example of how individual and collective interests can come in conflict with each other, and how such conflicts can be resolved in biological systems.

By integrating both proximate and ultimate perspectives on the evolution of sociality and combining both experimental work and theoretical modelling, this thesis contributes to our understanding of the evolution of division of labour in insect societies and should be a valuable addition to the broader field of social evolution.