The evolution of caste plasticity and caste dimorphism in insect societies

Social insects are a key model in the study of sociality and a prime example of “major transitions in evolution”, where previously independently breeding individuals teamed up and went on to live in highly advanced societies. Insect societies are characterized by a reproductive division of labour in which queens specialise in breeding and workers engage mostly in non-reproductive tasks. For comparative studies, Hymenopteran insects are an excellent model, as they comprise species that span the range from solitary to primitively eusocial and highly eusocial species. Indeed, explaining the transition to eusociality and the origin of a semi-permanent reproductive division of labour in insect societies has become a major endeavour in the field of social evolution. Complementary to this, there has been a great interest to understand what cognitive changes were associated with transitions in the evolution of sociality in insects. The social brain hypothesis, which proposes that high cognitive abilities are required to maintain social relationships, does not appear to apply to social insects. However, as insect societies can live in societies comprised of millions of individuals, yet at an individual level typically have only limited or highly specialized cognitive abilities. This is linked to the fact that insect societies gain much of their collective intelligence through decentralized, self-organized mechanisms and relatively simple behavioural rules of thumb, which has been referred to as the “distributed cognition” hypothesis. Hence, in contrast to what was originally thought to be the case, workers of social Hymenoptera living in larger colonies tend to show a reduced development and investment in important cognitive brain centers, e.g. having smaller mushroom bodies than their solitary ancestors. Social insect colonies also face a trade-off between producing a small number of larger and cognitively more capable workers versus producing a large number of smaller and cognitively more limited individuals. Likewise, at the individual level, there can be tissue trade-offs between the production of complex brain tissue that is energetically costly to develop and maintain and the production of other tissues, such as ovaries and oocytes for reproductively active individuals. In this project, we will compare primitively and highly eusocial species of social Hymenoptera to test how phenotypic plasticity evolved during and following the origin of sociality. In particular, we will formally test the mechanisms that were involved in the origin and maintenance of morphologically discrete queen-worker castes, caste plasticity and worker caste polymorphism in insect societies. The scientific research objectives that we will address are the following: the evolution of caste plasticity in function of reproductive role; the evolutionary origin of morphologically specialized reproductive castes; neuroethology and chemical ecology of task specialisation in ants. We will study the occurrence of caste plasticity in function of reproductive role across a wider range of primitively and advanced eusocial Palearctic and Neotropical ants, bees and wasps, investigating how tissue trade-offs affect investment in reproduction vs. in brain development. Comparative study of different species of bees and ants characterized by different degrees of sociality (primitively eusocial and highly eusocial) will enable us to characterize reproductive and behavioural plasticity in eusocial insects, quantify trade-offs that affect the expression and evolution of such plasticity, as well as document different mechanisms to obtain behavioural plasticity, e.g. through reversible changes in brain volume and brain functioning, or via changes in (highly modular) brain gene expression patterns. Furthermore, we will study another major leap in the evolution of eusociality, which is the origin of morphologically discrete queen-worker castes, i.e. the origin of an advanced eusocial lifestyle. Thanks to the recent assembly of the genome of the advanced eusocial common wasp Vespula vulgaris, we will now be able to test the reproductive ground plan hypothesis for the first time in the context of caste determination. Lastly, we will test how the evolution of morphologically specialized worker castes affects brain plasticity. This will be studied by comparing brain morphology and age cohort in the higher leafcutter ant Atta sexdens, which possesses a highly polymorphic and differentiated worker caste, with that in the lower leafcutter ant Mycocepurus goeldii, which has an undifferentiated, monomorphic worker caste.