Deciphering the genetic bases of social behavior in Hymenoptera

Ever since Darwin, biologists have been fascinated by the evolution of sociality. In its most extreme form, eusocial species exhibit a division of labor in which some individuals perform reproductive tasks such as egg-laying, while others perform non-reproductive roles such as foraging, nest building and defense.

This type of system forces individuals to give up some or all of their own reproductive success to aid in the reproduction of others in their group, a concept that at first glance seems inconsistent with key principles of evolution. (i.e. the training of natural selection on individuals). While the bee is perhaps the best-known example of a speciesthe complex bee society represents only one end of a spectrum of social structures that can be observed in Hymenoptera, which includes bees, wasps and ants.

At the other end are more rudimentary social structures involving, at the most basic level, the cooperation of only a few individuals and their offspring. While most insect research to date sociality focused on more complex social systems, understanding the evolution of these more rudimentary forms will likely help reveal early changes on the path to sociality.

The authors of a new study published in Biology and evolution of genomes, set out to fill this gap.

According to first author Emeline Favreau, “Our work was unique in that we focused on six bees and species of wasp which are not very social, but have more rudimentary forms of cooperation and are close relatives of highly social species.” Using machine learning algorithms to analyze gene expression Across six species that represent multiple origins of sociality, the authors uncovered a shared genetic “toolkit” for sociality that may form the basis for the evolution of more complex social structures.

The international team of researchers included Katherine S. Geist (co-first author) and Amy L. Toth from Iowa State University, Christopher DR Wyatt and Seirian Sumner from University College London, and Sandra M. Rehan from the York University in Toronto. The authors worked together on this article “because we all find it important to understand the origins of sociality”, explains Favreau.

“We had gone out into the field to observe the fantastic diversity of social life, such as large wasp nests busy with collective behavior or small carpenter bees organizing their broods in tiny tree branches. We kept asking ourselves: But how? Did these behaviors arise? With this paper, we have delved deep into evolutionary histories to uncover molecular evidence for the emergence of social organization.”

The study involved a comparative meta-analysis of data from three bee species and three wasp species that represent four independent origins of sociality: the halictid bee Megalopta genalis, the xylocopine bees Ceratina australensis and C. calcarata, the stenogastrin wasp Liostenogaster flavolineata and the polistine wasps Polistes canadensis and P. dominula.

“Using global gene expression data in the brains of different behavioral groups (reproductive and non-reproductive females), we found that there is a common set of genes associated with these fundamental social divisions in bees and bees. wasps,” says Favreau. “It’s exciting because it suggests that there may be common molecular ‘themes’ associated with cooperation between species.”

A number of functional groups associated with sociality in this study have also been linked to sociality in other social bees and ants. These include genes related to chromatin binding, DNA binding, regulation of telomere length, reproduction, and metabolism.

On the other hand, the study also identified many lineage-specific genes and functional groups associated with social phenotypes. According to the authors, these findings “reveal how taxon-specific molecular mechanisms complement a basic toolkit of molecular processes in the sculpting of traits related to the evolution of eusociality.”

Interestingly, Favreau notes that “a machine learning approach to these large datasets was the best method to uncover these similarities.” While the authors first attempted traditional methods to study differential gene expression, these largely grouped species by phylogeny and failed to identify sets of genes associated with sociality. In contrast, machine learning tools provided “a more nuanced and sensitive approach”, allowing the authors to identify gene expression similarities over a large evolutionary distance.

A remaining question is how the results of this study, which focused on species with rudimentary forms of sociality, might compare to an obligately eusocial species with morphologically distinct castes of breeding and non-breeding individuals. According to Favreau, “This is something we are currently working on and hope to be able to tackle in the near future. We are taking a broader approach to examining how genes and genomes change during social evolution.”

This includes the addition of transcriptomic data for 16 additional bee and wasp species, allowing “broader comparative study with wasp and bee species that are solitary, have rudimentary sociality, and complex sociality.”

Expanding the study, however, requires obtaining samples from around the world, a feat that has sometimes proven difficult. “It was actually a challenge to find many of these species, some of which had never been studied genetically before,” Favreau notes.

“Given the global diversity of taxa and the remote places many have been collected in it, we are happy to have been able to obtain all the specimens and genomes given the global pandemic and travel restrictions in recent years.

The team was ultimately able to acquire a number of samples through partnerships with other researchers and institutions, underscoring the essential role of collaboration in scientific discovery.