The Complexity of Social Complexity: A Quantitative Multidimensional Approach for Studies of Social Organization

Integrating computer vision and molecular neurobiology to bridge the gap between behavior and the brain

The past decade of social insect research has seen rapid development in automated behavioral tracking and molecular profiling of the nervous system, two distinct but complementary lines of inquiry into phenotypic variation across individuals, colonies, populations, and species. These experimental strategies have developed largely in parallel, as automated tracking generates a continuous stream of behavioral data, while, in contrast, 'omics-based profiling provides a single 'snapshot' of the brain. Better integration of these approaches applied to studying variation in social behavior will reveal the underlying genetic and neurobiological mechanisms that shape the evolution and diversification of social life. In this review, we discuss relevant advances in both fields and propose new strategies to better elucidate the molecular and behavioral innovations that generate social life.

Chemical and transcriptomic diversity do not correlate with ascending levels of social complexity in the insect order Blattodea

Eusocial insects, such as ants and termites, are characterized by high levels of coordinated social organization. This is contrasted by solitary insects that display more limited forms of collective behavior. It has been hypothesized that this gradient in sociobehavioral sophistication is positively correlated with chemical profile complexity, due to a potentially increased demand for diversity in chemical communication mechanisms in insects with higher levels of social complexity. However, this claim has rarely been assessed empirically. Here, we compare different levels of chemical and transcriptomic complexity in selected species of the order Blattodea that represent different levels of social organization, from solitary to eusocial. We primarily focus on cuticular hydrocarbon (CHC) complexity, since it has repeatedly been demonstrated that CHCs are key signaling molecules conveying a wide variety of chemical information in solitary as well as eusocial insects. We assessed CHC complexity and divergence between our studied taxa of different social complexity levels as well as the differentiation of their respective repertoires of CHC biosynthesis gene transcripts. Surprisingly, we did not find any consistent pattern of chemical complexity correlating with social complexity, nor did the overall chemical divergence or transcriptomic repertoire of CHC biosynthesis genes reflect on the levels of social organization. Our results challenge the assumption that increasing social complexity is generally reflected in more complex chemical profiles and point toward the need for a more cautious and differentiated view on correlating complexity on a chemical, genetic, and social level.

The diversity of social complexity in termites

Sociality underpins major evolutionary transitions and significantly influences the structure and function of complex ecosystems. Social insects, seen as the pinnacle of sociality, have traits like obligate sterility that are considered 'master traits', used as single phenotypic measures of this complexity. However, evidence is mounting that completely aligning both phenotypic and evolutionary social complexity, and having obligate sterility central to both, is erroneous. We hypothesize that obligate and functional sterility are insufficient in explaining the diversity of phenotypic social complexity in social insects. To test this, we explore the relative importance of these sterility traits in an understudied but diverse taxon: the termites. We compile the largest termite social complexity dataset to date, using specimen and literature data. We find that although functional and obligate sterility explain a significant proportion of variance, neither trait is an adequate singular proxy for the phenotypic social complexity of termites. Further, we show both traits have only a weak association with the other social complexity traits within termites. These findings have ramifications for our general comprehension of the frameworks of phenotypic and evolutionary social complexity, and their relationship with sterility.

Gene regulation supporting sociality shared across lineages and variation in complexity

Across evolutionary lineages, insects vary in social complexity, from those that exhibit extended parental care to those with elaborate divisions of labor. Here, we synthesize the sociogenomic resources from hundreds of species to describe common gene regulatory mechanisms in insects that regulate social organization across phylogeny and levels of social complexity. Different social phenotypes expressed by insects can be linked to the organization of co-expressing gene networks and features of the epigenetic landscape. Insect sociality also stems from processes like the emergence of parental care and the decoupling of ancestral genetic programs. One underexplored avenue is how variation in a group's social environment affects the gene expression of individuals. Additionally, an experimental reduction of gene expression would demonstrate how the activity of specific genes contributes to insect social phenotypes. While tissue specificity provides greater localization of the gene expression underlying social complexity, emerging transcriptomic analysis of insect brains at the cellular level provides even greater resolution to understand the molecular basis of social insect evolution.

Integration of information from multiple sources drives and maintains the division of labor in bumble bee colonies

Bumble bees are eusocial bees in which the division of labor (DoL) in reproduction and in task performance changes during their annual lifecycle. The queen monopolizes reproduction in young colonies, but at later stages, some workers start to challenge the queen and lay their own unfertilized eggs. The division of colony maintenance and growth tasks relates to worker body size. Reproduction and task performance are regulated by multiple social signals of the queen, the workers, and the brood. Here, we review recent studies suggesting that bumble bees use multiple sources of information to establish and maintain DoL in both reproduction and in task performance. Juvenile hormone (JH) is an important neuroendocrine signal involved in the regulation of DoL in reproduction but not in worker task performance. The reliance on multiple signals facilitates flexibility in face of changes in the social and geophysical environment.

Costs of reproduction are present but latent in eusocial bumblebee queens

BACKGROUND

The standard evolutionary theory of ageing proposes that ageing occurs because of a trade-off between reproduction and longevity. Eusocial insect queens exhibit positive fecundity-longevity associations and so have been suggested to be counter-examples through not expressing costs of reproduction and through remodelling conserved genetic and endocrine networks regulating ageing and reproduction. If so, eusocial evolution from solitary ancestors with negative fecundity-longevity associations must have involved a stage at which costs of reproduction were suppressed and fecundity and longevity became positively associated. Using the bumblebee (Bombus terrestris), we experimentally tested whether queens in annual eusocial insects at an intermediate level of eusocial complexity experience costs of reproduction, and, using mRNA-seq, the extent to which they exhibit a remodelling of relevant genetic and endocrine networks. Specifically, we tested whether costs of reproduction are present but latent, or whether a remodelling of relevant genetic and endocrine networks has already occurred allowing queens to reproduce without costs.

RESULTS

We experimentally increased queens' costs of reproduction by removing their eggs, which caused queens to increase their egg-laying rate. Treatment queens had significantly reduced longevity relative to control queens whose egg-laying rate was not increased. Reduced longevity in treatment queens was not caused by increased worker-to-queen aggression or by increased overall activity in queens. In addition, treatment and control queens differed in age-related gene expression based on mRNA-seq in both their overall expression profiles and the expression of ageing-related genes. Remarkably, these differences appeared to occur principally with respect to relative age, not chronological age.

CONCLUSIONS

This study represents the first simultaneously phenotypic and transcriptomic experimental test for a longevity cost of reproduction in eusocial insect queens. The results support the occurrence of costs of reproduction in annual eusocial insects of intermediate social complexity and suggest that reproductive costs are present but latent in queens of such species, i.e. that these queens exhibit condition-dependent positive fecundity-longevity associations. They also raise the possibility that a partial remodelling of genetic and endocrine networks underpinning ageing may have occurred in intermediately eusocial species such that, in unmanipulated conditions, age-related gene expression depends more on chronological than relative age.

Behavioral roles of biogenic amines in bumble bee males

To compare the behavioral roles of biogenic amines in the males of primitive and advanced eusocial bees, we determined the levels of dopamine- and octopamine-related substances in the brain, and the behavioral effects of these monoamines by drug injection in the primitive eusocial bumble bee, Bombus ignitus. The levels of dopamine and its precursors in the brain peaked at the late pupal stage, but the dopamine peak extended to adult emergence. The tyramine and octopamine levels increased from the mid-pupal to adult stages. The locomotor and flight activities, and light preference increased with age. Injection of octopamine and its receptor antagonist had significant effects on the locomotor and flight activities, whereas dopamine injection did not, indicating that these activities can be regulated by the octopaminergic system. We also determined the dynamics of dopamine-related substances in honey bee (Apis mellifera) drones. The changes in the dopamine level in the brains of honey bee drones exhibited two peaks from the pupal to adult stages, whereas the bumble bee males had only one peak. These are consistent with the behavioral functions of dopamine in honey bee drones and ineffectiveness of dopamine injection at the adult stage in bumble bee males.

Ecological Drivers of Non-kin Cooperation in the Hymenoptera

Despite the prominence of kin selection as a framework for understanding the evolution of sociality, many animal groups are comprised of unrelated individuals. These non-kin systems provide valuable models that can illuminate drivers of social evolution beyond indirect fitness benefits. Within the Hymenoptera, whose highly related eusocial groups have long been cornerstones of kin selection theory, groups may form even when indirect fitness benefits for helpers are low or absent. These non-kin groups are widespread and abundant, yet have received relatively little attention. We review the diversity and organization of non-kin sociality across the Hymenoptera, particularly among the communal bees and polygynous ants and wasps. Further, we discuss common drivers of sociality across these groups, with a particular focus on ecological factors. Ecological contexts that favor non-kin sociality include those dominated by resource scarcity or competition, climatic stressors, predation and parasitism, and/or physiological constraints associated with reproduction and resource exploitation. Finally, we situate Hymenopteran non-kin sociality within a broader biological context by extending insights from these systems across diverse taxa, especially the social vertebrates. Non-kin social groups thus provide unique demonstrations of the importance of ecological factors in mediating the evolutionary transition from solitary to group living.

Defense in Social Insects: Diversity, Division of Labor, and Evolution

All social insects defend their colony from predators, parasites, and pathogens. In Oster and Wilson's classic work, they posed one of the key paradoxes about defense in social insects: Given the universal necessity of defense, why then is there so much diversity in mechanisms? Ecological factors undoubtedly are important: Predation and usurpation have imposed strong selection on eusocial insects, and active defense by colonies is a ubiquitous feature of all social insects. The description of diverse insect groups with castes of sterile workers whose main duty is defense has broadened the purview of social evolution in insects, in particular with respect to caste and behavior. Defense is one of the central axes along which we can begin to organize and understand sociality in insects. With the establishment of social insect models such as the honey bee, new discoveries are emerging regarding the endocrine, neural, and gene regulatory mechanisms underlying defense in social insects. The mechanisms underlying morphological and behavioral defense traits may be shared across diverse groups, providing opportunities for identifying both conserved and novel mechanisms at work. Emerging themes highlight the context dependency of and interaction between factors that regulate defense in social insects.

Foraging gene expression patterns in queens, workers, and males in a eusocial insect

Reproductive division of labour is based on biased expression of complementary parental behaviours, brood production (egg-laying) by queens and brood care (in particular, brood-provisioning) by workers. In many social insect species, queens provision brood when establishing colonies at the beginning of a breeding season and reproductive division of labour begins with the emergence of workers. In many social insect species, the expression of foraging (for) mRNA is associated with the intensity of foraging behaviour and therefore brood-provisioning. However, only two studies have compared queen and worker for expression levels and neither accounted for transcript splice variation. In this study, we compare the expression level of the for-α transcript variant across four life stages of the queen caste, two behavioural groups of workers, and males of a eusocial sweat bee Lasioglossum laevissimum (Smith, 1853). Foundresses collected prior to the onset of the foraging season and males had the highest for-α expression levels. All active (post-hibernatory) queens and workers had similar for-α expression levels independent of behaviour. These results suggest that the for gene in L. laevissimum acts as a primer before foraging activity and that caste-specific expression patterns correlate with the timing of foraging activity in queens and workers.

Life History and the Transitions to Eusociality in the Hymenoptera

Although indirect selection through relatives (kin selection) can explain the evolution of effectively sterile offspring that act as helpers at the nest (eusociality) in the ants, bees, and stinging wasps (aculeate Hymenoptera), the genetic, ecological, and life history conditions that favor transitions to eusociality are poorly understood. In this study, ancestral state reconstruction on recently published phylogenies was used to identify the independent transitions to eusociality in each of the taxonomic families that exhibit eusociality. Semisociality, in which a single nest co-foundress monopolizes reproduction, often precedes eusociality outside the vespid wasps. Such a route to eusociality, which is consistent with groups consisting of a mother and her daughters (subsocial) at some stage and ancestral monogamy, is favored by the haplodiploid genetic sex determination of the Hymenoptera (diploid females and haploid males) and thus may explain why eusociality is common in the Hymenoptera. Ancestral states were also reconstructed for life history characters that have been implicated in the origins of eusociality. A loss of larval diapause during unfavorable seasons or conditions precedes, or coincides with, all but one transition to eusociality. This pattern is confirmed using phylogenetic tests of associations between state transition rates for sweat bees and apid bees. A loss of larval diapause may simply reflect the subsocial route to eusociality since subsociality is defined as females interacting with their adult daughters. A loss of larval diapause and a gain of subsociality may be associated with an extended breeding season that permits the production of at least two broods, which is necessary for helpers to evolve. Adult diapause may also lower the selective barrier to a first-brood daughter becoming a helper. Obligate eusociality meets the definition of a major evolutionary transition, and such transitions have occurred five times in the Hymenoptera.

Body Size and Behavioural Plasticity Interact to Influence the Performance of Free-Foraging Bumble Bee Colonies

Specialisation and plasticity are important for many forms of collective behaviour, but the interplay between these factors is little understood. In insect societies, workers are often developmentally primed to specialise in different tasks, sometimes with morphological or physiological adaptations, facilitating a division of labour. Workers may also plastically switch between tasks or vary their effort. The degree to which developmentally primed specialisation limits plasticity is not clear and has not been systematically tested in ecologically relevant contexts. We addressed this question in 20 free-foraging bumble bee (Bombus terrestris) colonies by continually manipulating colonies to contain either a typically diverse, or a reduced ("homogeneous"), worker body size distribution while keeping the same mean body size, over two trials. Pooling both trials, diverse colonies produced a larger comb mass, an index of colony performance. The link between body size and task was further corroborated by the finding that foragers were larger than nurses even in homogeneous colonies with a very narrow body size range. However, the overall effect of size diversity stemmed mostly from one trial. In the other trial, homogeneous and diverse colonies showed comparable performance. By comparing behavioural profiles based on several thousand observations of individuals, we found evidence that workers in homogeneous colonies in this trial rescued colony performance by plastically increasing behavioural specialisation and/or individual effort, compared to same-sized individuals in diverse colonies. Our results are consistent with a benefit to colonies of large and small specialists under certain conditions, but also suggest that plasticity or effort can compensate for reduced (size-related) specialisation. Thus, we suggest that an intricate interplay between specialisation and plasticity is functionally adaptive in bumble bee colonies.

Gene expression during larval caste determination and differentiation in intermediately eusocial bumblebees, and a comparative analysis with advanced eusocial honeybees

The queen‐worker caste system of eusocial insects represents a prime example of developmental polyphenism (environmentally‐induced phenotypic polymorphism) and is intrinsic to the evolution of advanced eusociality. However, the comparative molecular basis of larval caste determination and subsequent differentiation in the eusocial Hymenoptera remains poorly known. To address this issue within bees, we profiled caste‐associated gene expression in female larvae of the intermediately eusocial bumblebee Bombus terrestris. In B. terrestris, female larvae experience a queen‐dependent period during which their caste fate as adults is determined followed by a nutrition‐sensitive period also potentially affecting caste fate but for which the evidence is weaker. We used mRNA‐seq and qRT‐PCR validation to isolate genes differentially expressed between each caste pathway in larvae at developmental stages before and after each of these periods. We show that differences in gene expression between caste pathways are small in totipotent larvae, then peak after the queen‐dependent period. Relatively few novel (i.e., taxonomically‐restricted) genes were differentially expressed between castes, though novel genes were significantly enriched in late‐instar larvae in the worker pathway. We compared sets of caste‐associated genes in B. terrestris with those reported from the advanced eusocial honeybee, Apis mellifera, and found significant but relatively low levels of overlap of gene lists between the two species. These results suggest both the existence of low numbers of shared toolkit genes and substantial divergence in caste‐associated genes between Bombus and the advanced eusocial Apis since their last common eusocial ancestor.