Volume 92, No. 1 THE QUARTERLY REVIEW OF BIOLOGY March 2017 THE ROLE OF BROOD IN EUSOCIAL HYMENOPTERA Eva Schultner Centre of Excellence in Biological Interactions, Department of Biosciences, University of Helsinki 00014 Helsinki, Finland Tvärminne Zoological Station 10900 Hanko, Finland Institut für Zoologie, Universität Regensburg 93053 Regensburg, Germany e-mail: [email protected] Jan Oettler Institut für Zoologie, Universität Regensburg 93053 Regensburg, Germany e-mail: [email protected] Heikki Helanterä Centre of Excellence in Biological Interactions, Department of Biosciences, University of Helsinki 00014 Helsinki, Finland Tvärminne Zoological Station 10900 Hanko, Finland e-mail: [email protected] keywords eusocial Hymenoptera, brood, development, cooperation, conflict abstract Study of social traits in offspring traditionally reflects on interactions in simple family groups, with famous examples including parent-offspring conflict and sibling rivalry in birds and mammals. In contrast, studies of complex social groups such as the societies of ants, bees, and wasps focus mainly on adults and, in particular, on traits and interests of queens and workers. The social role of developing individuals in complex societies remains poorly understood. We attempt to fill this gap by illustrating The Quarterly Review of Biology, March 2017, Vol. 92, No. 1 Copyright © 2017 by The University of Chicago Press. All rights reserved. 0033-5770/2017/9201-0002$15.00 39 40 THE QUARTERLY REVIEW OF BIOLOGY Volume 92 that development in social Hymenoptera constitutes a crucial life stage with important consequences for the individual as well as the colony. We begin by describing the complex social regulatory network that modulates development in Hymenoptera societies. By highlighting the inclusive fitness interests of developing individuals, we show that they may differ from those of other colony members. We then demonstrate that offspring have evolved specialized traits that allow them to play a functional, coop- erative role within colonies and give them the potential power to act toward increasing their inclusive fitness. We conclude by providing testable predictions for investigating the role of brood in colony interactions and giving a general outlook on what can be learned from studying offspring traits in hymenopteran societies. Introduction can even display different levels of social ity IFE is social and it is the interactions depending on environmental conditions L among molecules, cells, or individuals (e.g., in bees; Michener 1974). Second, euso- that have created functional genomes from ciality within the Hymenoptera has evolved simple replicators, multicellular organisms several times independently ( Johnson et al. from a unicellular ancestor, and animal so- 2013). Finally, there is large variation in so- cieties from solitary organisms. Across these cial ecological complexity even within euso- evolutionary transitions (Maynard Smith and cial Hymenoptera, from the annual, simple Szathmáry 1995), similar ultimate factors family groups of bumble bees to the com- underlie how single entities form and main- plex nest networks of supercolonial wood tain cohesive social groups. On one hand, ants (Schultner et al. 2016), making this genetic relatedness between individual enti- taxon ideal for understanding the evolu- ties facilitates cooperation because partners tionary dynamics that govern the forma tion, gain indirect fitness from helping relatives maintenance, and cohesion of complex so- (Hamilton 1964). On the other hand, be- cial groups (Bourke 2011). cause partners are rarely genetically identical The evolution of eusociality in the Hy- and do not overlap perfectly in their fitness menoptera is anchored in simple family interests, potential for conflict remains and groups, with adult offspring staying in their complex control mechanisms are predicted parental nest to help their mother repro- to evolve in order to keep selfishness in check duce instead of dispersing and reproducing (Bourke 2011). Both cooperation and con- themselves (Hughes et al. 2008). A typical flict are thus crucial determinants of social eusocial Hymenoptera colony is charac- cohesion. terized by reproductive division of labor Insects have evolved an especially large between adult females of overlapping gen- range of social complexity (Figure 1), which erations—the reproductive queen(s) and reaches its apex in the colonies of eusocial her (facultatively) sterile daughter workers. insects that are characterized by reproduc- In primitively eusocial species, differences tive division of labor between individuals, between queens and workers are subtle, in- cooperative brood care, and the presence of volving mainly changes in adult physiology individuals from overlapping generations and/or behavior (O’Donnell 1998), and in- (Wilson 1971). The Hymenoptera—ants, dividuals are capable of switching between bees, and wasps—stand out across the range roles. Although the reproductive capacity of insect sociality for several reasons. First, of workers is constrained, for instance, via this group exhibits the entire range of soci- dominance hierarchies or queen phero- ality, from the solitary lifestyles of parasitic monal control (Van Oystaeyen et al. 2014), wasps over the semisocial groups of sweat they usually retain the ability to reproduce bees to the irreversibly eusocial superorgan- sexually, for example, in wasps (Suzuki ismal societies of ants and honey bees, with 1985; Chandrashekara and Gadagkar 1991) their morphologically separated queen and and bees (Michener 1990). When a queen worker castes and high levels of social co- dies, one of her daughters will take over the hesion (Helanterä 2016). The same species nest—in many cases her success depends March 2017 ROLE OF BROOD IN EUSOCIAL HYMENOPTERA 41 on her physical dominance over nestmates potent, leaving developing individuals with (Michener 1990; Kukuk and May 1991; more reproductive options (Khila and Abou- Kukuk 1994). In advanced eusocial species, heif 2008, 2010). Hymenoptera develop with reproductive division of labor is generally complete metamorphosis (so-called holo- permanent and queens and workers exhibit metabolous development) and individuals strong morphological differences. Queens undergo several developmental steps from are specialized for dispersal, colony found- egg to adult (Figure 2). An important con- ing, and egg laying, and workers are struc- sequence of development with complete turally adapted for cooperative tasks such metamorphosis is that growth occurs only as colony defense, nursing, and foraging during development so that morphologi cal (Wilson 1953, 1971; Oster and Wilson 1978; traits such as overall body size and size and Wheeler 1986). Workers are morphologi- allometry of specific body parts are irre- cally constrained in their reproductive op- versibly determined by the time individuals tions because they lack functional organs reach the adult stage. For advanced eusocial for sexual reproduction (e.g., loss of func- Hymenoptera species with morphol ogical tional sperm-storing organs in honey bee reproductive castes, holometabolous devel- workers, Gotoh et al. 2013, 2016; and most opment has particularly important conse- ants, Hölldobler and Wilson 1990; Gobin quences: whether a female egg develops into et al. 2008; Gotoh et al. 2016). Although a reproductive queen or a sterile worker is workers of many advanced eusocial species irreversibly determined during larval devel- have retained the ability to produce unfer- opment (Wheeler 1986). As a result, adult tilized eggs that develop into males (Bourke queens and workers are fixed in their re- 1988; Helanterä and Sundström 2007), productive roles when they reach the adult some species have lost worker reproductive stage. Development is similarly decisive for organs altogether (Hölldobler and Wilson males, who typically produce sperm only 1990; Boleli et al. 1999; Gotoh et al. 2016). during this life stage, after which the testes The presence of individuals from several degenerate (Hölldobler and Bartz 1985; generations within the same nest adds a Boomsma et al. 2005; Stürup et al. 2013). layer of social complexity to eusocial col- Within Hymenoptera colonies, develop- onies compared to subsocial or semisocial ing individuals embody future generations taxa (Figure 1), and social interactions in of sexuals and workers with individual fit- eusocial species involve parent-offspring ness interests. At the same time, they rep- and offspring-offspring interactions on resent the combined current reproductive several levels: between queens and work- investment of all colony members and much ers, between queens and their developing of a colony’s social life revolves around offspring, between workers and develop- brood care and the attempts of adult in- ing individuals, and among developing dividuals to follow their inclusive fitness in- individ uals. Nevertheless, study of eusocial terests by influencing offspring production Hymenoptera has largely concentrated on and development. Brood becomes a source the social interactions between adult queens of conflict within colonies when adults fol- and workers, and colony offspring produc- low contrasting fitness interests (Sundström tion has been seen as a simple consequence and Boomsma 2001; Beekman and Rat- of adult actions. This is perhaps not sur- nieks 2003; Beekman et al. 2003; Helanterä prising, since queens and workers share ex- and Ratnieks 2009), making them central pensive stakes