(3R,6E)-nerolidol, a fertility-related volatile secreted by the queens of higher termites (Termitidae: Syntermitinae)

The queens of advanced social insects maintain their reproductive monopoly by using exocrine chemicals. The chemistry of these “queen pheromones” in termites is poorly known. We show that primary queens of four higher termites from the subfamily Syntermitinae (Embiratermes neotenicus, Silvestritermes heyeri, Labiotermes labralis, and Cyrilliotermes angulariceps) emit significant amounts of the sesquiterpene alcohol (E)-nerolidol. It is the dominant analyte in queen body washes; it is present on the surface of eggs, but absent in kings, workers, and soldiers. In E. neotenicus, it is also produced by replacement neotenic queens, in quantities correlated with their fertility. Using newly synthesised (3R,6E)-nerolidol, we demonstrate that the queens of this species produce only the (R) enantiomer. It is distributed over the surface of their abdomen, in internal tissues, and in the haemolymph, as well as in the headspace of the queens. Both (R) and (S) enantiomers are perceived by the antennae of E. neotenicus workers. The naturally occurring (R) enantiomer elicited a significantly larger antennal response, but it did not show any behavioural effect. In spite of technical difficulties encountered in long-term experiments with the studied species, (3R,6E)-nerolidol remains among eventual candidates for the role in queen fertility signalling.

The Doublesex sex determination pathway regulates reproductive division of labor in honey bees

Eusociality, the ultimate level of social organization, requires reproductive division of labor, and a sophisticated system of communication to maintain societal homeostasis. Reproductive division of labor is maintained by physiological differences between reproductive and sterile castes, typically dictated by pheromonal queen fertility signals that suppress worker reproduction. Intriguingly, reproduction and pheromonal signalling share regulatory machinery across insects.The gene Doublesex (Dsx) controls somatic sex determination and differentiation, including the development of ovaries and secondary sexual characteristics, such as pheromonal signalling. We hypothesized that this regulatory network was co-opted during eusocial evolution to regulate reproductive division of labor. Taking advantage of the breakdown in reproductive division of labor that occurs in honey bees when workers commence to lay eggs in the absence of a queen, we knocked down Dsx to observe effects on ovary development and fertility signal production. As expected, treated workers had lower levels of egg yolk protein, for which Dsx is a cis-regulatory enhancer in other insects, and greatly reduced ovary development. Also as expected, while control workers increased their levels of pheromonal fertility signals, treated workers did not, confirming the role of Dsx in regulating pheromone biosynthesis. We further found that Dsx is part of a large network enriched for regulatory proteins, which is also involved during early larval development, and upregulated in queen-destined larvae. Thus, the ancient developmental framework controlling sex specification and reproduction in solitary insects has been exapted for eusociality, forming the basis for reproductive division of labor and pheromonal signalling pathways.Significance statementComplex social insect societies rely on division of reproductive labor among their members. Reproductive individuals (‘queens’) suppress ‘worker’ reproduction using pheromonal fertility signalling. We show that an ancient regulatory network that controls specification of sex and secondary sexual characteristics in solitary insects, has been co-opted for both both pheromonal signalling and ovary inactivation in honey bees. In addition, this network is also active during caste specification that takes place during the first few days of larval life. These results show that pheromonal signalling and ovary development share a common regulatory framework, potentially explaining why fertility signalling is ‘honest.’ Furthermore, they show that higher levels of biological complexity can arise by rewiring and elaborating ancestral gene regulatory networks.