Structural differences in the drone olfactory system of two phylogenetically distant Apis species, A. florea and A. mellifera.

Male insects that are attracted by sex pheromones to find their female mates over long distances have specialized olfactory subsystems. Morphologically, these subsystems are characterized by a large number of receptor neurons sensitive to components of the female's pheromones and hypertrophied glomerular subunits ('macroglomeruli' or 'macroglomerular complexes') in the antennal lobes, in which the axons of the receptor neurons converge. The olfactory subsystems are adapted for an increased sensitivity to perceive minute amounts of pheromones. In Apis mellifera, drones have 18,600 olfactory poreplate sensilla per antenna, each equipped with receptor neurons sensitive to the queen's sex pheromone, and four voluminous macroglomeruli (MG1-MG4) in the antennal lobes. In contrast, we show that drones of the phylogenetically distant species, Apis florea, have only 1,200 poreplate sensilla per antenna and only two macroglomeruli in their antennal lobes. These macroglomeruli are homologous in anatomical position to the two most prominent macroglomeruli in A. mellifera, the MG1 and MG2, but they are much smaller in size. The morphological and anatomical differences described here suggest major modifications in the sex-pheromone processing subsystem of both species: (1) less pheromone sensitivity in A. florea and (2) a more complex sex-pheromone processing and thus a more complex sex-pheromone communication in A. mellifera. Research in honey bee sex-pheromone communication dates back to the 1960s, when Gary (1962) demonstrated that in Apis mellifera the queen's mandibular gland secretion and especially its main component, 9-ODA (9-keto-2(E)-decenoic acid), is highly attractive to drones on their nuptial flight. Later, cross-species attraction experiments showed that other honey bee species, Apis florea, A. cerana, and A. dorsata probably also use the queen's mandibular gland secretion as a mating attractant (Butler et al. 1967; Sanasi et al. 1971). Besides its function in mating behavior, the queen's mandibular gland secretion is the main pheromone regulating queen-worker interactions (Free 1987). In this context, several studies have demonstrated the behavioral significance of single components (Slessor et al. 1988) and differences in the composition of the secretion between Apis species (Plettner et al. 1996, 1997; Keeling et al. 2000). Regarding the interspecific differences in the queen's signal, the question arises whether this variation is reflected in the olfactory system of drones and workers of the various species.