Mandibular gland component analysis in the head extracts of Apis cerana and Apis nigrocincta

Apidologie 32 (2001) 243–252 © INRA/DIB-AGIB/EDP Sciences, 2001 243 Original article Mandibular gland component analysis in the head extracts of Apis cerana and Apis nigrocincta Christopher I. KEELINGa*, Gard W. OTISb, Soesilawati HADISOESILOc, Keith N. SLESSORa a Department of Chemistry, Simon Fraser University, 8888 University Drive, Burnaby, British Columbia, V5A 1S6, Canada b Department of Environmental Biology, University of Guelph, Guelph, Ontario, N1G 2W1, Canada c Pusat Litbang Hutan dan Konservasi Alam, Jl. Gunung Batu No. 5, Bogor, Indonesia (Received 5 September 2000; revised 26 February 2001; accepted 28 February 2001) Abstract – Head extracts of workers and mated queens of the closely related species of Apis cerana and A. nigrocincta from Sulawesi, Indonesia were quantitatively analyzed by gas chromatographymass spectrometry for several mandibular gland components. The amounts of many compounds were significantly different between species for both queens and workers. Quantities of 10 of the 16 compounds quantified in queen bees differed significantly between the two species. Of the three known mandibular gland retinue pheromone components in A. cerana queens [(E)-9-oxodec-2-enoic acid (9-ODA), (E)-9-hydroxydec-2-enoic acid (9-HDA), and methyl p-hydroxybenzoate (HOB)], the amounts of 9-HDA and HOB were significantly different between species. Quantities of 6 of the 11 compounds quantified in worker bees differed significantly between the two species. This quantitative analysis supports the hypothesis that A. cerana and A. nigrocincta are indeed separate species. Apis cerana / Apis nigrocincta / honey bee / mandibular gland / pheromone 1. INTRODUCTION It has long been known that cavity-nesting honey bees inhabit Sulawesi and the surrounding islands of Indonesia. Maa (1953) recognized a unique species of honey bee * Correspondence and reprints E-mail: ckeeling @ sfu.ca from “The Celebes” (Sulawesi), Apis nigrocincta Fabr. Smith, 1861. However, most reports about honey bees from this region have presumed that the cavity-nesting bees there are forms of Apis cerana Fabr., the common hive bee that ranges over 244 C.I. Keeling et al. most of Asia. Recently two distinct populations of cavity-nesting honey bees were verified in Sulawesi (Hadisoesilo et al., 1995). Although sympatric in at least two areas of Sulawesi, they do not appear to hybridize, suggesting they are distinct species. Morphometric analyses have indicated that the smaller, darker morph corresponds to A. cerana and the larger morph with yellowish clypeus and legs corresponds to A. nigrocincta (Hadisoesilo et al., 1995; Damus and Otis, 1997). These species also have different drone cell cappings. It is well documented that A. cerana drone cells are sealed with a wax capping underlain by a hard conical cocoon structure that contains a central pore (Ruttner, 1988; Boecking et al., 1999). In contrast, sealed A. nigrocincta drone cells have only a thin wax capping and lack the hardened cap with pore (Hadisoesilo and Otis, 1998). The timing of the mating flights also differs between the two species, with flights of A. cerana drones preceding and only slightly overlapping those of A. nigrocincta drones (Hadisoesilo and Otis, 1996), thereby affecting almost complete reproductive isolation between the populations. These differences indicate that they are two distinct species and that we might expect to find other differences. The chemical composition of the mandibular glands of both honey bee queens and workers have previously been analyzed as a method to compare different species (Plettner et al., 1997). In the queen, these glands produce pheromone messages that elicit the retinue of workers, attract drones on mating flights, and are believed to control several aspects of colony functioning (Winston and Slessor, 1998). The functions of the components in the worker mandibular glands are attributed to food preservation and larval nutrition (Winston, 1987). In A. mellifera L., both female castes produce distinctive blends of compounds in their mandibular glands with functionalized aliphatic acids predominating in both castes (Plettner et al., 1996; Plettner et al., 1998). In mated queens, compounds functionalized at the penultimate (ω-1) position of the chain [(E)-9-oxodec-2-enoic acid (9-ODA) and the two enantiomers of (E)-9-hydroxydec-2enoic acid (9-HDA)] predominate. In workers, compounds functionalized at the terminal (ω) position [10-hydroxydecanoic acid (10-HDAA) and (E)-10-hydroxydec-2-enoic acid (10-HDA)] predominate. The mandibular glands of A. mellifera queens also produce several aromatic compounds of which methyl p-hydroxybenzoate (HOB) and 4-hydroxy-3-methoxyphenylethanol (HVA) account for the retinue attraction of the queen mandibular glands when combined with 9-ODA and 9-HDA (Slessor et al., 1988). These compounds comprise the queen mandibular pheromone for retinue attraction in A. mellifera but do not account for all of the retinue attraction or chemical communication attributed to the queen (Slessor et al., 1998; Winston and Slessor, 1998; Keeling et al., 2000a). Unfortunately, the queen mandibular pheromone of A. cerana has not been so well characterized. It is known that HVA is absent from queen mandibular glands and does not increase retinue attraction of A. cerana workers when added to the other three compounds (Plettner et al., 1997). There has been no analysis of the mandibular gland composition of A. nigrocincta. In this study, we quantitatively analyzed head extracts for several mandibular gland components to reveal any differences between the mandibular gland compositions of these closely related species. The compounds quantified in this study include those previously quantified in other studies (Plettner et al., 1997; Keeling et al., 2000b) as well as some of the compounds recently identified in A. mellifera queen mandibular glands (Engels et al., 1997; Matsuyama et al., 1997; Keeling and Slessor, unpublished observations). Apart from those mentioned above, the compounds quantified have not been reported to be Analysis of A. cerana and A. nigrocincta pheromone components in any honey bee species and may only represent differences in biosynthetic pathways of the different species and castes of honey bees. Most of the aliphatic acids are biosynthetically linked to the major queen and worker produced acids (Plettner et al., 1996; Plettner et al., 1998). The biosynthetic pathways of the aromatic compounds HOB and HVA have not been reported but the other aromatic compounds quantified in this study are potentially linked biosynthetically to these pheromone components. As we learn more about the mandibular gland biochemistry of honey bees and test compounds for biological activity in honey bee species other than A. mellifera, some of these compounds may be of importance in Apis semiochemistry. 2. MATERIALS AND METHODS 2.1. Collection of specimens Bees from both wild and ma (...truncated)