Journal of Chemical Ecology. Vol. 22. No. 10. 1996 AGGREGATION PHEROMONES OF Drosophila immigrans, D. phalerata, AND D. subobscura KATARINA HEDLUND, 1.3. * ROBERT J. BARTELT,2 MARCEL DICKE,' and LOUISE E. M. VET' 'Department ofEll1omology. Wageningen Agricullllral University P.O. Box 8031.6700 EH Wageningen TIle Netherlands 2USDA-ARS National Cell1er for Agricultural Utilization Research. Bioactive COll.5!illle1l1S Research 1815 North University Street. Peoria. lllinois 61604 3Presell1 address: Departmell1 ofEcology. Lund Universit.v Ecology Building. 5-223 62 Lund. Sweden (Received November 13. 1995; accepted May 16. 1996) Abstract-Aggregation pheromones of Drosophila immigrallS. D. phalerata and D. subobscura were demonstrated by testing attraction of adult flies to hexane extracts of the flies in a windtunnel bioassay. Extracts of adult males of all species attracted conspecific males and females. However. D. subob­ scura flies were attracted only when the extract was applied on food. GC-MS analysis identified (Z)-Il-octadecenyl acetate (cVA) in the extracts of adult male D. immigrans and D. phalerata. Both species were attracted to synthetic cVA. Male and female D. subobscura produced 5.9-pentacosadiene. 5-pen­ tacosene. 2-methylhexacosene and 5.9-heptacosadiene. while only male D. subobscura produced (Z)-5-tricosene and minor amounts of cVA. Key Words-Aggregation pheromones. (Z)-II-octadecenyl acetate. (Z)-5­ tricosene. 5.9-pentacosadiene. windtunnel. Drosophilidae. INTRODUCTION Drosophila flies that breed in ephemeral habitats such as decomposing fruits and mushrooms often have aggregated distributions (Rosewell et al., 1990; *To whom correspondence should be addressed. 1835 0098-0331/96/ I000·1835$09.50/0 1996 Plenum Publishing CorporJ.lion 1836 HEDLUND ET AL. Jaenike and James, 1991). Aggregative behavior in Drosophila flies is mediated by pheromones that can act in concert with odors of the habitat of the flies and indicate a suitable habitat for mating and oviposition (Bartelt et a!., 1986; Scha­ ner et a!., 1987). An aggregated distribution may reduce interspecific compe­ tition and facilitate coexistence between species (Atkinson and Shorrocks, 1981; rves, 1988), but should also be important for finding mates. Drosophila aggre­ gation pheromones are produced by males and attract flies of both sexes (Bartelt and Jackson, 1984; Bartelt et aL 1985; Jaenike et a!., 1992). The pheromones are transferred to females during mating and transferred further to an oviposition site by female flies (Bartelt et a!., 1985). It has been suggested that these pheromones have functions in addition to aggregation behaviors, such as pre­ venting matings of newly mated females and interactions between males (Zawis­ towski and Richmond, 1986). Aggregation pheromones ofDrosophila are generally volatile esters, ketones or unsaturated hydrocarbons (Bartelt et a!., 1985, 1988; Moats et a!., 1987). Several species produce the same compound; for instance, species of the Dro­ sophila melanogaster group all have (Z)-II-octadecenyl acetate as an aggre­ gation pheromone (Bartelt et a!., 1985; Schaner et a!., 1987, 1989). Such nonspecific pheromones can cause interspecific aggregates of related flies in a breeding habitat (Jaenike et a!., 1992). However, flies also have a range of cuticular hydrocarbons that are assumed to take part in regulating mating behav­ iors (Scott et aL 1988; Cobb and Jallon, 1990). A mixture of these compounds may give species specificity at close range. Apart from increasing the risk of intraspecific competition, aggregation pheromones can also affect the risk of parasitization as parasitoid wasps are attracted to the pheromones of adult Drosophila flies when searching for their larval hosts (Wiskerke et a!., 1993). However, the risk of an individual Dro­ sophila larvae to be attacked by a parasitoid should decrease at increasing den­ sities of flies according to the functional response of the parasitoid (Hertlein and Thorarinsson, 1987). Recently a study on parasitoid wasps showed new and unexpected results when both generalist and specialist larval parasitoids of dro­ sophilids were attracted to aggregation pheromones of adults of some host as well as some nonhost species (Hedlund et a!., 1996). To interpret the results of the latter study, the aggregation pheromones of the Drosophila species involved need to be chemically identified. From the study of parasitoid-host interactions (Hedlund et a!., 1996) we have chosen three species, Drosophila phalerata (Meigen) and Drosophila subobscura (Collin) with a European distribution and one cosmopolitan species, Drosophila immigrans (Sturtevant) (Wheeler, 1981). This study aims at iden­ tifying aggregation pheromones of these three species not only to increase the knowledge of communication of Drosophila but also that of parasitoid-host interactions. AGGREGATION PHEROMONES 1837 METHODS AND MATERIALS Fly Cultures Drosophila immigrans was reared on a medium ofapple, water, agar, sugar and yeast. The culture originated from flies reared out of fruits collected in orchards in The Netherlands. D. subobscura was reared in the same way as D. immigrans. The culture was established from flies reared from sap fluxes of trees originating from The Netherlands. D. phalerata was reared on decaying Agaricus hortensis mushrooms. The culture originated from flies reared out of Phallus impudicus mushrooms col­ lected in The Netherlands. The flies were reared in 250 ml glass vials at 20°C on a 16: 8 (L: D) photocycle. Fly pupae were washed out of the vial and transferred to 200 ml glass vials containing a layer of water agar. Fly Extracts. Flies were separated by sex 0-6 hr after emergence from the pupae and transferred to new glass vials with a layer of water agar and honey on a strip of filter paper. To get sexually immature and mature flies the flies were collected at day one and day seven, respectively, and soaked in hexane (l00 flies ml- I hexane) at room temperature. The crude extracts were removed after 24 hr and kept at -20°C until further analysis. GC-MS Analysis. The crude extracts were analyzed with gas chromatog­ raphy (GC) and mass spectrometry (MS). Male- and female-derived extracts were compared by capillary gas chromatography (DB-l column, 15 m length, 1 f.Lm film thickness, 0.25 mm ID; temperature programmed from 50°C to 300°C at lOoC per min, with a final hold of 10 min at 300°C). GC retention indices relative to n-alkanes were calculated for sex-specific compounds and other, relatively abundant compounds. The Hewlett Packard 5890 gas chro­ matograph was equipped with splitless injector, flame ionization detector, and Hewlett Packard 3396A integrator. Electron impact mass spectra were obtained for the detected sex-specific compounds on a Hewlett Packard 5970 Mass Selec­ tive Detector with capillary GC inlet. Additional analyses were performed after removal of fatty acids by washing samples with 5 % (w: w) Na2C03 solution, since the large broad, fatty acid GC peaks could have obscured minor sex­ specific compounds. Saturated hydrocarbons were identified from mass spectra and GC retention indices. Unsaturated hydrocarbons and esters were identified by mass spectra, GC retentions, and mass spectra of dimethyldisulfide (DMDS) adducts (Carlson et aI., 1989). The DMDS derivatives allowed location of double bonds, and double bond configuration was determined by GC retention in comparison with authentic standards. Sex-specific compounds were considered as possible pher- 1838 HEDLUND ET AL. omone components. When possible, synthetic versions of the sex-specific com­ pounds were evaluated in the bioassay to verify behavioral activity. Synthetic Compounds. (Z)-ll-octadecenyl acetate, also called cis-vaccenyl acetate (cVA) was purchased from Sigma (St Louis, USA). The 5-tricosenes were prepared by the method of Sonnet (1974) from pentanal and (octade­ cyl)triphenylphosphonium bromide, and the Z and E isomers were separated by AgNOrHPLC as described by Heath and Sonnet (1980). Bioassays. All bioassays were done in a cage (width 30 cm, depth 40 cm, height 35 cm) with nylon mesh (0.1 mm) along its long sides. A fan outside the cage generated an air How of 0.1 m S-I inside the cage. Each cage was stocked with 300 flies 16 hr before the tests. The flies were two to nine days old and of equal sex ratio. They were left with honey and water until the tests started. Fly extracts or synthetic compounds were applied to a filter paper (1 cm x 6 cm strip). The solvent was allowed to evaporate and the filter paper was inserted into a small glass vial (5.5 cm x 1.5 cm diameter). A drop of water was applied on the bottom of the vial. When a food substrate was present in the vial, either a 5 mm layer of an apple-yeast mixture (15: 1) (w: w) or pieces of decaying mushrooms (Agaricus hortensis) were placed at the bottom of the vial. In each test two vials were placed inside the cage and perpendicularly to the air flow. One vial contained a hexane control and the other an extract or a synthetic compound dissolved in hexane. Tests with D. phalerata and D. subobscura lasted for 5 min while tests with D. immigrans lasted for 10 min. During this period of time the flies were allowed to enter a vial. At the end of the test the vials were capped with a cotton plug. The number and sex of the flies were determined. Subsequent tests separated by 3 to 5 min between the tests were done with the same cage. The cage was stocked with fresh flies each experimental day. All tests were performed at 20°C. Each treatment (female extract, male extract etc.) was replicated 10 to 20 times and a series of treatments was carried out each experimental day (block design) so that any variation due to daily activities of the flies could be elimi­ nated. A treatment was compared to its control with a Wilcoxon two sample test (Sokal and Rohlf, 1981) and a Kruskal-Wallis (Hollander and Wolfe, 1973) test compared the number of flies in traps of different treatments.