Exocrine secretions of bees X. 3,7-Dimethyldeca-2,6-dien-1,10-diol: A sex-specific compound fromNomada annulata (Hymenoptera: Anthophoridae)

Journal of Chemical Ecology, Vol. 16, No. 4, 1990

EXOCRINE SECRETIONS OF BEES X. 3,7-Dimethyldeca-2,6-dien-1,10-diol: A Sex-Specific Compound from Nomada annulata (Hymenoptera: Anthophoridae)

R.M. D U F F I E L D , l C. SIMON-JORDAN, 2'3 E.W. RIDDICK, 1'4 and J.W. W H E E L E R 2 ~Department of Zoology 2Department of Chemistry Howard University Washington, D.C. 20059

(Received January 31, 1989; accepted June 12, !989) Abstraet--3,7-Dimethyldeca-2,6-dien-l,10-diol was isolated from mate cephalic extracts of the cleptoparasitic or "cuckoo" bee, Nomada annulata. The compoundis absent in femalehead extracts. This diol, previouslyknown only from a male danaid butterfly, is a new bee natural product and is not found in the volatile exocrinesecretionsof the host bee, Andrena macra. The role of this compound in this parasite-host system, including the chemical basis of Nomada-Andrena associations, is discussed. Key Words--Cleptoparasite, Nomada annulata, Nomadinae, Andrena macra, Andrenidae, Hymenoptera, mandibular gland, male specific, diol, host parasite.

INTRODUCTION There are approximately 5000 species of cleptoparasitic bees worldwide (Duffield et al., 1984), which include over 300 described species of N o m a d a in North America north of Mexico (Hurd, 1979). Fewer than 5% of these have been associated with their hosts. K n o w n hosts include species belonging to the Andrenidae, Anthophoridae, Halictidae, and Melittidae. The reasons that spe3present address: Rohm and Haas Company, Spring House, Pennsylvania. 4present address: Departmentof Entomology, SmithsonianInstitution, Washington, D.C. 1069 0098-0331/90/0400-1069506.00/0

9 1990 Plenum Publishing Corporation

1070

DUFFIELD ET AL.

cific Nomada choose a particular host bee are poorly understood. Teng6 and Bergstrrm (1975, 1976, 1977) have investigated the chemical basis of this problem with several European cleptoparasite-host pairs of bees. In each case the chemical profile of the volatile components released from the male cephalic secretion of Nomada is similar to that of the Dufour's gland of the female host. For example, in two host-parasite pairs, Melitta haemorrhoidalis (Melittidae)Nomadaflavopicta and M. leporina-N, flavopicta, the secretions are dominated by octadecyl butanoate (Teng6 and Bergstrrm, 1976). The Melitta Dufour's glands exhibit more volatile components than the Nomada cephalic extracts. Similarly for the pairs Andrena haemorrhoa (Andrenidae)-N. bifida and A. carantonica-N, marshamella, there is a common major component, all-trans-farnesyl hexanoate. In the other two pairs, A. helvola-N, panzeri and A. clarkellaN. leucophthalma, the cephalic and Dufour's gland secretions are dominated by geranyl octanoate (Teng6 and Bergstrrm, 1977). These authors do point out that the male cephalic extracts of three additional species of Nomada do not correspond with the Dufour's gland chemistry of the host Andrena. In each species of Nomada investigated by Teng6 and Bergstrrm, male and female cephalic extracts contained completely different chemical components. Teng6 and Bergstrrm (1977) postulated that during copulation, the volatile mandibular gland secretions of the male Nomada are sprayed onto the female partner. Gas chromatographic-mass spectroscopic analyses of extracts of females verified that the females had these chemicals after copulation. Teng6 and Bergstrrm (1977) speculated that chemical correspondence between the female Nomada and the host bee may allow the female Nomada entry into the host nest without being detected or repelled by the host female. In a different approach to the study of host-parasite associations between Andrena and Nomada species, Cane (1983) tested the effects of 13 olfactory stimuli on the searching behaviors of the cleptoparasites. Artificial holes were bored in the ground to mimic the appearance of the host nests. Stimuli were placed in the holes. The searching behaviors of the Nomada were compared with those of the host Andrena. The experiments were run in Ithaca, New York, at a nesting site containing Andrena alleghaniensis, A. regularis, and the halictids Agapostemon sericeus and Halictus ligatus. The dominant Nomada at the site was N. pseudops. Some of the stimuli tested included: (1) chilled Andrena with pollen, (2) frozen Nomada, (3) peony anthers, (4) Typhus pollen, (5) Andrena Dufour's gland, and (6) chilled Agapostemon with pollen. Cane found that Nomada appeared to locate host nests primarily by visually searching for entrance holes. The searching Nomada most frequently entered artificial host nests containing Andrena with pollen. In another chemical investigation of cleptoparasitic bees, Hefetz et al. (1982) found no similarities between the host bee Calliopsis andreniformis (Andrenidae) and its cleptoparasite, Holcopasites calliopsidis (Anthophoridae:

1071

EXOCRINE SECRETIONS OF BEES

Nomadinae). The Dufour's secretions of the host bee are dominated by hydrocarbons. Males of H. calliopsidis were not analyzed. During the past 10 years, we have been studying the biology of the solitary bee, Andrena macra, and one of its nest parasites, Nomada annulata. In an earlier investigation the chemistry of the Dufour's gland secretion of Andrena macra (reported as Andrenaflexa) was reported (Fernandes et al., 1981). Here we report the chemistry of the male cephalic extracts of Nomada annulata.

METHODS

AND MATERIALS

Collection of Bees. Specimens were collected by hand net at the Marine Training Base at Quantico, Virginia. The Nomada bees were collected during April and May, 1985-1987, at several Andrena macra nesting sites. Specimens were placed individually in glass shell vials and cooled in an ice chest for transport to the laboratory. Heads of specimens were removed with forceps and extracted with methylene chloride (100 heads/vial). Separate extracts were made of male and female specimens. One extract contained female heads and thoraces. An extract was prepared of 100 male N. annulata heads in 0.5 ml methylene chloride to be used for behavioral testing. Chemical Analyses. The solvent was drawn off the head extracts, dried over NazSO4, and concentrated by air evaporation. Extracts were analyzed on a Finnigan MAT 4500B gas chromatograph-mass spectrometer utilizing a 30m x 0.25-mm capillary fused silica column containing SP-2100 (0.25 t~m) temperature programmed from 60 ~ to 300~ at 10~ Retention times and mass spectra were compared with those of authentic compounds obtained commercially or synthesized by standard methods. Proton and [13C]NMR spectra were taken on a Varian 300 MHz spectrometer or a Bruker AM-300 spectrometer in deuterochloroform using tetramethylsilane as an internal standard. Behavioral Tests. The field tests were run on May 14, 1988, between 10:00 AM and 2:00 PM. The day was sunny and the temperature rose during the test period from 68 ~ to 75~ There was little to no breeze. Both male and female N. annulata and female Andrena were active at the nesting site. This particular site contained over 1000 A. macra nests. The procedure for the first series of tests consisted of placing filter paper squares (0.5 cm 2) on an insect pin and raised 2-3 cm above the ground. The test paper was coated with 5 ~I of a standard solution containing 3,7-dimethyldeca-2,6-dien-1,10-diol (1 mg/ml methylene chloride) and allowed to stand for 1 min after the test solution was added. This allowed the methylene chloride to evaporate before running the behavioral tests. The test filter paper square was placed in the A. macra nesting site. All N. annulata passing within 12 in. of

1072

DUFFIELDET AL.

the test filter paper were counted during the 10-rain test interval. Controls consisted of filter paper squares treated with 5 t~l of methylene chloride, mounted the same manner on an insect pin and allowed to evaporate for 1 min before each test was run. Data was reported in the same manner as described above. Each test was repeated three times. In a second series of experimental tests, 25 /~1 of a methylene chloride extract of male N. annulata heads were added to the filter paper squares, allowed to evaporate for 1 min, and placed in the nest site. Data were reported in the same manner as cited above. The test was repeated three times. Controls were the same as those in the first experiment.

RESULTS

Chemical Analyses. Separate male and female Nomada head extracts were analyzed by combined gas chromatography-mass spectroscopy. Extracts from both males and females contained a series of saturated and unsaturated hydrocarbons including: tridecane (MW 184) (0.7 %), heptadecane (MW 240) (2.5 %), nonadecane (MW 268) (32%), heneicosane (MW 296) (3%), tricosane (MW 324) (7.5 %), pentacosane (MW 352) (14 %), heptacosene (MW 378) (3 %), and hentriacontene (MW 434) (11.3%). In addition, both extracts showed hexadecanoic and octadecanoic acids in trace amounts. The male extract exhibited another component (26 % of total volatiles) that eluted between heptadecane and nonadecane. It had a very small ion at m/z 198 ( < 1%) with additional ions at 180(1), 167(1), 149(1), 139(1), 134(2), 121(7), 111(5), 95(100), 93(25), 85(28), 79(10), 69(20), 68(25), 67(55), 55(33), 53(15), 43(20), and 41(35). Chemical ionization mass spectroscopy confirmed that the ion at 198 was the molecular ion. Combined male head extracts were chromatographed on silicic acid with mixtures of pentane-methylene chloride-methanol. Elution with 10% CH3OH in CH2C12 gave the 198 compound. From approximately 600 male N. annulata heads, 1 mg of the 198 compound was isolated. Its proton NMR spectrum showed: 61.62,s,3H; 1.66,s,3H; 1.7,m,2H; 2.0-2.25,m,6H; 3.1,s,2H; 3.6,t(J = 7 Hz),2H; 4.14,d(J = 7 Hz),2H; 5.15,t(J = 7 Hz),IH and 5.4,t(J = 7 Hz), 1H. The protons of 63.1 exchanged with deuterium oxide. The ~3C spectrum gave additional information: 662.38,t,(C-1); 124.49,d,(C-2); 138.68,s,(C3); 29.89,t,(C-4); 25.64,t,(C-5); 123.85,d,(C-6); 135.07,s,(C-7); 29.68,t,(C8), 36.15,t,(C-9); 59.06,t,(C-10); 15.74,q,(C-11); 15.77,q,(C-12). The proton and [13C]NMR data indicated two allylic methyl groups, a saturated CH2, three allylic methylene groups, two hydroxyl groups, a methylene

1073

EXOCRINE SECRETIONS OF BEES

11 1

l

12 s

/

9

OH

FIG. 1. Structure of 3,7-dimethyldeca-2,6-dien-l,10-diol. group adjacent to oxygen, another methylene group both allylic and adjacent to an oxygen,and two vinyl hydrogens. 3,7-Dimethyldeca-2,6-dien-l,10-diol (Figure 1) seemed a suitable candidate. The proton NMR spectrum of our compound appeared compatible with the reported spectrum of synthetic diol prepared by Katzenellenbogen and Christy (1974) and more so with that reported by Masaki et al. (1985). The earlier reports of this compound did not have detailed proton NMR data (Meinwald et al., 1969; Miles et al., 1972; Masaki et al., 1980). The diol was synthesized from all-trans-farnesol 2 (Figure 2) by the method of Hanzlik (1977) as developed by van Tamelen and Sharpless (1967) and van Tamelen et al. (1982). Attempts to selectively epoxidize the terminal double bond with m-chloroperbenzoic acid gave a mixture of epoxides at the terminal and central double bonds (Simon-Jordan, 1988). Similar attempts to prepare the diol using osmium tetroxide were unsuccessful. The epoxide 5 (obtained from bromohydrin 4) was opened to the oxoacetate 6 with periodate (Boehm and Prestwich, 1986). Reduction of the oxoacetate 6 with lithium aluminum hydride gave the diol 1. The synthetic scheme is shown below in Figure 2. Behavioral Tests. The presence of 3,7-dimethyldeca-2,6-dien-l,10-diol at

HO~d~~~~ N BS

r----

---] AczO,PY AcO

3

A c o ~ O H Br

4

HIO4~-~ AcO

6

---~K2CO3

A

c CHO

O

~

5

LiAIH4

FIG. 2. Synthetic scheme for 3,7-dimethyldeca-2,6-dien-1,10-diol.

1074

DUFFIELD ET AL.

the Andrena macra nesting site did not cause any noticable change in the behavior of the male and female Nomada annulata. Responses for the three trials using standard diol were not significantly different from the control trials using a Mann-Whitney U test (Siegel, 1956). The trials employing male head extracts were also indistinguishable from the controls. Males were not attracted to the chemical source, nor were females. As individuals approached the chemical source, there was no noticeable increase in flight speed. Nomada did not land near the source. Neither males or females appeared to change direction or type of flight as they approached the chemical source.

DISCUSSION

3,7-Dimethyldeca-2,6-dien-1,10-diol (Figure 1) was first isolated from the danaid butterflies Danaus gilippus berenice and a subspecies D. g. strigosus along with an alkaloidal ketone (2,3-dihydro-7-methyl-lH-pyrrolizin-l-one) (Meinwald et al., 1969). While the ketone was shown to be a male pheromone (Schneider and Seibt, 1969), the only function attributed to the diol was that it imparted "the necessary stickiness to the carrier dust." It was suggested that the diol might have another function, mimicking the action of juvenile hormone (Pliske and Eisner, 1969). It is striking that diol 1 is present in males only of two such different insects as a butterfly and a Nomada bee. Its function remains to be determined in both cases. Our results showed several important differences from those of Teng6 and Bergstr6m (1976, 1977). Nomada annulata release only a few micrograms of secretion per individual. In contrast, Teng6 and Bergstr6m reported that both sexes produce relatively large amounts of cephalic secretion (on the order of 1 rag/individual). Secondly, the major compound isolated from male Nomada annulata is not found in the Dufour's gland secretions of its host, Andrena macra (reported as A. flexa by Fernandes et al., 1981). The Dufour's glands of A. macra are dominated by farnesyl hexanoate and several other esters of farnesol. Of the seven Nomada-Andrena associations investigated by Teng6 and Bergstr6m, the male cephalic compounds of three do not match the major compound in the Dufour's gland of the host Andrena (Teng6 and Bergstr6m, 1977). The studies reported here add to the list of those showing differences. Obviously, there must be other factors involved in these Nomada-Andrena parasite-host associations. We raise the question as to whether chemical congruency between male nomadine cleptoparasites and the Dufour's gland secretions of their respective hosts is the norm or the exception. Certainly some Nomada-Melitta associations (Teng6 and Bergstr6m, 1976) show this chemical congruency, but such is not the case for Holcopasites calliopsidis and Calliopsis andreniformis (Hefetz et al., 1982). Unfortunately, we have so few chemical and biological

EXOCRINE SECRETIONS OF BEES

1075

data on cleptoparasite-host associations that it is not yet possible to answer our question satisfactorily. Acknowledgments--This investigation has been supported in part by funds made available from the Howard University Faculty Research Support Grant Program and by grant RR-08016 from the Minority Biomedical Research Support (MBRS) Program, Division of Research Resources, National Institutes of Health. The Finnigan 4500 GC-MS was purchased with funds provided by MBRS supplemental Instrumentation Grant RR-08016-14-SI. The NMR spectra were taken by Dr. R.J. Highet of the NHLBI, National Institutes of Health and at Georgetown University. We thank Dr. Wallace E. LaBerge, Faunistic Surveys and Insect Identifications, Illinois Natural History Survey, Urbana, Illinois, for the identifications. We thank George C. Eickwort, Department of Entomology, Comell University, and Donna Maglott for critically reading the manuscript and for their helpful suggestions. Manuscript preparation was completed while the senior author was a visiting professor in the Department of Entomology at Cornell University, Ithaca, New York.

REFERENCES BOEHM, M.F., and PRESTWICIq, G.D. 1986. Synthesis of radioiodinated juvenile hormone analogues. J. Org. Chem. 51:5447-5450. CANE,J.H. 1983. Olfactory evaluation of Andrena host nest suitability by kleptoparasitic Nomada bees (Hymenoptera: Apoidea). Anim. Behav. 31:138-144. DUFFIELD,R.M., WHEELER,J.W., and EICKWORT, G.C. 1984. Sociochemicals of bees, pp. 387428, in W.J. Bell and R.T. Card6 (eds.). Chemical Ecology of Insects. Chapman and Hall, New York. FERNANDES, A., DUFF1ELD, R.M., WHEELER, J.W., and LABERGE, W.E. 1981. Chemistry of the Dufour's gland secretions of North American andrenid bees (Hymenoptera: Andrenidae). J. Chem. Ecol. 7:453-463. HANZLIK, R.P. 1977. Selective epoxidation of terminal double bonds. 10,11-epoxyfamesyl acetate, pp. 112-117, in G.H. Bfichi (ed.). Organic Syntheses, Vol. 56, John Wiley & Sons, New York. HEFETZ, A., EICKWORT, G.C., BEt;M, M.S., CANE, J., and BOHART, G.E. 1982. A comparative study of the exocrine products of cleptoparasitic bees (Holcopasites) and their hosts (Calliopsis) (Hymenoptera: Anthophoridae, Andrenidae). J. Chem. Ecol. 8:1389-1397. HURD, P.D., JR. 1979. The Apoidea, pp. 1741-2209, in K.V. Krombein, P.D. Hurd, Jr., D.R. Smith, and B.D. Burks (eds.). Catalog of Hymenoptera in America North of Mexico, Vol. II. Smithsonian Institution Press. Washington, D.C. KATZENELLENBOGEN,J.A., and CHRISTY, K.J. 1974. Stereoseleetivity of the rearrangement of allyl siloxyvinyl ethers. A highly stereoselective synthesis of a diol found in the pheromonal secretion of the queen butterfly. J. Org. Chem. 39:3315-3318. MASAKI, Y., SAKtJMA, K., and KAJf, K. 1980. A new stereoselective synthesis of a terpenoid diol component of the pheromonal secretion of the queen butterfly. Chem. Lett. 1061-1062. MASAKI, Y., SAKUMA,K., and KAJI, K. 1985. Regio- and stereoselective terminal allylic earboxymethylation of gem-dimethyl olefins. Synthesis of biologically important linear degraded terpenoids. Chem. Pharm. Bull. 33:1930-1940. MEINWALD, J., MEINWALD, Y.C., and MAZZOCCHI, P.H. 1969. Sex pheromone of the queen butterfly: Chemistry. Science 164:1174-1175. MILES, D.H., LOEW, P., JOHNSON, W.S., KLUGE, A.F., and MEINWALD, J. 1972. A short stereoselective synthesis of some terpenes from the pheromonal secretion of the queen and monarch butterflies. Tetrahedron Lett. 3019-3022.

1076

DUFFIELD ET AL.

PLISKE, T., and EISNER, T. 1969. Sex pheromone of the queen butterfly: Biology. Science 164:11701172. SCHNEIDER, D., and SEIBT, U. 1969. Sex pheromone of the Queen butterfly: Electroantennogram responses. Science 164:1173-1174. SIEGEL, S. 1956. Nonparametric Statistics for the Behavioral Sciences. McGraw-Hill, New York, 312 pp. SIMON-JORDAN, C. 1988. Synthesis of a sex specific compound identified in male Nomada bees. MS thesis. Howard University, Washington, D.C. TENG0, J., and BERGSTRGM,G. 1975. All-trans-famesyl hexanoate and geranyl octanoate in the Dufour gland secretion of Andrena (Hymenoptera: Apidae). J. Chem. Ecol. 1:253-268. TENGO, J., and BERGSTROM,G. 1976. Odor correspondence between Melitta females and males of their nest parasite Nomadaflavopicta K. (Hymenoptera: Apoidea). J. Chem. Ecol. 2:57-65. TEN~6, J. and BERGSTRGM,G. 1977. Cleptoparasitism and odor mimetism in bees: Do Nomada males imitate the odor of Andrena females? Science 196:1117-1119. VAN TAMELEN,E.E., and SHARPLESS,K.B. 1967. Positional selectivity during controlled oxidation of polyolefins. Tetrahedron Lett. 2655-2659. VAN TAMELEN, E.E., STORm, A., HESSLER, E.J., and SCHWARTZ,M.A. 1982. Cyclization studies with (+) 10,11-oxidofarnesyl acetate, methyl farnesate and methyl (+)-10,11-oxidofamesate. Bioorg. Chem. 11:133-170.