t J • ·l :f. ,1 j Journal of Chemical Ecology, Vol. 20, No. 10, 1994 PHEROMONE CHlRALITY OF AFRICAN PALM WEEVIL, Rhynchophorus phoenicis (F.) AND PALMETTO WEEVIL, Rhynchophorus cruentatus (F.) (COLEOPTERA: CURCULIONIDAE) ALICE L. PEREZ,I GERHARD GRIES,2 REGINE GRIES,2 ROBIN M. GIBLIN-DAVIS,3 and A. CAMERON OEHLSCHLAGERI.* 'Department of Chemistry 2Cenfre for Pest Management, Department of Biological Sciences Simon Fraser University Burnaby, British Columbia, Canada V5A lS6 J1nstitute of Food and Agricultural Sciences Fort Lauderdale Research and Education Centre University of Florida, Fort Lauderdale, Florida 3314 (Received January 10, 1994; accepted June 2, 1994) Abstract- There are four stereoisomers of both 3-methyl-octan-4-01, the aggregation pheromone of the African palm weevil , Rhynchophorus phoenicis (F.) and S-methyl-octan-a-ol. the aggregation pheromone of the palmetto wee- vil, Rhynchophorus cruentatus (F.). Synthetic stereoisomers of 3-methyl-octan- 4-01 and 5-methyl-octan-4-01 were baseline-separated on a Cyclodex-B fused silica column. Use of this column in gas chrornatographic-electroantenno- graphic detection (GC-EAD) and GC-mass spectrometric (GC-MS) analyses revealed that only one stereoisorner, (3SAS)-3-methyl-octan-4-01 and (4S,5S)- S-methyl-octan-a-cl, is produced by male R. phoenicis and male R. cruen- tatus, respectively, and elicits good antennal responses by conspecific male and female weevils. In field trapping experiments, with R. phoenicis in Cote d'Ivoire and R. cruentatus in Florida, (3SAS)-3-methyl-octan-4-01 and (4S,5S)-5-methyl-octan-4-01 strongly enhanced attraction of fresh palm tis- sue, whereas other stereoisomers were behaviorally benigno Stereoisomeric 3-methyl-octan-4-01 and S-methyl-octan-4-01 may be utilized to monitor andl or manage populations of these two palm weevils. Key Words-Coleoptera, Curculionidae, Rhynchophorus phoenicis, Rhyn- chophorus cruentatus, aggregation pherornone, pheromone chirality, (3S,4S)- 3-methy l-octan-4-ol, (3R ,4R)- 3-methyl-octan-4-01, (3S AR)- 3-methy l-octan-4- *To whom correspondence should be addressed. 2653 0098-0331194/1000-2653$07.0010 © 1994 Plcnum Publishing Corporation 2654 PEREZ ET AL. 01, (3R,4S)-3-methyl-octan-4-0I, (4S,5S)-5-methyl-octan-4-0I, (4R ,5R)-5- methyl-octan-4-ol, (4S,5R)-5-methyl-octan-4-0I, (4R,5S)-5-methyl-octan-4-01. INTROOUCTlON Palm weevils in the Rhynchophorinae produce methyl-branched, secondary alcohols as aggregation pheromones: (2E)-6-methyl-hepten-4-01 (rhynchopho- rol) [American palm weevil, Rhynchophorus palmarum (L.) (Rochat et al., 1991)]; 3-methyl-octan-4-01 (phoenicol) [African palm weevil, R. phoenicis (F.) (Gries et al., 1993, 1994; Rochat et al., 1993]; 4-methyl-nonan-5-01 (ferrugi- neol) [Asian palm weevils, R. ferrugineus (Oliv.), R. vulneratus (Panz) Hallett et al., 1993; Rochat et al., 1993) and R. bilineatus (Montr.) (Oehlschlager et al., 1994)], and 5-methyl-octan-4-01 (cruentol) [Palmetto weevil, R. cruentatus (F.) (Weissling et al., 1994)]. Racemic (rhynchophorol) and stereoisomeric mix- tures (phoenicol, ferrugineol, cruentol) of synthetic aggregation pheromones in combination with host material strongly attracted weevils in field experiments. Stereoselective production of and response to pheromone has been demonstrated in R. palmarum (Oehlschlager et al., 1992) and recently in the other Rhyn- chorphorus palm weevils (Perez et al., 1993). Male R. palmarum stereoselec- tively produce and both sexes respond to (S)-rhynchophorol, while the antipode is behaviorally benign (Oehlschlager et al., 1992). Male R. phoenicis produce one stereoisomer of phoenicol but electrophysiological and behavioral activity have not been investigated (Mori et al., 1993). In this study we report that R. phoenicis and R. cruentatus stereoselectively produce and respond to only one of the four possible stereoisomers of 3-methyl-octan-4-01 and 5-methyl-octan- 4-01, respectively. METHOOS ANO MA TERIALS Laboratory Analysis Male and female R. phoenicis were collected in oil palm plantations 40-50 km northeast of Abidjan, Cote d'Ivoire. Male and female R. cruenta tus were collected in a 300-ha pasture interspersed with Sabal palmetto (Walter) and saw palmetto, Serrenoa repens (Bartr.), 12 km south of La Belle, Florida. Male-produced phoenicol and cruentol were captured (Gries et al., 1993; Weiss- ling et al., 1994) and subjected to both gas chromatographic-electroantenno- graphic detection (GC-EAD) (Am et al., 1975) (Hewlett Packard 5890A) and GC-mass spectrometry (GC-MS) (Hewlett Packard 5985 B) on a fused silica, Cyclodex-B-coated column (30 m X 0.25 mm ID, J&W Scientific), which separates all four stereoisomers of phoenicol and cruentol. For GC-EAD record- ings, a weevil antenna was removed from the rostrum and suspended between • • ·, PHEROMONE CHlRALITY 2655 two glass capillary electrodes with the antennal base being inserted into one and the olfactory club impaled by the other electrode. Chemical ionization (CI, isobutane) GC-MS analysis was conducted in both full-scan and selected-ion monitoring mode (SIM). A full-scan mass spectrum of synthetic phoenicol or cruentol was obtained to select diagnostic ions. For GC-MSCI-SIM, synthetic phoenicol and cruentol, hexane, and concentrated weevil-produced pheromone were injected in split mode and analyzed by scanning for diagnostic ions. Instruments and General Procedures Nuclear magnetic resonance (NMR) spectroscopy was conducted on a Bru- ker AMX-400 spectrometer at 400.13 and 100.62 MHz for 'H and l3CNMR spectra, respectively. 'H che mical shifts are reported in parts per million (ppm, o) and relative to TMS (0.00 ppm). l3C spectra are referred to CDCI3 (77.0 ppm). Gas chromatographic analyses were performed on Hewlett-Packard 5880A and 5890 instruments equipped with a flame ionization detector and a fused silica, DB-1 coated column (15 m X 0.25 mm ID; 0.25 ,um film) (J&W Sci- entific). Elemental analyses were performed using a Carbo Erba model-1106 Elemental Analyzer. Diethyl ether (Et20), dichloromethane (CH2Cl2), and pen- tane were freshly distilled from sodium-benzophenone-ketyl, CaH, and P205, respectively. Chemicals obtained from commercial sources were used without further purification unless otherwise indicated. All moisture and air sensitive reactions were conducted under argon. Column chromatography refers to flash chromatography using Silica Gel 60 (230-400 mesh E Merck, Darmstadt) (Still et al., 1978). Thin-layer chromatography (TLC) was conducted on aluminum- backed plates precoated with Merck Silica Gel 60F-254 as the adsorbent, and visualized by treatment with an acidic solution of 1% Ce(S04h and l.5% molybdic acid followed by gentle heating. Synthesis of Phoenicol Stereoisomers (3R,4R)-, (3S,4S)-, (3R,4S)-, and (3S,4R)-3-methyl-octan-4-01 [(R,R)-, (S,S)-, (R,S)-, and (S,R)-phoenicol] were synthesized according to a method modified from Nakagawa and Mori (1984), which involved: (1) asymmetric epoxidation of (2Z-) or (2E)-2-penten-1-01 (Gao et al., 1987; Hill et al., 1985); (2) regioselective epoxide opening with trimethylaluminun (Pfaltz and Matten- berger, 1982; Suzuki et al., 1982; Takano et al., 1989; Vaccaro et al., 1992); (3) selective monotosylation; and (4) alkylation reaction using an organomag- nesium cuprate reagent. Synthesis of (3S,4S)-3-methyl-octan-4-01 exemplifies the synthetic procedure (Figure 1): (2S,3R)-2,3-Epoxy-pentan-I-ol (2a). Titanium(IV) isopropoxide (1l.4 mI, 10.87 g, 38 mmol) in 250 mI of dry CH2Cl2 was mixed under argon with 1 g of 4A powdered, activated molecular sieves. After cooling to -78 "C, diethyl 2656 PEREZ ET AL. 1 OH . v='V .. ~~~;;6Ur4. ~ ~'~~;;6U4 (+)-DET 47" 4S. (-)-DET 4)1. MS ,. Y. 4)1. MS 28 AOH AOH 2b 1. A'Mo31S5% 82% 11. AIMo3 2. NaF 88% ea 85% 99 2. NaF OH OH 3a .A.IYH 'OH 3b. Se OH 1LiAIH~~83%) Sd OH ~ -- HO/VV -- A(VV ~ 6 3R,45-3·methyt-octan-4-o1 (96% ee) 3S,4R-3-methyl-octan-4-o1 (95% ee) FIG. 2. Scheme for the synthesis of (3S,4R)- and (3R,4S)-3-methyl-octan-4-ol. o PHEROMONE CHlRALITY 2659 ., ~, (eDeI3): 00.86 (3 H, t, J = 8 Hz), 0.91 (3 H, d, J = 8 Hz), 1.18 (1 H, m), 1.40 (2 H, m), 3.12 (2 H, brs, D20 exchangeable), 3.40 (2 H, m), 3.55 (1 H, m); eI-MS miz (relative intensity): 119 (M+ + 1, 45). 8b: 1.28 g, 67% yield, 95% ee. (2R,3R)-3-Methyl-I-tosyloxy-pentan-2-01 (9a). 1.52 g, 60% yield, IH (eDel3): 00.84 (3 H, d, J = 8 Hz), 0.86 (3 H, t, J = 8 Hz), 1.22 (1 H, m), 1.45 (2 H, m), 2.0 (1 H, brs, D20 exchangeable), 3.64 (1 H, dt, J = 8,4 Hz), 3.94 (1 H, dd, J = 8, 2.5 Hz), 4.01 (-1H, dd, J = 8, 1.5 Hz); 7.30 (2 H, d, J = 8 Hz); 7.75 (2 H, d, J = 8 Hz); eI-MS miz (relative intensity): 273 (M++1, 100). 9b: 1.71 g, 62% yield. (3S,4R)-3-Methyl-octan-4-01 (5d). 0.68 g, 85% yield, 96% ee. IH (eDeI3): 00.81 (3H, d, J = 8.1 Hz), 0.84-1.0 (6H, m), 1.20 (2H, m), 1.22-1.60 (7H, m), 1.70 (lH, brs, D20 exchangeable), 3.45 (lH, m); 13e (eDel3) o 75.71, 40.50,33.07,28.25,24.55,22.76, 14.70, 13.11, 11.82 ppm. Anal. calcd. for e9H200: e, 74.92; H, 13.98, found: e, 74.76; H, 14.07. 5d: 0.76 g, 84% yield, 95% ee. Anal. calcd. for e9H200: e, 74.92; H, 13.98, found: e, 75.06; H, 14.01, Figure 2. Synthesis of Cruentol Stereoisomers (4R,5R)-, (4S,5S)-, (4R,5S)-, and (4S,5R)-5-methyl-octan-4-01 (R,R-, S,S-, R,S-, and S,R-cruentol) were synthesized according to a method modified from Nakagawa and Mori (1984), followed by Mitsunobu reaction (Mitsunobu, 1981) of the corresponding anti-isomers (Figure 3). (2S,3S)-2,3-Epoxy-hexan-I-ol (1la). This was prepared according to the procedure employed for 2a. Thus, 19 of 4A powdered, activated molecular sieves and 8.4 ml (8.79 g, 31 mmol) oftitanium(IV) isopropoxide in 250 ml of dry eH2Cl2 were cooled to -78°e in an acetone-Dry Ice bath. Then via syringe was added 6.3 ml (5.23 g, 25 mmol) of diethyl (2R,3R)-tartrate [L-( + )-DET] and 6.1 ml (5.2 g, 52 mmol) of (2E)-2-hexen-1-01 10 (Aldrich Chernical Co.), Stirring of the mixture was followed by dropwise addition of 18 ml (0.11 mol) of 6.2 M anhydrous tert-butyl hydroperoxide in eH2Cl2 (precooled to -20°C). The reaction was allowed to warm to - 200e with stirring and was stirred at this temperature for 3 hr while it was monitored by TLe (4: 6, hexane-ether; R¡ = 0.19). Ferrous sulfate/tartaric acid work-up (Gao et al., 1987) followed by column chromatography gave 11a (4.82 g, 80% yield, 95% ee) as a colorless liquid, which crystallized as white needles at -20°e. [l1a: IH (eDeI3): 00.96 (3 H, t, J = 7.6 Hz), 1.48 (2 H, m), 1.54 (2 H, m), 1.80 (1 H, brs, D20 exchangeable), 2.92 (2 H, m), 3.60 (1 H, dd, J = 5, 10 Hz), 3.90 (lH, dd, J = 10, 2.5 Hz); 13e (eDel3) o 61.76, 58.34, 55.81, 33.57, 19.23, 13.84 ppm.] 11b (4.94 g, 82% yield, 95% ee) was prepared following the same procedure but employing diethyl (2S,3S)-tartrate [D-( - )-DET]. 2660 PEREZ ET AL. 10 OH VV'V 80% / ~2'1095%ee 95%ee 118 OH 11b OH ~ ~ . 78% ¡ ¡ 70%95%ee 98%ee OH OH ~OH 12b = OHvyv 76'10¡ ¡ 78% OH OH ~OTS 13b ~ OTsvyv 78% ¡ ¡ 85% 98%ee 98%ee ~ 14b QH ~ 1 1. PhCOOH/Ph3P/DIAO 1 ••. 2. 15% KOH/MeOH •. OH ~ 98%ee 96%ee FIG. 3. Scheme for the synthesis of aH four stereoisomers of 5-methyl-octan-4-ol. (2R,3R)-3-Methyl-hexane-l,3-diol (12a). This was prepared according to the procedure employed for 3a. Thus, to 4.80 g (0.04 mol) of Ha in 250 ml of dry pentane cooled to -50°C was added dropwise 11.9 ml (8.64, 0.11 mol) of neat AIMe3. This was followed by 16 ml of 2.49 M n-butyllithium (0.04 mol). After stirring 20 min, the cooling bath was removed and the fiask allowed to warm to room temperature. The reaction was monitored by GC and TLC (2: 8, hexane-ethyl acetate, Rf = 0.33) and was complete after 30 mino After quenching with 3 M HCI at O°C and separation of the two phases, the aqueous layer was extracted with ether (3 X 40 ml), dried over anhyd. MgS04, and concentrated in vacuo. Purification by column chromatography afforded 12a (4.26 g, 78 % yield, 95 % ee) as a colorless liquid, which crystallized as a white sol id at -20°e. [12a: lH (CDCI3): 00.88 (3 H, t, J = 10 Hz), 0.90 (3 H, d, J = 10 Hz), 1.14 (1 H, m), 1.25 (1 H, m), 1.46 (1 H, m), 1.60 (1 H, m), 2.10 (1 H, brs, D20 exchangeable), 2.24 (1 H, brs, D20 exchangeable), 3.50 (2 H, m), 3.70 (1 H, m); l3C (CDCI3) o 76.28, 64.66, 35.94, 34.68, 20.06, 15.14,14.26); CI-MS miz (relative intensity): 119 (M++l, 40.] 12b: 3.91 g, 70% yield, 98% ee. (2R,3R)-3-Methyl-l-tosyloxy-hexan-2-ol (13a). After purification by col- PHEROMONE CHIRALITY 2661 umn chromatography (6: 4, pentane-ether, Rf = 0.45), 13a (6.54 g, 76% yield) was obtained as a pale yellow oil, IH (CDCl3): {j 0.86 (6 H, m), l.18 (2H, m), l.40 (2H, m), 1.60 (lH, m), l.90 (lH, brs D20 exchangeable), 2.48 (3H, s), 3.64 (lH, m), 3.98 (lH, dd, J = 12, 8 Hz), 4.10 (lH, dd, J = 12, 4 Hz), 7.38 (2H, d, J = 8 Hz), 7.80 (2H, d, J = 8 Hz); 13C(CDCl3) {j 144.99, 132.5, 129.92,127.93,73.43,72.66,35.54,34.15, 2l.26, 19.96, 15.10, 14.16 ppm. 13b: 6.09 g, 78 % yield. (4S,5R)-5-Methyl-octan-4-ol (l4a). This was prepared by the route used for Sa except that ethyl magnesium bromide (Aldrich Chemical Co.) (3 M solution in Et20) was used. After purification by column chromatography (9: 1, pentane-ether, R¡ = 0.08), I4a (2.74 g, 78% yield, 98% ee), was obtained as a colorless liquid, which crystallized as a white sol id at -20°C, IH (CDCI3): s 0.90 (3 H, t, J = 8 Hz), 0.92 (3 H, d, J = 8 Hz), 0.94 (3 H, t, J = 8 Hz), l.10 (l H, m), 1.24 (1 H, m), 1.32 (1 H, m), 1.40 (4 H, m), 1.50 (1 H, m), 1.70 (1 H, brs, D20 exchangeable), 3.48 (1 H, m); 13C(CDCI3) {j 75.82,38.61, 35.64, 34.17, 20.42, 19.28, 15.24, 14.34, 14.13 ppm; CI-MS miz (relative intensity): 127 (100) (M+ - H20): Anal. calcd. for C9H200: C, 74.92; H, 13.98, found: C, 75.16; H, 14.11. I4b: 2.73 g, 85% yield, 98% ee; Anal. calcd. for C9H200: C, 74.92; H, 13.98, found: C, 73.87; H, 14.08. [(4R,5R)-5-Methyl-4-octyl)]benzoate (ISa). Triphenylphosphine (9.97 g, 38 mmol) and I4a (2.74 g, 19 mmol) in 30 mi of dry benzene were added via cannula to diisopropyl azodicarboxyate (7.68 g, 7.5 rnl, 38 mmol) (Aldrich Chemical Co.) and benzoic acid (4.64 g, 38 mmol) in 45 mi dry benzene. After stirring ovemight at room temperature, pentane was added, at which point a white precipitate formed. The reaction mixture was filtered through a Florisil pad and concentrated under pressure. Purification by column chromatography (9: 1, pentane-ether, Rf = 0.61) afforded ISa (2.35 g, 50% yield) as a pale yellow liquid. Unreacted alcohol was recovered. [ISa: IH (CDCI3): {j 0.88 (3 H, t, J = 9 Hz), 0.98 (3 H, t, J = 9 Hz), 1.00 (3 H, d, J = 9 Hz), 1.10 (1 H, m), 1.38, (5 H, m), 1.58 (l H, m), 1.70 (l H, m), 1.80 (l H, m), 5.10 (l H, m), 7.40 (2 H, dd, J = 9, 2 Hz), 7.54 (l H, ddd, J = 9,2 Hz), 8.04 (2 H, dd, J = 9, 2 Hz); 13C(CDCl3) {j 166.37,132.62,130.98, 129.56, 128.29, 77.74,36.27, 35.41, 33.77, 20.34, 19.07, 14.48, 14.22, 14.01 ppm; CI-MS miz (relative intensity): 127 (M+ -C6H5-CO, 100).] ISb: 2.70 g, 57.3% yield. (4R,5R)-5-Methyl-octan-4-ol (I4c). To a 15% KOH solution of methanol was added ISa (1.30 g, 52 mmol). After stirring the mixture ovemight, it was quenched with water and extracted with Et20 (3 X 30 mi). The ether extracts were washed with dilute HCl and saturated NaCI and then dried over anhyd. MgS04' Concentration in vacuo and column chromatography (9: 1, pentane- ether, Rf = 0.13) gave I4c (0.71 g, 95% yield, 98% ee) as a colorless liquid. [I4c: IH (CDCl3): s 0.89 (3 H, d, J = 8 Hz), 0.92 (3 H, t, J = 8 Hz), 0.95 (3 H, t, J = 8 Hz), 1.12 (1 H, m), 1.24 (1 H, m), 1.33 (l H, m), 1.39 (5 H, 2662 PEREZ ET AL. m), 1.48 (1 H, m), 3.40 (1 H, m); 13C (CDC13) o 75.04,38.7036.73, 35.73, 20.51, 19.51, 15.33, 14.38, 13.65 ppm; Anal. calcd. for C9H200: C, 74.92; H, 13.98, found: C, 74.74; H, 13.84.] 14d: 0.75 g, 89% yield, 96% ee; Anal. calcd. for C9H200: C, 74.92; H, 13.98; found: C, 74.69; H, 13.81. The enantiomeric excesses of 5a (93%), 5b (87%), 5e (96%), 5d (95%), 14a (98%), 14b (98%), 14e (98%), and 14d (96%) and their correspond- ing intermedia tes except epoxides were determined by GC analyses on the Cyclodex-B column and by formation of the O-acetyllactyl methyl esters (Sles- sor et al., 1985). Enantiomeric excesses of epoxides 2a (85%), 2b (88%), 7a (96%), 7b (96%), 11a (95%), and 11b (95%) were determined by GC analysis of corresponding O-acetyllactyl methyl esters (Slessor et al., 1985) on a DB-23 column. Racemic 3-methyl-octan-4-o1 and 5-methyl-octan-4-01 were synthesized as previously described (Gries et al., 1993; Weissling et al., 1994). Field Experiments African Palm Weevil. A six-replicate, five-treatment field experiment in a lO-year-old oil palm stand (La Me Research Station, Cote d'Ivoire) tested attrac- tion of palm tissue (250 g) alone or in combination with either stereoisomeric, (S,S)-, (R,R)-, or (S,S)- plus (R,R)-phoenicol. Traps (Oehlschlager et al., 1993) were attached at breast height to oil palms in randomized blocks with traps at 27-m intervals and blocks 81 m aparto (S,S)- and (R,R)-phoenicol were released at 0.5 mg/day (at 25°C) from a 1.5-mI polyethylene centrifuge tube with two 2-mm holes below the topo Racemic phoenicol was dispensed at 2 mg/day (at 25°C) from four 1.5-mI polyethylene centrifuge tubes. Fresh palm tissue in each trap was treated with insecticidal (biodegradable) Evisect "S" (0.3% thiocy- clamhydrogenoxalate in water) to retain captured weevils (Gries et al., 1993, 1994). Trap catch data were subjected to analysis of variance followed by Scheffé test for comparisons of means (Zar, 1984). Palmetto Weevil. A 12-replicate, four-treatment experiment in the same location as for weevil collection tested attraction of Sabal palmetto tissue (1.5 kg) alone or in combination with either stereoisorneric, (S,S)- or (R,R)-cruentol. Traps (Weissling et al., 1994) were secured on the ground in randomized com- plete blocks with traps at 20-m intervals and blocks at least 50 m aparto Unlike Weissling's trap, a tapered, inverted white plastic container (4.9 liter) with a screened lid was suspended in the mouth of the bucket by a capped PVC pipe (1.3 cm diameter) from which pheromone release devices were hung. (S,S)- or (R,R)-cruentol were released at 0.06 mg/day (at 25°C) from one and stereoiso- meric cruentol from four bottom-sealed 1-1-'1microcapillary tubes (Drummond Scientific Co., Broomall, Pennsylvania) placed in bottom-sealed microhemato- crit tubes (length 75 mm, ID 1.1-1.2 mm; Fisher Scientific, Pittsburgh, Penn- PHEROMONE CHlRALITY 2663 sylvania). Hematocrit tubes were placed into polypropylene centrifuge tubes (Coming Glass Works, Coming, New York, with 6-mm holes drilled 1.8 cm from the top). Trap catch data were subjected to square root (x + 0.5) trans- formation and ANOVA (SAS Institute, 1990) followed by Waller-Duncan k-ratio t test to test differences between means (P ~ 0.05). RESULTS AND DISCUSSION Many coleopteran pheromones are optically active (Seybold, 1993; Leal ~ and Mochizuki, 1993; Bestmann and Vostrowsky, 1988; Evershed, 1988; Bor- den, 1985, and literature cited therein). Enantioselective production of and response to pheromones contribute to species specificity of semiochemical com- munication (Borden et al., 1976, 1980; Brand et al., 1979; Birch et al., 1980; Payne et al., 1982; OehIschlager et al., 1987; Pierce et al., 1987; Birch, 1984; Byers, 1989). The presence of nonnatural (non-beet1e-produced) enantiomers in synthetic pheromones has been demonstrated to interfere with optimal attraction. For instance, the male-produced aggregation pheromone in the southem pine beetle, Dendroctonus frontalis Zimm., (lR,5S, 7S)-( + )-endo-brevicomin, endo- 7-ethyl-5-methyl-6,8-dioxa-[3.2.1]octane, markedly enhances the response by both sexes to fema1e-produced fronta1in (l,5-dimethyl-6,8-dioxa-[3.2.1]octane), whereas the presence of the antipode in racemic endo-brevicomin interferes with optimal attraction (Vité et al., 1985). In the Japanese beetle, Popillia japonica (N .), female-produced Japonilure, (R,Z)-( - )-5-(l-decyl)oxacyclopentan-2-one, strongly attracts males, whereas the antipode inhibits responses (Tumlinson et al., 1977). In the scarab beetle, Anomala cuprea only the (R,Z)-5-( - )-(oct-l- enyl)oxacyclopentan-2-one attracts conspecifics, while the presence of the non- natural enantiomer reduced attraction (Leal and Mochizuki, 1993). Determina- tion of insect-produced pheromone enantiomer(s) and/or stereoisomers is required to fully elucidate the chemical communication system for a target insect and to implement efficient pheromone-based monitoring and/or management. In this study, we confirmed the chirality of weevil-produced phoenicol (Gries et al., 1993; Perez et al., 1993; Mori et al., 1993), determined chirality of weevil- produced cruentol (Weissling et al., 1994), and field tested weevil attraction to natural and nonnatural stereoisomers. Of several methods available to prepare the target chiral o-methyl second- ary alcohols, the Sharpless asymmetric epoxidation combined with diastereo- selective ring opening was the most appealing (Gao et al., 1987; Hill et al., 1985; for use of trimethylaluminum and organocuprates: Pfaltz and Mattenber- ger, 1982; Suzuki et al., 1982; Takano et al., 1989; Vaccaro et al., 1992; Miyashita et al., 1993). This strategy allowed the use of inexpensive reagents and the synthesis of all four stereoisomers from the same starting material. 2664 PEREZ ET AL. Sharpless asymmetric epoxidation has been previously used for the synthe- sis of the stereoisomers of the elm bark beetle pheromone, 4-methyl-heptan- 3-01 (Nakawaga and Mori, 1984). In contrast to this previous synthesis, prep- aration of phoenicol and cruentol used 0.5 equivalents of catalyst in the presence of a 4A molecular sieve coupled with addition of the oxidizing agent at -78 "C to increase the optical purity of the initial epoxide product. Epoxidations were maintained at -20°C until 97-98% conversion was obtained for 2a and 2b (two days) as well as for 7a, 7b, 11a, and 11b (3-4 hr). Although reactions are not reported for diisopropropyl tartrate, higher enantioselectivities were achieved with diethyl tartrate. Nonaqueous work-up (Gao et al., 1987) followed by flash chromatography was used in the synthesis of the C-5 epoxides. Separation of the tartrate from the C-5 epoxides required two or more chromatographic cycles, whereas ferrous sulfate/tartaric acid work-up (Gao et al., 1987) followed by a single chromatography cleanly gave the C-6 epoxides. Diasteroselective epoxide ring-opening was conducted with neat AlMe3 instead of a hexane solution of this reagent, as was employed by Nakawaga and Mori. This facilitated comple- tion of the reaction in less than 1 hr compared to two to three days. Work-up via addition of saturated NaF at -40°C (Suzuki et al., 1982) for 3a, 3b, 8a, and 8b, and 3 M HCl for 12a and 12b afforded the corresponding diols after flash chromatography. Quenching with NaF rather than HCl improved isolated yields of 3-methyl-l ,2-pentanediols, probably due to the high solubility of the diols in water. Products arising from breakage of the a-bond or retention of configuration during the cleavage of the {J-epoxide bond were not detected by GC or lH NMR analysis. The syn-isomers of phoenicol and cruentol were obtained with moderate optical purities from asyrnmetric epoxidation of the requisite Z-alkenols. In contrast, Mori and Brevet (1991) and Mori and Hara- shima (1993) generated chirally pure epoxides through crystallization of deriv- atives, a process that leads to yields in the range of 40%. The p-nitrobenzoates or 3,5-dinitrobenzoates of 2a and 2b were oils at room temperature and below. The syn isomers of 5-methyl-octan-4-ol were obtained in high enantiomeric excess through Mitsunobu (Mitsunobu, 1981; Hughes, 1992, and references cited therein) mediated inversion of configuration of the anti-isomers 14a and 14b. Use of p-nitrobenzoic acid-Ph.Pi-diethyl azocarboxylate (DEAD) in THF yielded les s than 25% of the corresponding p-nitrobenzoates. Successful Mit- sunobu conditions (- 51 % yields) employed benzoic acid-Ph-Pi-diisopropyl azocarboxylate (DIAD) and benzene as a solvent (Paquette and Sugimura, 1986; Dai et al., 1988; Dyer and Kishi, 1988). No epimerization or retention of configuration products were observed. Both stereoisomeric phoenicol and cruentol elute from a polar SP-lOOO- coated, fused silica column in two resolved components. The shorter eluting component coincided with the male-produced pheromone of each weevil and was hypothesized to be the syn diastereoisomer, consisting of coeluting S,S and PHEROMONE CHlRALITY 2665 R,R isomers. This assignment was made by analogy with the aggregation pher- omone of the smaller European elm bark beetle, Scolytus multistriatus (Mar- sham), 4-methyl-heptan-3-ol, that also has two stereogenic centers and exists as two diastereoisomeric forms that are separable by GC on a polar Carbowax 20 M column (Pearce et al., 1975). Analysis of stereoselectively prepared syn and anti stereoisomers of phoenicol and cruentol confirmed the assignments. Syn- thetic (R,R)-, (S,S)-, (R,S)-, and (S,R)-phoenicol and cruentol were separated with baseline resolution on a fused silica, Cyclodex-B column. These analyses revealed that male R. phoenicis and male R. cruentatus produce the S,S ster- eoisomer of phoenicol (Figure 4) and cruentol (Figure 5), respectively. S. mul- tistriatus also produces the S,S stereoisomer of 4-methyl-heptan-3-01 (Pearce et al., 1975), whereas the large European elm bark beetle, Scolytus scolytus (F.), produces both (3S,4S)- and (3R,4S)-4-methyl-heptan-3-01 (Blight et al., 1977, 1978, 1979; Wadhams et al., 1982). Coupled GC-EAD of synthetic phoenicol (Figure 6) and cruentol (Figure 7) revealed strong antennal responses to weevil-produced (S,S)-phoenicol and (S,S)-cruentol. Lack of or reduced response to later-eluting stereoisomers can hardly be explained by an antennal refractory periodo In GC-EAD recordings with the same Cyclodex-B column, antennae of two Asian palm weevils, R. ferrugineus (Oliv.) and R. vulneratus (Panz.), distinctively responded to both, closely eluting, (4S)- and (4R)-4-methyl-nonan-5-one (Perez et al., unpub- lished). Similarly, in GC-EAD analyses of oil palm volatiles, antennae of male Ion Chromatogram 01 3·methyl·octan·4·01 Dlastereoisomeric 3R.4R OH~ 3S,4R 9H .-~3S.~._ Produced by male R. phoenicis --,-- T T 17 18 19 20 21 22 Retention Time [mini FIG. 4. Selected ion miz 127 chromatogram (Hewlett Packard 5985B) of stereoisomeric and weevil-produced 3-methyl-octan-4-ol. miz 127 was the parent ion [(M+ -H) 143, (M+ -H-OH) 127] of the full-scan mass spectrum in Cl mode. (Cyclodex-B column; 90°C isothermal; linear flow velocity of carrier gas: 35 cm/sec; injector temperature: 220°C). 2666 PEREZ ET AL. Ion Chromalogram 015·melhyl·ocIan-4-01 Diastereoisomeric Produced by male R. cruentatus 14 15 16 17 18 19 Retention lime [min) FIG. 5. Selected ion mlz 127 chromatogram of stereoisomeric and weevil-produced 5-methyl-octan-4-ol. mIz 127 was the parent ion [(M+-H) 143, (M+-H-OH) 127] of the full-scan mass spectrum in CI mode (instrument and chromatographic conditions as in Figure 4). FID: Synthetic 3-methyl-octan-4-ol 3R.4R ?H 3RAS OH A)vv ~ , J('~~j[;tv:> oS ~ e o a. "'Q)a: ~ J!l Q) O EAD: Antenna 01 a male R. phoenicis ---l:: 22 26 30 34 36 Retenlion lime [mini FIG.6. Representative GC-EAD recording of a female Rhynchophorus phoenicis antenna responding to stereoisomers of 3-methyl-octan-4-ol (Hewlett Packard 5890A; split injec- tion; column and chromatographic conditions as in Figure 4). and female R. phoenicis responded within 2.5 min to four esters, two of which were barely baseline separated (Gries et al., 1994). Strong antennal activity of the S,S, and weak activity of S,R and R,S isomers of the pheromones (Figures 6 and 7) suggest that sensory recognition of the natural S,S stereoisomer is more dependent on the stereochemistry of the methyl than the hydroxy group. •• :: PHEROMONE CHrRALITY 2667 In field experiments (S,S)-phoenicol and (S,S)-cruentol strongly synergized attraction of weevils to palm tissue (Figures 8 and 9). Because racemic, ster- eoisomeric mixtures were as synergistic as S,S isomers, the weakly EAD-active S,R isomers (Figures 6 and 7) neither enhanced nor reduced behavioral activity in the stereoisomeric mixtures (Figures 8 and 9). Lack of strong antennal (Fig- ures 6 and 7) and any behavioral activity (Figures 8 and 9) of non natural isomers suggests that sympatric beetles are unlikely to utilize one or more stereoisomers o 13 "*o FID: Synlhetic s-metbyl-ocíen-a-o! EAD: Antenna 01 a femaJe R. cruentatus -v t rnv 18 22 26 30 34 38 Relenlion Time [min) FrG. 7. Representative GC-EAD reeording of a female Rhynchophorus cruentatus an- tenna responding to stereoisomers of 5-methyl-oetan-4-ol (Hewlett Paekard 5890A; split injection; column and ehromatographic conditions as in Figure 4). patm palm palm palm palm diastereoisomeric S,S S,S mixture R,R R,R Slereoisomers 01 3-methyl·octan-4-01 FrG. 8. Mean eounts (+ standard error) of male and female R. phoenicis in traps baited with 250 g of ehopped oil palm tissue alone and in combination with either stereoiso- merie, (3S,4S)-, (3R,4R)- or (3S,4S)- plus (3R,4R)-phoenieol. La Me Research Station, Cote d'Ivoire; May 6-10, 1993; six blocks. Bars superscripted by the same letter are not significantly different. ANOVA followed by Seheffé test, P < 0.05. 2668 PEREZ ET AL. - 50ur . Page 7 Titles ., Page 8 Titles ~ Images Image 1 Tables Table 1 Page 9 Page 10 Page 11 Page 12 Page 13 Titles .-~ Images Image 1 Image 2 Page 14 Titles •• :> ~ "' ~ , J(' ~~j[;tv ---l:: Images Image 1 Image 2 Page 15 Titles "* Images Image 1 Image 2 Image 3 Image 4 Page 16 Titles 2668 ~ ~ " FIG. 9. Mean counts (+ standard error) of male and female R. cruentatus in traps baited stereoisomeric, (4S,5S)- or (4R,5R)-cruentol. La Belle, Florida, USA, June 9-16, 1993; fI 12 blocks. Bars superscripted by the same letter are not significantly different. ANOV A P ~ 0.05. : Images Image 1 Page 17 Page 18 Titles 2670 Page 19 Titles 2671 /- i Images Image 1