J Chem Ecol (2009) 35:81–95 DOI 10.1007/s10886-008-9569-1

Identification of Sex Pheromone Components of the Hessian , Mayetiola destructor

Martin N. Andersson & Jenny Haftmann & Jeffrey J. Stuart & Sue E. Cambron & Marion O. Harris & Stephen P. Foster & Stephan Franke & Wittko Francke & Ylva Hillbur

Received: 15 May 2008 /Revised: 15 September 2008 /Accepted: 5 November 2008 /Published online: 10 December 2008 # Springer Science + Business Media, LLC 2008

Abstract Coupled gas chromatographic (GC)–electroan- (with respect to the main compound, (2S,10E)-10-tridecen- tennographic detection (EAD) analyses of ovipositor 2-yl acetate). The five-component blend was more attractive extract of calling , Mayetiola destructor, females to male than a three-component blend lacking the two revealed that seven compounds elicited responses from dienes. Furthermore, the five-component blend was more male antennae. Four of the compounds—(2S)-tridec-2-yl attractive than a blend with the same compounds but that acetate, (2S,10Z)-10-tridecen-2-yl acetate, (2S,10E)-10- contained one tenth the concentration of (2S,8E,10E)-8, tridecen-2-yl acetate, and (2S,10E)-10-tridecen-2-ol—were 10-tridecadien-2-yl acetate (more accurately mimicking the identified previously in female extracts. Two new EAD- ratios found in female extract). This suggests that the ratios active compounds, (2S,8Z,10E)-8,10-tridecadien-2-yl acetate emitted by females might deviate from those in gland and (2S,8E,10E)-8,10-tridecadien-2-yl acetate, were identi- extracts. In a field-trapping experiment, the five-component fied by GC–mass spectroscopy (MS) and the use of synthetic blend applied to polyethylene cap dispensers in a 100:10 μg reference samples. In a Y-tube bioassay, a five-component ratio between the main component and each of the other blend (1 ng (2S)-tridec-2-yl acetate, 10 ng (2S,10E)-10- blend components attracted a significant number of male tridecen-2-yl acetate, 1 ng (2S,10E)-10-tridecen-2-ol, 1 ng Hessian flies. Also, a small-plot field test demonstrated the (2S,8Z,10E)-8,10-tridecadien-2-yl acetate, and 1 ng attractiveness of the five-component blend to male Hessian (2S,8E,10E)-8,10-tridecadien-2-yl acetate) was as attractive flies and suggests that this pheromone blend may be useful to male Hessian flies as a similar amount of female extract for monitoring and predicting Hessian fly outbreaks in agricultural systems. M. N. Andersson : Y. Hillbur (*) Department of Plant Protection Biology, Keywords Mayetiola destructor . Hessian fly. Swedish University of Agricultural Sciences, . Diptera . Sex pheromone . (2S)-Tridec-2-yl 230 53 Alnarp, Sweden e-mail: [email protected] acetate . (2S,10E)-10-tridecen-2-yl acetate . . . : : (2S,10E)-10-Tridecen-2-ol (2S,8Z,10E)-8 J. Haftmann S. Franke W. Francke 10-Tridecadien-2-yl acetate . (2S,8E,10E)-8 . Institute of Organic Chemistry, University of Hamburg, 10-Tridecadien-2-yl acetate . GC–EAD . Y-tube bioassay. 20146 Hamburg, Germany Field trapping J. J. Stuart Department of Entomology, Purdue University, West Lafayette, IN 47907, USA Introduction S. E. Cambron USDA-ARS, Department of Entomology, Purdue University, The Hessian fly, Mayetiola destructor (Say) (Diptera: West Lafayette, IN 47907, USA Cecidomyiidae), is one of the most destructive pests of wheat (Triticum spp) in the US and North Africa (Gagné M. O. Harris : S. P. Foster Department of Entomology, North Dakota State University, 1989; El Bouhssini et al. 1999; Berzonsky et al. 2003; Fargo, ND 58105, USA Harris et al. 2003). In addition, it is considered a pest in 82 J Chem Ecol (2009) 35:81–95 many European countries (Barnes 1956). In common with pheromone extraction, coupled gas chromatographic– other species in the family Cecidomyiidae (gall midges), electroantennographic detection (GC–EAD) analyses, and Y- Hessian fly adults are short-lived, have a highly synchro- tube bioassays. Hessian flies used in the small-plot test were nized period of flight activity, and may be present in crops reared on hard red spring wheat, T. aestivum L. (genotype for brief periods of time only (Harris and Foster 1999). “Reeder”), in a greenhouse at North Dakota State University. Typically, outbreaks are sporadic, local, and difficult to These flies were also of the “Great Plains” biotype and predict (ibid.). As a consequence, Hessian flies are originated from approximately 5,000 puparia, obtained in normally discovered only after they have become a serious 2000, from the USDA-ARS laboratory at Purdue University. problem. Various methods are used to control Hessian flies Pheromone Extraction The pheromone gland in Hessian fly including delayed planting of winter wheat, use of resistant females is associated with the eighth and ninth abdominal wheat varieties, and application of broad-spectrum insecti- intersegmental membrane epidermis (Solinas and Isidoro cides (Buntin et al. 1992; El Bouhssini et al. 1999; Rausher 1996). Gland extracts were prepared by excising the 2001; Berzonsky et al. 2003). All of these methods might terminal abdominal segments of virgin calling females be improved by a sensitive method for the detection of early (Bergh et al. 1990; Foster et al. 1991a). Ovipositors were infestations (Buntin et al. 1990, 1992; Cox and Hatchett placed into a vial partially immersed in liquid nitrogen. 1986; Harris et al. 2003). One potential method for field When sufficient ovipositors (approximately 30) were monitoring might employ the use of sex pheromone-baited collected, the vial was allowed to warm to ambient traps (Harris and Foster 1999). Earlier work demonstrated temperature, and the tissue was extracted for 1–1.5 min in the presence of a sex pheromone in the Hessian fly (McKay distilled hexane (LabScan). Following extraction, the solvent and Hatchett 1984). One component of the female-produced was decanted into glass vials (2 ml) and stored at −18°C until sex pheromone, (2S,10E)-10-tridecen-2-yl acetate [2S-10E- use. All dissections were made between 09:00 and 11:30. 13:OAc], was identified (Foster et al. 1991b). However, when tested alone in a wind tunnel, this compound attracted GC–EAD Recordings GC–EAD (Arn et al. 1975) was used significantly fewer Hessian fly males than did female to analyze female gland extracts and synthetic compounds. extract (Harris and Foster, 1991). In a field study, traps A Hewlett-Packard 6890 GC (Palo Alto, CA, USA) with baited with the compound also failed to attract male flame ionization detection and an Innowax column (30 m× Hessian flies (Harris and Foster 1999). In the wind tunnel, 0.25 mm i.d., H-P), programmed from 80°C (held for male responses were tested to binary blends of 2S-10E-13: 2 min) to 220°C at 10°C/min, was used. Whole male bodies OAc and racemic mixtures of three other chemicals— were mounted in an antennal holder (JoAC, Lund, Sweden), (10Z)-10-tridecen-2-yl acetate, (10E)-10-tridecen-2-ol, and as described by Hillbur et al. (2001). Both antennae were tridec-2-yl acetate (Foster et al. 1991b; Harris and Foster exposed simultaneously to a charcoal-filtered and humidi- 1991)—found in female extract (Millar et al. 1991). fied air stream at a rate of approximately 0.3 m/s through a However, none of these blends attracted more males than glass tube (8-mm diameter). The signals from the antennae 2S-10E-13:OAc alone. In this study, we identified addi- were amplified (JoAC) before they were recorded and were tional compounds produced by female Hessian fly and analyzed with ElectroAntennoGraphy software (Syntech, formulated a blend that is attractive to male flies in the Hilversum, The Netherlands). field. Structure Elucidation Coupled GC–MS analyses of extracts of female Hessian fly were generally carried out Methods and Materials as described earlier (Hillbur et al. 2005). Separations were achieved with a 60 m, 0.25-mm i.d. fused silica capillary, Rearing Wheat plants (Triticum aestivum L., genotype DB-5/MS (J&W Scientific, Folsom, CA, USA) under “Blueboy”) that contained Hessian fly puparia of the “Great temperature program: 3 min at 60°C, then programmed to Plains” biotype (provided by the US Department of 280°C at a rate of 5°C/min. An additional fused silica Agriculture-Agricultural Research Service (USDA-ARS) column, 50 m, 0.25 mm i.d., Permabond FFAP (Macherey and the Department of Entomology, Purdue University, West & Nagel, Düren, Germany), was also used. Lafayette, IN, USA) were put into Plexiglas cages (29×34× 29 cm) held in an environmental chamber (25°C, 70% relative Enantioselective GC was carried out with a 25 m, 0.25-mm humidity (RH) and 12:12 L:D photoperiod; lights on 09:00) at i.d. fused silica capillary coated with a 1:1 mixture of Alnarp, Sweden. Infested plants were watered daily to avoid heptakis 6-O-tert.butyldimethylsilyl-2,3-di-O-methyl β- desiccation of developing flies. Typically, adults emerged 7– cyclodextrin and OV 17 at 100°C isothermal and using 14 days later, and the sexes were separated and used for hydrogen as the carrier gas. J Chem Ecol (2009) 35:81–95 83

1H nuclear magnetic resonance (NMR) spectra of hepten-2-ol. Reaction with 3,4-dihydro-2H-pyran and synthetic compounds were recorded on a Bruker AMX-400 ozonization followed by reductive workup produced instrument (Karlsruhe, Germany) using tetramethylsilan (5S)-5-(2-tetrahydropyranyloxy)hexan-1-ol (VII) via the as the internal standard. protected alkenol VI. Transformation of the primary hydrox- yl group to the bromide VIII and chain elongation of its Syntheses If not otherwise stated, starting material, Grignard product with (2E,4E)-2,4-heptadien-1-yl acetate reagents, and solvents were purchased from Aldrich and (Samain et al. 1978) employing cuprate coupling (Fouquet were of highest available grade. Syntheses were carried out and Schlosser 1974)yielded(2S,8E,10E)-2-(2-tetrahydro- under dry argon and synthetic compounds were purified by pyranyloxy)trideca-8,10-diene (IX). Acid-catalyzed column chromatography using silica 60 Å, 32–63 mesh deprotection and acetylation afforded the target compound (MP Eco Chrom) under 0.5-bar pressure. 2S-8E-10E-13:OAc. The stereogenic center in each of the six target compounds was introduced via a Grignard-type reaction 5-Bromo-1-tert.butyldimethylsilyloxypentane (II) Asolu- of a suitable precursor with commercially available (2S)-2- tion of 5.70 g (34.1 mmol) 5-bromopentane-1-ol (I; methyloxiran of 98% optical purity similar to the approach purchased from TCI Europe, Zwijendrecht, Belgium) and of Millar et al. (1991). The enantiomeric purity of the 6.18 g (90.9 mmol) imidazole in 5 ml of abs. dimethylfor- products corresponded with commercial (2S)-2-methyloxiran mamide was cooled to 0°C. After addition of 6.84 g (checked by enantioselective GC). The Grignard reagent was (45.5 mmol) tert.butyldimethylsilylchloride, the mixture produced from commercially available 1-bromodecane and was stirred for 3 h, warmed to 10°C, and stirred for an used to produce (2S)-tridec-2-yl acetate [2S-13:OAc]. Syn- additional 2 h. Workup started with the addition of 150 ml thesis of the mono-unsaturated compounds (i.e., 2S-10E-13: diethyl ether and 200 ml of saturated aqueous sodium OAc, (2S,10E)-10-tridecen-2-ol [2S-10E-13:OH], and bicarbonate, separation of the organic layer, and extraction (2S,10Z)-10-tridecen-2-yl acetate [2S-10Z-13:OAc]) in- of the aqueous layer (3×150 ml diethyl ether). The organic volved 7-decyn-1-ol as the educt (Yadav et al. 1995). solutions were combined, washed with brine, and dried Reduction of the ynol to (7E)-7-decen-1-ol (Rossi and over magnesium sulfate. After removal of the solvent, the Carpita 1977)orto(7Z)-7-decen-1-ol (hydrogenation over residue was chromatographed on silica using hexane– Lindlar catalyst), transformation to the corresponding diethyl acetate 95:5, yielding 7.25 g (25.7 mmol, 75%) of alkenyl bromides (Wiley et al. 1964), and formation of II as a colorless liquid. the Grignard products followed by chain elongation with 1 (2S)-2-methyloxiran furnished 2S-10E-13:OH and its (Z)- H-NMR (400 MHz, CDCl3) δ [ppm] = 0.85 (s, 9H, SiC isomer in good yields and purities of approximately 96– (CH3)3); 1.39–1.55 (m, 4H, 2-H, and 3-H); 1.79–1.88 (m, 98%. Standard laboratory acetylation produced the 2H, 4-H); 3.36 (t, 2H, J=6.87 Hz, 5-H); 3.57 (t, 2H, J= corresponding acetates. Analytical data of all these known 6.11 Hz, 1-H) compounds were in good accord with expected values and 13 those reported in the literature (Millar et al. 1991; Yadav C-NMR (100.6 MHz, CDCl3) δ [ppm] = −5.14 (q, 2C, et al. 1995). Si(CH3)2); 24.75 (t, 3-C); 26.12 (q, 3C, SiC(CH3)3); 32.07 The scope of the syntheses of (2S,8Z,10E)-8,10-trideca- (t, 2-C); 32.78 (t, 4-C); 33.95 (t, 5-C); 63.01 (t, 1-C) dien-2-yl acetate [2S-8Z-10E-13:OAc] and (2S,8E,10E)- 8,10-tridecadien-2-yl acetate [2S-8E-10E-13:OAc] is shown MS (70 eV) m/z [%] = 169 (40); 167 (41); 139 (40); 137 in Figs. 3 and 4, respectively. (42); 115 (6); 101 (10); 99 (7); 89 (10); 88 (5); 85 (6); 75 The synthesis of 2S-8Z-10E-13:OAc (Fig. 3) started (44); 73 (27); 70 (11); 69 (100); 61 (5); 59 (13); 58 (9); 57 from commercially available 5-bromopentan-1-ol (I), which (16); 55 (5); 47 (6); 45 (9); 43 (7); 41 (45); 39 (9) was converted to the silyl ether II and subsequently transformed to the Wittig reagent using triphenyl phos- (5Z,7E)-5,7-Decadienyloxy-tert.butyldimethylsilane (III) A phane. Wittig reaction with (2E)-2-butenal yielded III. mixture of 7.18 g (23.6 mmol) of the bromide II, 13.4 g Deprotection and transformation of the obtained alcohol to (51.1 mmol) triphenyl phosphane, and 813 mg (7.70 mmol) the corresponding bromide (Horner et al. 1959) produced sodium carbonate in 50 ml acetonitrile was refluxed for (5Z,7E)-5,7-decadien-1-yl bromide (IV). Chain elongation 24 h. The solvent was removed from the suspension, and the with (2S)-2-methyloxiran followed by acetylation afforded residue was dissolved in 50 ml dichloromethane. Filtration the target compound 2S-8Z-10E-13:OAc. of the oily solution over silica, using dichloromethane as the The synthesis of 2S-8E-10E-13:OAc (Fig. 4) started eluent, provided unreacted starting material, while the Wittig from commercially available 4-bromo-1-butene (V). Chain salt (11.0 g, 20.2 mmol, 79%) was isolated using a 1:1 elongation with (2S)-2-methyloxiran furnished (2S)-7- mixture of dichloromethane and methanol as the eluent. 84 J Chem Ecol (2009) 35:81–95

1 H-NMR (400 MHz, CDCl3) δ [ppm] = 0.85 (s, 9H, SiC diethyl ether). The combined organic layers were washed (CH3)3); 1.50–1.75 (m, 8H, 2-H to 5-H); 3.60 (t, 2H, with brine and dried over magnesium sulfate. Removal of J=6.1 Hz, 1-H); 7.71–7.91 (m, 15H, HPh) the solvent in vacuo and purification by silica chromatog- raphy (hexane–ethyl acetate 4:1) gave 1.52 g (9.85 mmol, 13 C-NMR (100.6 MHz, CDCl3) δ [ppm] = −5.54 (q, 2C, Si 96%) (5Z,7E)-5,7-decadien-1-ol. (CH3)2); 22.25/27.93/28.09 (t, 3-C to 5-C); 26.10 (q, 3C, 1 SiC(CH3)3); 32.07 (t, 2-C); 63.43 (t, 1-C); 131.32 (d, 6C, H-NMR (400 MHz, CDCl3) δ [ppm] = 0.68 (br.s, 1H, CAr); 134.57 (d, 9C, CAr); 136.02 (s, 3C, CAr) OH); 0.92 (t, 3H, J=7.3 Hz, 10-H); 1.28–1.41 (m, 4H, 2-H, and 3-H); 1.96–2.05 (m, 2H, 9-H); 2.08–2.16 (m, 2H, 4-H); A solution of 10.6 g (20.0 mmol) of the Wittig salt, 3.32 (t, 2H, J=6.1 Hz, 1-H); 5.27–5.33 (m, 1H, 5-H); 5.59– obtained in the previous step, in 100 ml abs. tetrahydrofu- 5.68 (m, 1H, 8-H); 6.05–6.15 (m, 1H, 6-H); 6.37–6.46 (m, ran was cooled to 0°C. Subsequently, 26.2 ml of a solution 1H, 7-H) of potassium hexamethyldisilazane (0.5 mmol in toluene, 13 equal to 3.26 g, 16.3 mmol KHMDS) was added slowly, C-NMR (100.6 MHz, CDCl3) δ [ppm] = 13.91 (q, 10-C); with stirring continued for 2 h. Subsequently, the mixture 26.26 (t, 9-C); 26.30 (t, 3-C); 27.81 (t, 4-C); 32.67 (t, 2-C); was cooled to −78°C and 1.34 g (16.0 mmol); (2E)-2- 62.54 (t, 1-H); 125.38 (d, 7-C); 129.59 (d, 6-C); 129.84 (d, pentenal, dissolved in 20 ml abs. tetrahydrofuran, were 5-C); 136.32 (d, 8-C) added dropwise. After stirring for another half hour, the mixture was warmed to 20°C, and 350 ml of a saturated MS (70 eV) m/z [%] = 154 (5); 136 (14); 121 (7); 111 (6); aqueous solution of ammonium chloride was added. The 108 (8); 107 (14); 98 (9); 97 (9); 95 (28); 94 (10); 93 (33); mixture was extracted five times with 150 ml diethyl ether. 91 (20); 84 (7); 83 (14); 82 (15); 81 (30); 79 (100); 78 (9); The combined organic solutions were washed with brine 77 (27); 72 (6); 71 (7); 70 (16); 69 (13); 68 (24); 67 (81); and dried over magnesium sulfate. Removal of the solvent 66 (9); 65 (12); 57 (8); 56 (5); 55 (33); 54 (8); 53 (14); 51 in vacuo and chromatography on silica using hexane–ethyl (5); 43 (6); 41 (24); 39 (14) acetate (97:3) yielded 2.77 g (10.3 mmol, 55%) of the protected dienol III as a colorless oil. A solution of 2.62 g (10.0 mmol) triphenyl phosphane and 680 mg (10.0 mmol) imidazole in a mixture of 50 ml 1 H-NMR (400 MHz, CDCl3) δ [ppm] = 0.94 (t, 3H, J= abs. diethyl ether and 2 ml abs. acetonitrile was cooled to 7.5 Hz, 10-H); 0.99 (s, 9H, SiC(CH3)3); 1.42–1.57 (m, 4H, 0°C. Subsequently, 1.62 g (10.0 mmol) of bromine was 2-H, and 3-H); 1.96–2.05 (m, 2H, 9-H); 2.11–2.25 (m, 2H, added dropwise, and a yellowish suspension was formed 4-H); 3.52 (t, 2H, J=6.3 Hz, 1-H); 5.31–5.38 (m, 1H, 5-H); that was stirred for 2 h at 20°C. A solution of 1.52 g 5.59–5.68 (m, 1H, 8-H); 6.06–6.13 (m, 1H, 6-H); 6.38– (9.85 mmol) of (5Z,7E)-5,7-decadien-1-ol, obtained in the 6.48 (m, 1H, 7-H) previous step, in 20 ml abs. diethyl ether was added slowly, and the mixture was stirred overnight. Workup started with 13 C-NMR (100.6 MHz, CDCl3) δ [ppm] = -5.13 (q, 2C, Si the addition of 100 ml diethyl ether and 100 ml of a (CH3)2); 13.91 (q, 10-C); 26.20 (q, 3C, SiC(CH3)3); 26.28 saturated aqueous solution of ammonium chloride, separa- (t, 3-C); 26.44 (t, 9-C); 27.82 (t, 4-C); 32.77 (t, 2-C); 63.13 tion of the layers, and extraction of the aqueous layer (6× (t, 1-H); 125.42 (d, 7-C); 129.61 (d, 6-C); 129.92 (d, 5-C); 100 ml diethyl ether). The combined organic solutions were 136.27 (d, 8-C) washed with brine and dried over magnesium sulfate. After removal of the solvent, the residue was chromatographed MS (70 eV) m/z [%] = 211 (41); 155 (33); 135 (20); 129 on silica (hexane–ethyl acetate 97:3) yielding 1.57 g (5); 115 (8); 107 (13); 101 (13); 99 (5); 95 (9); 93 (25); 91 (7,23 mmol, 73%) of the bromide IV as a colorless oil. (8); 89 (12); 81 (10); 79 (28); 77 (11); 76 (9); 75 (100); 73 1 (26); 69 (7); 67 (13); 59 (11); 57 (6); 55 (10); 41 (13) H-NMR (400 MHz, CDCl3) δ [ppm] = 0.92 (t, 3H, J= 7.5 Hz, 1-H); 1.17–1.27 (m, 2H, 8-H); 1.43–1.53 (m, 2H, (5Z,7E)-5,7-Decadienyl bromide (IV) To a solution of 9-H); 1.92–2.05 (m, 4H, 2-H, and 7-H); 2.91 (t, 3H, J= 2.77 g (10.3 mmol) III in 50 ml tetrahydrofuran was added 6.8 Hz, 10-H); 5.13–5.21 (m, 1H, 3-H); 5.58–5.67 (m, 1H, 14.4 ml (14.4 mmol) of a 1-M solution of tetra-n- 6-H); 6.01–6.10 (m, 1H, 4-H); 6.29–6.39 (m, 1H, 5-H) butylammonium fluoride in tetrahydrofuran, and the mix- 13 ture was stirred for 6 h at 20°C. Workup started with the C-NMR (100.6 MHz, CDCl3) δ [ppm] = 13.88 (q, 1-C); addition of 100 ml diethyl ether and 100 ml of a saturated 26.28/27.0 (t, 2-C, and 7-C); 28.37 (t, 8-C); 32.46 (t, 9-C); aqueous solution of ammonium chloride, separation of the 33.41 (t, 10-C); 125.15 (d, 5-C); 129.02 (d, 3-C); 129.88 (d, layers, and extraction of the aqueous layer (6×100 ml 4-C); 136.64 (d, 6-C) J Chem Ecol (2009) 35:81–95 85

MS (70 eV) m/z [%] = 218 (11); 216 (11); 109 (9); 107 (7); aqueous layer (five times with 50 ml diethyl ether). The 96 (7); 95 (100); 93 (13); 91 (10); 82 (17); 81 (48); 80 (7); combined organic solutions were washed with 50 ml of an 79 (26); 77 (15); 69 (7); 68 (23); 67 (66); 65 (7); 55 (18); aqueous saturated solution of copper-II-sulfate and 50 ml of 53 (9); 41 (23); 39 (13). brine and dried over magnesium sulfate. After removal of the solvent in vacuo, the crude product was chromato- 2S-8Z-10E-13:OAc To a mixture of 353 mg (14.5 mmol) graphed on silica (hexane–ethyl acetate 95:5), yielding freshly crushed magnesium turnings and 30 ml abs. 1.29 g (5.35 mmol, 36%) of 2S-8Z-10E-13:OAc as a tetrahydrofuran kept at 60°C, a solution of 1.57 g of IV in colorless oil. 30 ml tetrahydrofuran was added dropwise. The solution of 1 the Grignard product was added dropwise to a suspension H-NMR (400 MHz, CDCl3) δ [ppm] = 0.93 (t, 3H, of 420 mg (7.23 mmol) of (2S)-2-methyloxiran and 140 mg J=7.4 Hz, 13-H); 1.08 (d, 3H, J=6.3 Hz, 1-H); 1.12–1.35

(0.72 mmol) copper-I-iodide in 10 ml abs. tetrahydrofuran (m, 7H, 3-Ha, and 4-H to 6-H); 1.49–1.61 (m, 1H, 3-Hb); that was cooled to −78°C. After warming to room tempera- 1.72 (s, 3H, COCH3); 1.96–2.06 (m, 2H, 12-H); 2.10–2.19 ture, the mixture was stirred for another hour. Workup started (m, 2H, 7-H); 4.92–5.01 (m, 1H, 2-H); 5.30–5.38 (m, 1H, with the addition of 50 ml diethyl ether and 100 ml of a 11-H); 5.60–5.70 (m, 1H, 8-H); 6.07–6.15 (m, 1H, 10-H); saturated aqueous solution of ammonium chloride, separation 6.39–6.50 (m, 1H, 9-H) of the layers, and extraction of the aqueous layer (4×50 ml 13 hexane–ethyl acetate 1:1). The combined organic solutions C-NMR (100.6 MHz, CDCl3) δ [ppm] = 13.91 (q, 13-C); were dried over magnesium sulfate and concentrated in 20.08 (q, 1-C); 20.95 (q, COCH3); 25.66 (t, 12-C); 26.28 (t, vacuo. The obtained (2S,8Z,10E)-8,10-tridecadien-2-ol was 12-C); 27.98/29.38/29.99 (t, 4-C to 6-C); 36.26 (t, 3-C); purified by column chromatography on silica (hexane–ethyl 70.70 (d, 2-C); 125.41 (d, 9-C); 129.54 (d, 11-C); 129.82 acetate 1:1) yielding 654 mg (3.33 mmol, 46%) of (d, 10-C); 136.30 (d, 8-C); 169.77 (s, CO) (2S,8Z,10E)-trideca-8,10-dien-1-ol. MS (70 eV) m/z [%] = 178 (14); 149 (13); 136 (5); 135 1 H-NMR (400 MHz, CDCl3) δ [ppm] = 0.93 (t, 3H, J= (11); 122 (6); 121 (11); 111 (5); 110 (43); 109 (18); 108 7.4 Hz, 13-H); 0.99 (d, 3H, J=6.1 Hz, 1-H); 1.15–1.39 (m, (18); 107 (24); 97 (8); 96 (50); 95 (61); 94 (37); 93 (52); 92 9H, 3-H to 6-H and OH); 1.96–2.07 (m, 2H, 12-H); 2.13– (5); 91 (16); 87 (8); 83 (8); 82 (50); 81 (65); 80 (28); 79 2.21 (m, 2H, 7-H); 3.45–3.55 (m, 1H, 2-H); 5.32–5.42 (m, (95); 78 (9); 77 (18); 71 (6); 69 (10); 68 (21); 67 (100); 66 1H, 11-H); 5.60–5.70 (m, 1H, 8-H); 6.09–6.15 (m, 1H, (12); 65 (8); 55 (35); 54 (8); 53 (11); 43 (92); 42 (6); 41 10-H); 6.42–6.39 (m, 1H, 9-H) (32); 39 (10). For a plotted spectrum, see Fig. 2B.

13 C-NMR (100.6 MHz, CDCl3) δ [ppm] = 13.94 (q, 13-C); (2S)-2-(2-Tetrahydropyranyloxy)hept-6-ene (VI) Starting 23.81 (q, 1-C); 26.03 (t, 12-C); 26.31 (t, 7-C); 28.07/29.61/ from 10 g (74.1 mmol) 4-bromo-1-butene, dissolved in 30.14/39.70 (t, 3-C to 6-C); 67.68 (d, 2-C); 128.22 (d, 9-C); 10 ml abs. tetrahydrofuran and 2.71 g (111 mmol) freshly 129.42 (d, 11-C); 130.08 (d, 10-C); 136.26 (d, 8-C) cut magnesium turnings, a Grignard reagent was prepared at 60°C. This was added dropwise to a suspension of 4.3 g MS (70 eV) m/z [%] = 149 (7); 135 (7); 125 (13); 122 (5); (74.1 mmol) (2S)-2-methyloxiran and 1.41 g (7.41 mmol) 121 (13); 114 (8); 111 (8); 110 (16); 109 (13); 108 (14); copper-I-iodide in 50 ml abs. tetrahydrofuran that was 107 (25); 100 (9); 97 (14); 96 (52); 95 (59); 94 (18); 93 cooled to −78°C. After warming to room temperature, the (83); 91 (20); 83 (12); 82 (52); 81 (63); 80 (26); 79 (91); 78 mixture was stirred for an additional hour. Workup started (11); 77 (23); 71 (14); 69 (11); 68 (17); 67 (100); 66 (11); with the addition of 50 ml ethyl acetate and 100 ml of a 65 (10); 57 (6); 55 (31); 54 (9); 53 (13); 45 (48); 43 (19); saturated aqueous solution of ammonium chloride, separa- 41 (33); 39 (14) tion of the layers, and extraction of the aqueous layer (five times with a 1:1 mixture of hexane and ethyl acetate). The A solution of 641 mg (3.26 mol) of the aforementioned combined organic layers were washed with brine and dried tridecadien-2-ol and 10 mg of N,N-dimethyl-4-aminopyr- over magnesium sulfate. After removal of the solvent in idine in 10 ml abs. pyridine was cooled to 0°C. Subse- vacuo, the crude product was purified by silica column quently, 916 mg (9.09 mmol) acetic anhydride, dissolved in chromatography (hexane–ethyl acetate 1:1), yielding 8.46 g 5 ml tetrahydrofuran, were added dropwise. After warming (74.1 mmol, 100%) of (2S)-6-hepten-2-ol which was to room temperature, the mixture was stirred for 2 h. immediately used for the next step. A solution of 8.46 g Workup started with the addition of 50 ml diethyl ether and (74.1 mmol) of (2S)-6-hepten-2-ol and 1.80 g (9.92 mmol) 100 ml of an ice-cold saturated aqueous solution of sodium p-toluene sulfonic acid monohydrate in 70 ml dichloro- bicarbonate, separation of the layers, and extraction of the methane was cooled to −10°C. Subsequently, a solution of 86 J Chem Ecol (2009) 35:81–95

13 9.67 g (115 mmol) 3,4-dihydro-2H-pyran in 20 ml C-NMR (100.6 MHz, CDCl3) δ [ppm] = 20.10 (t, 4′-C); dichloromethane was added dropwise with the temperature 21.97 (q, 6-C); 25.88 (t, 3-C); 25.91 (t, 5′-C); 31.16 (t, 2- kept below 0°C. Stirring was continued for 24 h at 4°C. C); 31.61 (t, 3′-C); 39.45 (t, 4-C); 62.78 (t, 6′-C); 67.90 (t, Workup started with the addition of 200 ml hexane and 1-C); 68.39 (d, 5-C); 99.32 (d, 2′-C) 200 ml saturated aqueous solution of sodium bicarbonate, separation of the layers, and extraction of the aqueous MS (70 eV) m/z [%] = 101 (45); 86 (5); 85 (100); 84 (10); layer (4×200 ml of a 4:1-mixture of hexane and ethyl 83 (51); 67 (14); 57 (12); 56 (15); 55 (39); 45 (7); 43 (16); acetate). The combined organic layers were washed with 42 (9); 41 (28); 39 (9) brine and dried over magnesium sulfate. After removal of the solvent in vacuo, the product was purified by silica (5S)-5-(2-Tetrahydropyranyloxy)hexyl bromide (VIII) A column chromatography (hexane–ethyl acetate 97:3), solution of 7.29 g (27.8 mmol) of triphenyl phosphane yielding 8.09 g (40.8 mmol, 55%) of the THP derivative and 1.89 g (27.8 mmol) imidazole, in a mixture of 120 ml VI. abs. diethyl ether and 40 ml abs. acetonitrile, was cooled to 0°C. Subsequently, 4.44 g (27.78 mmol) bromine were 1 H-NMR (400 MHz, CDCl3) δ [ppm] = 1.30 (d, 3H, J= added slowly until a yellowish suspension was formed, and 6.35 Hz, 1′-H); 1.45–1.70 (m, 8H, 3′-H, 4′-H, 3-Ha, 4-Ha, stirring was continued at 20°C for 2 h. After the addition of and 5-H); 1.70–1.95 (m, 1H, 3-Hb); 1.85–1.95 (m, 1H, 4- 5.11 g (25.3 mmol) (5S)-5-(2-tetrahydropyranyloxy)hexan- Hb); 2.08–2.20 (m, 2H, 5′-H); 3.52–3.61 (m, 1H, 6-Ha); 1-ol, dissolved in 15 ml abs. diethyl ether, the mixture was 3.75–3.90 (m, 1H, 2′-H); 3.92–4.30 (m, 1H, 6-Hb); 4.68– stirred for another 1 h. Workup started with the addition of 4.80 (m, 1H, 2-H) 150 ml water, separation of the organic layer, and extraction of the aqueous layer (6×120 ml diethyl ether). The 13 C-NMR (100.6 MHz, CDCl3) δ [ppm] = 19.87 (t, 4-C); combined organic extracts were washed with brine and 21.70 (q, 1′-C); 25.32 (t, 4′-C); 25.72 (t, 5-C); 31.36 (t, 3- dried over magnesium sulfate. After removal of the solvent C); 37.12 (t, 3′-C); 62.95 (t, 6-C); 71.04 (d, 2′-C); 98.79 (d, in vacuo, the crude product was purified upon column 2-C); 114.50 (t, 7′-C); 139.08 (d, 6′-C) chromatography over silica (hexane–ethyl acetate 96:4). As a result, 5.07 g (19.1 mmol, 75%) of the target bromide MS (70 eV) m/z [%] = 101 (18); 96 (20); 86 (5); 85 (100); VIII were obtained as a colorless oil. 84 (5); 81 (7); 67 (12); 57 (10); 56 (25); 55 (52); 54 (9); 45 1 (5); 43 (14); 41 (29); 39 (12) H-NMR (400 MHz, CDCl3) δ [ppm] = 1.23 (d, 3H, J= 6.1 Hz, 1′-H); 1.40–1.60 (m, 8H, 3-Ha, 4-Ha, 5-H, 3′-H, and (5S)-5-(2-tetrahydropyranyloxy)hexan-1-ol (VII) Asolu- 4′-H); 1.65–1.75 (m, 1H, 3-Hb); 1.78–1.95 (m, 3H, 4-Hb, tion of 8.09 g (40.8 mmol) of VI in 130 ml of a 1:1 and 5′-H); 3.42 (t, 2H, J=6.87, 6′-H); 3.45–3.54 (m, 1H, 6- mixture of dichloromethane and methanol was cooled to Ha); 3.70–3.85 (m, 1H, 2′-H); 3.85–3.95 (m, 1H, 6-Hb); −78°C. At this temperature, ozone was bubbled through the 4.58–4.72 (m, 1H, 2-H) solution until it turned slightly bluish (about 1 h). Subse- 13 quently, oxygen was bubbled through the solution for C-NMR (100.6 MHz, CDCl3) δ [ppm] = 20.25 (t, 4-C); 10 min. After the addition of 3.11 g (81.6 mmol) solid 21.98 (q, 1′-C); 24.83 (t, 4′-C); 25.94 (t, 5-C); 31.60 (t. 5′- sodium borohydride and warming to room temperature, C); 31.64 (t, 3-C); 34.18 (t, 6′-C); 36.01 (t, 3′-C); 63.04 (t, stirring was continued for 12 h. Workup started with the 6-C); 71.23 (d, 2′-C); 99.20 (d, 2-C) careful addition of 100 ml water, separation of the layers, and extraction of the aqueous layer (9×50 ml ethyl acetate). MS (70 eV) m/z [%] = 165 (12); 163 (12); 129 (8); 101 The combined organic layers were dried over magnesium (20); 86 (6); 85 (100); 84 (5); 83 (33); 67 (8); 57 (11); 56 sulfate, and the solvent was removed in vacuo. The crude (31); 55 (30); 45 (6); 43 (14); 41 (23); 39 (6) product was chromatographed on silica (hexane–ethyl acetate 4:1) yielding 5.11 g (25.3 mmol, 66%) of the 2S-8E-10E-13:OAc Starting from a mixture of 440 mg alcohol VII as a colorless oil. (18.1 mmol) freshly crushed magnesium turnings and 40 ml abs. tetrahydrofuran, a Grignard reagent was prepared at 1 H-NMR (400 MHz, CDCl3) δ [ppm] = 1.20 (d, 3H, J= 60°C using 2.93 (14.8 mmol) of the bromide VIII, 6.11 Hz, 6-H); 1.35–1.65 (m, 11H, 2-H to 4-H, 3′-Ha,4′-Ha, dissolved in 10 ml abs. tetrahydrofuran. At −20°C, this 5′-H, and OH); 1.65–1.75 (m, 1H, 3′-Hb); 1.75–1.90 (m, solution was added dropwise to a cold solution of 1.52 g 1H, 4′-Hb); 3.45–3.55 (m, 1H, 6′-Ha); 3.65 (t, 2H, J= (9.87 mmol) (2E,4E)-2,4-heptadienyl acetate and 3.95 ml 6.35 Hz, 1-H); 3.70–3.80 (m, 1H, 5-H); 3.80–3.95 (m, 1H, (396 mmol) of a 1-M solution of lithium tetrachlorocuprate

6′-Hb); 4.55–4.77 (m, 1H, 2′-H) in tetrahydrofuran. After the addition was complete, stirring J Chem Ecol (2009) 35:81–95 87 was continued for 5 h at 0°C. Workup started with the MS (70 eV) m/z [%] = 149 (5); 135 (5); 125 (8); 121 (8); addition of 100 ml diethyl ether and 60 ml of a saturated 110 (6); 109 (8); 108 (11); 107 (16); 97 (11); 96 (49); 95 aqueous solution of ammonium chloride, separation of the (52); 94 (11); 93 (37); 91 (13); 83 (11); 82 (49); 81 (56); 80 organic layer, and extraction of the organic layer (5×40 ml (22); 79 (77); 78 (8); 77 (17); 71 (14); 69 (10); 68 (16); 67 diethyl ether). The combined organic solutions were (100); 66 (10); 65 (7); 57 (5); 55 (37); 54 (10); 53 (11); 45 washed with brine and dried over magnesium sulfate. The (36); 43 (21); 41 (35); 39 (12) crude product was purified by silica–10% silver nitrate column chromatography (Ikan 1982) and hexane–ethyl A solution of 900 mg (4.59 mmol) of (2S,8E,10E)-8,10- acetate 98:2 as the eluent. A final distillation at 10 Torr per tridecadien-2-ol and 10 mg N,N-dimethyl-4-aminopyridine 117°C yielded 1.40 g (4.99 mmol, 51%) of the protected in 10 ml abs. pyridine was cooled to 0°C. Subsequently, dienol IX. 1.29 g (12.8 mmol) acetic anhydride, dissolved in 10 ml tetrahydrofuran, were added dropwise. After warming to 1 H-NMR (400 MHz, CDCl3) δ [ppm] = 1.00 (t, 3H, room temperature, the mixture was stirred for 2 h. Workup J=7.27 Hz, 13′-H); 1.20 (d, 3H, J=6.27 Hz, 1′-H); 1.25– started with the addition of 80 ml diethyl ether and 150 ml

1.45 (m, 6H, 4′-H to 6′-H); 1–45–1.65 (m, 6H, 3-Ha, 4-Ha, of an ice-cold saturated aqueous solution of sodium 5-H, and 3′-H); 1.65–1.75 (m, 1H, 3-Hb); 1.75–1.90 bicarbonate, separation of the layers, and extraction of the (m, 1H, 4-Hb); 2.00–2.13 (m, 4H, 7′-H, and 12′-H); 3.45– aqueous layer (5×50 ml diethyl ether). The combined 3.52 (m, 1H, 6-Ha); 3.65–3.81 (m, 1H, 2′-H); 3.85–3.95 (m, organic solutions were washed with 50 ml of an aqueous 1H, 6-Hb); 4.60–4.75 (m, 1H, 2-H); 5.50–5.70 (m, 2H, 8′-H, saturated solution of copper-II-sulfate and 50 ml brine and and 11′-H); 5.95–6.0 (m, 1H, 9′-H); 6.0–6.05 (m, 1H, 10′-H) dried over magnesium sulfate. After removal of the solvent in vacuo, the crude product was chromatographed on silica 13 C-NMR (100.6 MHz, CDCl3) δ [ppm] = 14.03 (q, 13′-C); (hexane–ethyl acetate 97:3), yielding 960 mg (4.03 mmol, 20.15 (t, 4-C); 21.96 (q, 1′-C); 25.71 (t, 12′-C); 25.97 (t, 5- 88%) of 2S-8E-10E-13:OAc. C); 26.11 (t, 6′-C); 29.67 (t, 4′-C); 29.78 (t, 5′-C); 31.65 (t, 1 3-C); 32.94 (t, 7′-C); 36.85 (t, 3′-C); 62.82 (t, 6-C); 71.51 H-NMR (400 MHz, CDCl3) δ [ppm] = 0.99 (t, 3H, J= (d, 2′-C); 99.01 (d, 2-C); 129.80 (d, 9′-C); 130.80 (d, 10′- 7.38 Hz, 13-H); 1.20 (d, 3H, J=6.36 Hz, 1-H); 1.25–1.40

C); 132.79 (d, 8′-C); 134.31 (d, 11′-C) (m, 6H, 4-H to 6-H); 1.43–1.51 (m, 1H, 3-Ha); 1.52–1.62 (m, 1H, 3-Hb); 2.02 (s, 3H, COCH3); 2.02–2.12 (m, 4H, 7- MS (70 eV) m/z [%] = 109 (13); 101 (5); 97 (5); 96 (13); 95 H, and 12-H); 4.84–4.93 (m, 1H, 2-H); 5.50–5.68 (m, 2H, (32); 93 (7); 86 (8); 85 (100); 83 (5); 82 (8); 81 (11); 79 (13); 8-H, and 11-H); 5.95–5.99 (m, 1H, 9-H); 6.00–6.04 (m, 1H, 69 (7); 67 (30); 57 (9); 56 (7); 55 (16); 43 (10); 41 (19) 10-H)

13 Deprotection was carried out by stirring a solution of C-NMR (100.6 MHz, CDCl3) δ [ppm] = 14.01 (q, 13-C); 1.40 g (4.99 mmol) of IX and 290 mg (1.56 mol) p-toluene 20.33 (q, 1-C); 21.75 (q, COCH3); 25.64/25.95/29.36/ sulfonic acid monohydrate in 100 ml methanol for 30 min 29.67/32.84 (t, 4-C to 7-C and 12-C); 36.25 (t, 3-C); 71.40 at 50°C. Subsequently, 835 mg (10.0 mmol) of sodium (d, 2-C); 129.75 (d, 9-C); 130.87 (d, 10-C); 132.53 (d, 8-C); bicarbonate were added. The solvent was removed in vacuo, 134.35 (d, 11-C); 171.14 (s, CO) and the residue was dissolved in 100 ml diethyl ether. The organic solution was dried over magnesium sulfate and MS (70 eV) m/z [%] = 178 (8); 149 (8); 135 (6); 121 (8); concentrated in vacuo. Silica column chromatography 110 (14); 109 (9); 108 (14); 107 (15); 97 (6); 96 (43); 95 (hexane–ethyl acetate 4:1) yielded 900 mg (4.59 mmol, (44); 94 (16); 93 (33); 91 (11); 87 (7); 83 (6); 82 (33); 92%) of (2S,8E,10E)-8,10-tridecadien-2-ol as a colorless oil. 81 (41); 80 (19); 79 (66), 78 (7); 77 (15); 69 (6); 68 (12); 67 (74); 66 (9), 65 (7); 55 (32); 54 (7); 53 (11), 43 1 H-NMR (400 MHz, CDCl3) δ [ppm] = 1.00 (t, 3H, J= (100); 42 (6); 41 (37); 39 (13). For a plotted spectrum, 7.52 Hz, 13-H); 1.18 (d, 3H, J=6.02 Hz, 1-H); 1.29–1.50 see Fig. 2C. (m, 9H, 3-H to 6-H and OH); 2.02–2.14 (m, 4H, 7-H, and 12-H); 3.74–3.84 (m, 1H, 2-H); 5.50–5.68 (m, 2H, 8-H, Y-tube Bioassay The attractiveness of various synthetic and 11-H); 5.95–6.0 (m, 1H, 9-H); 6.0–6.05 (m, 1H, 10-H) pheromone blends and female gland extract were studied in a glass Y-tube olfactometer (modified from Jönsson et al. 13 C-NMR (100.6 MHz, CDCl3) δ [ppm] = 14.02 (q, 13-C); 2005). The arms of the Y-tube were 14 cm long, with the 23.90 (q, 1-C); 25.55/26.01/29.55/29.76/32.89 (t, 4-C to 7- stem and inner diameter 12.5 and 2.2 cm, respectively. C and 12-C); 39.69 (t, 3-C); 68.53 (d, 2-C); 129.76 (d, 9- Charcoal-filtered and humidified air was pumped (Micro C); 130.84 (d, 10-C); 132.63 (d, 8-C); 134.35 (d, 11-C) pump NMP 30 KNDC, 12 V, KNF Neuberger, Germany) 88 J Chem Ecol (2009) 35:81–95 through Teflon tubes and entered each arm via a 9.5-cm-long Seven blocks were run on separate days (September 6–10 glass tube. Experiments were conducted between 09:30 and and 12–13, 2006), and each block contained six treatments 11:30 at 25°C and 70% RH. The airflow through each arm of with randomized positions. The three-component blend the Y-tube was 500 ml/min (approximately 2 cm/s; BA-4AR, (2S-10E-13:OAc, 2S-10E-13:OH, and 2S-13:OAc) and the Kytölä, Muurame, Finland). For testing chemical stimuli, the five-component blend (2S-10E-13:OAc, 2S-10E-13:OH, stimulus (in hexane) was applied to a piece of filter paper 2S-13:OAc, 2S-8E-10E-13:OAc, and 2S-8Z-10E-13:OAc) (1.5×0.5 cm) attached to a steel wire (approximately 1.5 cm were tested on two dispenser materials, polyethylene long); the solvent was allowed to evaporate, and the filter (PE; PE stoppers, 10 mm i.d., Semadeni, Ostermundigen, paper was positioned at the tube center. In bioassays in which Switzerland) and cotton (Dental Rolls No. 3 (cut in half), virgin calling females were used as a stimulus, females were IVF Hartmann AG, Neuhausen, Switzerland). Compounds held in small glass tubes (5 cm long, 2.1 cm o.d.; both ends were applied to the dispensers in a hexane solution. The closed by a fine mesh) and placed inside the distal part of one ratio between 2S-10E-13:OAc and the other components of the Y-tube arms. During tests, a single male was taken was100:10µg(seealsoFig.6). Empty cotton and PE from a rearing cage and placed into a glass tube (5 cm long, dispensers were used as controls. Lures and traps were 2.1 cm o.d.; a fine mesh on the distal end) that was then used only once. Newly emerged adult males were collected immediately placed 5 cm into the stem of the Y-tube. Each between 16:00 and 21:00 the night before the test and male was given 5 min to respond. A male was regarded as a placed in cylindrical plastic cages (10-cm diameter, 20 cm responder if it progressed 7 cm up one of the arms of the Y- high, and with a ceiling of cotton netting material) placed tube within 5 min. If a male moved 7 cm up one side arm but over a pot that contained moist potting soil (for further then moved downwind and then upwind into the other arm, details, see Harris and Foster 1991). At each time of the first arm was regarded as its choice. Males that did not collection, males were evenly distributed between four to move into one of the two arms were not included in statistical five cages. Traps were placed directly on the grass in a analyses. line running south to north at 07:00. Since males were Binary-choice bioassays were performed to evaluate differ- released from point sources (cages), we reduced the ences in attractiveness of the main pheromone component, likelihood of potential position effects by maintaining a 2S-10E-13:OAc (abbreviated 1), a three-component blend small distance (i.e., 30 cm) between traps. At 07:15, cages (blend abbreviated 3), consisting of 2S-10E-13:OAc, 2S- weretakenoutsideandplaced60cmapartinaline 10E-13:OH, and 2S-13:OAc, two different four-component parallel to the traps. Cages were placed 3 m east (the blends, consisting of the compounds in blend 3 and with prevailing wind was from the west) of the trap line with either 2S-8Z-10E-13:OAc (blend abbreviated 4Z) or 2S-8E- the opening facing the traps. Males were released when all 10E-13:OAc (blend abbreviated 4E) added, and two five- cages were in position. The following numbers of males component blends, both consisting of all compounds were released on the 7 days when tests were run: 30, 50, included in the other blends but with 2S-8E-10E-13:OAc in 30, 30, 70, 45, and 30, respectively. Each day, traps were different amounts (low amount of 2S-8E-10E-13:OAc collected at 18:00, and the number of males in each trap abbreviated 5L, high amount 5H). Synthetic pheromone was counted. Temperatures ranged from 7°C when males blends were applied in ratios of 10 ng of the main component, were released to 23°C when traps were collected. Because 2S-10E-13:OAc, and 1 ng of the other components in all temperatures were low and the air humid, males that were blends except for 5L in which 0.1 ng of 2S-8E-10E-13:OAc not attracted to traps probably survived for more than was used. Synthetic blends were compared with a female 1 day. Thus, males caught in a trap could have been from pheromone gland extract of comparable concentration (with previous days’ releases. respect to the main component) as well as to five calling females. A blank–blank (hexane only) treatment was also Field Test A field test was performed in fields of wheat bioassayed to check for non-odor-mediated directional stubble (10–25 cm high) outside Wichita, KS, USA, 15–26 preferences. A maximum of six consecutive males (or 10- September 2006. A total of 190 delta traps with sticky min time limit) was tested before filter papers were changed. inserts (same type as in small-plot test) were used, divided Potential non-odor-mediated preferences were eliminated by into ten blocks with 19 treatments. Traps within a block switching sides of the two stimuli during a bioassay. All were placed 5 m apart, and blocks were separated by 10 m. glass equipment were cleaned by heating at 320°C for 8 h The traps were placed <1 cm above ground level. Three before tests. synthetic pheromone treatments were tested: the main component (2S-10E-13:OAc), the three-component blend Small-Plot Test A small-plot test was done on a fly-free (2S-10E-13:OAc, 2S-10E-13:OH, and 2S-13:OAc), and the lawn (grass being 6–10 cm high) in Fargo, ND, USA, using five-component blend (2S-10E-13:OAc, 2S-10E-13:OH, delta traps with sticky inserts (PheroNet, Alnarp, Sweden). 2S-13:OAc, 2S-8E-10E-13:OAc, and 2S-8Z-10E-13:OAc). J Chem Ecol (2009) 35:81–95 89

The ratio between the main component and each of the GC–MS analyses, facilitated structural assignments since other components was 10:1. Three different doses of each the target compounds belonged to the same chemical blend were tested, 1, 10, and 100 µg, based on the main classes. Structure assignments based on GC–MS analyses component (see also Fig. 7). All blends were tested on PE in Hamburg were scrutinized and verified by GC–EAD in and cotton dispensers (same types as in small-plot test; Alnarp. Compound 2 proved to be 2S-13:OAc, compound dispenser preparation is also the same). A blank unbaited 3, 2S-10Z-13:OAc, compound 4, 2S-10E-13:OAc, and trap was included as negative control. Treatments were compound 5, 2S-10E-13:OH (Fig. 1). All four compounds placed randomly within blocks and positions were main- had been identified previously (Foster et al. 1991b; Millar tained throughout the experiment. Trap catches were et al. 1991). Compounds 1 and 7 were present in checked daily and sticky inserts changed. insufficient quantities for MS analysis. In contrast, the mass spectrum of compound 6 (Fig. 2a) had a molecular Statistics Y-tube bioassay data were analyzed by chi-square ion at m/z 238 and signals at m/z 178 (M+-60, i.e., loss of tests. The data from the small-plot test and field test did not acetic acid) and m/z 61 (protonated acetic acid), as well as fulfill the requirements for parametric testing due to intense signals at m/z 43 (acetyl fragment), m/z 79, and m/z unequal variances (Levene’s test, P<0.001); catch propor- 67 (indicating an unsaturated system). These data suggested tions within blocks were, thus, analyzed by the Kruskal– that the compound was an acetic acid ester of a doubly Wallis test. Mann–Whitney U tests were used in pairwise unsaturated alcohol with 13 carbon atoms. The diagnostic comparisons. The Bonferroni method was used to calculate fragment at m/z 87 revealed the acetate moiety to keep experiment-wise error rate with differences at P<0.005 for position 2 along the chain (α-cleavage). The relatively large the small-plot test and P<0.001 for the field test, being abundance of the molecular ion (approximately 5–6%) regarded as significant. suggested a conjugated double-bond system (Löfstedt and Odham 1984). As the main pheromone component, compound 4 has a double bond in position 10, we assumed Results that compound 6 was a 2-acetoxytridecadiene with double bonds in positions 8,10 or 10,12. Because of the relatively GC–EAD recordings of female extracts showed that male small difference in retention times between the diene 6 and antennae responded to seven components. Comparison of the monoene 4, the 8,10 system seemed a more likely mass spectra and retention times of synthetic compounds candidate compared to the terminally unsaturated 10,12 with corresponding data of the natural products allowed system, which, due to its higher polarity, would be expected structure elucidation of six volatile substances. The fact that to have a much longer retention time. This was supported the two laboratories involved in the investigation used GC by the presence of a pronounced signal at m/z 149, columns with slightly different selectivity did not present indicating a terminal ethyl group (M+-acetic acid-ethyl). problems. The known pheromone component 2S-10E-13: In addition, compound 6 was hypothesized to keep (2S)- OAc served as an orientation marker and, supported by configuration, as other compounds identified in the Hessian

Fig. 1 Coupled flame ionization (FID) and electroantennographic tridecen-2-yl acetate [2S-10E-13:OAc], 5 (2S,10E)-10-tridecen-2-ol detection (EAD) of female Hessian fly extract (20 female [2S-10E-13:OH], 6 (2S,8Z,10E)-8,10-tridecadien-2-yl acetate [2S- equivalents). Male antennal responses were recorded to seven 8Z-10E-13:OAc], and 7 (2S,8E,10E)-8,10-tridecadien-2-yl acetate compounds: 1 unknown, 2 (2S)-tridec-2-yl acetate [2S-13:OAc], 3 [2S-8E-10E-13:OAc] (2S,10Z)-10-tridecen-2-yl acetate [2S-10Z-13:OAc], 4 (2S,10E)-10- 90 J Chem Ecol (2009) 35:81–95

A

B

C

Fig. 2 The 70-eV mass spectra of a compound 6 present in Hessian fly females, b synthetic (8Z,10E)-8,10-tridecadien-2-yl acetate, and c synthetic (8E,10E)-8,10-tridecadien-2-yl acetate

fly pheromone gland (Foster et al. 1991b; Millar et al. The mass spectra of the two synthetic compounds were 1991). Consequently, (2S,8Z,10E)-8,10-tridecadien-2-yl very similar. However, a significant difference in the acetate [2S-8Z-10E-13:OAc] and its (8E,10E)-isomer [2S- intensity of the signal at m/z 110 was found. The mass 8E-10E-13:OAc] were synthesized for comparison. spectrum and retention time of synthetic 2S-8Z-10E-13: The preparative routes to the synthetic products are OAc matched perfectly the corresponding data of the summarized in Figs. 3 and 4. The compounds were obtained natural compound 6. In addition, the synthetic compound in overall yields of approximately 10%. With respect to the elicited behavioral (see below) and electrophysiological conjugated double-bond system, the stereochemical purity activities. A small impurity in the synthetic sample of 2S- of 2S-8Z-10E-13:OAc was 96%, containing approximately 8Z-10E-13:OAc (regarded as the 8E,10E-isomer, an 4% of the (E,E)-isomer while that of 2S-8E-10E-13:OAc expected by-product of the Wittig reaction) matched the was 98% and contained approximately 2% of its stereo- retention time of compound 7. Because the concentration of isomers. Enantioselective gas chromatography proved 2S- this compound was very low, a mass spectrum of the 8Z-10E-13:OAc to show an excess of the (S)-enantiomer natural product could not be obtained. However, by using over the (R)-enantiomer of at least 98%, while 2S-8E-10E- an Innowax column, the GC retention time of synthetic 2S- 13:OAc showed an enantiomeric excess of 96%. 8E-10E-13:OAc corresponded precisely to the retention

Fig. 3 Synthetic route to 2S-8Z- a b, c 10E-13:OAc. a: TBDMSCl, Br OH Br OTBDMS OTBDMS imidazole, 0°C, DMF; b: PPh3, III III Na2CO3, reflux, CH3CN; c: (2E)-2-pentenal, KHMDS, 0°C, THF; d:BuN+F−, 20°C, THF; 4 d, e f, g, h e: PPh3,Br2, imidazole, 0°C, Br DCM; f: Mg, −78°C, CuI, THF; g:(2S)-2-methyloxiran, −78°C; IV OAc 2S-8Z-10E-13:OAc h:Ac2O, DMAP, 0°C, pyr J Chem Ecol (2009) 35:81–95 91

Fig. 4 Synthetic route to 2S-8E- OTHP OTHP OTHP 10E-13:OAc. a: Mg, −78°C, a, b, c d, e f CuI, THF; b:(2S)-2-methylox- Br HO Br iran, −78°C; c: p-TsOH, DHP, V VI VII VIII DCM; d:O3, −80°C, DCM/ MeOH; e: NaBH4,20°C.f:PPh3, Br2, imidazole, 0°C, DCM; g: OTHP OAc Mg, 60°C, Li2CuCl4,THF;h: g, h i, j (2E,4E)-2,4-heptadienyl acetate; i: p-TsOH, 60°C, MeOH; j: IX 2S-8E-10E-13:OAc Ac2O, DMAP, 0°C, pyr

time of the compound that elicited antennal response number responses of males to 5H or five calling females (5FE) did 7 (Fig. 1). As GC–EAD recordings were not carried out on not differ significantly either (χ2=2.00, P>0.05). an enantioselective GC column, the enantiomeric composi- tion of the tentatively identified tridecadien-2-yl acetate Small-Plot Test During the 7 days of testing, a total of 285 remains unknown. However, the synthetic (S)-enantiomer males were released; 218 were caught in the traps. Traps elicited behavioral responses from males (see below) as containing the five-component blend caught more male well as electrophysiological activity. We, therefore, postu- Hessian flies than blank traps (Fig. 6; Z=−3.343, P<0.001 late 2S-E8-E10-13:OAc to be the natural substance causing for both dispenser types) or the three-component blend (Z= response 7 in Fig. 1. −3.144, P<0.001 for both dispenser types). Traps baited with the three-component blend did not catch significantly Y-tube Bioassay The proportion of males responding was more males than blank traps. The attractiveness of the five- high in all bioassays (86–100%), except in the blank–blank component blend did not differ between dispenser types. treatment (53%). The blank–blank bioassay demonstrated that males did not have non-odor-mediated directional Field Test During the 9 days of field testing, a total of 963 preferences; ten of the responding males chose the right male Hessian flies were caught. Of these, only eight arm and 11 chose the left arm (χ2=0.048, P>0.05). When a individuals were caught in the cotton dispenser traps. blank was tested against the main pheromone compound Therefore, we excluded results from traps with this type 2S-10E-13:OAc (attractant 1), males were more attracted to of dispenser from further analysis. In traps with PE the latter component (χ2=33.92, P<0.001; Fig. 5). How- dispensers, a total of 848 males were caught at the largest ever, when 2S-13:OAc and 2S-10E-13:OH were added to dose (100 µg) of the five-component blend (blend 9, the main component, males were more attracted to the Fig. 7), while 93 males were caught in traps containing the three-component blend (attractant 3) than to the main same blend, but at the lower dose (10 μg, blend 8, Fig. 7). component alone (1; χ2=10.29, P<0.01). A tendency for Mean trap catches of these treatments differed significantly increasing numbers of males responding to the blend was from catches in the blank (Z=−3.963, P<0.001 for both observed when either 2S-8E-10E-13:OAc or 2S-8Z-10E-13: doses) and from each other (Z=−3.780, P<0.001; Fig. 7). OAc was added to the three-component blend (attractants To ensure the specificity of this system, sticky inserts with 4E and 4Z, respectively), but these blends did not attract trapped midges were characterized for species and sex; all significantly more males than blend 3 (3 vs. 4E: χ2=1.72, midges caught were Hessian fly males (R. Gagné, personal P>0.05; 3 vs. 4Z: χ2=1.53, P>0.05). There was also no communication). Reference males are stored in the North difference between the responses to 3 and the five- Dakota State University insect collection (reference number component blend with the low amount (0.1 ng) of 2S-8E- 6543, lot # 0610677). 10E-13:OAc (attractant 5L; χ2=0.022, P>0.05). However, when 2S-8E-10E-13:OAc was present in the same amount (1 ng) as the other minor components (attractant 5H), the Discussion five-component blend attracted more males than the three- component blend (3; χ2=8.40, P<0.01). 4Z and 5H were We have shown that a synthetic blend of five out of seven both compared with gland extract. The results from these electrophysiologically active compounds found in Hessian comparisons showed that gland extract was more attractive fly female sex pheromone gland extract was highly than the four-component blend (4Z; χ2=17.47, P<0.001) attractive to laboratory-reared males and to males from but that there was no difference between the gland extract natural populations. A sixth compound, the Z-isomer of the and the five-component blend (5H; χ2=1.28, P>0.05). The main pheromone component, 2S-10E-13:OAc, was not 92 J Chem Ecol (2009) 35:81–95

Fig. 5 Responses of male 100 100 100 Hessian flies in binary-choice bioassays. Abbreviations: 1: 2S-10E-13:OAc (10 ng), 3:2S- 80 80 80 10E-13:OAc, 2S-13:OAc, and 2S-10E-13:OH (10:1:1 ng), 4E: 2S-10E-13:OAc, 2S-13:OAc, 2S-10E-13:OH, and 60 60 60 2S-8E-10E-13:OAc (10:1:1:1 ng), 4Z:2S-10E-13: ns OAc, 2S-13:OAc, 2S-10E-13: 40 *** 40 ** 40 OH, and 2S-8Z-10E-13:OAc (10:1:1:1 ng), 5L:2S-10E-13: OAc, 2S-13:OAc, 2S-10E-13: OH, 2S-8Z-10E-13:OAc, and Proportion of males (%) 20 20 20 2S-8E-10E-13:OAc (10:1:1:1:0.1 ng), 5H:as5L but (10:1:1:1:1 ng) and 5FE: five 0 0 0 calling females. N=40–60, blank 1 1 3 3 4E *=P<0.05, ** = P<0.01, and *** = P<0.001 100 100 100

80 80 80

60 60 ns 60 ns 40 40 40 **

Proportion of males (%) 20 20 20

0 0 0 3 4Z 35L 35H

100 100 100

80 80 80

60 60 60 ns ns 40 *** 40 40

Proportion of males (%) 20 20 20

0 0 0 4Z extract 5H extract 5H 5 FE included in the blend as a separate synthetic chemical. female extract (Millar et al. 1991). Since an earlier wind However, due to the isomeric purity of the synthetic 2S-10E- tunnel study had demonstrated that different E/Z ratios of 13:OAc (98%), a small amount of the Z-isomer was present the main component did not influence male attraction in the blend. A low ratio of the Z-isomer was also found in (Harris and Foster 1991), no other ratios were tested. J Chem Ecol (2009) 35:81–95 93

Fig. 6 Mean male Hessian fly 70 trap catch (%)+SE in the small- b plot test. Numbers represent 60 microgram amounts per com- pound. Bars with different let- ters are significant at P<0.001. 50 Co = cotton, PE = b polyethylene 40

30

20

Mean trap catch (%) + SE a 10 a aa 0 2S-10E-13:OAc 12- - 100 µg 100 456100 100 2S-10E-13:OH - - 10 10 10 10 2S-13:OAc - - 10 10 10 10 2S-8Z-10E-13:OAc - - - - 10 10 2S-8E-10E-13:OAc - - - - 10 10 Dispenser Co PE Co PE Co PE

The five-component blend was as active as female systems, it is desirable to reach sufficient attraction at low extract in the Y-tube bioassays and sufficiently active for doses. For this reason, it would be of interest to identify the monitoring field populations of Hessian fly. Since the seventh EAD-active compound (response 1 in Fig. 1), as highest catch was in traps with the highest dose, we do not nothing is known regarding its capacity to mediate behavior. know whether we reached the optimal dose for attraction Studies of various moth species have shown that addition of by this blend. However, in pheromone-based monitoring minor components lowers the behavioral response threshold

Fig. 7 Mean male Hessian fly 100 trap catch (%)+SE in traps with c polyethylene dispensers in the field test. Numbers represent 80 microgram amounts per com- pound. Catch in blank traps is also shown. Bars with different letters are significant at P< 60 0.001

40

Mean trap catch (%) + SE 20 b

a aaaaa a a 0 Blend 0123456789

2S-10E-13:OAc - 1 µg 10 100 1 10 100 1 10 100 2S-10E-13:OH - - - - 0.1 1 10 0.1 1 10 2S-13:OAc - - - - 0.1 1 10 0.1 1 10 2S-8Z-10E-13:OAc ------0.1 1 10 2S-8E-10E-13:OAc ------0.1 1 10 94 J Chem Ecol (2009) 35:81–95 to the pheromone blend (reviewed by Linn and Roelofs The sex pheromone blend that was identified in this study 1995), and it is possible that attractiveness to pheromone could be used as a tool for detecting the presence and traps could be achieved at a lower dose, should this abundance of Hessian flies in the field. The Hessian fly is a component be added. However, this minor compound is major pest of winter wheat. A common approach to reduce present in very low levels in gland extracts, as is the infestation levels in the field is to delay crop planting until compound corresponding to response 7 (see Fig. 1)tenta- Hessian fly activity has ceased in autumn (Buntin et al. 1990). tively identified as 2S-8E-10E-13:OAc. Despite the minute Since flight activity is highly variable from year to year, a amounts of 2S-8E-10E-13:OAc in the extract, the Y-tube pheromone-based detection method could enable farmers to experiment indicates that this compound must be present in better judge when Hessian fly activity is over for the season. higher amounts (the same amount as for the other minor The traps that were used in the small-plot test and field assay compounds) in order to increase the attractiveness of the were placed at (or almost at) ground level. The fact that male pheromone blend. The ratios of compounds found in the Hessian flies were caught at this height suggests that they fly gland extract might thus not be reflective of the compound close to the ground, at least for within-field dispersal. This ratios released by calling females, suggesting that collection has also been demonstrated for both the swede midge of volatiles released from females might allow the collection (Hillbur et al. 2005) and the pea midge (Wall et al. 1991). of greater quantities of this compound and facilitate structure An important characteristic of a pheromone-based mon- confirmation. This may also apply to the unknown com- itoring system is that it is specific for the target species and pound associated with response 1 (Fig. 1). does not catch that are morphologically similar. In The structures of Hessian fly sex pheromone compo- our field trial, the only midges caught in traps baited with our nents are similar to other gall midge pheromones in that synthetic pheromone blend were Hessian fly males. With this they are chiral compounds with odd-numbered carbon specificity and the highly synchronized emergence of chains and a functional group at the C-2 position (Foster Hessian fly adults (Harris and Foster 1999), it appears that et al. 1991b; Harris and Foster 1991; Gries et al. 2000, the five-component blend may be a useful tool for 2002, 2005; Hillbur et al. 1999, 2000, 2001, 2005; Choi et monitoring and predicting Hessian fly outbreaks in crops. al. 2004). In contrast, the attractive sex pheromone blends from other cecidomyiids are composed of either one (Gries Acknowledgements We thank Thiago Benatti and Rajat Aggarwal (Dept. Entomology, Purdue University, West Lafayette, IN, USA) for et al. 2000, 2002; Choi et al. 2004) or three compounds insect rearing. We also thank Elin Isberg for assistance in rearing and (Hillbur et al. 1999, 2005; Gries et al. 2005), compared to Y-tube bioassays. Many thanks also to Gary Cramer (Sedgwick the complex Hessian fly blend. In the laboratory, the main County Extension, Wichita, KS, USA) for field assistance and Ray Hessian fly pheromone component attracted 56% of males Gagné (Dept. Entomology, Smithsonian Institute, Washington, DC, USA) for the identification of insects. This study was partly funded by when tested in a wind tunnel (Harris and Foster 1991). In USDA-NRI, grant #2004-03099 (J. J. Stuart) and by the Royal our study, more than 90% of the responding males were Swedish Academy of Agriculture and Forestry, grant FS-234 (M. N. attracted to this component when tested against a blank in Andersson). W. Francke acknowledges the support of the Fonds der the Y-tube. When tested against more complex blends, the Chemischen Industrie. main component alone was not attractive. The attractive- ness of the main Hessian fly sex pheromone component probably is reflective of the different challenges faced by References male flies when orienting to the pheromone in highly controlled laboratory conditions vs. variable field condi- ARN, H., STÄDLER, E., and RAUSCHER, S. 1975. The electroantenno- tions. In the field, synthetic pheromone lures presumably graphic detector—a selective tool in the gas chromatographic compete with calling females. The presence of calling analysis of insect pheromones. Z. Naturforsch. 30c:722–725. BARNES, H. F. 1956. Gall Midges of Economic Importance, Vol. VII, females around traps in the field might have contributed to Gall midges of cereal crops. Crosby Lockwood, London. the low attractiveness of the main component and the three- BERGH, J. C., HARRIS, M. O., and ROSE, S. 1990. Temporal patterns component blend. The only discrepancy between the trap of emergence and reproductive behavior of the Hessian fly catches in the small-plot test and in the field test was that (Diptera: Cecidomyiidae). Ann. Entomol. Soc. Am. 83:998–1004. BERZONSKY, W. A., DING, H., HALEY, S. D., HARRIS, M. O., LAMB, cotton dispensers were as attractive as PE dispensers in R. J., MCKENZIE, R. I. H., OHM, H. W., PATTERSON,F.L., the small-plot test. Chemicals are typically released from PEAIRS, F. B., PORTER, D. R., RATCLIFFE, R. H., and SHANOWER, cotton dispensers at a faster rate than from polyethylene T. G. 2003. Breeding wheat for resistance to insects. Plant Breed. ones. In the small-plot test, we were able to change the Rev. 22:221–296. BUNTIN,G.D.,BRUCKNER,P.L.,andJOHNSON,J.W.1990. dispensers daily but did not do so in the field trial. This Management of Hessian fly (Diptera: Cecidomyiidae) in Georgia may explain the differences between the two types of by delayed planting of winter wheat. J. Econ. Entomol. 83:1025– tests. 1033. J Chem Ecol (2009) 35:81–95 95

BUNTIN, G. D., OTT, S. L., and JOHNSON, J. W. 1992. Integration of the attraction of male pea midges, Contarinia pisi, to synthetic sex plant resistance, insecticides, and planting date for management pheromone components. J. Chem. Ecol. 26:1941–1952. of the Hessian fly (Diptera: Cecidomyiidae) in winter wheat. HILLBUR, Y., BENGTSSON, M., LÖFQVIST, J., BIDDLE, A., PILLON, O., J. Econ. Entomol. 85:530–538. PLASS, E., FRANCKE, W., and HALLBERG, E. 2001. A chiral sex CHOI, M.-Y., KHASKIN, G., GRIES, R., GRIES, G., ROITBERG, B. D., pheromone system in the pea midge, Contarinia pisi. J. Chem. RAWORTH, D. A., KIM, D.-H., and BENNETT,R.G.2004.(2R,7S)- Ecol. 27:1391–1407. diacetoxytridecane: sex pheromone of the aphidophagous gall HILLBUR, Y., CELANDER, M., BAUR, R., RAUSHER, S., HAFTMANN, J., midge, Aphidoletes aphidimyza. J. Chem. Ecol. 30:659–670. FRANKE, S., and FRANCKE, W. 2005. Identification of the sex COX, T. S., and HATCHETT, J. H. 1986. Genetic model for wheat/ pheromone of the swede midge, Contarinia nasturtii. J. Chem. Hessian fly (Diptera: Cecidomyiidae) interaction: strategies for Ecol. 31:1807–1828. deployment of resistance genes in wheat cultivars. Environ. HORNER, L., OEDIGER, H., and HOFFMANN, H. 1959. Reaktionen mit Entomol. 15:24–31. Triphenylphosphan-dihalogeniden. Liebigs Ann. Chem. 626:26– EL BOUHSSINI, M., NSARELLAH, M. N., NACHIT, M. M., BENTIKA, A., 34. BENLAHBIB, O., and LHALOUI, S. 1999. First source of resistance IKAN, R. 1982. Chromatography in Organic Microanalysis. Academic, in durum wheat to Hessian fly (Diptera: Cecidomyiidae) in New York. Morocco. Genet. Resour. Crop Evol. 46:107–109. JÖNSSON, M., LINDKVIST, A., and ANDERSON, P. 2005. Behavioral FOSTER, S. P., BERGH, J. C., ROSE, S., and HARRIS, M. O. 1991a. responses in three ichneumonid pollen beetle parasitoids to Aspects of pheromone biosynthesis in the Hessian fly, Mayetiola volatiles emitted from different phenological stages of oilseed destructor (Say). J. Insect Physiol. 37:899–906. rape. Entomol. Exp. Appl. 115:363–369. FOSTER, S. P., HARRIS, M. O., and MILLAR, J. G. 1991b. Identification LINN, C. E. Jr., and ROELOFS, W. L. 1995. Pheromone communication of the sex pheromone of the Hessian fly, Mayetiola destructor in moths and its role in the speciation process, pp. 263–300, in D. (Say). Naturwiss. 78:130–131. M. Lambert, and H. G. Spencer (eds.). Speciation and the FOUQUET, G., and SCHLOSSER, M. 1974. Improved carbon-carbon Recognition Concept: Theory and Application. Johns Hopkins linking by controlled copper catalysis. Angew. Chem. Intern. Ed. University Press, Baltimore. 13:82–83. LÖFSTEDT, C., and ODHAM, G. 1984. Analysis of moth pheromone GAGNÉ, R. J. 1989. The Plant-feeding Gall Midges of North America. acetates by selected ion monitoring using electron impact and Cornell University Press, Ithaca. chemical ionization mass spectrometry. Biomed. Mass Spectrom. GRIES, R., GRIES, G., KHASKIN, G., KING, S., OLFERT, O., KAMINSKI, 11:106–113. L.-A., LAMB, R., and BENNETT, R. 2000. Sex pheromone of MCKAY, P. A., and HATCHETT, J. H. 1984. Mating behavior and orange wheat blossom midge, Sitodiplosis mosellana. Naturwiss. evidence of a female sex pheromone in the Hessian fly, 87:450–454. Mayetiola destructor (Say) (Diptera: Cecidomyiidae). Ann. GRIES, R., KHASKIN, G., GRIES, G., BENNETT, R. G., KING, G. G. S., Entomol. Soc. Am. 77:616–620. MOREWOOD, P., SLESSOR, K. N., and MOREWOOD, W. D. 2002. MILLAR, J. G., FOSTER, S. P., and HARRIS, M. O. 1991. Synthesis of (Z,Z)-4,7-tridecadien-(S)-2-yl acetate: sex pheromone of Douglas- the stereoisomers of the female sex pheromone of the Hessian fly, fir cone gall midge, Contarinia oregonensis. J. Chem. Ecol. Mayetiola destructor. J. Chem. Ecol. 17:2437–2447. 28:2283–2297. RAUSHER, M. D. 2001. Co-evolution and plant resistance to natural GRIES, R., KHASKIN, G., BENNETT, R. G., MIROSHNYCHENKO, A., enemies. Nature. 411:857–864. BURDEN, K., and GRIES, G. 2005. (S,S)-2,12-, (S,S)-2,13-, and (S, ROSSI, R., and CARPITA, A. 1977. Insect pheromones: Stereoselective S)-2,14-diacetoxyheptadecanes: sex pheromone components of reduction of β-orω-alkynols to the corresponding (E)-alkenols red cedar cone midge, Mayetiola thujae. J. Chem. Ecol. 31:2933– by lithium tetrahydroaluminate. Synthesis 8:561–562. 2946. SAMAIN, D., DECOINS, C., and LANGLOIS, Y. 1978. Stéréochimie de HARRIS, M. O., and FOSTER, S. P. 1991. Wind tunnel studies of sex l’alkylation des acétates diéniques conjugées en 2.4 par les pheromone-mediated behavior of the Hessian fly (Diptera: organo-magnésiens en présence de sels de cuivre. Nouveau J. Cecidomyiidae). J. Chem. Ecol. 17:2421–2435. Chim. 2:249–254. HARRIS, M. O., and FOSTER, S. P. 1999. Gall midges, pp. 27–49, in J. SOLINAS, M., and ISIDORO, N. 1996. Functional anatomy of female Hardie, and A. K. Minks (eds.). Pheromones of Non-Lepidopteran sex pheromone gland of Mayetiola destructor Say (Diptera: Insects Associated with Agricultural Plants. CAB International, Cecidomyiidae). Entomologica (Bari) 30:43–54. Oxford, UK. WALL, C., BIDDLE,A.J.,BLOOD SMYTH,J.,FOOT, G. E., GARTHWAITE, HARRIS, M. O., STUART, J. J., MOHAN, M., NAIR, S., LAMB, R. J., and D. G., MORRIS,N.,OWEN,C.R.,STURGEON, D. M., and ROHFRITSCH, O. 2003. Grasses and gall midges: plant defense WEYMAN, G. 1991. Local spatial and temporal distribution of and insect adaptation. Annu. Rev. Entomol. 48:549–577. pea midge, Contarinia pisi. Asp. Appl. Biol.27:355–360. HILLBUR, Y., ANDERSON, P., ARN, H., BENGTSSON, M., LÖFQVIST, J., WILEY, G. A., HERSHKOWITZ, R. L., REIN, B. M., and CHUNG,B.C. BIDDLE, A. J., SMITT, O., HÖGBERG, H.-E., PLASS, E., FRANKE, 1964. Studies in organophosphorus chemistry I. Conversion of S., and FRANCKE, W. 1999. Identification of sex pheromone alcohols and phenols to halides by tertiary phosphine dihalides. components of the pea midge, Contarinia pisi (Diptera: Cecido- J. Am. Chem. Soc. 86:964–965. myiidae). Naturwiss. 86:292–294. YADAV,J.S.,CHANDRASEKHAR,S.,andKACHE,R.1995.Short HILLBUR,Y.,EL-SAYED,A.,BENGTSSON, M., LÖFQVIST,J.,BIDDLE,A., and stereoselective syntheses of pheromone components of PLASS,E.,andFRANCKE, W. 2000. Laboratory and field study of Aproaerema modicella. Synth. Commun. 25:4035–4043.