Introduction

Planococcus minor (Maskell), also known by the common names passionvine mealybug, Pacific mealybug, or guava mealybug, is a serious pest of more than 250 host plants in the Afrotropical, Australasian, Nearctic, Neotropical, and Oriental regions (Biswas and Ghosh 2000; USDA 2000; Venette and Davis 2004). In Taiwan, it is a major pest of economically important crops, including banana, citrus, mango, celery, melon, pumpkin, soybean, betel nut, star fruit, guava, and passionvine.

Planococcus minor and the closely related mealybug, Planococcus citri (Risso), are similar morphologically (Venette and Davis 2004). In a study that tested the synthetic sex pheromone of P. citri (Bierl-Leonhardt et al. 1981) in Taiwan, Tu et al. (1988) reported that some mealybugs thought to be P. citri were not attracted to synthetic sex pheromone. These unresponsive mealybugs were later identified as P. minor (Tu et al. 1988).

The sex pheromones of two Planococcus species, P. citri and Planococcus ficus Signoret, have been identified and are used for management of these pests (Moreno et al. 1984; Millar et al. 2002; Walton et al. 2006). Considering the importance of P. minor as a pest, identification of its sex pheromone would also be useful. We report in this paper the identification of the sex pheromone of P. minor as (E)-2-isopropyl-5-methyl-2,4-hexadienyl acetate.

Methods and Materials

Insects

Mealybugs, originally collected from star fruit (Averhoa carambola L.) in the central area of Taiwan, were reared on pumpkins, Cucurbita moschata Duch. Poiret, at 25 ± 1°C, 70 ± 5% relative humidity under a photoperiod of 12L:12D. The identity of the species was confirmed by Dr. Doug Miller and colleagues at the United States Department of Agriculture in Beltsville, Maryland, USA.

Pumpkins infested with ovisacs (60–70 ovisacs per pumpkin) were placed on a shallow plastic pan in a walk-in incubator. After 16 to 20 d, male nymphs of P. minor were separated by the method of Negishi et al. (1980) and Arai (2000). Briefly, male nymphs were allowed to crawl off the pumpkins and pupate on paper towels that covered the pumpkins. The paper towels were then placed in a container (21.5 cm ID. × 6.5 cm in depth) in an incubator, and adult males allowed to emerge. Adult males, 2–3 d after emergence, were used in bioassays. Virgin females, after the last molt, were transferred to a new pumpkin with a fine paintbrush so as to prevent any mating with males that had not moved onto the paper towels. Each pumpkin was infested with about 1000–2000 females. Females were used for collection of pheromone 5–7 d after the final molt.

Collection of Pheromone Extract

Pumpkins infested with virgin females were placed in a 30 × 30 × 30 cm3 chamber, with purified air passing through the chamber. Because the males’ responses to females were strongest during the first 2 hr of the photophase (Hwang and Chu 1987), the headspace volatiles were collected for the first 4 hr of the photophase on a tube (2.5 cm × 8 mm ID) of Porapak-Q held in place with glass wool plugs. A total of 10,000–40,000 female-day equivalents (hereafter, FDE) of sex pheromone was collected in each tube. Aerations of uninfested pumpkins served as controls. Chemicals adsorbed on the Porapak-Q were extracted with pentane.

Analysis of Pheromone Extracts

Extracts were analyzed by gas chromatography-mass spectrometry (GC-MS) with a ThermoQuest Trace GC interfaced to a Finnigan Trace mass spectrometer (electron impact ionization, 70 eV). The GC oven was held at 40°C for 1 min, then programmed at 10°C min−1 to 250°C. DB-5MS, DB-1, and DB-23 columns (all 30 m × 0.25 mm ID, J & W Scientific, Folsom, CA, USA) were used for analyses. The injection was splitless, with helium as carrier gas. NMR spectra were recorded on Bruker Avance 300 and Bruker DRX500 spectrometers. High-resolution mass spectra (HRMS) were recorded on a JEOL JMS-700 mass spectrometer.

Isolation of Pheromone

High-performance liquid chromatography (HPLC) was employed to isolate the sex pheromone by using an ECOM® Beta 10 Gradient Pump (Prague, Czech Republic), a silica gel column (250 × 4.6 mm, 5 μm, LDC/Milton Roy, Riviera Beach, FL, USA), a Rheodyne Model 7125 injector, and a DAD230 photo diode array detector (set at 200–650 nm) (ChromTech, Singapore). Extracts from collections of infested and uninfested pumpkins were eluted with a solvent gradient of pentane and diethyl ether (1.0 ml/min), starting with 100% pentane for 5 min, and then decreasing to 85% pentane over 75 min with a concomitant increase of diethyl ether. Fractions were collected every minute with a fraction collector. Fractions were concentrated under a stream of nitrogen and analyzed by GC-MS.

Hydrolysis of the Unknown

The fraction containing the unknown (ca. 1 μg) was dissolved in 300 μl of ethanol, and two drops of 1 N NaOH were added. The reaction mixture was kept in a water bath at 35°C for about 30 min before 1 N HCl was added to neutralize the solution. The product was extracted with hexane. The hexane solution was dried by passage through a short plug of anhydrous MgSO4. After concentration under a stream of nitrogen, the product was analyzed by GC-MS.

Hydrogenation of the Unknown

One microgram of the isolated unknown was reduced in a septum-capped vial by catalytic reduction with 5% Pd on carbon and a balloon of H2. After about 30 min, the reaction mixture was filtered through powdered MgSO4 to remove the catalyst. The filtrate was analyzed by GC-MS.

Chemicals

Triphenylphosphine, ethyl 3-methyl-2-oxobutyrate, 1-bromo-3-methyl-2-butene, n-butyl lithium (n-BuLi), tetrahydrofuran (THF), dimethylaminopyridine (DMAP), triethylamine, and acetic anhydride were obtained from Sigma-Aldrich Co. (Milwaukee, WI, USA). Diisobutylaluminum hydride (DIBAL-H) was from Strem Chemicals, Inc. (Newburyport, MA, USA), and 2-isopropyl-5-methylhexyl acetate was purchased from Tokyo Kasei Kogyo Co., Ltd. (Tokyo, Japan).

Laboratory Bioassay

A turntable olfactometer (Tashiro et al. 1969) was used with the modifications described below. The turntable was set up in a room with lighting of 940 ± 120 lx and temperature at 25 ± 2°C, and rotated at 0.67 rpm to reduce any positional effects. Fifteen to 20 sticky traps (6 × 6 cm2) were placed on the wheel of a 2-m-diameter turntable, with treatments or controls placed at the center of the sticky traps. Males to be used in bioassays were kept in the dark from 17:00 of the previous day until 09:00 the day of the bioassay, with bioassays carried out during the first 4 hr of the photophase. Males (300–500 for each replicate) were released from a glass dish (20 cm i.d.) in the center of the turntable. Pieces of Parafilm-wrapped pumpkins (1.5 × 1.5 × 1 cm), with only the pumpkin skin exposed, were put in a plastic tube (3 cm i.d. × 1.5 cm deep) as neutral controls. Pumpkins wrapped in the same way as the neutral controls but infested with 20 females were used as positive controls.

Synthesis of 2-Isopropyl-5-methyl-2,4-hexadienyl acetate (3)

The geometric isomers 3A and 3B were synthesized as shown in Scheme 1. Details of the procedure are described below.

Scheme 1
scheme 1

Synthesis of geometric isomers 3A and 3B

Full size image

Preparation of (3-methyl-2-butenyl)Triphenyl phosphonium Bromide

A mixture of triphenylphosphine (2.62 g, 1 mmol) and 1-bromo-3-methyl-2-butene (1.49 g, 1 mmol) was refluxed overnight in dry toluene (30 ml) under N2. After cooling, the precipitate was filtered and washed with toluene. The crystalline product was dried under vacuum to give the phosphonium salt (4.11 g, yield 100%) as a white solid: mp 236–237°C (mp 235–236°C, Armesto et al. 1995). 1H NMR (CDCl3, 300 MHz): δ 1.26 (3H, d, J = 4.2 Hz), 1.64 (3H, d, J = 6.0 Hz), 4.62 (2H, m), 5.12 (1H, m), 7.88–7.64 (15H, m).

Preparation of Ethyl 2-isopropyl-5-methyl-2,4-hexadienoate (1)

n-BuLi (1.6 M, 6.3 ml, 10 mmol) was added slowly to a solution of (3-methyl-2-butenyl)triphenylphosphonium bromide (4.11 g, 10 mmol) in a 40-ml dry THF under N2 at 0°C, until the solution turned deep red. The reaction mixture was brought to reflux. Ethyl 3-methyl-2-oxobutyrate (1 g, 6.9 mmol in 10 ml of dry THF) was added slowly and the mixture refluxed for another 15 hr. Water (10 ml) was added to the cooled reaction mixture and then 10% aqueous NH4Cl to neutralize the solution. The mixture was extracted with ether (20 ml × 3), and the ether extracts dried with anhydrous MgSO4, filtered, and concentrated. Purification of the residue on a silica gel column (hexane–CH2Cl2 3:1) afforded a 1.3:1 mixture of stereoisomers of ethyl 2-isopropyl-5-methyl-2,4-hexadienoate (1, 0.71 g, yield 72%) determined by 1H NMR spectroscopy. 1H NMR spectrum (CDCl3, 300 MHz) gave two sets of closely related signals: δ 1.10 (7.8H, d, J = 6.9 Hz), 1.20 (6H, d, J = 6.9 Hz), 1.31 (3.9H, t, J = 7.2 Hz), 1.33 (3H, t, J = 7.2 Hz), 1.82 (3.9H, s), 1.84 (3.9H, s), 1.88 (3H, s) 1.89 (3H, s), 2.79 (1.3H, septet, J = 6.9 Hz), 3.03 (1H, septet, J = 6.9 Hz), 4.12–4.29 (4.6H, m), 6.21 (1H, d, J = 12.0 Hz), 6.52 (1.3H, d, J = 12.0 Hz), 6.56 (1.3H, d, J = 12.0 Hz), 7.31 (1H, d, J = 12.0 Hz). The 13C NMR spectrum gave 24 resonances, as expected for the mixture of two geometric isomers: δ 14.0 (2 × C), 17.8, 18.4, 20.8 (2 × C), 21.8 (2 × C), 26.4, 26.6, 27.0, 31.4, 59.6, 59.7, 120.1, 121.9, 129.3, 133.0, 134.1, 135.5, 140.6, 143.6, 167.9, and 168.3.

Preparation of 2-Isopropyl-5-methyl-2,4-hexadienol (2)

DIBAL-H (20% in hexane, 3.6 ml, 4.0 mmol) was added to a solution of the 1.3:1 mixture of ester 1 (294 mg, 1.5 mmol) in CH2Cl2 (15 ml) at −78°C under N2. The reaction mixture was stirred for 30 min and then quenched with −78°C MeOH. Celite (1 g) and Na2SO4·10H2O (1 g) were added and the suspension warmed to room temperature and stirred for another 2 hr (Takano et al. 2001). Filtration and concentration of the resulting mixture, followed by flash chromatography on silica gel (CH2Cl2), gave a 1.3:1 mixture of the geometric isomers of alcohol 2 (217 mg, 94%) as a colorless oil. 1H NMR (CDCl3, 300 MHz): δ 1.08 (13.8H, d, J = 6.9 Hz), 1.78 (7.8H, s), 1.82 (6H, s), 2.49 (1.3H, septet, J = 6.9 Hz), 2.99 (1H, septet, J = 6.9 Hz), 4.16 (2H, s), 4.24 (2.6H, s), 6.23–6.07 (4.6H, m). 13C NMR (CDCl3, 75 MHz): δ 18.1 (2 × C), 21.4 (2 × C), 21.9 (2 × C), 26.4, 26.5, 28.1, 33.8, 59.4, 64.3, 119.9, 120.1, 121.1, 122.2, 135.7, 135.9, 143.2, 144.0.

Preparation of 2-Isopropyl-5-methyl-2,4-hexadienyl acetate (3A and 3B)

To a solution of compound 2 (123 mg, 0.8 mmol) in CH2Cl2 (10 ml) was added acetic anhydride (204 mg, 2.0 mmol), triethylamine (202 mg, 2.0 mmol), and 4-dimethylaminopyridine (1.3 mg, 0.01 mmol) at room temperature. The solution was stirred for 30 min and then quenched with water. The reaction mixture was extracted with CH2Cl2 (10 ml × 3). Extracts were combined, dried over MgSO4, filtered, and concentrated in vacuo (Takano et al. 2001). Purification on a silica gel column afforded compound 3 (139 mg, 89%) as a mixture of two isomers. The mixture was separated by HPLC, with the same gradient program as that used for isolation of pheromone from the aeration collections, collecting fractions every 0.25 min. Compound 3A gave a 1H NMR (CDCl3, 300 MHz) spectrum as follows: δ1.07 (6H, d, J = 6.9 Hz, 2-CH(CH 3)2), 1.79 (3H, s, 5-CH 3), 1.82 (3H, s, H-6), 2.07 (3H, s, 1-OCOCH 3), 2.44 (1H, septet, J = 6.9 Hz, 2-CH(CH3)2), 4.72 (2H, s, H-1), 6.08 (1H, d, J = 11.7 Hz, H-4), 6.29 (1H, d, J = 11.7 Hz, H-3). 13C NMR (CDCl3, 75 MHz) of compound 3A: δ18.2 (5-CH3), 21.1 (1-OCOCH3), 21.8 (2-CH(CH3)2), 26.5 (C-6), 33.8 (2-CH(CH3)2), 61.1 (C-1), 120.0 (C-4), 124.4 (C-3), 136.9 (C-5), 138.3 (C-2), 171.3 (1-OCOCH3). MS (m/z; %): 196 (5, M+), 121 (100); HRMS m/z 196.1462 [M]+ (calculated for C12H20O2 196.1463); UV (CH2Cl2) λ max (log ɛ) 248 (4.9) nm. Compound 3B 1H NMR (CDCl3, 300 MHz): δ1.07 (6H, d, J = 6.9 Hz, 2-CH(CH 3)2), 1.79 (3H, s, 5-CH 3), 1.82 (3H, s, H-6), 2.07 (3H, s, 1-OCOCH3), 3.01 (1H, septet, J = 6.9 Hz, 2-CH(CH3)2), 4.60 (2H, s, H-1), 6.08 (1H, d, J = 11.7 Hz, H-4), 6.23 (1H, d, J = 11.7 Hz, H-3). 13C NMR (CDCl3, 75 MHz): δ 18.2 (5-CH3), 21.2 (1-OCOCH3 and 2-CH(CH3)2), 26.5 (C-6), 28.1 (2-CH(CH3)2), 66.6 (C-1), 119.8 (C-4), 125.1 (C-3), 137.3 (C-5), 137.6 (C-2), 170.9 (1-OCOCH3). MS m/z (%): 196 (5, M+), 121 (100); HRMS m/z 196.1460 [M]+ (calculated for C12H20O2 196.1463); UV (CH2Cl2) λ max (logɛ) 248 (4.4) nm. Neat compound 3 gave diagnostic IR absorptions at 1707 (C=O), 1637 cm−1 (C=C).

Results

Analysis

Comparison of the extracts of volatiles from mealybug-infested and uninfested pumpkins revealed one significant insect-produced compound, which was isolated as a single component of the fraction eluting at 50 min by HPLC. The Kovats Indices (KI) of this unknown were 1687 on the DB-23 column, 1347 on the DB-1 column, and 1350 on the DB-5MS column. In the bioassay, a 100-ng sample of the isolated compound attracted a mean (N = 6) of 25.3 male mealybugs, whereas 20 females attracted a mean (N = 6) of 3.2 male mealybugs, suggesting that the unknown was active. The mass spectrum of the compound (Fig. 1) showed significant fragments at m/z (%): 43 (18), 69 (7), 91 (13), 93 (18), 105 (13), 121 (100), 123 (11), 136 (16), with a small molecular ion at 196 (5). The presence of significant peaks at m/z 43 (CH3C≡O+) and 136 (M+-CH3COOH) and the absence of other prominent fragmentations between m/z 196 and 136, suggested that the unknown was an acetate ester. After base hydrolysis, a compound with molecular weight m/z 154 was found in the reaction mixture. The mass spectrum (Fig. 2) of the hydrolyzed product showed fragment ions with m/z (%): 43 (34), 55 (38), 69 (34), 93 (46), 111 (100), and 123 (54), and a molecular ion of 154 (50). The molecular weight of the hydrolyzed product was 42 mass units less than that of the unknown, supporting our hypothesis that the unknown was an acetate ester.

Fig. 1
figure 1

EI mass spectrum (70 eV) of the active component in pheromone collections from female Planococcus minor

Full size image
Fig. 2
figure 2

EI mass spectrum (70 eV) of hydrolyzed active component in pheromone collections from female Planococcus minor

Full size image

The mass spectrum (Fig. 3) of the hydrogenated product showed fragment ions with m/z (%): 43 (100), 61 (8), 69 (96), 84 (58), 97 (38), 112 (11), 140 (14), and 157 (6). Comparison of the mass spectrum of the hydrogenated product with the NIST mass spectral database suggested a possible structure for the hydrogenated product as 2-isopropyl-5-methylhexyl acetate. This was confirmed by comparing the GC retention time and the mass spectrum of the hydrogenated compound from the insect with that of an authentic standard. These results indicated that the unknown had a carbon skeleton of 2-isopropyl-5-methylhexyl acetate with two sites of unsaturation.

Fig. 3
figure 3

EI mass spectrum (70 eV) of the hydrogenation product of the active component in pheromone collections from female Planococcus minor

Full size image

HPLC collection yielded about 3 μg of the unknown from 37,000 FDE of crude extract. The 1H NMR spectrum of the unknown showed signals at δ 1.79 (3H, s), 1.82 (3H, s), 2.07 (3H, s), 4.60 (2H, s), 6.08 (1H, d, J = 11.7 Hz), and 6.23 (1H, d, J = 11.7 Hz). The six-proton singlet at δ 1.79 and 1.82 (two allylic methyl groups), typical acetyl protons (δ 2.07), oxygenated methylene protons (δ 4.60), and two vinylic protons mutually coupled with a coupling constant of 11.7 Hz, suggested a skeleton of 2-substituted-5-methyl-2,4-hexadienyl acetate. In addition, an isopropyl moiety was revealed by the signals at δ 1.07 (6H, d, J = 6.9 Hz) and 3.01 (1H, septet, J = 6.9 Hz). On the basis of the above, the unknown compound was identified as 2-isopropyl-5-methyl-2,4-hexadienyl acetate.

Synthesis

The stereochemistry of the unknown compound was determined by a non-stereoselective synthesis followed by separation of the (Z)- and (E)-2-isopropyl-5-methyl-2,4-hexadienyl acetates (3A and 3B). Thus, Wittig reaction between ethyl 3-methyl-2-oxobutyrate and the ylide from (3-methyl-2-butenyl)triphenylphosphonium bromide (formed by addition of n-BuLi) produced an isomeric mixture of 1 in a 1.3:1 ratio. GC-MS analysis of the mixture indicated that the two components had KIs of 1639 and 1651, respectively, on the DB-23 column. The two compounds exhibited similar mass fragmentation patterns, with fragments of m/z (%) 81 (28), 91 (21), 107 (49), 123 (100), 135 (12), 153 (25), 181 (28), and a molecular ion at 196 (41). Reduction of compound 1 with DIBAL-H yielded alcohols 2, with KIs of 1691 and 1733, respectively, on the DB-23 column. The two isomers gave similar mass fragmentation patterns, with m/z (%) of 69 (40), 81 (34), 91 (36), 93 (42), 111 (100), 125 (55), and 154 (M+, 53). The mixture was acetylated to produce the acetates 3 as a (Z)- and (E)- mixture with KIs of 1666 and 1687, respectively, on the DB-23 column. Compound 3A eluted at 49.6 min and compound 3B eluted at 50.3 min from the silica HPLC column. The KI, mass spectrum, and 1H NMR data of compound 3B were the same as that of the active compound isolated from the insect extracts. The stereochemistry of 3A and 3B, and the full assignment of 1H and 13C NMR signals, were obtained from 2D-COSY, NOESY, HMQC, and HMBC spectra as described below.

In 3A, HMBC correlations were observed between C-1 (δ 61.1) and the proton at δ 6.29, as well as C-1 and the methine proton (δ 2.44) of the isopropyl group, indicating that the more downfield vinylic proton at δ 6.29 was H-3. In addition, the 2D NOESY NMR spectrum of 3A (Fig. 4) indicated that the vinylic proton with a chemical shift of δ 6.08 had cross couplings with H-1 (δ 4.72) and H-6 (δ 1.82). The vinylic proton with a chemical shift of δ 6.29 had cross coupling with the isopropyl group (δ 1.07 and δ 2.44) and the 5-methyl group (δ 1.79). These NOE correlations suggested the configuration of the double bond in this isomer was Z. Thus, the structure of compound 3A was assigned as (Z)-2-isopropyl-5- methyl-2,4-hexadienyl acetate. The NOE effects are summarized in Fig. 4.

Fig. 4
figure 4

2D-NOESY NMR spectrum and the key NOE correlations (shown with dashed curves) of compound 3A

Full size image

Similarly, in 3B, HMBC correlations were observed between C-1 (δ 66.6) and the proton at δ 6.23, as well as C-1 and the methine proton at δ 3.01 of the isopropyl group, suggesting that the more downfield vinylic proton at δ 6.23 was H-3. In addition, the 2D NOESY NMR spectrum of 3B (Fig. 5) indicated that the vinylic proton with chemical shift of δ 6.08 had cross couplings with the methine proton (δ 3.01) of the isopropyl group and H-6 (δ 1.82), but not with the methyls of the isopropyl group; presumably, the two methyl groups were turned away from the vinyl proton because of steric repulsion. These data indicated that the alkenyl group (C4-C6) was cis to the isopropyl group. The vinylic proton with a chemical shift of δ 6.23 had cross couplings with the 5-methyl group (δ 1.79) and H-1 (δ 4.6). Thus, the downfield vinyl proton must be cis to the methylene proton. Thus, the structure of compound 3B was assigned as (E)-2-isopropyl-5-methyl-2,4-hexadienyl acetate. The NOE effects are summarized in Fig. 5.

Fig. 5
figure 5

2D-NOESY NMR spectrum and the key NOE correlations (shown with dashed curves) of compound 3B

Full size image

As a result of the matching NMR spectra, KIs on GC, and mass spectra, the pheromone of the mealybug P. minor was identified as (E)-2-isopropyl-5- methyl-2,4-hexadienyl acetate.

Bioassay of Synthetic Pheromone

The results for bioassays of the synthetic materials vs. the compound from volatiles of female mealybugs, live females, and controls are shown in Table 1. Synthetic (E)-2-Isopropyl-5-methyl-2,4-hexadienyl acetate was more attractive (P < 0.05) to male mealybugs than any of the other treatments. Interestingly, the other stereoisomer appeared to antagonize attraction, with the mixture of isomers (59% of 3A [Z-isomer] and 41% of compound 3B [E-isomer]) being unattractive (i.e., not significantly different from the number attracted to the uninfested pumpkin control) to males.

Table 1 Number of male mealybugs attracted to various treatments
Full size table

Discussion

We identified the sex pheromone for the passionvine mealybug, P. minor, a major pest of economically important crops. Male mealybugs were attracted to the (E) isomer of 2-isopropyl-5-methyl-2,4-hexadienyl acetate. That male bugs were attracted in greater numbers to the pure (E)-isomer, rather than to a mixture of (E)- and (Z)-isomers, suggests that a stereospecific synthesis of the (E)-isomer will be required for practical use of this pheromone.

The structure of the sex pheromone compound of P. minor represents another example of the irregular terpenoid motif of the sex pheromones of pseudococcid species. The sex pheromone of Pseudococcus viburni (Millar et al., 2005) is a monoterpene incorporating a cyclopentane unit. The sex pheromones of Pseudococcus cryptus (Arai et al. 2003) and P. citri (Bierl-Leonhardt et al. 1981) are monoterpenes with a cyclobutane moiety. The sex pheromone of P. ficus (Hinkens et al. 2001) is an acyclic monoterpene with the same basic carbon skeleton as the compound we identified here. All of these species appear to use single components as attractants.

Further work will seek to develop a stereospecific synthesis of (E)- 2-isopropyl-5-methyl-2,4-hexadienyl acetate, as well as determine optimum doses and longevity of synthetic pheromone lures.