PHEROMONES of MILKWEED BUGS (HETEROPTERA: LYGAEIDAE) ATTRACT WAYWARD PLANT BUGS: Phytocoris MIRID SEX PHEROMONE1

P1: GCR Journal of Chemical Ecology [joec] pp900-joec-467953 June 20, 2003 17:36 Style ﬁle version June 28th, 2002

Journal of Chemical Ecology, Vol. 29, No. 8, August 2003 (C 2003)

PHEROMONES OF MILKWEED BUGS (HETEROPTERA: LYGAEIDAE) ATTRACT WAYWARD PLANT BUGS: Phytocoris MIRID SEX PHEROMONE1

QING-HE ZHANG2,3 and JEFFREY R. ALDRICH3,

2Department of Entomology University of Maryland College Park, Maryland 20742, USA 3USDA-ARS Chemicals Affecting Insect Behavior Laboratory B-007, BARC-West Beltsville, Maryland 20705, USA

(Received November 21, 2002; accepted March 24, 2003)

Abstract—The synthetic aggregation pheromone of the large milkweed bug, Oncopeltus fasciatus (Dallas) (Lygaeinae), also attracted males of the plant bug, Phytocoris difficilis Knight (Miridae). Field testing partial blends against the six-component blend comprising the Oncopeltus pheromone showed that cross- attraction of P. difficilis males was due to synergism between (E)-2-octenyl acetate and (E,E)-2,4-hexadienyl acetate. Hexyl acetate was abundant in the metathoracic scent gland (MSG) secretion of P. difficilis males, but because female P. difficilis could not initially be found in the field, further combinato- rial tests were guided by prior research on the pheromones of two Phytocoris species in the western United States. The combination of hexyl, (E)-2-hexenyl, and (E)-2-octenyl acetates was as attractive to P.difficilis males as the milkweed bug pheromone, yet no milkweed bugs were drawn to this blend. Gas chromatographic (GC)-electroantennographic detection (EAD) and GC-mass spectromet- ric (MS) analyses of female P. difficilis MSGs determined that their secretion contained predominantly hexyl, (E)-2-hexenyl, and (E)-2-octenyl acetates (all strongly EAD-active)—the latter two compounds found only in trace amounts from males—plus five minor female-specific compounds, three of which were EAD-active. (E,E)-2,4-Hexadienyl acetate was not detected from P.difficilis females or males. The blend of the three major components, hexyl, (E)-2-hexenyl, and (E)-2-octenyl acetates (2:1.5:1 by volume), was as attractive as the blend of all six EAD-active compounds identified from females, indicating that this ternary blend constitutes the sex pheromone of P. difficilis. Hexyl acetate with (E)-2-octenyl acetate also attracted males of another species, P. breviusculus

1Mention of commercial products does not constitute an endorsement by USDA. To whom correspondence should be addressed. E-mail: [email protected]

1835

0098-0331/03/0800-1835/0 C 2003 Plenum Publishing Corporation P1: GCR Journal of Chemical Ecology [joec] pp900-joec-467953 June 20, 2003 17:36 Style ﬁle version June 28th, 2002

1836 ZHANG AND ALDRICH

Reuter, but addition of (E)-2-hexenyl acetate and/or (E,E)-2,4-hexadienyl acetate inhibited attraction of P.breviusculus males. Attraction of P.difﬁcilis males occurred mainly during the ﬁrst half of scotophase. The possible neurophysio- logical basis for this asymmetrical cross-attraction is discussed.

Key Words—Miridae, Phytocoris, milkweed bug, sex pheromone, sex-attractant, metathoracic scent gland, hexyl acetate, (E)-2-hexenyl acetate, (E)-2-octenyl acetate, (E,E)-2,4-hexadienyl acetate, cross-attraction.

INTRODUCTION

In the course of testing synthetic pheromone blends for the large milkweed bug [Oncopeltus fasciatus (Dallas), Lygaeidae: Lygaeinae] (Aldrich et al., 1999), an interesting discovery was made: males of the plant bug Phytocoris difficilis Knight (Miridae) were also highly attracted. Most of these mirids were caught in traps near the edge of an oak-pine mixed forest in midsummer and early fall. This finding revealed an interesting cross-attraction between species in distantly related families (Schuh and Slater, 1995), and provided strong evidence for existence of a female- produced sex pheromone in P. difficilis. The Miridae, known as plant bugs, comprise the largest family of the Heteroptera, with over 1200 genera, including about 10,000 described species (Schuh and Slater, 1995). In North America alone, 223 genera and 1930 species have been catalogued (Henry and Wheeler, 1988). Although some mirids are preda- cious (Stonedahl, 1988; Wheeler, 2001), most plant bugs are phytophagous, feed- ing on flowers, fruit, and meristematic tissue, causing plant deformation or fruit abscission. Because of their broad host range, many phytophagous mirids are se- rious pests of a wide variety of crops throughout the world. Yet, in spite of their importance as pests, methods for sampling and monitoring mirid populations are relatively primitive, relying on the time- and labor-intensive methods of beating- tray- or sweep-net sampling (Ho and Millar, 2002). Pheromone-baited traps are an important tool for detection and monitoring of many agricultural and forest insect pests. The identification of pheromones for mirids would faciliate the scouting and timing of control measures required in crops attacked by these bugs (McBrien et al., 1994a,b). Mirid sex pheromones are also potentially useful for direct control through mating disruption or mass trapping (McBrien et al., 1996, 1997). Although repeated efforts have been made to identify the sex pheromones of various mirids, especially lygus bugs (Aldrich et al., 1988; Ho and Millar, 2002), sex pheromones have only been identified for six species thus far: Campylomma verbasci (Meyer) (Smith et al., 1991), Phy- tocoris relativus Knight (Millar et al., 1997), P. californicus Knight (Millar and Rice, 1998), Trigonotylus caelestialium (Kirkaldy) (Kakizaki and Sugie, 2001), Distantiella theobroma Distant, and Sahlbergella singularis Haglund (Downham et al., 2002, and personal communication). P1: GCR Journal of Chemical Ecology [joec] pp900-joec-467953 June 20, 2003 17:36 Style file version June 28th, 2002

Phytocoris PHEROMONE 1837

The objectives of the work described here were 1) to determine which components from the milkweed bug pheromone were responsible for the cross-attraction of P. difficilis, 2) to collect live Phytocoris females for pheromone analyses, and 3) to monitor the diurnal flight activity of P. difficilis males.

METHODS AND MATERIALS

Adult Insects and Preparation of Extracts. Phytocoris difficilis males were collected from milkweed bug pheromone-baited bucket traps deployed in an oak- pine mixed forest, on the South Farm of the Beltsville Agricultural Research Center (BARC), Prince George’s County, Maryland (Aldrich et al., 1999). Two female adults were collected in the same woods from a white sheet (1.5 m 10 m) illuminated by a battery-driven black-light on September 9 and October 2, 2002. A few males of another mirid species, Phytocoris breviusculus Reuter, were also collected from the lighted sheet. All bugs were dissected within 24 hr of capture for extraction of the metathoracic scent gland (MSG) and for electrophysiological and chemical analyses. MSGs from both males and females of P.difficilis and from P. breviusculus males were excised from CO2-anesthetized bugs submerged in tap water, and the glands were extracted individually in 30 l of methyl-tert-butyl ether. No aerations of live insects were carried out because of the low number of females available. Gas Chromatographic–Electroantennographic Detection (GC-EAD) Analy- sis. MSG extracts were analyzed using an Hewlett-Packard 6890 GC equipped with a DB-WaxETR column (30 m 0.25 mm 0.25 m; splitless mode; J & W Scientific, Folsom, CA), and a 1:1 effluent splitter that allowed simultaneous flame ionization detection (FID) and EAD of the separated volatile compounds. Helium was used as the carrier gas, and the injector temperature was 220C. The column temperature was 50C/2 min, rising to 240Cat10C/min, then held for 10 min. The outlet for the EAD was held in a humidified airstream flowing at 0.5 m/sec over a male antennal preparation. A glass capillary indifferent electrode was filled with Beadle–Ephrussi Ringer (Zhang et al., 2000), grounded via a silver wire, and inserted into the open side of the severed mirid head. A similar recording electrode connected to a high-impedance DC amplifier with automatic baseline drift compensation was inserted over the distal ends of both antennae (the tip of each antenna was cut off). The antennal signals were stored and analyzed on a PC equipped with a serial IDAC interface box and the program EAD ver. 2.5 (Syntech, Hilversum, The Netherlands). In addition, EAD responses for males of each mirid species to a synthetic mixture (3 l/injection), containing seven acetates and two butyrates (10 ng/l each) were recorded. For each sample (both extracts and synthetic mixture), at least three pairs of antennae from each mirid species were tested. Only those that elicited electrophysiological responses P1: GCR Journal of Chemical Ecology [joec] pp900-joec-467953 June 20, 2003 17:36 Style file version June 28th, 2002

1838 ZHANG AND ALDRICH

obviously different from background noise in at least two antennal preparations were considered as stimulatory to test antennae. Compound Identification. The following authentic standards were obtained from commercial sources or were synthesized: 1-hexanol, (E)-2-hexenol, nonanal, decanal, hexyl acetate (HA), octyl acetate (OA), and octyl butyrate (Aldrich Chem- ical Co., Milwaukee, WI); (E)-2-octenyl acetate (2OA), hexyl butyrate (HB), (E)-2-hexenyl butyrate (2HB), (E)-2-heptenyl acetate (2HPA), (E)-2-hexenyl butyrate (2HB), and (E)-2-nonenol (Bedoukian Research, Inc., Danbury, CT); hep- tyl acetate (Eastman Organics); 3-hydroxy-2-butanone (TCI America) and (Z)-3- octenyl acetate (J. G. Millar, University of California, Riverside). Butyl butyrate, (E)-2-hexenyl acetate (2HA), (E,E)-2,4-hexadienyl acetate (2,4HA), (E)-2,5- hexadienyl acetate (2,5HA), and (E)-2,7-octadienyl acetate (2,7OA) were synthesized as previously described (Aldrich et al., 1997, 1999). MSG samples were analyzed by gas chromatography-mass spectrometry (GC-MS) on an HP 6890 GC series coupled with an HP 5973 Mass Selective Detector by using the same type of GC column and conditions as described above. Compounds were identified by comparison of retention times with those of authentic standards and with mass spectra of standards. Field Trapping. All field-trapping experiments were carried out in an oak- pine mixed forest on the BARC South Farm from early July through the fall of 2002, using Jackson Delta sticky traps (Agrisense, Fresno, CA) baited with 15–30 l of individual or mixed neat test compounds loaded onto grey rubber septa (5 mm sleeve-type, The West Co., Lititz, PA). Rubber septa dispensers were replaced every week. Traps were hung on the trunk or branches of either oak or pine trees in lines at a height of 1.8–2.0 m, with a spacing of about 10 m between traps within each trap line, and about 15 m between trap lines. For each experiment, two sets of traps were deployed with their initial trap positions randomized. The trap positions were then systematically rotated within the same trap set after each replicate, based on a procedure of Latin-square design (Byers, 1991), so that each trap appeared in each location at least once. The bug collections and trap rotations were carried out when 1 mirid was caught in the best traps, to minimize positional effect. Each replicate lasted 1–3 days, depending on flight activity (i.e., numbers of bugs caught in the best traps). Experiments 1–3 were initiated before the Phytocoris females were analyzed, on the basis of earlier cross-attraction data (J. R. Aldrich, unpublished) and the published results on the pheromones of other Phytocoris species (Millar et al., 1997; Millar and Rice, 1998), whereas experiments 4–5 were carried out after analysis of the first female P. difficilis MSG extract. Experiment 1 was conducted to determine in a subtractive manner which components in the milkweed bug pheromone blend, consisting of 2HA, 2,4HA, 2,5HA, 2HPA, 2OA, and 2,7OA, were responsible for the cross-family attraction. Experiment 2 tested the potential activity of hexyl acetate, (E)-2-hexenyl acetate, and (E)-2-octenyl acetates in a three-way factorial design, i.e., three individual P1: GCR Journal of Chemical Ecology [joec] pp900-joec-467953 June 20, 2003 17:36 Style file version June 28th, 2002

Phytocoris PHEROMONE 1839

components and all their possible binary/ternary blends (in the same ratio). Hexyl acetate was used because it is an essential pheromone component of two Phy- tocoris spp. in the western United States, and is produced by both sexes of each species (Millar et al., 1997; Millar and Rice, 1998). (E)-2-Hexenyl acetate and (E)- 2-octenyl acetate are female-specific compounds of the western Phytocoris spp. (Millar et al., 1997; Millar and Rice, 1998), and are also present in the milkweed bug pheromone (Aldrich et al., 1999). Experiment 3 was similar to experiment 2, but with one more milkweed bug pheromone compound, (E,E)-2,4-hexadienyl acetate, involved in the full factorial design. In addition, a milkweed bug pheromone blend (30 l) was also added to this experiment as a positive control. Exper- iment 4 tested the complete reconstruction of the EAD-active female-specific MSG volatiles (“full blend”), subtraction of each EAD-active component from the full blend, and a blend of three major strongly EAD-active components, and the milkweed bug pheromone blend. In experiment 5, the diurnal flight activity of P. difficilis males was monitored with attractant-baited traps on three occasions in late September and the beginning of October 2002 by checking the traps each hour from late afternoon until midnight. Statistical Analysis. Because of strong heterogeneity of variances among treatments, trap catch data (number of males caught /trap/replicate) were analyzed using the nonparametric Kruskal–Wallis ANOVA on rank test, followed by the Student–Newman–Kuels all pairwise comparison to separate means (Zar, 1984).

RESULTS

Field Experiments before Analysis of Female P. difficilis. In experiment 1, a total of 262 P. difficilis males were caught from July 10 to September 18, 2002, and no P. breviusculus were trapped. Traps baited with the six-component pheromone blend of milkweed bugs caught significantly more mirid males than the blank control (P < 0.05; Figure 1). Subtraction of (E)-2-hexenyl acetate, (E)- 2,5-hexadienyl acetate, (E)-2-heptenyl acetate, or (E)-2,7-octadienyl acetate from the six-component blend had no effect on the trap catches. However, deletion of (E, E)-2,4-hexadienyl acetate or (E)-2-octenyl acetate from the blend eliminated attraction of P. difficilis males (Figure 1). In experiment 2 conducted from July 20 to September 26, 2002, none of the three individual components tested alone (hexyl acetate, (E)-2-hexenyl acetate, and (E)-2-octenyl acetate) were attractive to P. difficilis males, nor were their binary combinations. However, the ternary blend attracted significant numbers of male bugs (Figure 2). A binary combination of hexyl acetate and (E)-2-octenyl acetate also attracted males of another mirid species, P. breviusculus, and addition of (E)-2-hexenyl acetate to this binary blend reduced the attraction of this mirid to a level not different from the blank (Figure 2). P1: GCR Journal of Chemical Ecology [joec] pp900-joec-467953 June 20, 2003 17:36 Style file version June 28th, 2002

1840 ZHANG AND ALDRICH

FIG. 1. Captures of male Phytocoris difﬁcilis in traps baited with the synthetic pheromone of the large milkweed bug (MWB), Oncopeltus fasciatus, and deletion blends. Means followed by the same letter are not signiﬁcantly different (P > 0.05), Kruskal–Wallis ANOVA on ranks, followed by Student–Newman–Keuls all pairwise comparison. 2HA = (E)-2-hexenyl acetate; 2,4HA = (E, E)-2,4-hexadienyl acetate; 2,5HA = (E)-2,5-hexadienyl acetate; 2HPA = (E)-2-heptenyl acetate; 2OA = (E)-2-octenyl acetate; 2,7OA = (E)-2,7-octadienyl acetate.

In experiment 3 (September 6–October 8, 2002), the combination of hexyl acetate, (E)-2-hexenyl acetate, and (E)-2-octenyl acetate again resulted in significant attraction to P. difficilis males, with trap catches similar to those baited with the milkweed bug pheromone blend (MWB blend) (Figure 3). Individual compounds and binary blends were not attractive except for the mixture of (E,E)-2,4- hexadienyl acetate and (E)-2-octenyl acetate. All three- or four-component blends containing (E)-2-octenyl acetate caught significantly more P. difficilis males than the blank control, and there were no differences in trap catches among these treatments. However, any blends that were missing (E)-2-octenyl acetate were not attractive. Interestingly, (E, E)-2,4-hexadienyl acetate, one of the two key components responsible for the cross-attraction in experiment 1, showed a strong synergism with (E)-2-octenyl acetate. In fact, (E, E)-2,4-hexadienyl acetate could substitute for either hexyl acetate or (E)-2-hexenyl acetate or both in the three- component blend to achieve the same level of attraction (Figure 3). As shown in P1: GCR Journal of Chemical Ecology [joec] pp900-joec-467953 June 20, 2003 17:36 Style file version June 28th, 2002

Phytocoris PHEROMONE 1841

FIG. 2. Captures of male Phytocoris difﬁcilis and P. breviusculus in traps baited with HA, 2HA, or 2OA, and their binary and ternary blends. Means followed by the same letter are not signiﬁcantly different (P > 0.05), Kruskal–Wallis ANOVA on ranks, followed by Student–Newman–Keuls all pairwise comparison. Abbreviations as in Table 1 and Figure 1.

experiment 2, the binary blend of hexyl acetate and (E)-2-octenyl acetate was again attractive to P. breviusculus males, and addition of (E)-2-hexenyl acetate and/or (E, E)-2,4-hexadienyl acetate to this blend totally shut off attraction (Figure 3). Milkweed bugs were captured only in the traps baited with the milkweed bug pheromone. GC-EAD and GC-MS Analyses. In general, antennal preparations of both mirid species remained active for at least 2 hr, which allowed time for 3–4 GC-EAD experiments per preparation. GC-EAD analyses of MSG extracts from two female P. difficilis showed an identical pattern in GC profiles and male EAD responses. Antennae of P.difficilis males responded strongly to three major components of female MSG extracts that were identified by GC-MS as hexyl acetate, (E)-2-hexenyl acetate, and (E)-2-octenyl acetate (with ratio of ca. 2:1.5:1) (Figure 4 and Table 1). Weak but reproducible responses to several minor or trace components (e.g., hexyl butyrate, (E)-2-heptenyl acetate, octyl acetate, and (E)-2,7-octadienyl acetate) were also detected (Figure 4). In male MSG extracts, there were only two major volatile components, hexyl acetate and hexyl butyrate, each of which elicited strong responses from male antennae (Figure 4, Table 1). Consistent EAD responses were also detected to three minor or barely FID-detectable components in extracts from males, (E)-2-hexenyl acetate, (E)-2-hexenyl butyrate, and (E)-2-octenyl acetate. Females of P. breviusculus were not available for comparison; however, P1: GCR Journal of Chemical Ecology [joec] pp900-joec-467953 June 20, 2003 17:36 Style file version June 28th, 2002

1842 ZHANG AND ALDRICH

FIG. 3. Captures of male Phytocoris difficilis and P. breviusculus in traps baited with HA, 2HA, 2OA, or 2,4HA, and all possible two- to four-way combination blends. Milkweed bug pheromone (MWB) blend was included as a positive control. Means followed by the same letter are not significantly different (P > 0.05), Kruskal–Wallis ANOVA on ranks, followed by Student–Newman–Keuls all pairwise comparison. ***Significantly different from all other treatments. Abbreviations as in Table 1 and Figure 1.

GC-EAD/MS analyses of MSG extracts from males of this species showed a similar volatile profile to that of P.difficilis males, with hexyl acetate and hexyl butyrate being the major, strongly EAD-active volatile components (Table 1). No antennal responses to other minor components were detected. Analysis of a synthetic mixture of seven acetates and two butyrates indicated that male antennae of both mirid species responded strongly to all nine compounds at a dose of about 15 ng for each compound. This mixture included not only those compounds found in the mirid MSG (males or females or both), but also two milkweed bug pheromone components, (E, E)-2,4-hexadienyl acetate and (E)-2,5- hexadienyl acetate, which were not detected from these mirids (data not shown). Further recordings of this mixture coinjected with (E)-2-hexenyl (E)-2-hexenoate showed that antennae of both species strongly responded to acetates and butyrates, but not to the hexenoate ester. Field Experiments after Analysis of Female P. difficilis. In experiment 4 (September 21–October 8, 2002), the reconstructed EAD-active six-component “full blend” of female MSG volatiles attracted significant numbers of P. difficilis males. However, the number of mirid males caught in traps baited with the full blend was not different from captures in traps baited with either the three-component blend of hexyl acetate, (E)-2-hexenyl acetate, and (E)-2-octenyl acetate, or the P1: GCR Journal of Chemical Ecology [joec] pp900-joec-467953 June 20, 2003 17:36 Style file version June 28th, 2002

Phytocoris PHEROMONE 1843

FIG. 4. GC-EAD responses of Phytocoris difﬁcilis male antennae to metathoracic scent gland extracts of conspeciﬁc males and females. HB = hexyl butyrate, 2HB = (E)-2- hexenyl butyrate. Abbreviations of other chemical names as in Table 1 and Figure 1.

MWB blend (Figure 5). Deletion of any of the three major components, hexyl acetate, (E)-2-hexenyl acetate, or (E)-2-octenyl acetate, from this full blend resulted in reduction of trap catches to the level of the blank control, indicating that all three compounds are components of the P. difficilis sex pheromone. In contrast, subtraction of any or all of the minor components, (E)-2-heptenyl acetate, octyl acetate, or (E)-2,7-octadienyl acetate, had no effect on the attractiveness of baits (Figure 5). Only the blend from which (E)-2-hexenyl acetate had been deleted was attractive to P. breviusculus males (Figure 5). No milkweed bugs were caught in traps baited with partial, full, or three-component blends attractive to Phytocoris spp., while the MWB blend was still attractive to milkweed bugs. Diurnal Flight Activity of P. difficilis. During late September and early October 2002, attractant-baited traps in the experimental area were checked at 1-hr intervals from late afternoon until midnight. Data from 3 days’ observations were pooled to calculate the cumulative relative catch (Figure 6). Male flight of P. difficilis to sex-attractant traps started around 18:00 when sunset began, with peak flight occurring between 20:00 and 22:00, and flight stopped sometime after P1: GCR Journal of Chemical Ecology [joec] pp900-joec-467953 June 20, 2003 17:36 Style file version June 28th, 2002

1844 ZHANG AND ALDRICH c b P. breviusculus 8 2 1 6 2 4 5 3 ...... Strongly active (0.15– 1 2 0 0 1 0 0 0 SE activity AND 6 9 3 5 9 9 8 3 ...... 2 1 2 (%) 27 57 c Phytocoris difficilis Male EAD Relative amounts Male EAD 3) activity 24 48 03 05 01 00 0306 0 5 09 0 03 01 . . . . . = ...... 0 1 0 1 0 0 0 0 0 0 0 N SE 7 0 . . 04 23 60 13 10 04 03 23 08 ...... DENTIFIED FROM I P. difficilis P. breviusculus males Repeatably active at level of 0.05–0.14 mV; 2) Males ( = 15 29 604 0 0105 68 05 1505 0 0103 0 0 01 05 ...... 0 1 0 0 0 1 0 0 0 0 0 N OMPONENTS 7 C . Relative amount (%) 10 48 41 45 05 04 35 45 35 15 ...... Females ( LAND G CENT S a ETATHORACIC Chemicals 1. M Tentatively identified based on MS data; standard not obtained. Compounds in bold type are EAD-active. = ABLE )-2-Hexenol 0 )-3-Octenyl acetate 0 Occasionally (weak) active, not repeatable (0–0.02 mV); )-2-Heptenyl acetate (2HPA))-2-Hexenyl butyrate (2HB))-2-Octenyl acetate (2OA))-2, 0 7-Octadienyl acetate (2,7OA) 0 22 1 )-2-Hexenyl acetate (2HA) 31 (?) T Females not found. 0.5 mV). Z E E E E E E Hexyl acetate (HA) 43 Decanal Butyl butyrate3-Hydroxy-2-butanone Hexyl propionate (?)( Nonanal ( ( Octyl acetate (OA)( ( ( Unidentified 0 0 b 0 c 0 ( 1-HexanolHeptyl acetateHexyl butyrate (HB) 0 0 Octyl butyrate 0 0 a P1: GCR Journal of Chemical Ecology [joec] pp900-joec-467953 June 20, 2003 17:36 Style file version June 28th, 2002

Phytocoris PHEROMONE 1845

FIG. 5. Captures of male Phytocoris difﬁcilis and P. breviusculus in traps baited with complete and partial blends of EAD-active female MSG volatiles. A three-component blend (HA, 2HA, and 2OA) and a milkweed bug pheromone (MWB) blend were included as positive controls. Means followed by the same letter are not signiﬁcantly different (P > 0.05), Kruskal–Wallis ANOVA on ranks, followed by Student–Newman–Keuls all pairwise comparison. Abbreviations as in Table 1.

midnight (Figure 6). No ﬂights to the traps were observed during daytime. Only two P. difﬁcilis females were collected from the black-light trap in the same experimental forest; each was captured between 20:00 and 22:00. In addition, a few of the P. breviusculus males were caught in the traps baited with hexyl acetate and (E)-2-octenyl acetate during the same period.

DISCUSSION

The results indicate that the sex pheromone of Phytocoris difficilis consists of hexyl acetate, (E)-2-hexenyl acetate, and (E)-2-octenyl acetate. These compounds are produced by adult females in the metathoracic scent gland (MSG), the gland that in most heteropterans is responsible for chemical defense (Aldrich, 1988). Although females of P. breviusculus were not available for identification, P1: GCR Journal of Chemical Ecology [joec] pp900-joec-467953 June 20, 2003 17:36 Style file version June 28th, 2002

1846 ZHANG AND ALDRICH

FIG. 6. Diurnal ﬂight activity of Phytocoris difﬁcilis males monitored by traps baited with attractive synthetic blends. Pooled data from 3 days’ observation (with a total of 103 males captured) in late September and early October 2002 are presented as cumulative proportional catches.

our field trapping results strongly suggest that hexyl acetate and (E)-2-octenyl acetate are sex pheromone components for this species. Thistlewood et al. (1989) suspected the MSG as a possible pheromone gland in the mullein bug, Campy- lomma verbasci, and Millar et al. (1997) also found that the sex pheromone of P. relativus was produced in the thorax, but neither group definitively located the pheromone-producing glands. In fact, attractant pheromones are released from the MSG of either males or females (depending upon the species) in various Lygaeidae (Aldrich et al., 1997, 1999) and Alydidae (Leal et al., 1995; Aldrich et al., 2000). In the Miridae, all evidence suggests that females are the attractive sex (McBrien and Millar, 1999; Kakizaki and Sugie, 2001, and references therein). Including the two mirids reported here, identifications of sex pheromone or sex-attractant components have now been achieved for eight mirid species (Table 2). Their sex pheromones consist of blends of two or three simple esters (acetates, butyrates, or hexanoates), usually with one of the components being common to both sexes and one or two components being abundantly produced only by females. Plant bugs in the genus Phytocoris have similar volatile profiles, and the chemical constituents are parsimoniously distributed among species (Table 2). Only four esters are involved in the pheromonal communication systems of the four species thus far investigated, with hexyl acetate (the only compound that is produced by both sexes in large amounts) being common to all species, and (E)-2-octenyl acetate being produced by females in at least three species (Table 2). (E)-2-Octenyl P1: GCR Journal of Chemical Ecology [joec] pp900-joec-467953 June 20, 2003 17:36 Style file version June 28th, 2002

Phytocoris PHEROMONE 1847 IRIDAE M PP PP OF (I) P P P NHIBITORS I TTRACTION A AND (SA), TTRACTANTS A EX (P), S OMPONENTS verbasci relativus P. californicus P. difﬁcilis P. breviusculus caelestianlium theobroma singularis C Campylomma Phytocoris Trigonotylus Distantiella Sahlbergella HEROMONE P NOWN 2. K )-2-butenoic acid of HHB Chemicals E )-hydroxybutyrate (HHB) ABLE R T : 1. Smith et al., 1991; 2. Millar et al., 1997; 3. Millar and Rice, 1998; 4. Current paper; 5. Kakizaki and Sugie, 2001; 6. Downham et al., 2002; and )-2-Butenyl butyrate)-2-Hexenyl acetate)-2-Octenyl acetate)-2-Octenyl butyrate)-2-Hexenyl hexanoate P P P I P P I SA E E E E E Note personal communication. Butyl butyrate( Hexyl acetate( ( Octyl butyrate ( Hexyl hexanoate ( Hexyl (3 3-Ester with ( References P P P 1 P 2 SA 3 4 4 5 6 6 P1: GCR Journal of Chemical Ecology [joec] pp900-joec-467953 June 20, 2003 17:36 Style ﬁle version June 28th, 2002

1848 ZHANG AND ALDRICH

butyrate and (E)-2-hexenyl acetate, produced by females of P. relativus and P. difficilis, respectively, are antagonistic to their corresponding sympatric congeners, P. californicus (Millar et al., 1997; Millar and Rice, 1998) and P. breviusculus (herein). These antagonists appear to play an important role in maintaining re- productive isolation of the two pairs of sympatric species. In addition to the three pheromone components of P.difficilis, several other esters were also detected from the female MSG in minor or trace amounts, including (E)-2,7-octadienyl acetate (Figure 4, Table 1). This unsaturated acetate is a key aggregation pheromone component in many milkweed bugs (Lygaeinae), but had not been found previously in any other species outside of the Lygaeinae (Aldrich et al., 1997, 1999). The major constituents of male volatiles, whether from aeration or from MSG dissection, are almost identical among these four Phytocoris species, with hexyl butyrate being the dominant component, followed by hexyl acetate) (Millar et al., 1997; Millar and Rice, 1998) (Table 1). During the course of our study, only two P. difficilis females were collected, which made any attempts at aeration impractical. How- ever, the GC profiles from our MSG extracts were similar to those obtained from aeration extracts of the two western Phytocoris spp. (Millar et al., 1997; Millar and Rice, 1998), suggesting that there is a direct correspondence between pheromone production in and release from the MSG in these mirids. What remains unknown is how the production of the pheromone is regulated by the MSG, which also produces defensive secretions. Our GC-EAD analyses showed that antennae of males of both mirid species responded strongly to the acetates and butyrates from the MSG extracts, but not to the hexenoate ester tested. However, not all antennal stimulatory compounds were behaviorally active and, in fact, none of them were attractive when tested alone. The synergism among the three pheromone components of P. difficilis suggests that each component may target a different antennal receptor neuron (Todd and Baker, 1999). Integration of neuronal input, either in the antennal lobe or in the protocerebrum, is likely responsible for the synergism ultimately leading to the modulation of behavior (Hansson and Anton, 2000). The cross-family attraction of P.difficilis males to the Oncopeltus pheromone blend resulted from the coexistence of (E)-2-octenyl acetate, one of its key pheromone components, and a nonpheromone compound, (E, E)-2,4-hexadienyl acetate. Thus, (E, E)-2,4-hexadienyl acetate could substitute for both hexyl acetate and (E)-2-hexenyl acetate in the P. difficilis pheromone blend. Furthermore, adding (E, E)-2,4-hexadienyl acetate to the three-component blend did not show any synergistic effect on trap catches of P. difficilis males as might have been expected if this compound were a true pheromone component. GC-EAD experiments revealed that (E, E)-2,4-hexadienyl acetate was highly stimulatory to the antennae of P. difficilis males. This may be due to either a high diversity or a low specificity of receptor neurons on the antennae of P. difficilis. It seems possible that (E, E)-2,4-hexadienyl acetate may activate one or both types of neurons (or P1: GCR Journal of Chemical Ecology [joec] pp900-joec-467953 June 20, 2003 17:36 Style file version June 28th, 2002

Phytocoris PHEROMONE 1849

binding sites) that are specially tuned to hexyl acetate and (E)-2-hexenyl acetate. Further study of the diversity and specificity of receptor neurons to these behaviorally active compounds using the single-cell recording technique might clarify this situation. Our strategy of using subtraction and factorial design methods for field screening proved effective. By these methods, we were able to pinpoint the two compounds responsible for the cross-family attraction and, guided by published data on the two western Phytocoris species (Millar et al., 1997; Millar and Rice, 1998), we arrived at the correct sex pheromone components of P. difficilis even before females were available for analysis. Once females of P. difficilis were collected and their MSGs were analyzed, a simple and clear picture of its sex pheromone system became obvious, verifying our deduction. The two mirids studied are both conifer-associated species, with P. difficilis being reported to feed on pine trees (Pinus virginana) (Knight, 1927), whereas P. breviusculus is reportedly a predator of scale insects on cedar or juniper trees (Wheeler and Henry, 1977). Our observations indicate that both of these Phyto- coris species are night fliers, with calling and mating events occurring mainly during the first half of scotophase. The economic importance of P. difficilis and P. breviusculus is unclear. Nevertheless, the results from this study enrich our knowledge of pheromone production and chemistry in the Miridae, and provide a simple model system for further study of this little-known but important group of insects.

Acknowledgments—We thank Dr T. J. Henry (Miridae), Systematic Entomology Laboratory (SEL), USDA-ARS, Beltsville, Maryland, for identiﬁcation of these two mirid species and for reviewing the manuscript. Dr R. Carlson, SEL, Beltsville, Maryland, facilitated the identiﬁcation process. We are also grateful to Dr J. G. Millar, Department of Entomology, University of California at Riverside, for chemical standards and for reviewing the manuscript. Field assistance by Nathan Zahm, University of Maryland, is highly appreciated.

REFERENCES

ALDRICH, J. R. 1988. Chemical ecology of the Heteroptera. Annu. Rev. Entomol. 33:211–238. ALDRICH, J. R., LEAL, W. S., NISHIDA, R., KHRIMIAN,A.P.,LEE, C.-J., and SAKURATANI, Y. 1997. Semiochemistry of aposematic seed bugs (Heteroptera: Lygaeidae). Entomol. Exp. Appl. 84:127– 135. ALDRICH, J. R., LUSBY, W. R., KOCHANSKY, J. P., HOFFMANN,M.P.,WILSON, L. T., and ZALOM, F. G. 1988. Lygus bug pheromones, vis-a-vis` stink bugs. Proc. Beltwide Cotton Prod. Res. Conf. 213–216. ALDRICH, J. R., OLIVER, J. E., TAGHIZADEH, T., FERREIRA, J. T. B., and LIEWEHR, D. 1999. Pheromones and colonization: Reassessment of the milkweed bug migration model (Heteroptera: Lygaeidae: Lygaeinae). Chemoecology 9:63–71. ALDRICH, J. R., ZHANG, A., and OLIVER, J. E. 2000. Attractant pheromone and allomone from the metathoracic scent gland of a broad-headed bug (Hemiptera: Alydidae). Can. Entomol. 132:915– 923. P1: GCR Journal of Chemical Ecology [joec] pp900-joec-467953 June 20, 2003 17:36 Style ﬁle version June 28th, 2002

1850 ZHANG AND ALDRICH

BYERS, J. A. 1991. BASIC algorithms for random sampling and treatment randomization. Comput. Biol. Med. 21:69–77. DOWNHAM, M., CORK, A., FARMAN, D., HALL, D., INNOCENZI, P., PHYTHIAN, S., PADI, B., LOWER, S., and SARFO, J. E. 2002. Sex pheromone components of the cocoa mirids, Distantiella theobroma and Sahlbergella sinfularis (Heteroptera: Miridae). p. 167, in Abstract Book, 19th Annual Meeting of International Society of Chemical Ecology, Hamburg, Germany, August 3–7, 2002. HANSSON,B.S.andANTON, S. 2000. Function and morphology of the antennal lobe: New develop- ments. Annu. Rev. Entomol. 45:203–231. HENRY,T.J.andWHEELER, A. G., JR. 1988. Family Miridae Hahn, 1833 (= Capsidae Burmeister, 1835), The plant bugs, pp. 251–507, in T. J. Henry and R. C. Froeschner (eds.). Catalog of the Heteroptera, or True Bugs, of Canada and the Continental United States. E. J. Brill, New York. HO,H.Y.andMILLAR, J. G. 2002. Identification, electroantennogram screening, and field bioassays of volatile chemicals from Lygus hesperus Knight (Heteroptera : Miridae). Zool. Stud. 41:311– 320. KAKIZAKI, M. and SUGIE, H. 2001. Identification of female sex pheromone of the rice leaf bug, Trigonotylus caelestialium. J. Chem. Ecol. 27:2447–2458. KNIGHT, H. H. 1927. Descriptions of twelve new species of Miridae from the District of Columbia and vicinity (Hemiptera). Proc. Biol. Soc. Wash. 40:9–18. LEAL, W. S., HIGUCHI, H., MIZUTANI, N., NAKAMORI, H., KADOSAWA, T., and ONO, M. 1995. Mul- tifunctional communication in Riptortus clavatus (Heteroptera: Alydidae): Conspecific nymphs and egg parasitoid Ooencyrtus nezarae use the same adult attractant pheromone as chemical cue. J. Chem. Ecol. 21:973–985. MCBRIEN, H. L., JUDD, G. J. R., and BORDEN, J. H. 1994a. Campylomma verbasci (Heteroptera: Miridae): Pheromone-based seasonal flight patterns and prediction of nymphal densities in apple orchards. J. Econ. Entomol. 87:1224–1229. MCBRIEN, H. L., JUDD, G. J. R., and BORDEN, J. H. 1996. Potential for pheromone-based mating disruption of the mullein bug, Campylomma verbasci (Meyer) (Heteroptera: Miridae). Can. Entomol. 128:1057–1064. MCBRIEN, H. L., JUDD, G. J. R., and BORDEN, J. H. 1997. Population suppression of Campylomma verbasci (Heteroptera: Miridae) by atmospheric permeation with synthetic sex pheromone. J. Econ. Entomol. 90:801–808. MCBRIEN, H. L., JUDD,G.J.R.,BORDEN, J. H., and SMITH, R. F. 1994b. Development of sex pheromone- baited traps for monitoring Campylomma verbasci (Heteroptera: Miridae). Environ. Entomol. 23:442–446. MCBRIEN,H.L.andMILLAR, J. G. 1999. Phytophagous bugs, in J. Hardie and K. Minks (eds.). Non-Lepidopteran Insects Associated with Agricultural Plants. CABI Publishing, New York. MILLAR,J.G.andRICE, R. E. 1998. Sex pheromone of the plant bug Phytocoris californicus (Heteroptera: Miridae). J. Econ. Entomol. 91:132–137. MILLAR, J. G., RICE, R. E., and WANG, Q. 1997. Sex pheromone of the mirid bug Phytocoris relativus. J. Chem. Ecol. 23:1743–1754. SCHUH,R.T.andSLATER, J. A. 1995. True Bugs of the World (Hemiptera: Heteroptera). Classification and Natural History. Comstock Publishing Associates, Cornell University Press, Ithaca, New York. SMITH,R.F.,PIERCE, H. D., JR., and BORDEN, J. H. 1991. Sex pheromone of the mullein bug, Campy- lomma verbasci (Meyer) (Heteroptera: Miridae). J. Chem. Ecol. 17:1437–1447. STONEDAHL, G. M. 1988. Revision of the mirine genus Phytocoris Fallén (Heteroptera: Miridae) for western North America. Bull. Am. Mus. Nat. Hist. 188:1–257. THISTLEWOOD, H. M. A., BORDEN, J. H., SMITH, R. F., PIERCE, H. D., JR., and MCMULLEN, R. D. 1989. Evidence for a sex pheromone in the mullein bug, Campylomma verbasci (Meyer) (Heteroptera: Miridae). Can. Entomol. 121:737–744. P1: GCR Journal of Chemical Ecology [joec] pp900-joec-467953 June 20, 2003 17:36 Style file version June 28th, 2002

Phytocoris PHEROMONE 1851

TODD,J.L.andBAKER, T. C. 1999. Function of peripheral olfactory organs, pp. 67–96, in B. S. Hansson (ed.). Insect Olfaction. Springer, Berlin. WHEELER, A. G., JR. 2001. Biology of the Plant Bugs (Hemiptera: Miridae): Pests, Predators, Oppor- tunists. Cornell University Press, Ithaca, New York. WHEELER, A. G., JR. and HENRY, T. J. 1977. Miridae associated with Pennsylvania conifers. 1. Species on arborvitae, false cypress, and juniper. Trans. Am. Entomol. Soc. 103:623–656. ZAR, J. H. 1984. Biostatistical AAnalysis. Prentice Hall, Englewood Cliffs, New Jersey. ZHANG, Q. H., SCHLYTER, F., and BIRGERSSON, G. 2000. Bark volatiles from nonhost angiosperm trees of spruce bark beetle, Ips typographus L. (Coleoptera: Scolytidae): Chemical and electrophysiological analysis. Chemoecology 10:69–80.