Journal of Chemical Ecology, Vol. 17, No. 11, 1991

SEX ATTRACTANTS A N D SEX PHEROMONE COMPONENTS OF NOCTUID MOTHS Euclidea cuspidea, Caenurgina distincta, A N D GEOMETRID MOTH Eupithecia

annulata 1

J.G. MILLAR, 2'* M. GIBLIN, 3 D. BARTON, 3 J.W. W O N G , 2'4 and E.W. UNDERHILL 3 2Department of Entomology University of California Riverside, California 92521 3Plant Biotechnology Institute 110 Gymnasium Road Saskatoon SASK S7N OW9, Canada. (Received April 26, 1991; accepted July 1, 1991)

Abstract--Pheromone components and sex attractant blends consisting of 3Z,6Z,9Z-triene hydrocarbons and racemic and chiral forms of 3Z,6Z-cis9,10-epoxydienes have been elucidated for two noctuid and one geometrid moth species. Male Euclidea cuspidea moths were attracted to blends of 3Z,6Z,9Z-heneicosatriene (3Z,6Z,9Z-21 :H) with 3Z,6Z-cis-9,10-epoxyheneicosadiene (3Z,6Z-cis-9,10-epoxy-21 : H). In addition to these compounds, 3Z,6Z,9Z-20 : H, and two regioisomeric Cz~ epoxides were tentatively identified in pheromone gland extracts. Caenurgina distincta moths were attracted by an 8 : 1 blend of 3Z,6Z,9Z-20 : H with 3Z,6Z-cis-9,10-epoxy-20 : H. Eupithecia annulata moths were attracted by either 3Z,6Z-cis-9,10-epoxy-20:H or 3Z,6Z-cis-9,10-epoxy-21 : H, and by the 9S, 10R enantiomer of each epoxide. 3Z,6Z,9Z-21 :H and 3Z,6Z-cis-9,10-epoxy-21 :H were tentatively identified from pheromone glands. Pheromone components were identified by a combination of coupled gas chromatography-electroantennography, gas chromatography-mass spectrometry, and field bioassays. Key Words--Pheromone, sex attractant, Noctuidae, Geometridae, epoxide, * To whom correspondence should be addressed. ~Issued as NRCC #32477. 4Present address: Pharmaceutical Production Research Facility, 3553 31st Street N.W., Discovery Place 1, Calgary, ALTA T2L 2K7, Canada. 2095 00984)331/91/1100-2095506.50/0• 1991 PlenumPublishingCorporation

2096

MILLAR ET AL. triene hydrocarbon, 3Z,6Z,9Z-eicosatriene, 3Z,6Z,9Z-heneicosatriene, 3Z,6Z-

cis-9,10-epoxyeicosadiene, 3Z,6Z-cis-9,10-epoxyheneicosadiene. INTRODUCTION

The first diene monoepoxide pheromone component, 3Z,6Z-cis-9,10-epoxyheneicosadiene (3Z,6Z-cis-9,10-epoxy-21 : H; subsequent abbreviations follow this model), in combination with 9Z,12Z-octadecadienal (9Z,12Z-18:A1) and 9Z, 12Z, 15Z-18:A1, was identified in pheromone gland extracts and effluvia of the saltmarsh caterpillar moth, Estigmene acrea Drury (Lepidoptera: Arctiidae) (Hill and Roelofs, 1981). The biologically active epoxide enantiomer was subsequently determined to have the 9S, IOR configuration (Moil and Ebata, 1981). Since that time, this compound, or homologs of it, have been identified as sex attractants or pheromone components for a number of lepidopteran species. For example, 3Z,6Z-cis-9,10-epoxy-21:H has been reported as a sex pheromone component of the arctiid species Creatonotos transiens L., C. gangis Walker (Bell and Meinwald, 1984), Antichloris viridis (Meyer, 1984), Hyphantrea cunea Drury (Hill et al., 1982; Einhorn et al., 1982), and Phragmatobia fuliginosa L. (Descoins and Fr6rot, 1984). This compound also has been identified in female pheromone gland extracts from the arctiids Tyriajacobaeae and Cymbalophora pudica Esper, and it was shown to elicit strong electroantennogram responses from male moth antenna, although no data on its attractiveness to male moths was reported (Fr6rot et al., 1988a,b). This C21 epoxide, or the C18_ 20 homologs, also have been reported as sex attractants for several geometrid and noctuid moth species (Wong et al., 1985; Kovalev and Nikolaeva, 1986). Racemic and chiral regioisomers of the cis-9,10-epoxydienes, 3Z,9Z-cis6,7- and 6Z,9Z-cis-3,4-epoxydienes, also have been reported as sex attractants and sex pheromone components for a number of geometrid and noctuid species (Becker et al., 1990; Millar et al., 1987, 1990 a-d; Hansson et al., 1990). The behavioral response to the compounds is often synergized or antagonized by the corresponding 3Z,6Z,9Z-trienes. Preliminary data on the attraction of male moths of a number of noctuid and geometrid species to lures containing blends of 3Z,6Z,9Z-trienes, alone or in combination with mixtures of racemic cis-monoepoxydienes (combined monoepoxides, CME-3Z,6Z,9Z-X : H, X = 19-21), or racemic and enantiomeric forms of 19- to 21-carbon 3Z,6Z-cis-9,10-epoxydienes were published several years ago (Wong et al., 1985). Since that time, we have investigated the pheromone chemistry of several of these species in more detail. We report here: (1) the identification of sex pheromone components for the geometrid species Eupithecia annulata Hulst and the noctuid species Euclidea cuspidea HiJbner, by a combination of coupled gas chromatography-electroantennogram detection (GC-EAD), gas chromatography-mass spectrometry (GC-MS), and

NOCTUID ATTRACTANTS AND PHEROMONES

2097

comparisons with synthetic standards; (2) the elucidation and field testing of sex pheromone and sex attractant blends for the above two species, and for Caenurgina distincta Cram; and (3) the determination of the chirality of the epoxydienes that are attractive to the above species. METHODS

AND MATERIALS

Insects, Electroantennography, and Pheromone Identification. Female moths of unknown mating status were collected from the field by sweep-netting or black-light trapping and used as a source of pheromone. Males were captured as described above or in sticky wing-traps (Phero Tech, Vancouver, British Columbia; traps are similar to Pherocon 1C traps from Scentry, Inc., Buckeye, Arizona). Preparation and analysis of pheromone gland extracts by coupled GC-EAD and coupled GC-MS was carried out as previously described (Millar et al., 1987, 1990d). GC-MS analyses were carried out with a Finnigan 4000E instrument interfaced to an INCOS 2300 data system, in electron impact (70 eV) or chemical ionization (isobutane) modes. DB-5 (50 m x 0.32 mm; J&W Scientific, Folsom, California) and Ultra-2 (Hewlett-Packard, Avondale, Pennsylvania) capillary columns were used with He carrier gas, programmed from 40 to 250~ GC-EAD studies were carried out with a Hewlett-Packard 5910 GC equipped with either a DB-1701 or a DB-5 capillary column (30 m • 0.32 mm, J&W Scientific). The column effluent was split 7 : 3 in favor of the FID detector, and signals from the EAD and FID signals were recorded simultaneously with a matched pair of Hewlett-Packard 3392A integrators. Injections were made in splitless mode, programmed from 40 to 225~ Insect Trapping. Field experiments were carried out in a mixed forest area (black spruce, jack pine, willow, aspen) approx. 100 km northeast of Saskatoon, Saskatchewan. Phero Tech wing-traps were used. Details of lure preparation (red rubber septa), and the setup of field survey traps and field experiments were as previously described (Millar et al., 1990d). Trap captures were transformed [(x + 1) 1/2] and subjected to analysis of variance. The homogeneity of the variances of the transformed means was confirmed with Bartlett's test (Sokal and Rohlf, 1981). Treatments with zero captures overall were not included in the ANOVA, as replicates within a treatment must have some variance to satisfy the assumptions of the ANOVA test. Significantly different means were separated by Duncan's (1955) multiple-range test. Synthetic Chemicals. The syntheses of the 3Z,6Z,9Z-trienes (Underhill et al., 1983) and the enantiomerically enriched forms of the 3Z,6Z-cis-9,10-epoxydienes (Wong et al., 1985) have been reported previously. The chemical purity of all compounds used was >98% by capillary GC. The enantiomerically enriched 3Z,6Z-cis-9,10-epoxydienes had chiral purities of 93% ee (9R,10S series) and 92 % ee (9S, 10R series) respectively (Wong et al., 1985).

2098

MILLAR ET AL. RESULTS

Euclidea cuspidea Habner. The preliminary report by W o n g et al. (1985) suggested that males o f this species were attracted by blends o f 3Z,6Z,9Z-21 : H in combination with one or more C21 monoepoxydienes. Analysis o f a female pheromone gland extract by G C - E A D elicited three strong responses from a male moth antenna at retention times corresponding to 3Z,6Z,9Z-20:H, 3Z,6Z,9Z-21 : H, and 3Z,6Z-cis-9,10-epoxy-21 : H (Figure 1). The concurrent flame ionization detection (FID) trace showed peaks at these retention times, and two further peaks at retention times corresponding to 3Z,9Z-cis-6,7-epoxy21 : H and 3Z,9Z-cis-3,4-epoxy-21 : H . The five compounds were present in the ratio o f 3Z,6Z,9Z-20: H (1.6), 3Z,6Z,9Z-21 : H (100), 3Z,6Z-cis-9,10-epoxy-

~

~ nn

m

rn

or) a

oq 0 Q_ oO

~

w

L

ID

I---

F,Q. w r

ILl bJ ._J

Euc/idia cuspidea

W 1.1_ EAD 20

I 0 . 2 mV. i

i

24

28

I

5 :> TIME

36

40

44

(rain)

FIG. 1. Simultaneously recorded flame ionization detector (FID) and electroantennographic detector (EAD) responses of Euclidea cuspidea male antennae in GC-EAD experiments. Upper trace is an FID trace of a mixture of C18-22 3Z,6Z,9Z-trienes (designated by the letter A and preceded by the carbon chain length) and their monoepoxydiene analogs (designated with the letter B); the latter eluted from the 30-m DB-1701 column in the order 6,7, 3,4, and 9,10 monoepoxydiene. Internal standards (heptadecane and tetracosane) are shaded. Lower pair of traces are in response to a pheromone gland extract from conspecific females.

NOCTUID ATTRACTANTS AND PHEROMONES

2099

21 :H (2), 3Z,9Z-cis-6,7-epoxy-21 : H (0.4), and 6Z,9Z-cis-3,4-epoxy-21 :H (0.4). The presence of all five components in pheromone gland extracts was confirmed with both selected ion monitoring (SIM) and full-scan chemical ionization (isobutane) mass spectrometry. Exact retention time matches were obtained for all five compounds versus synthetic standards, and full-scan C1 mass spectral matches were obtained for 3Z,6Z,9Z-21 :H and 3Z,6Z-cis-9,10-epoxy-21 :H versus standards (Figures 2 and 3). The isobutane C1 mass spectrum of 3Z,6Z,9Z-21 : H (Figure 2) was characterized by a base peak at m/z 289 (M + H) +, and adducts at m/z 347 (M + C4H9) + , and 291 (M + H) +. Fragments at m/z 277, 263,249, and 235 may be attributed to sequential losses of CnH2, (n = 5-8) from the m/z 347 adduct ion. Ions at m/z 123, 137, 151, 165, and 179 may be attributed to a series of homologous fragments (CnH2n_ 4 + H +, n = 10-14) containing the three unsaturations. Fragments at m/z 211 and 223 were not readily assignable. The isobutane C1 spectrum of 3Z,6Z-cis-9,10-epoxy-21 :H (Figure 3) was much simpler, with a base peak at m/z 307 (M + H) + and a large fragment at m/z 289 (M + H - H20) +. Diagnostic ions arising from rearrangement and cleavage of the epoxide functionality were present at m/z 123 (C9H15) + and 183 (C12H230) +, locating the epoxide at the 9,10 position. The potential M + 57 adduct at m/z 363 was beyond the scan range used (120-350 amu). The full-scan isobutane C1 spectra of the other three components were not as good due to the low levels at which they were present, but the spectra clearly matched those of the synthetic standards. In particular, the peak corresponding to 3Z,6Z,9Z-20 : H gave strong fragment ions at m/z 333 (M + 57), 277 (M + 1), and 275 ( M - 1 ) , and the peaks corresponding to 3Z,9Z-cis-6,7-epoxy-21:H and 6Z,9Z-cis-3,4-epoxy-21 :H gave strong ions at m/z 307 ( M + I ) and m/z 289 (M + 1 - 18). Wong et al. (1985) had demonstrated that 3Z,6Z,9Z-21:H and CME3Z,6Z,9Z-21 :H were not active as single components, so our field tests began with blends of 3Z,6Z,9Z-21 : H with epoxides. In the first experiment (replicated three times), lures containing 3Z,6Z,9Z-21 : H + 3Z,6Z-cis-9,10-epoxy-21 : H (450:20/~g) attracted 21 moths, while the corresponding blends with the cis6,7- and cis-3,4-epoxide regioisomers caught none, demonstrating that the 9,10 epoxide was the attractive regioisomer. Furthermore, lures containing 3Z,6Z,9Z21 : H in combination with 3Z,6Z,9S, 10R-epoxy-21 : H (450 : 10 ~g) attracted 42 moths, while the corresponding lure containing the 9R,10S enantiomer attracted no moths, indicating that the attraction was enantiospecific. Experiments conducted in 1985 confirmed these data (Table 1). Mixtures of 3Z,6Z,9Z-21:H with the enantiomers of 3Z,6Z-cis-9,10-epoxy-21:H, in a series of ratios bracketing the blend ratio found in pheromone gland extracts were tested, and the blends with the 9S, 10R enantiomer were significantly more

2100

MILLAR

I00-

289

ET AL.

A

\

347

50-

,7

l

151

bJ

(.)

z

h L,,,ll, ,E,

r~

150

z

I

263 2~

211 234 223

165 179

,,,, 2()0

,, ]L i

329

I,I

I

250

360

350

I00-

289 \

m 0.05).

21 : H and 3Z,6Z,9S, lOR-epoxy-21 : H (450: 10 #g) spiked with varying amounts o f 3Z,6Z,9Z-20 : H (1, 10, 50, or 100,/zg), and the enantiomers o f 3Z,9Z-cis6,7-epoxy-21 : H (5 or 50 /zg). There were no significant differences in trap captures between the basic lure and the adulterated lures; that is, the C20 triene and the 3Z,9Z-cis-6,7-epoxy-21 : H enantiomers have no apparent biological activity as attractants for this species despite being present in the pheromone gland. Caenurgina distincta Cram. Blends o f 3Z,6Z,9Z-20: H with 3Z,6Z-cis-9,10e p o x y - 2 0 : H have been implicated as potential sex pheromone components for this species (Wong et al., 1985). G C - E A D analysis o f a pheromone gland extract from a field-collected female moth showed strong antennal responses from a male antenna at retention times corresponding to 3Z,6Z,9Z-19:H, 3Z,6Z,9Z2 0 : H , 3Z,6Z,9Z-21:H, 3Z,6Z-cis-9,10-epoxy-20:H, and 3Z,6Z-cis-9,10epoxy-21 : H (Figure 4). The concurrent F I D trace showed discernible peaks only at the retention times o f the C2o and C21 trienes, indicating that the quantities o f the other components in the extract eliciting antennal responses were very small. W h e n challenged with synthetic standards, a male antenna demonstrated strong responses to C19-C21 trienes, 3Z,6Z-cis-9,10-epoxydienes, and 3Z,9Z-cis-6,7-epoxydienes (Figure 4, upper traces). A number o f two-component blends o f 3Z,6Z,9Z-20:H with 3Z,6Z,9Z21 : H were found to be minimally attractive and not significantly different from 3Z,6Z,9Z-20:H alone. In like fashion, blends o f 3Z,6Z,9Z-20:H with the shorter-chain E A G - a c t i v e homolog 3Z,6Z,9Z-19:H had been demonstrated to have no synergistic effects (Wong et al., 1985). A field test in 1985 (replicated four times) demonstrated that 3Z,6Z-9S, 10Repoxy-20 : H was the more attractive o f the two enantiomers, as a 1 : 8 blend o f epoxide-3Z,6Z,9Z-20:H attracted 59 male moths, while the corresponding

NOCTUID ATTRACTANTSAND PHEROMONES

2103

m --

03 O n"

od

0

FIO

a Z

I r I

I--

bJ O0 Z 0 003 bJ n," fig W a n," 0 (...) bJ w"

] E

Caenurglna distincta F i

FID

r~ iX bJ

r

I

hi ..J [ 0.2 m V

h i EAD LL

E

I

I

I

I

24

28

52

56

TIME

I

40

44

(min)

FIG. 4. Response of Caenurgina distincta male antennae in GC-EAD experiments using a DB-1701 column. Upper pair of traces are in response to a mixture of C~8-22 synthetic standards; lower pair are in response to a pheromone gland extract from a conspecific female. Number and letter designations are identical to those of Figure 1. Internal standards (heptadecane and tetracosane) are shaded.

2104

MILLAR ET AL.

blend with the 9R, 10S enantiomer attracted only five moths. Concurrently, lures containing racemic 3Z,6Z-cis-9,10-epoxy-20: H in combination with 3Z,6Z,9Z2 0 : H (1:4 blend) attracted 55 moths, suggesting that the 9R,10S enantiomer was not antagonistic. Because a strong antennal response had been elicited by 3Z,9Z-cis-6,7epoxy-20: H, the effect of adding the enantiomers of this compound to an attractive lure blend (3Z,6Z,9Z-20: H + 3Z,6Z-cis-9,10-epoxy-20 : H, 400 : 100/zg) was tested. Additions of 5 or 50 txg of either 6,7-epoxide enantiomer had no discernible effect on the attractiveness of the basic blend. Eupithecia annulata Hulst (Geometridae). The preliminary report of sex attractants for males of this species had implicated monoepoxydienes of chain lengths C19_21, but moths were attracted by a variety of lures, and there was no clear indication of a specific attractant or attractants (Wong et al., 1985). Further evidence for the role of C19-Zl monoepoxydienes was obtained from electroantennogram studies with the enantiomers of the cis-3,4-, -6,7-, and -9,10-monoepoxydienes of chain length C18_22. Only the cis-9,10 epoxides elicited responses greater than 1 mV. The Cm_22 3Z,6Z-9R, lOS-monoepoxydienes elicited male antennal responses of 0.9 ___ 0.1 mV (N = 2), 3.9 + 1.7 mV (N = 8), 1.7 + 0.6 mV (N = 8), and 0.6 _ 0.2 mV (N = 2) respectively. The corresponding 9S,10R series elicited responses of 2.1 _ 1.1 mV (N = 3), 3.0 _ 0.8 mV (N = 8), 3.6 _ 1.1 mV (N = 8), and 1.4 ___ 0.6 mV (N = 3). GC-EAD studies were conducted with pheromone gland extracts from three field-collected female moths of unknown mating status. The extract from the first female elicited a male antennal response at the retention time of 3Z,6Z-cis9,10-epoxy-21 :H. This was corroborated by analysis of the extracts from the second (Figure 5) and third females, and additional antennal responses were seen at the retention times of 3Z,6Z,9Z-21 : H (both extracts) and 3Z,6Z-cis9,10-epoxy-20: H (second extract only). An FID peak was detected concurrently for the triene in the second extract (Figure 5), but the region of the FID chromatogram where the epoxides eluted was obscured by late-eluting compounds from the previous extract. 3Z,6Z,9Z-20:H was not detected in the extracts either by the FID detector or the antennae; the small peak seen on the FID trace (Figure 5) in the vicinity of the 3Z,6Z,9Z-20 : H had a retention time sufficiently different from that of 3Z,6Z,9Z-20:H to be discernible (0.13 min; run-to-run variability of retention times of standards was determined to be -< 0.02 min). When a male antenna was challenged with synthetic standards, strong responses were elicited by the Cm-Cz~ cis-9,10 epoxides, with weaker responses to the C2o and C21 trienes and also to the 19-21 carbon 3Z,9Z-cis-6,7-monoepoxydienes (Figure 5, top traces). Corroborating evidence for the presence of 3Z,6Z,9Z-21 : H and 3Z,6Z-cis9,10-epoxy-21 :H in the pheromone gland extract from another field-collected

NOCTUID ATTRACTANTS AND PHEROMONES

W

GO

a

E3 r'r