Sex pheromone and related compounds in the Ishigaki and Okinawa strains of the tussock moth Orgyia postica (Walker) (Lepidoptera: Lymantriidae)

Article abstract of DOI:10.1271/bbb.69.957

Two distinct electroantennographycally active (EAG-active) components, A and B, and a weakly active component C were found in a solvent extract from virgin females of the Ishigaki strain of the tussock moth, Orgyia postica (Walker). Components A, B, and C were found in the extract of the females at 4.0, 0.5, and 4.0 ng/female respectively. Components A, B, and C were identified as (6Z,9Z,11S,12S)-11,12-epoxyhenicosa-6,9-diene [(11S,12S)-1: posticlure], (6Z)-henicos-6-en-11-one (2), and (6Z,9Z)-henicosa-6,9-diene (3), respectively. Component B was absent in the extract from the Okinawa strain, in which components A and C were present at 2.0 and 1.5 ng/female respectively. (11S,12S)-1 and the racemic mixture showed attractiveness for both the Okinawa and Ishigaki strains, whereas (11R,12R)-1 did not. The addition of 2 significantly reduced the trap catches with (11S,12S)-1 on the Okinawa strain which lacked 2, while there was no significant inhibitory effect on the Ishigaki strain. The addition of 3 to (11S,12S)-1 did not significantly affect trap catches at Ishigaki or Okinawa. This confirmed that the attractant pheromone of O. postica of the Ishigaki strain is also (11S,12S)-1.

Full text of DOI:10.1271/bbb.69.957

Biosci. Biotechnol. Biochem., 69 (5), 957–965, 2005
Sex Pheromone and Related Compounds in the Ishigaki and Okinawa Strains
of the Tussock Moth Orgyia postica (Walker) (Lepidoptera: Lymantriidae)
y
Sadao WAKAMURA,1; Norio ARAKAKI,2 Masanobu YAMAMOTO,3 Syuntaro HIRADATE,4
Hiroe YASUI,1 Kunio KINJO,5 Tetsuya YASUDA,1; Hiroyuki YAMAZAWA,
and Tetsu ANDO
3
*
3;**
1Laboratory of Insect Behavior, National Institute of Agrobiological Sciences (NIAS),
Ohwashi 1-2, Tsukuba 305-8634, Japan
2Okinawa Prefectural Agricultural Experiment Station, 4-222 Sakiyama, Naha 903-0814, Japan
3Graduate School of Bio-Applications and Systems Engineering (BASE),
Tokyo University of Agriculture and Technology, Koganei, Tokyo 184-8588, Japan
4National Institute for Agro-Environmental Sciences (NIAES), Tsukuba 305-8604, Japan
5Yaeyama Branch, Okinawa Prefectural Plant Disease and Insect Control Station, 1178-6 Hirae,
Okinawa 907-0003, Japan
Received November 30, 2004; Accepted February 18, 2005
Two distinct electroantennographycally active (EAG-
active) components, A and B, and a weakly active
component C were found in a solvent extract from
virgin females of the Ishigaki strain of the tussock moth,
Orgyia postica (Walker). Components A, B, and C were
found in the extract of the females at 4.0, 0.5, and
4.0 ng/female respectively. Components A, B, and C
were identified as (6Z,9Z,11S,12S)-11,12-epoxyhenicosa-
6,9-diene [(11S,12S)-1: posticlure], (6Z)-henicos-6-en-
11-one (2), and (6Z,9Z)-henicosa-6,9-diene (3), respec-
tively. Component B was absent in the extract from the
Okinawa strain, in which components A and C were
present at 2.0 and 1.5 ng/female respectively. (11S,12S)-
1 and the racemic mixture showed attractiveness for
both the Okinawa and Ishigaki strains, whereas
(11R,12R)-1 did not. The addition of 2 significantly
reduced the trap catches with (11S,12S)-1 on the
Okinawa strain which lacked 2, while there was no
significant inhibitory effect on the Ishigaki strain. The
addition of 3 to (11S,12S)-1 did not significantly affect
trap catches at Ishigaki or Okinawa. This confirmed
that the attractant pheromone of O. postica of the
Ishigaki strain is also (11S,12S)-1.
and litchi in Okinawa Prefecture, Japan.1) Larvae are
covered with bristles that can irritate human skin. They
occasionally defoliate plants and cause considerable
damage. The female moths are wingless, so they release
a sex pheromone to attract males. They copulate and lay
eggs on the cocoon from which they have emerged. Sex
pheromone would provide a useful tool for a pest control
agent.
In gas chromatography–electroantennographic detec-
tion (GC–EAD) analyses of the extracts of the abdomi-
nal tips of female O. postica, geographical divergence
has been found in the populations of Ishigaki and
Okinawa Islands in Okinawa Prefecture.2) These two
islands are separated by 380 km of water. Two EAG
responses were distinct in the Ishigaki strain, but only
one was observed in the Okinawa strain.2) The major
EAG-active and attractive component in the Okinawa
strain was identified as (6Z,9Z,11S,12S)-11,12-epoxy-
henicosa-6,9-diene [(11S,12S)-1], and named postic-
lure.3) In this paper, we report identification of the three
EAG-active components found in the Ishigaki strain.
The attractiveness of these components to the males on
both islands is also discussed in the context of the
activity and possible evolution of pheromonal poly-
morphism in this species.
Key words: sex pheromone; Orgyia postica; trans-epox-
ide; (6Z,9Z,11S,12S)-11,12-epoxyhenicosa-
6,9-diene; (6Z)-henicos-6-en-11-one
Materials and Methods
The tussock moth Orgyia postica (Walker) (Lepido-
ptera: Lymantriidae) is one of the pest species on mango
Insects. O. postica larvae were collected in green-
houses for the cultivation of mango in Nago on Okinawa
y
To whom correspondence should be addressed. Fax: +81-29-838-6205; E-mail: swaka@affrc.go.jp
Present address: Pest Management Laboratory, National Agricultural Research Center (NARC), Tsukuba 305-8666, Japan
Present address: State Analysis Laboratory, National Food Research Institute, Tsukuba 305-8602, Japan
*
**
Abbreviations: DMDS, dimethyl disulfide; FE, female equivalent; GC–EAD, gas chromatography and electroantennographic detection; EAG-
active, electoroantennographically active; GC–MS, gas chromatography–mass spectrometry; HPLC, high performance liquid chromatography; KI,
´
Kovats retention indices; THF, tetrahydrofuran; TP, temperature program

958
S. WAKAMURA et al.
Island in February 1998 (Okinawa strain), and at the
Yaeyama Branch of the Okinawa Prefectural Agricul-
tural Experiment Station on Ishigaki Island in March
1998 (Ishigaki strain). They were reared on an artificial
diet (Insecta LF, Nihon Nosan Kogyo, Kanagawa,
Japan) in the laboratory at 25 ꢀC and L16:D8 until
emergence. Pupae were taken from cocoons by dissolv-
ing the latter with 1% sodium hypochlorite solution, and
washed with tap water in order to remove the poisonous
hairs of the larval skin. Pupae were sexed and
maintained in separate containers until emergence.
Okinawa strain were used for this purpose throughout
the study, since males of the Okinawa and Ishigaki
strains showed the same EAG responses.2)
Coupled gas chromatography–mass spectrometry
(GC–MS). GC–MS analysis was conducted with a JEOL
JMS SX-102A double-focusing magnetic sector mass
spectrometer (EI mode, 70 eV) interfaced to an HP6890
gas chromatograph and operated with an HP Model
715/64 computer. The carrier gas was helium. An HP-1
column or HP-INNOWax column was operated in a
constant flow mode (1.6 ml/min, about 65 cm/s), and
the initial column head pressure was 35 kPa. The TP for
the column oven was 50(1)-20-150(0)-5-240(5), unless
otherwise stated. The GC was equipped with a cool on-
column injector and a split/splitless injector. The
products of microozonolysis were injected in splitless
mode at 220 ꢀC (splitless delay: 1 min), whereas the
other solutions were injected through the cool on-
column injector. In certain analyses, n-alkanes with an
even number of carbon atoms (between 18 and 28) were
Extraction. Solvent extracts of sex pheromone glands
were obtained from 1-d-old females at around 2 h after
the start of the dark period, since calling behaviors were
frequently observed at this time. The tip containing the
pheromone gland was excised from the female abdomen
with fine tweezers and soaked in hexane (5–10 tips/
500 ml) for about 15 min. The tips were filtered off
through a small wad of cotton to obtain the extract. They
were rinsed twice with the same volume of hexane and
the rinses were added to the extract.
´
added as internal standards for calculation of Kovats
retention indices (KI).5)
Column chromatography. Fractionations were done
on 300 mg of Florisilꢀ (100–200 mesh, purchased from
Wako Pure Chemical Industries, Osaka, Japan) in a
Pasteur pipette. The eluting solvents were 1.5 ml of
hexane, and 5%, 15%, and 50% ether in hexane.
Epoxide and ketone components were separated with
repeated column chromatography on 300 mg of silica gel
(Wako gel C-200, particle size 75–150 mm, Wako Pure
Chemical Industries). The solvents were 1.5 ml of
hexane, and 0.2%, 0.5%, 1%, 2%, and 5% ether in
hexane. Unsaturated hydrocarbon components were
separated by column chromatography on 300 mg of
silica gel impregnated with 16.7% silver nitrate. The
eluting solvents were 1.5 ml of hexane, and 1%, 2%, 3%,
5%, 7%, 10%, 20%, and 50% ether in hexane.
High performance liquid chromatography (HPLC). A
Jasco PU-980 liquid chromatograph equipped with an
integrator (System Instrument Chromatorecorder 21), an
RI detector (Labo System RI-98SCOPE), and a Chiral-
pak AD column (0.46 cm ID ꢁ 25 cm, Daicel Chemical,
Tokyo) was used, eluting with 0.075% 2-propanol in
hexane (0.5 ml/min). Injection was made through a
Rheodyne 7125 loop injector in 10 ml of hexane. To
confirm the compounds eluted, the eluate was fraction-
ated at 0.50-min intervals from 10.00 to 14.00 min, and
fractions were checked by GC. The amounts of each
component were determined by comparison of the GC
peak areas with those of authentic compounds.
Dimethyl disulfide (DMDS) derivatization. According
to the method of Buser et al. (1983),6) 20 ml of DMDS
and a very small crystal of iodine were added to about
10 ml of hexane solution of the test material in a 100-ml
capillary tube with one end sealed. The other end of the
tube was sealed and kept at 55 ꢀC for 2 h. After mixing
the solution with 50 ml of 5% aqueous sodium thiosulfate
solution, the products were extracted three times with
100 ml of hexane. The pooled extract was passed through
a short column of powdered anhydrous sodium sulfate
(100–200 mg) in a Pasteur pipette, and then concentrated
to about 5 ml and analyzed by GC–MS through the on-
column injector.
Gas chromatography (GC) and electroantennograph-
ic detection (EAD). GC analysis was conducted on a
Hewlett-Packard (HP) 5890II gas chromatograph equip-
ped with a flame ionization detector (FID). An HP-1
or HP-INNOWax fused silica column (15 m or 30 m ꢁ
0.25 mm ID ꢁ 0.25 mm film thickness, HP) was used at
a column head pressure of 55 kPa for the 15-m column
or 110 kPa for the 30-m column, with helium carrier gas
and a cool on-column injector held at 53 ꢀC, and
controlled at oven temperature plus 3 ꢀC after injection.
The temperature program (TP) for the column oven was
50 ꢀC for 1 min, 50 ꢀC to 150 ꢀC at 20 ꢀC/min, 150 ꢀC to
260 ꢀC (or 240 ꢀC for the INNOWax column) at 5 ꢀC/
min, and then the final temperature for 5 min [abbrevia-
tion for this TP: 50(1)-20-150(0)-5-260(5), or -240(5)].
In certain GC analyses, an electroantennographic (EAG)
response was measured simultaneously with the FID
recording. The EAG detection system (EAD) was set up
according to Struble and Arn (1984).4) Males of the
Diimide reduction. Polyene compounds were partially
or completely hydrogenated with diimide according to
the method described by Yamaoka et al. (1976).7)
A
solution of polyene compound in hexane, concentrated
to about 10 ml in a 1-ml taper-bottomed vial, was
complemented with 10 ml each of hydrazine and hydro-

trans-Epoxide Sex Pheromone of Orgyia postica
959
gen peroxide solution in ethanol [N2H4: 0.3 ml of
hydrazine hydrate in 10 ml of ethanol; H2O2: 0.04 ml
of 31% hydrogen peroxide in 10 ml of ethanol]. The
mixture was maintained at 55 ꢀC for 80 min for partial
hydrogenation. For GC or GC–MS analysis, reaction
mixture was directly injected without cleanup. In cases
that required further analysis, the mixture was added to
0.1 ml of 3% hydrochloric acid and immediately
extracted with 0.2 ml of hexane three times. The hexane
extract was passed through a short column of anhydrous
sodium sulfate and then concentrated to about 10 ml.
DMDS adducts were analyzed with GC–MS equipped
with an apolar HP-1 column in which TP was 50(1)-20-
150(0)-5-270(5).
Dose-response relations of EAG responses. Dose-
response curves were measured with the GC–EAD
system. A single antenna was used for a series of
measurements that ranged from low to high dosages of
synthetic component or extract. Since considerable
variation was observed in raw EAG voltage when
different antennae were used, the EAG responses were
normalized to 1.0 at 10 ng of 1 (racemic mixture), which
was injected immediately after a series of measurements
from low to high concentrations. Measurements were
replicated six times using different male antennae.
Field attraction. A tent-shaped sticky trap (white
type, SE trap of Sankei Chemicals, Kagoshima, Japan,
29 ꢁ 32 ꢁ 8 cm high) with a 24-cm ꢁ 30-cm sticky
plate was used for field attraction. Synthetic chemicals
were blended and impregnated in rubber septa (gray
septa, West, Singapore) by applying about 0.3 ml of the
hexane solution. Rubber septum lures were suspended
about 2 cm above the sticky plate in the traps. The traps
were placed on steel rods about 1 m above the ground at
7-m intervals in the field. Field tests were conducted in
Naha, Okinawa, Japan in June 2000 and from May to
July 2002, and on Ishigaki in June 2002.
Microozonolysis. Microozonolysis was conducted as
in Beroza and Bierl (1967).8) Oxygen containing ozone
was blown into about 3 ml of hexane solution cooled to
below ꢂ80 ꢀC through a glass capillary tube. The
solution (about 2 ml) was injected into GC–MS without
further treatment in splitless mode at 230 ꢀC for 1 min.
The GC was equipped with an HP-INNOWax column
and TP was 50(1)-10-240(5).
1=2
Chemicals. Both enantiomers of (6Z,9Z)-11,12-epoxy-
henicosa-6,9-diene [(11S,12S)-1: posticlure] were stereo-
selectively synthesized using Sharpless epoxidation9) of
trans-allylic alcohol derived from decanal as a key
reaction. Preparative HPLC with a chiral column (Chiral-
pak AD) furnished optically pure samples. Lots used for
field tests were more than 99% pure in GC analysis. A
detailed synthesis will be published elsewhere by
Yamamoto et al. [for the summarized procedure, see
Wakamura et al. (2001)3)].
Trap data (X) were transformed to ðX þ 0:5Þ and
subjected to two-way-layout analysis of variance where
zero data (mean ¼ 0) were omitted. The means were
ranked by Tukey’s method when analysis of variance
was significant at the 5% level. In the figures and the
table, means accompanied with the same letter are not
significantly different, and the same letter was given to
zero data when the confidence range of the mean
contained zero [¼ ð0 þ 0:5Þ1=2]. When trap catches were
small, data were pooled for 2 or 3 nights before
statistical analysis.
(6Z)-Henicos-6-en-11-one (2) was synthesized via
Grignard reaction of (5Z)-undec-5-enal with decylmag-
nesium bromide in THF and subsequent oxidation with
pyridinium chlorochromate in dichloromethane. (5Z)-
Undec-5-enal was prepared by acetylene coupling
reaction of 1-bromopentane with 5-hexynyl pyranyl
ether in hexamethylphosphoric triamide, according to
the method described by Hendry et al. (1975),10) and
subsequent hydrogenation on Lindlar catalyst and
oxidation with pyridinium chlorochromate. Crude 2
was purified to more than 99.5% in respect to positional
and geometric isomerism with chromatography on
16.7% silver nitrate-impregnated silica gel. From this
chromatography, (6E)-isomer of 2 was successfully
separated in the earlier fractions with the same purity.
(6Z,9Z)-Henicos-6,9-diene (3) was synthesized by
Grignard coupling reaction of (9Z,12Z)-octadeca-9,12-
dienyl tosylate and propylmagnesium bromide in
THF,11) and purified by HPLC equipped with Excelpak
SIL-C18 5B column (0.46 cm ID ꢁ 15 cm, Yokogawa,
Tokyo). The lot of this compound used for field tests
was more than 99.5% pure in respect to positional and
geometric isomerism.
Results
GC–EAD analysis
Crude extracts were accumulated for 490 and 590
females from the Ishigaki and Okinawa strains respec-
tively. When the crude extract of Ishigaki females was
injected into the GC–EAD system, two distinct EAG
responses were observed at tR ¼ 25:84 min and tR ¼
26:34 min (components A and B respectively) (Fig. 1).
In addition, a weak response was also observed at
tR ¼ 23:28 min (component C). Components A and B
were eluted with 5% ether in hexane, whereas compo-
nent C was eluted with hexane from Florisil column
chromatography (Fig. 1). The other signals were not
reproducible in repeated analysis. The amounts of
components A, B, and C were determined to be
approximately 4.0, 0.5, and 4.0 ng per female respec-
tively. In contrast, component B was absent in the crude
extract and the 5%-ether-in-hexane fraction of the
Okinawa strain (A: 2.0 ng/female and C: 1.5 ng/female,
Fig. 1).

960
S. WAKAMURA et al.
was a diunsaturated C21 hydrocarbon. This component
was isolated by chromatography on silica gel impreg-
nated with silver nitrate. Component C was found as a
major compound in the 20%-ether-in-hexane fraction.
Structure elucidation and identification of compo-
nent A
Component A showed primarily single peaks in the
GC analyses on HP-1 and HP-INNOWax columns. In a
subsequent GC–MS analysis, mass spectrum of compo-
nent A showed a distinct molecular ion at m=z 306 (Mþ,
21%, base m=z 155) and diagnostic ions for an epoxide
at m=z 290 (½M ꢂ 16ꢃþ, 3.2%) and 288 (½M ꢂ 18ꢃþ,
2.6%) (Fig. 2A) which was identical to that of posticlure
[(11S,12S)-1].3) The chemical structure was further
confirmed by chemical derivatization and subsequent
GC–MS analysis, which yielded identical results to
those of the Okinawa strain.3)
To confirm the absolute configuration, 0.5% fraction
was injected into the chiral HPLC, and the HPLC
fractions were checked by GC (Fig. 3). The natural
component A showed only one peak with a retention
time very close (tR ¼ 13:31 min) to that for (11S,12S)-1
(tR ¼ 13:23 min), but no peak corresponding to
(11R,12R)-1 (tR ¼ 11:45 min). The absolute configura-
tion was further confirmed by measuring dose-response
curves of EAG responses of male antennae to (11S,12S)-
1 and (11R,12R)-1. The response to the (11S,12S)-1 was
significantly greater than to the (11R,12R)-1 (Fig. 4).
Component A isolated from the Okinawa and Ishigaki
strains showed almost identical dose-response curves to
that of (11S,12S)-1. Thus component A isolated from
the Ishigaki strains was also identified as posticlure
[(11S,12S)-1].
Fig. 1. GC–EAD Profiles of Crude Extract of Orgyia postica
Females of the Ishigaki Strain (upper figure) and 5%-Ether-in-
hexane Fractions from the Ishigaki and Okinawa Strains (lower
figure).
Upper figure: HP-1, 15 m, TP: 50(1)-25-100(0)-5-260(5), column
head pressure (CHP): 55 kPa. Arrows indicate the corresponding
FID peaks. Lower figure: HP-INNOWax, 30 m, TP: 50(1)-20-
150(0)-5-240(5), CHP: 110 kPa.
Structure elucidation and identification of compo-
nent B
Isolation of components A, B, and C
When about 12 FE (about 2.5 ng) of component B
was injected into a GC–MS equipped with a polar HP-
INNOWax column, it was eluted at tR ¼ 13:53 min (KI
2571) and showed a similar mass spectrum (Fig. 2B) to
that of (6Z)-henicos-6-en-11-one (2).12) A molecular ion
was observed at m=z 308 (intensity 17%, base m=z 124),
and important fragment ions were m=z 290 (4%,
M ꢂ 18), m=z 169 (85%, [C10H21C=O]þ), m=z 167
(29%, [C10H19C=O]þ) and m=z 124 (base ion derived
via a McLafferty rearrangement with charge retention on
the hydrocarbon fragment; see Smith et al. (1978)12)).
This interpretation suggests that the component was 2.
To confirm the olefin position, about 120 FE (25 ng)
of the 1% fraction was subjected to DMDS derivatiza-
tion and the product was analyzed with GC–MS. In the
mass spectrum, the molecular ion at m=z 402 (1.8%,
base m=z 87) and diagnostic fragment ions at m=z
131 (57%, [C5H11CHSCH3]þ) and m=z 271 (39%,
[CH3SCH(CH2)3COC10H21]þ) confirmed the olefin
bond at the 6-position. To determine the geometric
isomerism, authentic 2 and its (6E)-isomer were
Components A and B in the Ishigaki strain extracts
were eluted as a mixture in the 5%-ether-in-hexane
fraction from Florisil column chromatography. This
fraction was further chromatographed on silica gel.
Components A and B were eluted with 0.5% and 1%
ether in hexane (0.5% and 1% fractions) respectively.
The latter fraction still contained component A. Com-
ponent B was further purified by repeating the same
chromatography on silica gel. Although the purified
fraction still contained about 30% of other impurities, it
was subjected to chemical derivatization and subsequent
GC–MS analysis without further purification to avoid
the risk of loss (the amount of compound B in this
fraction was about 100 ng).
Component C in the hexane fraction from Florisil
chromatography of the extracts from the Ishigaki and
Okinawa females showed a mass spectrum in which the
molecular ion was observed at m=z 292 and a series of
fragment ions was spaced at 14 mass units in the GC–MS
analyses (Fig. 2C). This suggested that component C

trans-Epoxide Sex Pheromone of Orgyia postica
961
Fig. 2. EI Mass Spectra of EAG-Active Components A, B, and C in the Extract of Abdominal Tips of Ishigaki Orgyia postica Females.
JEOL JMS SX-102A double focusing magnetic sector mass spectrometer, 70 eV, 200 ꢀC.
Fig. 4. Dose Response Relationships of EAG Response of Orgyia
postica Males to Compound A from Okinawa and Ishigaki Females
and Authentic Enantiomers of (6Z,9Z)-trans-11,12-Epoxyhenicosa-
6,9-diene (1).
Fig. 3. Separation of Enantiomers of (6Z,9Z)-trans-11,12-Epoxy-
henicosa-6,9-diene (1) on Chiral HPLC and GC Quantification of 1
in the HPLC Fractions.
Values were normalized: 1.0 for 3 ng of a racemic mixture of 1.
Values accompanied by the same letters at the same doses are not
significantly different by one-way-layout ANOVA and subsequent
ranking by Tukey’s method (P < 0:01).
analyzed comparatively with component B by GC–MS
equipped with polar HP-INNOWax and apolar HP-1
GC-columns. The observed KI values for component B
were 2571 and 2224, and these values were almost equal
to those for authentic 2 (KI 2572 and 2224) but different
from those for its (6E)-isomer (KI 2577 and 2234) on a
polar HP-INNOWax and an apolar HP-1 column
respectively. Thus component B was identified as 2.

962
S. WAKAMURA et al.
Table 1. Captures of Male Orgyia postica with (11S,12S)-1,
(11R,12R)-1 and Their Racemic Mixture on Ishigaki and Okinawa
Structure elucidation and identification of compo-
nent C
Component C was partially hydrogenated by diimide
Islands
reduction to obtain a mixture of saturated and monoenyl
and dienyl compounds. The mixture was injected into a
GC–MS equipped with a polar HP-INNOWax column.
A saturated compound was eluted at tR ¼ 8:06 min (KI
2100), and the mass spectrum was identical to that of
authentic henicosane. Hence, the original component C
was determined to have a straight carbon chain.
To determine the position of two double bonds in
component C, the 20% fraction (about 45 FE) was
subjected to reductive ozonolysis and subsequent GC–
MS analysis. Two products were found and identified as
hexanal (tR ¼ 1:65 min) and dodecanal (tR ¼ 8:55 min),
since the retention times and mass spectra were identical
to those of authentic compounds. Thus component C
was 6,9-henicosadiene.
Dose (mg/septum)
Male catches (mean ꢄ SE)
(11S,12S)-1
(11R,12R)-1
Ishigakia
Okinawab
500
250
100
0
0
0
2:4 ꢄ 0:84 a
4:0 ꢄ 0:62 a
1:9 ꢄ 0:46 ab
—
1:1 ꢄ 0:40 bc
0
—
0 c
500
250
100
500
250
50
0 c
0
0 c
—
0 c
0
—
1:6 ꢄ 0:48 abc
0:3 ꢄ 0:18 bc
—
500
250
50
0
—
3:4 ꢄ 0:90 ab
1:1 ꢄ 0:67 bc
0 c
0
0 c
aValues are means and standard errors of pooled catches for three nights for
one trap from 25 May to 14 June 2002 (¼ 7 replications).
bMeans for one trap for seven nights from 15 to 19 and from 26 to 29 June
2000 (¼ 7 replications). Means followed by the same letters are not
significantly different (P < 0:05). See text for details of statistical analyses.
To determine geometric isomerism in component C,
retention times for the DMDS derivatives of partial
hydrogenation of component C were measured on an
apolar HP-1 column. The retention indices (KI values)
of authentic DMDS derivatives of (Z)-6-, (E)-6-, (Z)-9-,
and (E)-9-henicosenes were 2705, 2714, 2692 and 2701
respectively. The values of derivatives of 6- and 9-
henicosene from component C were 2704 and 2693,
which were almost equal to those for (Z)-6- and (Z)-9-
henicosenes respectively. Consequently, the original
component C was determined to have two double bonds
of (Z)-configuration. Authentic (6Z,9Z)-henicos-6,9-di-
ene (3) showed the same mass spectra and the same KI
value (2162) on the HP-INNOWax column in GC–MS
analysis. Almost identical results were obtained from
chemical analysis of component C of the Okinawa
strain. Hence, component C was identified as 3 in both
the Ishigaki and the Okinawa strains.
Attraction of O. postica males in the field
Field tests were conducted in Naha, Okinawa Island
in June 2000 and in May and June 2002, and on Ishigaki
Island in June 2002 (Table 1). Males were captured with
(11S,12S)-1 but not with (11R,12R)-1 on both Okinawa
and Ishigaki. Trap catches with the racemic mixture
were comparable to those with (11S,12S)-1 on Okinawa
but likely inferior on Ishigaki. This confirmed that
(11S,12S)-1 is a component of the sex attractant
pheromone of O. postica.
Trap catches with the seven possible blends of
(11S,12S)-1, 2, and 3 at approximate natural ratios
(500:50:500 mg/rubber septum) were examined on
Okinawa and Ishigaki (Fig. 5). On Okinawa, trap catch
with (11S,12S)-1 was significantly reduced when blend-
ed with 2. The reduction by blending with this
compound was not significant on Ishigaki. Blending 3
with (11S,12S)-1 did not significantly affect the trap
catches. No males were captured with a blend of 2 and 3
or their single applications.
Fig. 5. Captures of Male Orgyia postica with Blends of (6Z,9Z,11S,
12S)-11,12-Epoxyhenicosa-6,9-diene [(11S,12S)-1], (6Z)-Henicos-
6-en-11-one (2), and (6Z,9Z)-Henicosa-6,9-diene (3) on Okinawa
and Ishigaki Islands.
Values are means and SE for one trap for 12 nights between 23
June and 08 July 2002 for Okinawa (pooled for every 2 nights, 6
repetitions), and for one trap for 15 nights from 25 May to 14 June
2002 for Ishigaki (pooled for every 3 nights, 5 repetitions). Means
accompanied by the same letters are not significantly different
(P < 0:05). See text for details of statistical analyses.

trans-Epoxide Sex Pheromone of Orgyia postica
963
tionary process. This hypothesis, however, raises more
questions. Why has male inhibition by 2 evolved in
Okinawa strain, and why has neither strain abandoned
3? It is therefore interesting to observe sex pheromone
composition in Taiwan, where lymantriid moth species
are highly diverse.18)
Discussion
Three EAG-active components, A, B, and C, in the
extracts of abdominal tips of O. postica females were
identified as (6Z,9Z,11S,12S)-11,12-epoxyhenicosa-6,9-
diene [(11S,12S)-1], (6Z)-henicos-6-en-11-one (2), and
(6Z,9Z)-henicosa-6,9-diene (3) respectively. (11S,12S)-
1 was found to be an essential component for male
attraction in the field experiments, while the antipode
(11R,12R)-1 was found not to be attractive (Table 1). In
contrast, 2 was apparently inhibitory for males of the
Okinawa strain (Fig. 5). In this strain, females lacked
this component in the pheromone gland (Fig. 1).
Pheromone polymorphism has been suggested for the
Ishigaki and Okinawa populations.2) Two distinct EAG-
active components were found in the extract of female
abdominal tips of the Ishigaki strain but only one in
those of the Okinawa strain. Virgin Ishigaki females
were also found to be generally inferior to those of the
Okinawa strain in attraction of male moths on both
islands. Two possibilities are proposed to explain this
phenomenon: (1) the attractiveness potentials are not
significantly different between Okinawa and Ishigaki
strains, but the Okinawa females release their sex
pheromone for a longer time than the Ishigaki females,
and (2) the second EAG-active component B (2), found
only in the Ishigaki strain, has some particular signifi-
cance in this context. The fact that 2 has an inhibitory
effect on male attraction in the Okinawa strain supports
the latter possibility.
O. postica is distributed in India, Indonesia, New
Guinea, China, Taiwan, and the Ryukyu Islands in
Japan.13) This distribution suggests that the main and
original habitats of this species are in continental East
Asia. Several hypotheses about the geo-history of the
Ryukyu Islands have been proposed by Kizaki and
Oshiro,14,15) Ujiie,16) and Ota.17) A dominant hypothesis
is that the Ryukyu Islands were connected to the Asian
Continent by a land bridge in the Miocene epoch, and
then became separated by rising sea levels during the
Pliocene or Pleistocene. O. postica perhaps have dis-
persed from the Asian Continent to the Ryukyu Islands
over this land bridge and then diversified through insular
isolations.
Okinawa and Ishigaki are separated by about 380 km
of ocean, so gene flow of O. postica between the two
populations must have become extremely scant after
separation. The differences in the pheromone compo-
nents between the strains suggest that the populations of
the two islands diverged to some extent. Recently, the
genetic distance between the Okinawa and Ishigaki
strains was determined by analysis of mitochondrial
DNA sequences. The distance is considered to corre-
spond to about 5 millions years isolation (Nomura et al.,
personal communication). Hence we hypothesize that
the loss of a component with no value in pheromonal
communication in Okinawa might have been an evolu-
In the Taiwan population, Chow et al. (2001)19)
mentioned that 1 and 2 were identified as sex pheromone
components, but they did not assign geometric and
positional isomerism. Although they presented a gas
chromatogram of the crude extract [Fig. 1 in Chow et al.
(2001)19)], it is hard to estimate the contents or ratio of 1
and 2 because of their incorrect peak assignments. They
assigned 1 for the peak that followed the peak of 2, but
this order was never reproducible in our follow-up GC-
analyses. Nevertheless, we reinterpret their chromatog-
ram and speculate that the content of 2 would be greater
in Taiwan than in Ishigaki. Their identification result is
probably correct, and 1 in Taiwan may also have a
(11S,12S)-configuration. An examination of the effect of
2 in the Taiwan population is a future effort.
Compound 2 has been identified as a sex pheromone
component only in genus Orgyia, in O. pseudotsugata
(McDonough)20) and O. thyellina Butler.21) Compound 3
has been found as a sex pheromone component in Mocis
latipes Guenee22) (Noctuidae), Panaxia quadripunctaria
Poda23) (Arctiidae), Lobophora nivigerata Walker24)
(Geometridae), and other species [see Ando (2003);25)
Arn et al. (1992; 1997)26,27)]. Furthermore, 3 has been
found to be an attractant for males of two geometrids,
Agathia visenda visenda Prout11) and Synegia esther
Butler.28) Many mono-epoxy compounds derived from
(3Z,6Z,9Z)-trienes and (6Z,9Z)-dienes, which are bio-
synthesized from linolenic and linoleic acids, have been
identified as sex pheromone components secreted by
females in several families of Lepidoptera.29) Males of
many more species have been found to be attracted to
synthetic epoxides having
a
cis configuration.28)
(11S,12S)-1 is unique in possessing a trans-epoxy ring
in the structure, and is a new member of the ‘‘linolenic
and linoleic acid’’ family of pheromone components. It
should provide an interesting focus for biosynthetic
study.
The field population level was low for both strains
from 2001 to 2004, which made further detailed field
tests by us fruitless. For a more detailed understanding
of component functions, further field tests should be
conducted when the field populations increase. Further-
more, the synthesis of (11S,12S)-1 is not easily attained
at present, although synthesis trials were accomplished
by Muto and Mori (2001)30) and Fernandes and Kumar
(2002)31) after our rapid paper.3) Nevertheless, synthetic
pheromone would be useful for population monitoring
and mating disruption of this pest. Economical methods
of synthesis and stabilization of this compound need to
be developed before utilization for pest control becomes
possible.

964
S. WAKAMURA et al.
13) Inoue, H., Lymantriidae. In ‘‘Moths of Japan’’ (in
Acknowledgments
Japanese), eds. Inoue, H., Sugi, S., Kuroko, H., Moriuti,
S., and Kawabe, A., Kodan-sha, Tokyo, pp. 628–638
(1982).
We thank Dr. Masashi Nomura of Chiba University
for invaluable personal information on comparative
mitochondrial DNA analyses of O. postica. We also
thank Mr. Serge Glushkoff for editing the manuscript.
14) Kizaki, K., and Oshiro, I., Paleogeography of the
Ryukyu Islands. Marine Science Monthly (in Japanese),
9, 542–549 (1977).
15) Kizaki, K., and Oshiro, I., The origin of the Ryukyu
Islands. In ‘‘Ryu¯kyu¯ no Shizenshi’’ (in Japanese), ed.
Kizaki, K., Tsukiji-shokan, Tokyo, pp. 8–37 (1980).
16) Ujiie, H., Geographical history of the Ryukyu Island
Arc. In ‘‘Nature of Okinawa: Geomorphology and
Geology’’ (in Japanese), ed. Ujiie, H., Hirugisha, Naha,
pp. 251–255 (1990).
References
1) Arakaki, N., Kawasaki, K., and Yoshimatsu, S., List of
insects and other pests of the litchi plant in Okinawa.
Okinawa No¯gyo¯ Shikenjo Ho¯koku (in Japanese), 18, 29–
38 (1997).
2) Arakaki, N., and Wakamura, S., Different electroanten-
nograms and field responses in males to virgin females
between Okinawa and Ishigaki strains of the tussock
moth, Orgyia postica (Lepidoptera: Lymantriidae).
Entomol. Sci., 3, 421–426 (2000).
3) Wakamura, S., Arakaki, N., Yamamoto, M., Hiradate,
S., Yasui, H., Yasuda, T., and Ando, T., Posticlure: A
novel trans-epoxide as a sex pheromone component of
the tussock moth, Orgyia postica (Walker). Tetrahedron
Lett., 42, 687–689 (2001).
4) Struble, D. L., and Arn, H., Combined gas chromatog-
raphy and electroantennogram recording of insect
olfactory responses. In ‘‘Techniques in Pheromone
Research’’, eds. Hummel, H. E., and Miller, T. A.,
Springer-Verlag, New York, pp. 161–178 (1984).
17) Ota, H., Geographic pattern of endemism and speciation
in amphibians and reptiles of the Ryukyu archipelago,
Japan, with special reference to their paleogeographical
implications. Res. Popul. Ecol., 40, 189–204 (1998).
18) Wang, Y. W., ‘‘Illustrations of Lymantriidae in Taiwan’’
(in Chinese), Scientific Publishers, Inc., Taipei, Taiwan,
p. 118 (1993).
19) Chow, Y. S., Tsai, R. S., Gries, G., Gries, R., and
Khaskin, G., Preliminary identification of the sex
pheromone of Orgyia postica (Walker) (Lepidoptera:
Lymantriidae). Entomol. Sin., 8 (Suppl.), 13–20 (2001).
20) Smith, R. G., Daterman, G. E., and Daves, G. D., Jr.,
Douglas-fir tussock moth: Sex pheromone identification
and synthesis. Science, 188, 63–64 (1975).
21) Gries, G., Clearwater, J., Gries, R., Khaskin, G., King,
S., and Schaefer, P., Synergistic sex pheromone compo-
nents of white-spotted tussock moth, Orgyia thyellina.
J. Chem. Ecol., 25, 1091–1104 (1999).
´
5) Kovats, E., Gas chromatographic characterization of
organic substances in the retention index system. Adv.
Chromatogr., 1, 229–247 (1965).
6) Buser, H.-R., Arn, H., Guerin, P., and Rausher, S.,
Determination of double bond position in mono-unsat-
urated acetates by mass spectrometry of dimethyl
disulfide adducts. Anal. Chem., 55, 818–822 (1983).
7) Yamaoka, R., Fukami, H., and Ishii, S., Isolation and
identification of the female sex pheromone of the potato
tuberworm moth, Phthorimaea operculella. Agric. Biol.
Chem., 40, 1971–1977 (1976).
8) Beroza, M., and Bierl, B. A., Rapid determination of
olefin position in organic compounds in microgram
range by ozonolysis and gas chromatography. Anal.
Chem., 39, 1131–1135 (1967).
9) Katsuki, T., and Sharpless, K. B., The first practical
method for asymmetric epoxidation. J. Am. Chem. Soc.,
102, 5974–5976 (1980).
10) Hendry, L. B., Korzeniowski, S. H., Hindenlang, D. M.,
Kosarych, Z., Mumma, R. O., and Jugovich, J., An
economical synthesis of the major sex attractant of the
oak leaf roller—cis-10-Tetradecenyl acetate. J. Chem.
Ecol., 1, 317–322 (1975).
11) Ando, T., Ohsawa, H., Ueno, T., Kishi, H., Okamura, Y.,
and Hashimoto, S., Hydrocarbons with a homoconju-
gated polyene system and their monoepoxy derivatives:
Sex attractants of geometrid and noctuid moths dis-
tributed in Japan. J. Chem. Ecol., 19, 787–798 (1993).
12) Smith, L. M., Smith, R. G., Loehr, T. M., and Daves, G.
D., Jr., Douglas fir tussock moth pheromone: Identifica-
tion of a diene analogue of the principal attractant and
synthesis of stereochemically defined 1,6-, 2,6-, and 3,6-
heneicosadien-11-ones. J. Org. Chem., 43, 2361–2366
(1978).
22) Descoins, C., Lalanne-Cassou, B., Malosse, C., and
Milat, M.-L., Analysis of the sex pheromone produced
´ ´
by the virgin female of Mocis latipes (Guenee),
Noctuidae, Catocalinae, from Guadeloupe (French
Antilla). C. R. Acad. Sci. Ser. III, 302D, 509–512 (1986).
23) Schneider, D., Schulz, S., Priesner, E., Ziesmann, J., and
Franke, W., Autodetection and chemistry of female and
male pheromone in both sexes of the tiger moth Panaxia
quadripunctaria. J. Comp. Physiol., A182, 153–161
(1998).
24) Millar, J. G., Giblin, M., Barton, D., and Underhill, E.
W., Sex pheromone components of the geometrid moths
Lobophora nivigerata and Epirrhoe sperryi. J. Chem.
Ecol., 18, 1057–1068 (1992).
25) Ando, T., Sex Pheromones of Moths. Available online at
http:www.tuat.ac.jp/~antetsu/index-e.html (2003) (veri-
fied on 20 November 2004).
´
26) Arn, H., Toth, M., and Priesner, E., ‘‘List of Sex
Pheromones of Lepidoptera and Related Attractants’’
2nd ed., International Organization for Biological Con-
trol, Montfavet, p. 179 (1992).
´
27) Arn, H., Toth, M., and Priesner, E., List of sex
pheromones of Lepidoptera and related attractants,
Supplement 1992–1996, Technology transfer in mating
disruption. IOBC/WPRS Bull., 20, 257–293 (1997).
28) Ando, T., Kishi, H., Akashio, N., Qin, X.-R., Saito, N.,
Abe, H., and Hashimoto, S., Sex attractants of geometrid
and noctuid moths: Chemical characterization and field
test of monoepoxides of 6,9-dienes and related com-
pounds. J. Chem. Ecol., 21, 299–311 (1995).
29) Millar, J. G., Polyene hydrocarbons and epoxides: A

trans-Epoxide Sex Pheromone of Orgyia postica
965
second major class of lepidopteran sex attractant
pheromones. Annu. Rev. Entomol., 45, 575–604 (2000).
30) Muto, S., and Mori, K., Synthesis of posticlure [(6Z,9Z,
11S,12S)-11,12-epoxyhenicosa-6,9-diene], the female
sex pheromone of Orgyia postica. Eur. J. Org. Chem.,
24, 4635–4638 (2001).
31) Fernandes, R. A., and Kumar, P., Asymmetric di-
hydroxylation and one-pot epoxidation route to (þ)-
and (ꢂ)-posticlure: a novel trans-epoxide as a sex
pheromone component of Orgyia postica (Walker).
Tetrahedron, 58, 6685–6690 (2002).