Parasitic wasp, Dinarmus basalis, utilizes oviposition-marking pheromone of host azuki bean weevils as host-recognizing kairomone

Article abstract of DOI:10.1023/A:1026425407150

A host-recognizing kairomone responsible for the stinging behavior of the parasitic wasp, Dinarmus basalis, was studied. Fresh azuki beans coated with an acetone extract of the azuki beans, from which both emerged wasps and their host weevils were removed, elicited stinging behavior from female wasps. The kairomone is a mixture of saturated hydrocarbons and diacylglycerols, both of which are required for activity. The kairomone is composed of normal and methyl-branched hydrocarbons with carbon numbers ranging from 25 to 35, most of which are known as the hydrocarbon constituents of an oviposition-marking pheromone of the host azuki bean weevils, Callosobruchus chinensis. This indicates that D. basalis utilizes the oviposition-marking pheromone of its host weevils as a host-recognizing kairomone.

Full text of DOI:10.1023/A:1026425407150

Journal of Chemical Ecology, Vol. 26, No. 12, 2000
PARASITIC WASP, Dinarmus basalis, UTILIZES
OVIPOSITION-MARKING PHEROMONE OF HOST AZUKI
BEAN WEEVILS AS HOST-RECOGNIZING KAIROMONE
MOTONARI KUMAZAKI,1,3 SHIGERU MATSUYAMA,1
TAKAHISA SUZUKI,1,* YASUMASA KUWAHARA,1,4
and KOICHI FUJII2
1Institute of Applied Biochemistry
2Institute of Biological Sciences
University of Tsukuba
Tsukuba, Ibaraki 305-8572, Japan
3Present address: Takeda Chemical Ind. Ltd., Jusohonmachi
2-Chome, Yodogawa-ku, Osaka 532-8686, Japan
4Present address: Graduate School of Agriculture,
Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan.
(Received November 16, 1999; accepted July 29, 2000)
Abstract—A host-recognizing kairomone responsible for the stinging
behavior of the parasitic wasp, Dinarmus basalis, was studied. Fresh azuki
beans coated with an acetone extract of the azuki beans, from which both
emerged wasps and their host weevils were removed, elicited stinging behavior
from female wasps. The kairomone is a mixture of saturated hydrocarbons
and diacylglycerols, both of which are required for activity. The kairomone is
composed of normal and methyl-branched hydrocarbons with carbon numbers
ranging from 25 to 35, most of which are known as the hydrocarbon
constituents of an oviposition-marking pheromone of the host azuki bean
weevils, Callosobruchus chinensis. This indicates that D. basalis utilizes the
oviposition-marking pheromone of its host weevils as a host-recognizing
kairomone.
Key Words—Parasitic wasp, Dinarmus basalis, kairomone, host recognition,
azuki bean weevil, Vigna angularis.
*To whom correspondence should be addressed; e-mail: tksuzuki@agbi.tsukuba.ac.jp
2677
0098-0331 00 1200-2677$18.00 0  2000 Plenum Publishing Corporation
/
/
/

2678
KUMAZAKI ET AL.
INTRODUCTION
Dinarmus basalis (Rondani) (Hymenoptera: Pteromalidae) was first discovered
in the Punjab region of India parasitizing its host, the azuki bean weevil, Calloso-
bruchus chinensis (L.). The wasp was introduced into Japan in 1980. Females
are ectoparasites on larvae and pupae of several bruchid weevils including azuki
bean weevil, cowpea weevil, C. maculatus (F.), C. analis, Mexican bean weevil,
Zabrotes subfasciatus Bohman, and maize weevil, Sitophilus zeamais (Motschul-
sky) (Fujii and Wai, 1990; Verma, 1990; Caubet and Jaisson, 1991). These and
subsequent biological studies (Verma, 1991; Caubet et al., 1992; Islam, 1994;
Nishimura and Johon, 1996; Nishimura, 1997) showed that the fourth-instar lar-
vae of C. chinensis were the most suitable host for reproduction of next gener-
ation D. basalis and that both sexes of D. basalis emerged from eggs laid by
mated females, whereas only males emerged from eggs laid by unmated females.
Female wasps are 2.5 mm in length with black abdomens, while males are
ca. 2.0 mm in length with white abdomens. Thus, the sexes can easily be dis-
tinguished. When a female locates an azuki bean containing a host larva, she
examines the surface of the bean by antennal tapping, bends her abdomen ver-
tically, stings her ovipositor through the surface of the bean (stinging behavior),
and lays an egg on the surface of the larva. Of the sequential behaviors described
above, the stinging behavior is observed only when females locate beans con-
taining host larvae (or pupae).
Several studies that utilized D. basalis as natural enemies against these
legume bruchids have been made in South Asia (Islam and Kabir, 1995) and
West Africa (Sanon et al., 1998), but little is known about the chemical ecol-
ogy. We now report the characterization of the host-recognizing kairomone, from
the azuki beans from which emerged wasps and host weevils were removed, that
is responsible for the stinging behavior.
METHODS AND MATERIALS
Insects. Azuki bean weevils and D. basalis were obtained from original
colonies kept at Fujii’s laboratory (Tsukuba University). Several adult weevils
(both sexes) were transferred to plastic Petri dishes (9 cm ID × 15 mm in height)
packed to half the height with azuki beans, Vigna angularis. Weevils were reared
at 30 C, 55 ∼ 60% relative humidity, and 16L : 8D. After one week, adults were
removed by sifting with a sieve. New weevils began to emerge after 24 days.
D. basalis were reared as follows; several wasps (both sexes) were released into
a plastic Petri dish containing azuki beans in which host larvae (16–17 days after
laying, the fourth instar) were present. Wasps were also reared under the same
environmental conditions as the bruchids. Wasps began to emerge 12 days after

HOST-RECOGNIZING KAIROMONE FOR D. basalis
2679
being exposed to infested beans. The newly emerged wasps (not sexed) were
transferred similarly to other dishes containing azuki beans with the fourth-instar
larvae for reproduction. Mated females were used for bioassay after starvation
of one or two days.
Bioassay. In preliminary studies, the active kairomone(s) eliciting the sting-
ing behavior could be extracted with acetone but not with hexane, ether, or
methanol. The behavior was observed when the extract was applied to a fresh
azuki bean but not to a glass bead or a filter paper.
Based on these observations, the bioassay was conducted as follows: a fresh
azuki bean treated with 5 ml [2.5 mg bean equivalent (BE)] of the test sample
(0.5 g BE ml) was covered with a glass container (18 mm ID × 13 mm in
/
height; mini P, Tsukuba Flat Co.) after evaporation of the solvent. Similarly, a
fresh azuki bean treated with 5 ml of the solvent was used as a control. A mated
female wasp (2 ∼ 3 days old) was released into the container, and her stinging
behavior was observed for 15 min. One assay set was composed of eight pairs
of tests. At least three sets (24 pairs) were repeated for each sample. The total
number of females showing the stinging behavior at least once during 15 min
was counted (N 24, 32, or 40). The assay was done in a room controlled at
25 C and 60% R.H. The activity was evaluated by Fisher’s exact test (P < 0.05).
Analytical Methods. Gas chromatographic (GC) analyses were carried out
by using two capillary columns: fused silica CP-Sil 19 CB column (25 m ×
0.25 mm ID, Chrompack, Middelburg, The Netherlands) with a Hewlett-Packard
5890 (Palo Alto, California) or a Hitachi 263-30 (Hitachi, Ltd., Tokyo, Japan)
gas chromatograph, and an FFAP column (25 m × 0.25 mm ID, quadrex, New
Haven) with a Yanaco G-180F (Yanagimoto Co. Ltd., Kyoto, Japan). Nitro-
gen was the carrier gas. The column temperatures were: 200–270 C at 4 C min
/
for analyses of hydrocarbons (HP 5890), isothermally at 230 C for measuring
Kovats retention indices (Hitachi 263-30), and isothermally at 160 C for anal-
yses of fatty acid methyl esters (Yanaco G-180 F). Mass spectra (MS) were
measured with a Hitachi M-80B GC-mass spectrometer by using an Al-clad
methylsilicone (nonpolar) capillary column (25 m × 0.22 mm ID, quadrex)
(EI, 70 eV). Mass spectra of diacylglycerols were obtained by the direct inlet
method, and those of other compounds were obtained by GC-MS. The col-
umn temperatures were: 200–270 C at 4 C min for analyses of saturated hydro-
/
carbons, 135–250 C at 4 C min for analyses of fatty acid methyl esters, and
/
200–350 C at 4 C min for analyses of dimethyl disulfide (DMDS) adducts of
/
the methyl esters. Helium was used as the carrier gas. Nuclear magnetic reso-
nance (NMR) spectra were recorded on a Bruker AM-500 (German) or a Jeol
EX 270 (Jeol Ltd., Tokyo, Japan) spectrometer with TMS as an internal standard
in CDCl3. Thin-layer chromatography (TLC) was done with silica gel (SiO2)
60F254 plates (E. Merck, Darmstadt, Germany; precoated plates, 5 × 20 cm, 0.2
mm thickness) and AgNO3-impregnated SiO2 plates made by dipping the plates

2680
KUMAZAKI ET AL.
FIG. 1. Extraction procedure of the kairomones and purification of the active saturated
hydrocarbons. The activity when used alone ([ ]) or mixed (< >) with the methanol
fraction, respectively. : inactive; +: active; Et2O: ethyl ether; Hex: hexane; MeOH:
methanol.
in 5% AgNO3–acetonitrile solution and activation for 2 hr at 105 C. Spots were
detected by UV (245 or 365 nm) or by exposing to iodine vapor the SiO2 60F254
plates and by spraying with 50% sulfuric acid and heating at 105 C or with
0.02% rhodamine 6G ethanol solution followed by UV (365 nm) (fluorescent
spots) for AgNO3-impregnated SiO2 plates.
The double bond positions of unsaturated fatty acids (after methylation with
diazomethane) were determined based on the diagnostic ions derived from their
DMDS adducts (Francis and Veland, 1981; Vincenti et al., 1987).
Extraction and Purification of Kairomone(s). As described above, the
kairomone(s) were extracted with acetone from the beans containing host lar-
vae. Since female wasps also showed the stinging behavior for the azuki beans
from which both emerged wasps and host weevils were removed, the extrac-
tion was conducted by using azuki beans (1.4 kg; ca. 70 mg bean) from which
/
emerged host weevils were removed (Figure 1). The acetone extract was fil-
tered through cotton, and the filtrate was concentrated to dryness. The residue
(3.30 g) was partitioned with ether and water, and the active ether layer (2.60
g) was purified as shown in Figure 1. No fraction showed any activity, but the
activity was recovered by mixing the hexane and the methanol fractions. The
hexane fraction (0.94 g) was fractionated by using a 5% AgNO3-impregnated

HOST-RECOGNIZING KAIROMONE FOR D. basalis
2681
SiO2 column. Only the hexane eluate (0.79 g) comprising saturated hydrocar-
bons displayed the activity by mixing it with the methanol fraction. The saturated
hydrocarbons were analyzed by GC, GC-MS, and Kovats retention indices.
The methanol (MeOH) fraction (0.12 g) was separated into acidic, neu-
tral, and basic fractions, as follows: it was evaporated to dryness and dissolved
in ether (50 ml). This was extracted successively with 2 NHCl (25 ml × 3),
2 N NaOH (25 ml × 3), and water (25 ml × 2) to give a neutral fraction (0.02
g). The acidic aqueous layer (75 ml) was treated with 2 N NaOH (80 ml) and
extracted with ether (75 ml × 3). The ether layer (225 ml) was washed with
water to yield the basic fraction (almost no residue). The aqueous basic layer
(75 ml) was acidified with 2 N HCl (80 ml), extracted with ether (75 ml × 3),
and washed with water to give the acidic fraction (0.19 g). Only the neutral frac-
tion showed activity when mixed with the hexane fraction. However, the small
amount of the neutral fraction prevented further chemical analyses. Therefore,
the extraction was scaled up by using azuki beans (18 kg) from which emerged
insects were removed, and the extract was treated by a method similar to the first
extraction, except elution with 10% ether in hexane was omitted this time. The
active methanol fraction [4.30 g; hereafter, showing kairomone activity when
mixed with the hexane fraction (12.02 g)] was separated into acidic, neutral,
and basic fractions as described above, in which only the neutral fraction (0.28
g) was active. This was submitted to SiO2 column chromatography (the first
SiO2 column chromatography) according to the procedure shown in Figure 2.
Both 1% (17.25 mg) and 5% EtOAc–benzene (Bz) (5.55 mg) fractions
obtained by the third SiO2 column chromatography were active and further stud-
ied. The former fraction was less polar and gave two spsots (Rf 0.63 and a
tailing spot) in TLC (Bz–EtOAC 1 : 1, v v). The latter spot was also observed
/
in the Bz fraction and showed no activity. Therefore, the former spot (Rf 0.63)
was separated from the 1% EtOAc–Bz fraction by SiO2 chromatography. The
5% EtOAc–Bz eluate (7.75 mg) yielded a single spot (Rf 0.57, named EB-1).
EB-1 was further separated into two spots by AgNO3–SiO2TLC (CHCl3–MeOH
95 : 5). EB-1A (Rf 0.61) and EB-1B (Rf 0.53) were separated by preparative
TLC by using AgNO3-impregnated SiO2 plates.
Another active 5% EtOAc-Bz fraction (5.55 mg) from the third SiO2 col-
umn chromatography yielded one spot (Rf 0.57; CHCl3–MeOH 1 : 1). Prepar-
ative TLC with AgNO3–SiO2 plates of this spot gave the active compound (2.18
mg; EB-5). The chemical structures of EB-1 and EB-5 were analyzed by MS
and NMR.
Analysis of Saturated Hydrocarbons of Azuki Beans and Larvae or Adult
Weevils. Since we employed azuki beans from which both emerged insects were
removed, it is possible that the source of the saturated hydrocarbons is derived
from secretions (adults and larvae), excrement (larvae), and or azuki beans them-
/
selves. To confirm the source of the saturated hydrocarbons, adult weevils (500

2682
KUMAZAKI ET AL.
FIG. 2. Large scale extraction of the kairomones and purification of the active diacylglyc-
erols. The activity when used alone ([ ]) or mixed (< >) with the hydrocarbon fraction,
respectively. : inactive; +: active; Et2O: ethyl ether; Hex: hexane; EtOAc: ethyl acetate;
Bz: benzene.
individuals) and their larvae (112 individuals) were extracted separately with
hexane for 3 min. Each hexane extract was treated with SiO2 chromatography
followed by AgNO3-impregnated SiO2 column chromatography to give cuticu-
lar saturated hydrocarbons. Fresh azuki beans (300 g, crushed) were extracted
with acetone for 24 hr, and upon evaporation of the solvent, the extract was par-

HOST-RECOGNIZING KAIROMONE FOR D. basalis
2683
titioned with ether and water. The ether layer was treated by the same method
to yield hydrocarbons.
Synthesis of Diacylglycerols. Several 1,2-diacylglycerols were prepared by
the procedure of Mattson and Volpenhein (1961). Hydrolysis of triacylglyc-
erols with pancreas lipase type II (EC 3.1.1.3.; Sigma, St. Louis, MO, USA)
in the presence of bile salts (Difco bile salts No. 3) and CaCl2 in Tris buffer
solution (pH 8.0) at 40 C, yielded 2-acylglycerols. These were treated with
desired, equimolar acyl chlorides to give the target 1,2-diacylglycerols. The cor-
responding 1,3-diacylglycerols were obtained as byproducts from isomerization
of the above reactions. For example, 2-oleoyl-1-palmitoylglycerol (2O1P) and
1-oleoyl-3-palmitoyl-glycerol (1O3P) were prepared as follows: a mixture of tri-
oleoylglycerol (1O2O3O; 3.02 g, 3.41 mmol, purity >99%; Wako Pure Chemical
Industries, Ltd., Osaka, Japan), Tris buffer (1 M, 36 ml, pH 8.0), CaCl2aq. solu-
tion (45%, 2 ml), and bile salt (1% aq. solution, 0.8 ml) wsa stirred at 40 C in
a test tube, to which pancreas lipase (500 mg suspended in 4 ml of 1 M Tris
buffer) was added. This mixture was vigorously stirred at 40 C for 45 min. The
reaction mixture was treated with ethanol (40 ml) and 6 N-HCl (20 ml) to stop
hydrolysis. After dilution with water, the 2-monooleoylglycerol (2O) formed was
extracted with CH2CL2 (150 ml). The CH2Cl2 layer was washed successively
with saturated NaHCO3 (×4) and saturated NaCl and dried over Na2SO4.
The crude 2O was purified by SiO2 column chromatography with stepwise
elution of the ether–hexane solvent system. Pure 2O (380.7 mg, 1.06 mmol) was
obtained from the ether eluate, which was treated with palmitoyl chloride (359.3
mg, 1.27 mmol; Nakalai Tesque, Inc., Kyoto, Japan) in the presence of pyridine
(85 ml, 1.06 mmol) in anhyd. CHCl3(10 ml) for three days at room temperature.
The reaction mixture was washed successively with water, 2 N HCl, and water
(×2), and dried over Na2SO4. Upon evaporation, the residue was purified by
SiO2 column chromatography. A mixture of 2O1P and 1O3P (191.4 mg, 3.22 ×
1
10 mmol, yield 9.4%) was obtained from the 40% ether–hexane eluate. From
a part of this mixture, pure 2O1P (5.90 mg) and 1O3P (6.76 mg) were obtained
from further SiO2 column chromatography. Similarly, 1-oleoyl-2-palmitolyglyc-
erol (1O2P), 2-oleoyl-1-stearoylglycerol (2O1S), and 1-oleoyo-3-stearoyl-glyc-
erol (1O3S) were prepared, but the yields were low, because solid tripalmitoyl-
glycerol (1P2P3P, mp 66–67 C) and tristearoylglycerol (1S2S3S, mp 68 C) were
not liquefied and barely hydrolyzed by lipase in the reaction mixture at 40 C, and
large amounts of them were recovered. These synthetic diacylglycerols showed
one spot in TLC (hexane–ether–formic acid 25 : 25 : 1, v v).
/
1
The MS, H NMR and 13C NMR spectra of the synthetic diacylglycerols
were measured. MS (relative abundance): 2O1P: 576(M+ H2O, 10), 339(M+
OCOC15H31, 18), 313(M+ OCOC17H34, 59), 264(30), 239(M+ COC15H31,
25), 43(100); 1O3P: 576(9), 339(14), 313(44), 264(21), 239(19), 43(100); 1O2P:
576(13), 339(20), 313(66), 264(31), 239(28); 2O1S: 604(M+ H2O, 6), 341(M+

2684
KUMAZAKI ET AL.
OCOC17H34, 36), 339(M+
OCOC17H35, 16), 267(M+
COC17H35, 14),
264(21), 43(100); 1O3S: 604(7), 341(33), 339(16), 267(14), 264(18), 43(100).
Kairomonal Activity of Candidate Diacylglycerols. The six candidate di-
acylglycerols, 1O2P, 2O1P, 2O1S, 1O3S, 1O3O, 1O3P; three related mono-
acylglycerols, 1-monopalmitoylglycerol (1P), 1-monooleoylglycerol (1O), and
1-monostearoylglycerol (1S); and four triacylglycerols, 1P2P3P, 1S2S3S,
1O2O3O, and triacetylglycerol (1A2A3A) were used for bioassay. The former
four diacylglycerols were synthetic compounds, and the others were purchased
from Wako Pure Chemical Industries, Ltd.). These acylglycerols were assayed
by mixing with the saturated hydrocarbon fraction (the first extraction). A fixed
3
quantity (2.5 BE 1.41 × 10 mg) of hydrocarbons was used for all the sample
7
mixtures. The quantities of the acylglycerols used were at doses of 3 × 10 or
8
9
7
3 × 10 6, 3 × 10 and 3 × 10 mg. The dose of 3 × 10 mg was decided on
as follows: the combined diacylglycerols (9.93 mg) of EB-1 (7.75 mg) and EB-5
(2.18 mg) were composed of five diacylglycerols [1O2P (or 2O1P) and 1O2S
(or 2O1S) for EB-1 and 1O3P, 1O3S, and 1O3O for EB-5], and the mean con-
tent per diacylglycerol was 1.99 mg (obtained from 18 kg of the azuki beans).
Thus, 1.1 × 10 7( 1.99 18 × 106) mg of the diacylglycerols were involved in 1
/
7
mg BE, and 2.5 mg BE were used in the bioassay corresponded to 2.78 × 10
(rounded up to 3 × 10 7) mg.
RESULTS
Structural Determination of Saturated Hydrocarbon Fraction. Individual
fractions were inactive, and the activity was recovered only when the hexane
fraction was mixed with the methanol fraction (Table 1). The hexane fraction
gave the same GC profiles (Figure 3) before and after passing through a AgNO3-
impregnated SiO2 column, indicating that these fractions were composed of sat-
urated hydrocarbons. This was confirmed by GC and GC-MS analysis as well as
by the Kovats retention indices (CP-Sil column, 230 C, isothermal). The authen-
tic saturated hydrocarbons (n-C24–n-C34) were used for measuring the indices.
At least 29 saturated hydrocarbons, with carbon numbers ranging from 25 to
35, were detected from the hexane fraction in which the major peak (7) was
identified as n-heptacosane (M+m z 380). 3-Methylalkanes yielded clear frag-
/
ments at m z M+
29, along with the fragments at m z M+
15, and M+
/
/
57, and the increment of Kovats retention index due to the 3-methyl posi-
tion was ca. 30 (Kojima, 1983) compared with the corresponding n-alkane.
Similarly, internally branched methylalkanes gave fragments at m z M+
15
/
and ion doublets according to the positions of methyl-branching (Lockey, 1978;
Suzuki et al., 1988). The next abundant peak (8) was determined as 11-methyl-
heptacosane, as judged by the diagnostic ions at m z 379 (M+ 15), 253, 252
/

HOST-RECOGNIZING KAIROMONE FOR D. basalis
2685
TABLE 1. KAIROMONAL ACTIVITY OF EACH FRACTION AND THEIR MIXTURES
(FIRST EXTRACTION)
No. of females
a
responding
Dose
Sample
(mg BE)
Sample
Control
P
Activity
Acetone extract
Ether layer
Aqueous layer
Fractions and their mixtures
separated by SiO2 gel column
chromatography
2.5
2.5
2.5
11 24
0 24
<0.001
<0.001
1.000
+
+
/
/
13 24
0 24
/
/
0 24
0 24
/
/
Hex. fr. (1)
2.5
25.0
2.5
2.5
2.5
2.5
25.0
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
3 32
0 32
0.119
1.000
1.000
1.000
1.000
0.245
0.305
0.020
1.000
0.500
0.500
0.013
1.000
0.500
0.756
1.000
1.000
0.119
/
/
0 24
0 24
/
/
10% Et2O–Hex. fr. (2)
50% Et2O–Hex. fr. (3)
Et2O fr. (4)
0 24
0 24
/
/
0 24
0 24
/
/
0 24
0 24
/
/
MeOH fr. (5)
0 24
0 24
/
/
3 24
1 24
/
/
Mixture of (1)–(5)
Mixture of (1) and (2)
Mixture of (1) and (3)
Mixture of (1) and (4)
Mixture of (1) and (5)
Mixture of (2) and (3)
Mixture of (2) and (4)
Mixture of (2) and (5)
Mixture of (3) and (4)
Mixture of (3) and (5)
Mixture of (4) and (5)
Mixtures of MeOH fr. (5) and
each fraction separated by
5% AgNO3–SiO2 column
chromatography
6 24
0 24
+
+
/
/
0 24
0 24
/
/
2 24
1 24
/
/
1 24
0 24
/
/
6 32
0 32
/
/
0 24
0 24
/
/
0 24
1 24
/
/
1 24
1 24
/
/
0 24
0 24
/
/
0 24
0 24
/
/
3 24
0 24
/
/
Mixture of (5) and Hex. fr. (1 )
Mixture of (5) and 1% Et2O–Hex.
2.5
2.5
2.5
9 32
0 32
0.002
0.500
0.500
+
+
′
/
/
fr. (2 )
1 24
/
0 24
/
′
Mixture of (5) and 3% Et2O–Hex.
fr. (3 )
1 24
/
0 24
/
′
Mixture of (5) and 5% Et2O–Hex.
fr. (4 )
2.5
2.5
2.5
1 24
0 24
0.500
1.000
0.003
′
/
/
Mixture of (5) and Et O fr. (5 )
0 24
0 24
′
2
/
/
Mixture of (5) and (1 )–(5 )
7 24
0 32
′
′
/
/
a
A total number of female wasps showing stinging behavior at least one time during 15 min (N
24 or 32). BE: bean equivalent; Et2O: ethyl ether; Hex.: hexane, MeOH: methanol.

2686
KUMAZAKI ET AL.
FIG. 3. Gas chromatogram of the active saturated hydrocarbons. Condition: CP-Sil 19
CB capillary column; 200–270 C, 4 C min. Arabic figures indicate GC peak numbers.
/
(ion doublets), 169, and 168 (ion doublets), and the increment ( 70). Dimeth-
ylalkanes afforded fragments at m z M+ 15, and the increment was about 45.
/
Similarly, 13 other hydrocarbons were determined by GC-MS and Kovats indices
as shown in Table 2. Percent composition of each hydrocarbon component of
the hexane fraction is shown in Table 3.
Chemical Analysis of EB-1 and EB-5 in Methanol Fraction. As with the
first extraction, no kairomonal activity was observed when each primary fraction
was tested alone, and the activity was recovered only after mixing the hexane
and the methanol fractions. In addition, the hexane fractions from the first and
second extractions gave almost identical GC profiles. Table 4 displays the bioas-
say results from the mixtures of the hexane fraction and the subfractions of the
methanol fraction (the second extraction). The fixed quantity (2.5 BE 1.67 ×
3
10 mg) of the hexane fraction was used for all the test mixtures.
The active EB-5 (2.18 mg) and EB-1 (7.75 mg, a mixture of EB-1A and
EB1B) were analyzed by 1H NMR (500 MHz), 13C NMR (125.7 MHz), and MS
(direct inlet method).
EB-5:
A
13C NMR spectrum gave carbon signals as follows (ppm): 14.1
(CH3), 22.7 (CH2), 24.9 (2 CH2), 25.0 (2 CH2), 27.2 (2 CH2), 29.1 (3 CH2),
29.2 (CH2), 29.3 (2 CH2), 29.4 (CH2), 29.5 (2 CH2), 29.6 (CH2), 29.7 (2 CH2),

HOST-RECOGNIZING KAIROMONE FOR D. basalis
2687
TABLE 2. SATURATED HYDROCARBON COMPONENTS IN HEXANE FRACTION
Kovats
GC
Peak
retention
index
MS diagnostic fragment and base peak
(relative abundance)
a
Compound
n-C25
1
2
3
4
5
6
7
8
2497
2571
2597
2631
2655
2670
2699
2731
352(M+, 0.5), 43(100)
351(M+–15, 0.2), 337(5), 309(0.3), 57(100)
3-Me-C25
Me-C26
di-Me-C28(?)
3-Me-C26(?)
n-C27
380(M+, 0.5), 43(100)
379(M+–15, 0.5) 253(8), 252(4), 169(4),
168(10), 57(100)
11-Me-C27
9
10
11
12
13
2747
2756
2772
2800
2829
di-Me-C28
3-Me-C27
n-C28
10-Me-C28
+ 11-Me-C28
di-Me-C29(?)
379(M+–15, 0.2), 365(6.5), 337(0.5), 57(100)
394(M+, 0.8), 57(100)
408(M+, 0.2), 281(0.4), 280(1), 267(0.3), 266
(0.5), 169(0.4), 168(4), 155(6), 154(8), 57(100)
14
15
16
17
2855
2897
2930
n-C29
11-Me-C29
+ 13-Me-C29
408(M+, 0.3), 57(100)
407(M+–15, 0.5), 281(0.7), 280(3), 253(1.7),
252(2), 197(0.8), 196(1.8), 169(4), 168(10),
57(100)
18
2955
9,13-di-Me-C29
436(M+, 0.1), 323(2), 295(0.3), 253(0.3), 252
(0.4), 211(4), 183(2), 141(5), 140(6.5), 57
(100)
19
20
2999
3027
n-C30 (?)
11-Me-C30
+ 12-Me-C30
di-Me-C30
295(1.1), 294(0.6), 281(1), 280(0.7), 183(1),
182(4), 169(3.5), 168(6), 57(100)
435(M+–15, 0.5), 57(100)
21
22
23
3055
3100
3129
11-Me-C31
+ 13-Me-C31
9,13-di-Me-C31
450(M+, 0.1), 309(2), 308(2), 281(1.5), 280
(2), 197(2), 196(3), 169(4), 168(8), 57(100)
449(M+–15, 0.2), 351(2), 323(0.5), 281
(0.7), 280(0.7), 211(4), 141(3), 140(8), 57
(100)
24
3155
25
26
27
28
29
3196
3251
3349
11,15-di-Me-C33
478(M+, 0.1), 351(3), 281(2), 280(2), 239
(4), 169(4), 168(8), 57(100)
a
GC peak number corresponds to that of Figure 3.

2688
KUMAZAKI ET AL.
TABLE 3. COMPOSITION OF SATURATED HYDROCARBON COMPONENTS IN HEXANE
FRACTION AND EXTRACTS FROM ADULTS AND LARVAE OF WEEVILS
Composition (%)
a
GC peak
Hexane fraction
Adult
Larvae
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
1.74
2.13
0.94
0.29
t
0.44
1.53
0.77
0.40
0.29
0.37
17.08
15.08
0.53
1.92
14.38
1.94
1.90
0.84
t
2.90
11.96
9.43
0.91
0.98
0.93
0.49
2.27
4.25
0.39
t
1.80
0.66
1.87
t
16.07
14.02
0.72
1.29
8.89
2.73
2.24
0.67
t
3.01
19.31
9.17
0.82
1.14
1.00
0.42
4.41
4.37
t
47.31
3.98
t
3.51
5.22
1.15
12.22
11.93
4.97
t
t
t
1.76
3.62
t
0.68
t
3.94
0.97
t
7.05
a
GC peak number corresponds to that of Table 2 and Figure 3. t: trace amount.
29.8 (CH2), 31.9 (2 CH2), 34.1 (2 CH2), 34.3 (2 CH2), 61.6. 62.0, 72.2, 129.7
(CH ), 130.1 (CH ), 173.4, and 173.8. TLC behavior and a 13C NMR spectrum
revealed that EB-5 was a 1,2-diacylglycerol. Two carbon signals at 173.4 and
173.8 ppm indicataed the presence of two ester carbonyl groups. Three carbon
signals at 61.2, 62.0, and 72.1 ppm were assigned to two methylene and one
methyne carbons connected with oxygen atoms of glycerol, respectively. The
1H NMR spectrum yielded the proton signals (ppm): 0.88 (6H, t, J 6.88 Hz),
1.27 (br.), 1.61 (2H, quintet, J 6.97 Hz), 1.63 (2H, quintet, J 6.97 Hz),
2.01 (4H, q, J 6.50 Hz), 2.32 (2H, t, J 7.54 Hz), 2.34 (2H, t, J 7.56

HOST-RECOGNIZING KAIROMONE FOR D. basalis
2689
TABLE 4. KAIROMONAL ACTIVITY OF MIXTURES OF HEXANE FRACTION AND
SUBFRACTION OF METHANOL FRACTION (SECOND FRACTION)
No. of females
respondinga
Dose
Sampleb
(mg BE)
Sample
Control
P
Activity
Acetone extract
2.5
12 24
2 24
0.002
+
/
/
Mixtures of Hex. fr. (a) and
three sub-frs. of MeOH fr.
Acidic fr.
Neutral fr.
Basic fr.
2.5
2.5
2.5
3 32
0 32
0.118
0.002
0.245
/
/
8 24
0 24
+
/
/
2 24
0 24
/
/
Mixtures of (a) and each fraction
separated by first SiO2
column chromatography
CHCl3 fr.
1% MeOH–CHCl3 fr.
3% MeOH–CHCl3 fr.
5% MeOH–CHCl3 fr.
MeOH fr.
Mixture of above five frs.
Mixtures of (a) and each fraction
separated by the second SiO2
column chromatography
Bz fr.
10% EtOAc–Bz fr.
20% EtOAc–Bz fr.
50% EtOAc–Bz fr.
EtOAc fr.
2.5
2.5
2.5
2.5
2.5
2.5
8 24
0 24
0.002
0.037
0.005
0.024
0.025
0.005
+
+
+
+
+
+
/
/
8 24
2 24
/
/
7 32
0 32
/
/
7 24
1 24
/
/
5 24
0 24
/
/
7 24
0 24
/
/
2.5
2.5
2.5
2.5
2.5
2.5
6 32
2 32
0.129
0.014
0.025
0.245
0.500
0.025
/
/
8 32
1 32
+
+
/
/
5 32
0 32
/
/
2 24
0 24
/
/
1 24
0 24
/
/
Mixture of above five frs.
Mixtures of (a) and each fraction
separated by the third SiO2
column chromatography
Bz fr.
5 32
0 32
+
/
/
2.5
25.0
2.5
25.0
2.5
2.5
2.5
25.0
2.5
25.0
2.5
0 24
0 24
1.000
0.074
0.120
<0.001
0.034
0.027
0.178
0.009
0.500
0.074
0.034
/
/
7 32
2 32
/
/
1% EtOAc–Bz fr.
3 32
0 32
/
/
12 32
1 32
+
+
+
/
/
2% EtOAc–Bz fr.
5% EtOAc–Bz fr.
10% EtOAc–Bz fr.
10 40
3 40
/
/
7 32
1 32
/
/
4 32
1 32
/
/
10 24
2 24
+
+
+
/
/
Et2O fr.
2 24
1 24
/
/
7 32
2 32
/
/
Mixture of above six frs.
Mixtures of (a) and EB-5 separated
by preparative TLC
10 40
3 40
/
/
EB-5
2.5
5 32
/
0 32
/
0.027
a See Table 1 (N 24, 32, or 40). BE: bean equivalent; Hex.: hexane; Et2O: ethyl ether; MeOH: methanol;
CHCl3: chloroform; Bz: benzene; EtOAc: ethyl acetate.
3
b Fixed quantity (2.5 BE 1.67 × 10 mg) of Hex. fr. (a) was used for all the sample mixtures.

2690
KUMAZAKI ET AL.
Hz), 3.73 (2H, ddd, J 5.54, 5.54, and 1.88 Hz), 4.24 (1H, dd, J 11.89, and
5.66 Hz), 4.32 (2H, dd, J 11.91, and 4.55 Hz), 5.08 (1H, dddd, J 4.97,
4.97, 4.97, and 4.97 Hz), and 5.34 (2H, m). The four multiplets at 3.73, 4.24,
4.32, and 5.08 ppm were assigned to CIIIHaHb, C1Hb, C1Hq and CIIH protons
attached to glycerol carbons (CI–CIII), respectively, when compared with the
authentic 1,2-dioleoylglycerol. The MS spectrum yielded the fragments at m z
/
576 (M+
H2O, 6%), 339 (M+
OCOC15H31, 6), 313 (M+
OCOC17H33,
22), 265 (COC17H33, 34), and 239 (COC15H31, 22), indicating that EB-5 was a
mixture of 1,2-diacylglycerol connected with two fatty acids, octadecenoic, and
hexadecanoic acids.
To determine the acids attached to position 1 of the 1,2-diacylglycerols,
EB-5 was treated with lipase (Mattson and Volpenhein, 1961) to give a mixture
of palmitic, stearic, and oleic acids (4.3 : 1.0 : 5.6). In addition, a spot of 1,3-di-
acylglycerol was detected in TLC (developing solvent: hexane–ether–formic acid
25 : 25 : 1) from the recovered acylglycerols. This indicated that the 1,2-diacyl-
glycerols were partly isomerized to the 1,3-diacylglycerols during the hydroly-
sis by lipase. Hydrolysis of EB-5 with 5% KOH ethanol afforded a mixture of
palmitic, stearic, and eleic acids in a ratio of 4.4 : 1.0 : 5.0. Based on these results,
we conclude that EB-5 is a mixture of 1 (or 2)-oleoyl-2 (or 1)-palmitolylglycerol,
and 1 (or 2)-oleolyl-2 (or 1)-stearoylglycerol (ratio 4.4 : 1.0). Almost no differ-
ence was found between EB-5 and the synthetic 1-oleolyl-2-palmitolylglycerol
or 2-oleoyl-1-palmitoylglycerol.
EB-1: EB-1 was thought to be 1,3-diacylglycerols, based on TLC behav-
ior (less polar than EB-5) and a 13C NMR spectrum showing two carbon sig-
nals at 65.1 and 66.5 ppm that were assigned to one methyne carbon and two
methylene carbons attached to oxygen atoms of glycerol, respectively. Three
1
multiple proton signals at 4.08, 4.13, and 4.19 ppm in the H NMR spectrum
were assigned to the protons in the glycerol skeleton, compared with the authen-
tic 1,3-oleolyglycerol. Hydrolysis of EB-1 with 5% KOH ethanol gave a mix-
ture of palmitic, stearic, and oleic acids (2.9 : 1.0 : 7.9). EB-1 was composed of
at least two components, EB-1A (Rf
0.61) and EB-1B (Rf
0.53), as dis-
closed by AgNO3–SiO2 TLC, and MS. The MS spectrum gave fragments at
m z 602 (M+ H2O, 0.7%), 576 (M+ H2O, 8), 339 (22), 313 (25), 264 (16),
/
and 239 (24). EB-1A and EB-1B were isolated by preparative TLC by using a
AgNO3–SiO2 plate. The MS spectrum of EB-1B yielded fragments at m z 602
/
(M+ H2O, 6%), 339 (M+ OCOC17H34, 50), and 265 (M+ COC17H34, 28).
EB-1A and EB-1B were separately hydrolyzed by alkali. EB-1A yielded a mix-
ture of palmitic, stearic, and eleic acids in a ratio of 4.4 : 1.0 : 6.9. If EB-1A is
composed of a mixture of 1-oleoyl-3-palmitoylglycerol and 1-oleoyl-3-stearoyl-
glycerol in a ratio of 4.4 : 1.0, the theoretical ratio of palmitic, stearic, and oleic
acids obtained by hydrolysis would be 4.4 : 1.0 : 5.4 ( 4.4 + 1.0). The analytical
ratio of these acids was in acceptable agreement with the theoretical ratio. The

HOST-RECOGNIZING KAIROMONE FOR D. basalis
2691
content of oleic acid was higher than that of the theoretical one, which may be
due to contamination of EB-1B.
Hydrolysis of EB-1B with alkali afforded a mixture of oleic and palmitic
acids in a ratio of 10.6 : 1.0. Detection of palmitic acid suggested the presence
of 1-oleoyl-3-palmitoylglycerol or 1,3-dipalmitoylglycerol, but the Rf value of
the former was 0.61 and the latter, >0.61. Since the Rf value of EB-1B was
0.53, palmitic acid may be a contaminant of EB-1A. Thus, we conclude EB-1B
is primarily 1,3-dioleoylglycerol.
Based on these data, EB-1 was thought to be a mixture of 1-oleoyl-
3-palmitoyl-glycerol, 1-oleoyl-3-stearoylglycerol, and 1,3-dioleoylglycerol in a
ratio of approximately 2.9 : 1.0 : 2.0 [(7.9
2.9
1.0) 2], which was sup-
/
ported by MS analysis of EB-1; the fragments at m z 576 and 602 were
/
assigned to the dehydrated ions from the molecules of 1-oleoyl-2-palmitoylglyc-
erol and 1,3-dioleolylglycerol, respectively. Fragments at m z 239 (COC15H31),
/
265 (COC17H33), 313 (M+ OCOC17H33), and 339 (M+ OCOC15H31) were
also observed.
Saturated Hydrocarbons from Azuki Beans. The saturated hydrocarbons
(0.06 g) from the azuki bean extract showed no significant peaks in GC. In
addition, no activity was found by mixing the hexane fraction with the methanol
fraction (the first extraction), supporting the idea that the active saturated hydro-
carbons were insect-derived.
Saturated Hydrocarbons from Adults and Larvae of Azuki Bean Weevils.
The GC profile and the composition of the saturated hydrocarbons of the adult
weevils were nearly identical to those of the first extraction (Table 3). Further-
more, the adult hydrocarbons displayed the kairomone activity when mixed with
the methanol fraction (data not shown). On the other hand, the larval hydrocar-
bons contained higher amounts of n-heptacosane (ca. 50%) than that of adults
(Table 3), and their GC profiles were qualitatively similar, although the con-
stituents were present in different ratios.
Kairomonal Activity of Glycerols. Not only the separated diacylglycerols
but also the six authentic diacylglycerols showed the kairomone activity in com-
bination with the saturated hydrocarbons (Table 5). Of these diacylglycerols,
8
1O3S showed the activity at a lower dose (3 × 10 mg). Three monoacylglyc-
erols were also tested, because other active fractions of neutral fraction, con-
siderng their polarities, may be composed of monoacylglycerols (Figure 2 and
Table 4). They were active. Triacylglycerols are known as components of the
oviposition-marking pheromone of the host azuki bean weevils (Sakai et al.,
1986). Four triacylglycerols that were easily commercially obtained were also
assayed. All triacylglycerols showed kairomonal activity, and even triacetylglyc-
erol (1A2A3A), a nonnatural compound, had activity. 1O2O3O was the most
8
active of the triacylglycerols tested (3 × 10 mg). Therefore, mono- and tri-
acylglycerols, as well as diacylglycerols, also showed kairomone activity.

TABLE 5. KAIROMONAL ACTIVITY OF CANDIDATE DIACYLGLYCEROLS AND RELATED MONO- AND TRIACYLGLYCEROLS
b
No. of females responding
6
7
8
9
3 × 10 mg
3 × 10 mg
3 × 10 mg
3 × 10 mg
a
Sample
S
C
P
S
C
P
S
C
P
S
C
P
1O
1P
1S
16 32
2 32
<0.001(+)
0.027(+)
<0.001(+)
0.003(+)
0.247( )
0.027(+)
<0.001(+)
0.120( )
<0.001(+)
0.002(+)
0.074( )
<0.001(+)
0.006(+)
4 32
0 32
0.057( )
0.754( )
0.057( )
0.120( )
/
/
/
/
7 32
1 32
1 32
1 32
/
/
/
/
13 32
1 32
4 32
0 32
/
/
/
/
1O2P
2O1P
2O1S
1O3O
1O3P
1O3S
1A1A3A
1P2P3P
1O2O3O
1S2S3S
8 32
0 32
3 32
0 32
/
/
/
/
9 32
2 32
0.022(+)
2 32
0 32
/
/
/
/
7 32
1 32
1 32
0 32
0.500( )
0.057( )
/
/
/
/
16 32
2 32
4 32
0 32
/
/
/
/
12 32
/
1 32
/
<0.001(+)
3 32
0 32
/
/
14 32
2 32
10 32
1 32
0.003(+)
0.120( )
4 32
1 32
0.178( )
0.247( )
/
/
/
/
/
/
11 32
1 32
3 32
0 32
/
/
/
/
9 32
/
0 32
/
0.002(+)
7 32
2 32
/
/
15 32
0 32
7 32
0 32
0.006(+)
0.0099( )
2 32
/
0 32
/
/
/
/
/
7 32
0 32
5 32
1 32
/
/
/
/
3
a
Fixed quantity (2.5 BE 1.41 × 10 mg) of the saturated hydrocarbons (the first extraction) was used for the all sample mixtures.
b
See Table 1 (N 32). IP: 1-monopalmitoylglycerol; 1O: 1-monooleoylglycerol; 1S: 1-monostearoylglycerol; 1O2P: 1-oleoyl-2-palmitoylglycerol;
2O1P: 2-oleoyl-1-palmitoylglycerol; 2O1S: 2-oleoyl-1-stearoylglycerol; 1O3O: 1,3-dioleoylglycerol; 1O3P: 1-oleoyl-3-palmitoylglycerol; 1O3S:
1-oleoyl-3-stearoylglycerol; 1A2A3A: triacetylglycerol; 1P2P3P: tripalmitoylglycerol; 1O2O3O: trioleoylglycerol; 1S2S3S: tripalmitoyl-glycerol; S:
sample; C: control.

HOST-RECOGNIZING KAIROMONE FOR D. basalis
DISCUSSION
2693
Host selection by parasitic wasps is separated into the following processes
according to Doutt (1959): (1) host habitat finding; (2) host finding; (3) host
acceptance; and (4) host suitability. Of these processes, the stinging behavior of
D. basalis that was used in this study is thought to be involved in step 3. Vin-
son (1985) indicated that these processes may be mediated mainly by chemical
stimuli and not by sight. In D. basalis, however, stinging behavior was observed
only when the extract was applied to a fresh azuki bean but not to a glass bead or
a filter paper, indicating the necessity of physical conditions as well as chemical
stimuli. The encyrtid parasitoid, Anicetus beneficus Ishii et Yasumatsu, which
attacks the scale, Ceroplastes rubens Maskell, also shows ovipositional behav-
ior for a hemispherical shape but not for a filter paper treated with the scale
extract (Takahashi and Takabayashi, 1984).
Azuki bean weevils secrete a copulation release pheromone, electin, which
elicits males to extrude their genital organs and to attempt copulation (Tanaka
et al., 1981), and they also secrete an oviposition-marking pheromone to pre-
vent multiple ovipositions in a bean (Sakai et al., 1986). The former pheromone
is composed of hydrocarbons and (E)-3,7-dimethyl-2-octene-1,8-dioic (calloso-
bruchusic acid), which act synergistically. The GC profile of the hexane fraction
of electin, although it was obtained by using a 1% OV-1 packed column, was
similar to that of Figure 3 in the present study. In the hexane fraction of electin,
mono-methylalkanes (3-Me-C25, 11- and 3-Me-C27, 11- and 15-Me-C29, and 13-
Me-C31) and dimethylalkanes (11,15-di-Me-C29, 9,13-di-Me-C31, and 11,15-di-
Me-C33) had strong activity. The latter pheromone is a mixture of hydrocar-
bons, triacylglycerols, and fatty acids. A total of 17 straight and methyl-branched
hydrocarbons (C26–C35), of which 11 hydrocarbons (3-Me-C25, n-C27, 11-Me-
C27, 3-Me-C27, n-C28, n-C29, 11-Me-C29, 13-Me-C29, 13-Me-C31,9,13-di-Me-
C31, and 11,15-di-Me-C33) were also determined in the present study, were char-
acterized from the latter pheromone.
Both pheromones involve common saturated hydrocarbons. The host-recog-
nizing kairomone responsible for the stinging behavior of D. basalis was demon-
strated to be a mixture of saturated hydrocarbons (C25–C35, straight chain and
methyl-branched) and diacylglycerols. The active hydrocarbon kairomone con-
sisted of almost the same components as those of the pheromones and of the
cuticular hydrocarbons of adult weevils. Nearly no adult weevils were present
in the azuki beans employed for extracting the kairomones (but it was pos-
sible that some host larvae were present in the azuki beans). Thus, the active
hydrocarbon kairomone may be derived from the hydrocarbon components of
these two pheromones marked by host weevils. This suggests that D. basalis
utilizes one of the oviposition-marking pheromones of its host azuki bean wee-
vils as a host-recognizing kairomone. Thus, beans marked with the pheromone

2694
KUMAZAKI ET AL.
by their host weevils suggest to female wasps the presence of host larvae inside
beans.
n-Heptacosane (16.07%) was a major component of the active hydrocar-
bons, but mixtures of the authentic n-heptacosane or n-alkanes (n-C25, n-C27, and
n-C29, total 23.55%) and the methanol fraction were inactive (data not shown),
suggesting the necessity of the branched hydrocarbons for specific recognition.
However, it was not determined whether both groups of the hydrocarbons were
required for bioactivity or whether only mono-methylalkanes showed the activ-
ity, due to lack of authentic methylalkanes. The content of mono-methylal-
kanes (total >47.73%, 11-Me-C27 was the main one) was higher than that of
n-alkanes. 13-Methylhentriacontane has been identified from larvae of corn ear-
worm, Heliothis zea, as a host-seeking stimulant for the parasitic wasp, Micropli-
tis croceipes (Jones et al., 1971).
The amounts of the diacylglycerols obtained were small compared with those
of the hydrocarbon fraction (∼1 × 10 6), but no activity was obtained from the
hydrocarbons alone. Three diacylglycerols (1O3P, 1O3S, and 1O3O) were char-
acterized, and 1(or 2) O2(or 1)P and 1(or 2)O2(or 1)S were also determined as the
kairomonal components. The neutral fraction of the methanol fraction contained
other active compounds (Figure 2 and Table 4). These fracations (1%, 3%, 5%
and MeOH–CHCl3 and MeOH fractions, etc., in the first SiO2 column chromatog-
raphy in Figure 2) may be composed of other di- or monoacylglycerols, consid-
erng their polarities, but remain to be further studied. In fact, some monoacylglyc-
erols induced the stinging behavior. Neither diacylglycerols nor monoacylglyc-
erols have been described as components of the oviposition-marking pheromone
(Sakai et al., 1986) or copulation release pheromone (Tanaka et al., 1981) of host
azuki bean weevils. Triacylglycerols were also active (Table 5). Surprisingly, even
triacetylglycerol (1A2A3A) showed the activity. This is a nonnatural glycerol
with a short-chain fatty acid, which means that D. basalis responds to all types
of glycerols. Azuki bean weevils secrete a large quantity of triacylglycerols as
oviposition-marking pheromone, but individual triacylglycerol has not been iden-
tified from this marking pheromone. It is thought that triacylglycerols were eluted
in the 10% Et2O-Hex fraction (the first extraction, Figure 1) or in the 50% Et2O-
Hex fraction (the second extraction, figure 2), considering their polarities, but both
of these were inactive. The reason is unclear and remains to be further studied.
The two components of the kairomone probably act as short-range attrac-
tants (hydrocarbons) and chemical contact stimuli (diacylglycerols) for D.
basalis and may be useful in biological control of the weevils.
REFERENCES
CAUBET, Y., and JAISON, P. 1991. A post-eclosion early learning involved in host recognition by
Dinarmus basalia Rondani (Hymenoptera: Pteromalidae). Anim. Behav. 42:977–980.

HOST-RECOGNIZING KAIROMONE FOR D. basalis
2695
CAUBET, Y., JAISON, P., and LENOIR, A. 1992. Preimaginal induction of adult behaviour in insects.
Q. J. Exp. Psychol. 44B:165–178.
DOUTT, R. L. 1959. The biology of parasite Hymenoptera. Annu. Rev. Entomol. 4:161–182.
FRANCIS, G. W., and VELAND, K. 1981. Alkylthiolation for the determination of double bond position
in linear alkenes. J. Chromatogr. 219:379–384.
FUJII, K., and WAI, K. M. 1990. Sex-ratio determination in three wasp species ectoparasitic on bean
weevil larvae, pp. 331–340, in K. Fujii, A. M. R. Gatehouse, C. D. Johnson, R. Mitchel and
T. Yoshida (eds.). Bruchids and Regumes: Economics, Ecology and Coevolution. Kluwer Aca-
demic Publishers, Dordrecht, 407 pp.
ISLAM, W. 1994. Effect of host age on rate of development of Dinarmus basalis (Rond.) (Hym.,
Pteromalidae). J. Appl. Entomol. 118:392–398.
ISLAM, W., and KABIR, M. H. 1995. Biological control potential of Dinarmus basalis (Rond.)
(Hymenoptera: Pteromalidae), a larval-pupal ectoparasitoid of the pulse beetle Callosobruchus
chinensis (L.). Crop Protection 14:439–443.
JONES, R. L., LEWIS, M. C., BOWMAN, M. C., BEROZA, M., and BERL, B. A. 1971. Host-seeking stim-
ulant for parasite of corn earworm: Isolation, identification, and synthesis. Science 173:842–843.
KOJIMA, T. 1983. Qualitative Analysis, pp. 131–171, in Y. Masada, and T. Kojima (eds.). High
Resolution Gas Chromatography. Kagaku-dojin, Tokyo, Japan, 271 pp. (in Japanese).
LOCKEY, K. H. 1978. Hydrocarbons of adult Tribolium castaneum Hbst. and Tribolium confusum
Duv. (Coleoptera: Tenebrionidae). Comp. Biochem. Physiol. 61B:401–407.
MATTSON, F. H., and VOLPENHEIN, R. A. 1961. The use of pancreatic lipase determining distribution
of fatty acids in partial and complete glycerides. J. Lipid Res. 2:58–62.
NISHIMURA, K. 1997. Host selection by virgin and inseminated females of the parasitic wasp, Din-
armus basalis (Pteromalidae, Hymenoptera). Funct. Ecol. 11:336–341.
NISHIMURA, K., and JOHON, G. C. 1996. Sex allocation of three solitary ectoparasitic wasp species
on bean weevil larvae: Sex ratio change with host quality and local mate competition. J. Ethol.
14:27–33.
SAKAI, A., HONDA, H., and YAMAMOTO, I. 1986. Oviposition marking pheromones of two bean
weevils, Callosobruchus chinensis and Callosobruchus maculatus. J. Pestic. Sci. 11:163–168.
SANON, A., QUEDRAOGO, A. P., TRICAULT, Y., CREDLAND, P. F., and HUIGNARD, J. 1998. Biological
control of bruchids in cowpea stores by release of Dinarmus basalis (Hymenoptera: Pteroma-
lidae) adults. Environ. Entomol. 27:717–715.
SUZUKI, T., NAKAKITA, H., and KUWAHARA, Y. 1988. Defensive secretions and hydrocarbons of
two Tribolium species and their hybrids (Coleoptera: Tenebrionidae). Appl. Entomol. Zool.
23:329–337.
TAKAHASHI, S., and TAKABAYASHI, J. 1984. Host selection behavior of Anicetus beneficus Ishii et
Yasumatsu (Hymenoptera: Encyrtidae) II. Bioassay of oviposition stimulants in Ceroplastes
rubens Maskell (Hemiptera: Coccidae). Appl. Entomol. Zool. 19:117–119.
TANAKA, K., OHSAWA, K., HONDA, H., and YAMAMOTO, I. 1981. Copulation release pheromone,
electin, from the azuki bean weevil (Callosobruchus chinensis L.). J. Pestic. Sci. 6:75–82.
VERMA, R. 1990. Host habitat location and host location by Dinarmus basalis, a parasite of bruchids
of stored legumes. Ind. J. Exp. Biol. 28:179–184.
VERMA, R. 1991. Life-history and some aspects of biology of Dinarmus basalis (Hymenoptera,
Pteromalidae), a parasite of stored legume bruchids. Acta Entomol. Bohemoslov. 88:359–366.
VINCENTI, M., GUGLIELMETTI, G., CASSANI, G., and TONINI, C. 1987. Determination of double bond
positions in diunsaturated compounds by mass spectrometry of dimethyl disulfide derivatives.
Anal. Chem. 59:694–699.
VINSON, S. B. 1985. The behavior of parasitoids, pp. 417–469, in G. A. Kerkut and L. I. Gilbert
(eds.). Comprehensive Insect Physiology, Biochemistry and Pharmacology. Vol. 9. Behaviour.
Pergamon Press, Oxford, 735 pp.