Host-recognizing kairomones for parasitic wasp, Anisopteromalus calandrae, from larvae of azuki bean weevil, Callosobruchus chinensis.

Article abstract of DOI:10.1023/A:1016285801824

Host-recognizing kairomones for the stinging behavior of the parasitic wasp, Anisopteromalus calandrae, were identified on host azuki bean weevil larvae, Callosobruchus chinensis (L.). The kairomones were extracted with acetone from Chinese green beans, from which emerged wasps and host weevils had been removed. The kairomones are a mixture of triacylglycerols and fatty acids, each of which is separately active, and with no observable synergistic effect between them. These compounds are known to be constituents of an oviposition-marking pheromone of host azuki bean weevils. However, they differ from the previously reported saturated hydrocarbons and diacylglycerols of the kairomone that another parasitic wasp, Dinarmus basalis, uses for the host recognition of C. chinensis. Thus, A. calandrae and D. basalis selectively utilize different constituents of the oviposition-marking pheromone of C. chinensis as host-recognizing kairomones.

Full text of DOI:10.1023/A:1016285801824

C
Journal of Chemical Ecology, Vol. 28, No. 6, June 2002 ( 2002)
HOST-RECOGNIZING KAIROMONES FOR PARASITIC
WASP, Anisopteromalus calandrae, FROM LARVAE
OF AZUKI BEAN WEEVIL, Callosobruchus chinensis
JUNKO ONODERA,1,3 SHIGERU MATSUYAMA,1
TAKAHISA SUZUKI,1, and KOICHI FUJII2
1Institute of Applied Biochemistry
2Institute of Biological Sciences
University of Tsukuba
Tsukuba, Ibaraki 305-8572, Japan
(Received March 14, 2001; accepted February 2, 2002)
Abstract—Host-recognizing kairomones for the stinging behavior of the par-
asitic wasp, Anisopteromalus calandrae, were identified on host azuki bean
weevil larvae, Callosobruchus chinensis (L.). The kairomones were extracted
with acetone from Chinese green beans, from which emerged wasps and host
weevils had been removed. The kairomones are a mixture of triacylglycerols
and fatty acids, each of which is separately active, and with no observable syn-
ergistic effect between them. These compounds are known to be constituents of
an oviposition-marking pheromone of host azuki bean weevils. However, they
differ from the previously reported saturated hydrocarbons and diacylglycerols
of the kairomone that another parasitic wasp, Dinarmus basalis, uses for the host
recognition of C. chinensis. Thus, A. calandrae and D. basalis selectively utilize
different constituents of the oviposition-marking pheromone of C. chinensis as
host-recognizing kairomones.
Key Words—Anisopteromalus calandrae, parasitic wasp, oviposition-marking
pheromone, host-recognizing kairomone, azuki bean weevil, Callosobruchus
chinensis.
INTRODUCTION
The azuki bean weevil, Callosobruchus chinensis (L.) (Coleoptera: Bruchidae)
is a serious pest of stored Vigna legumes, such as azuki beans (V. angularis),
cowpea beans (V. catiang), and Chinese green beans (V. radiata). The physiology
To whom correspondence should be addressed. E-mail: tksuzuki@agbi.tsukuba.ac.jp
Present address: Advantec Toyo Kaisha Ltd., Kotobuki 1-5-10, Taito-ku Tokyo 111-0042, Japan.
3
1209
C
0098-0331/02/0600-1209/0 2002 Plenum Publishing Corporation

1210
ONODERA, MATSUYAMA, SUZUKI, AND FUJII
and ecology of this multivoltine insect has been widely researched, and it had
also been used as a test insect against insecticides. Consequently, the chemical
ecology of this weevil has been well studied, including that of oviposition-marking
pheromones (Ohshima et al., 1973; Sakai et al., 1986; Honda and Ohsawa, 1990),
a copulation-release pheromone (Tanaka et al., 1981), and oviposition stimulants
(Ueno et al., 1990; Matsumoto et al., 1994; Tebayashi et al., 1995).
The parasitic wasp, Anisopteromalus calandrae (Howard) (Hymenoptera:
Pteromalidae), is an important natural parasite of larvae of the maize weevils,
SitophiluszeamaisMotschulsky, amajorpestofstoredcorn, ZeamaysL. (Williams
and Floyd, 1971; Arbogast and Mullen, 1990; Smith, 1992). This wasp also attacks
several other bruchid weevils, including azuki bean weevils (Utida and Nagasawa,
1949; Okamoto, 1971, 1972), rice weevils, S. oryzae (L.) (Press et al., 1984; Cline
et al., 1985), granary weevils, S. granarius (L.) (Ghani and Sweetman, 1955;
van den Assem et al., 1984), and Mexican bean weevils, Zabrotes subfasciatus
Bohman.
Adults of A. calandrae are sexually dimorphic; females are entirely black
and ca. 2.5 mm in length while males have partly milk-white abdomens and are
ca. 2.0 mm in length. When an azuki bean or Chinese green bean containing host
larvae is presented to a female, the following sequential behaviors are observed:
the female walks and examines the surface of the bean with antennal tapping,
bends her abdomen vertically, stings her ovipositor through the surface of the bean
(stinging behavior), and lays an egg on the surface of the larva. Stinging behavior
is only observed when females locate beans that contain host larvae.
We previously reported that another parasitic wasp, Dinarmus basalis
(Rondani) (Hymenoptera: Pteromalidae), utilizes the oviposition-marking phero-
mone of its host azuki bean weevil as a host-recognizing kairomone (Kumazaki
et al., 2000). This prompted us to search for similar kairomone(s) in A. calandrae,
which paratitizes the same larvae. Here, we report that A. calandrae utilizes dif-
ferent constituents of the oviposition-marking pheromone of its host azuki bean
weevils as host-recognizing kairomones.
METHODS AND MATERIALS
Host and Wasp. Azuki bean weevils and A. calandrae were obtained from
colonies maintained at Fujii’s laboratory, but which had been subsequently reared
for several generations on Chinese green beans, Vigna radiata (L.) in Suzuki’s
laboratory. Although Chinese green beans are smaller than azuki beans (ca. 60 mg/
bean), azuki bean weevils developed well in this host. The rearing method of the
weevil host and A. calandrae is reported elsewhere (Kumazaki et al., 2000). Both
sexes of wasps were introduced into plastic Petri dishes (9 cm ID 20 mm high)
containing Chinese green beans in which host larvae were present (ca. 18 days
old). Both insect species were reared at 25 C, 60% relative humidity, and 16L:8D

HOST-RECOGNIZING KAIROMONES FOR A. calandrae
1211
photoregime. Wasps began to emerge 16–18 days after being exposed to infested
beans. The newly emerged wasps (not sexed) were either assayed or used for
further reproduction.
Prior to the bioassay, 1- to 3-day-old mated, female wasps were individually
released into glass Petri dishes (18 mm ID 13 mm high; mini P, Tsukuba Flat Co.)
on an acrylic plate and conditioned overnight. Only females that showed stinging
behavior (2 hr) when presented to a bean with host larvae were bioassayed.
Bioassay. In preliminary studies, the active kairomone(s) eliciting the stinging
behavior could be extracted with acetone from beans containing host larvae. In
addition, the behavior was observed only when the extract was applied to a fresh
bean. Based on these observations, the bioassay in this study was conducted as
follows. Two fresh green beans were used for each trial; one was treated with 5
[1 mg bean equivalent (BE)] of the test sample, and the other was treated with 5
l
l
of acetone as a control. After evaporation of the solvent, the beans were placed ca.
1 cm apart on a filter paper (3 3 cm) inside an 18-mm glass Petri dish. A mated,
2- to 4-day-old female wasp (described above) was released. Her stinging behavior
was observed for 15 min, and the number of stings on each bean was recorded.
The assay was repeated 15 times for each sample with a different female. Samples
were diluted 10-fold with acetone when necessary. Total number of the stinging
behaviors for each sample was compared to the control using Wilcoxon’s sign test
(P < 0.05) (Sokal and Rohlf, 1973).
Analytical Methods. Gas chromatography (GC) was performed with three
capillary columns: a fused silica CP-Sil 19 CB column (25 m 0.25 mm, 0.2 m;
Chrompack, Middelburg, The Netherlands), a heat-resisting MP65HT column
(25 m
0.25 mm, 0.1 m; Quadrex, New Haven, Connecticut, USA) with
a Hewlett-Packard 5890 (Palo Alto, California, USA) gas chromatograph, or a
bonded-phase FFAP column (25 m 0.25 mm, 0.25 m; Quadrex) with a Hitachi
263-30 gas chromatograph (Hitachi, Ltd., Tokyo, Japan), with nitrogen as the car-
rier gas. The column temperatures were raised from 100 to 230 C at 4 C/min for
analyses of fatty acid methyl esters (FFAP and CP-Sil columns; the injector and
detector temperature was set at 250 C) and from 300 to 365 C (held for 10 min)
at 5 C/min for analyses of triacylglycerols (MP65HT column; the injector and de-
tector temperature was set at 400 C). Mass spectra were measured with a Hitachi
M-80B GC mass spectrometer (MS) coupled with a Hitachi 367-0200 gas chro-
matograph equipped with an Al-clad methylsilicone capillary column (25 m
0.22 mm, 0.10 m, Quadrex) (EI, 70 eV). Mass spectra of methyl esters were
obtained by GC-MS (column temperature: 120–270 C programmed at 4 C/min)
and those of triacylglycerols were measured by the direct inlet method, using he-
lium as the carrier gas. Thin layer chromatography (TLC) was performed with
precoated silica gel 60F254 plates (E. Merck, Darmstadt, Germany). The develop-
ing solvent was a mixture of hexane–ether (7:1, v/v) or hexane–ether–acetic acid
(70:30:1, v/v). Spots were detected by exposing the sample to iodine vapor or

1212
ONODERA, MATSUYAMA, SUZUKI, AND FUJII
spraying with 0.02% rhodamine 6G ethanol solution, followed by UV (365 nm;
fluorescent spots). The infrared (IR) spectrum was measured with a JASCO IRA-1
spectrometer (Japan Spectroscopic Co. Ltd., Tokyo, Japan) as a liquid film.
Extraction and Fractionation of Kairomone(s). As described above, the
kairomone(s) was (were) extracted with acetone only from beans that had con-
tained host larvae. The selection of beans previously infected by host weevils is
important as extracts from uninfected beans were not active. The kairomone(s)
was (were) extracted from the green beans (2.6 kg, 30 mg/bean) with acetone
(2.5 liters 2) at room temperature for two days. The extract was dried, and the
major part of the residue (40 of 50 g) was partitioned with ether and water. Half of
the active layer (14.8 of 30 g) was chromatographed on a silica gel (SiO2) column
(125 g, 3.5 cm ID 37 cm, Wako-gel C-200, Wako, Tokyo, Japan) with a stepwise
elution of 1.5 liters each of hexane, hexane–ether (90:10 and then 50:50 v/v), ether,
and methanol. Only the 10% ether in hexane fraction (10.8 g) was active, and this
was checked by TLC (hexane–ether 7:3, v/v) to give three spots (R f s = 0.85, 0.64,
and 0.11), in which the last two spots (the R f 0.11 spot was tailing) were considered
to be triacylglycerols (TGs) and fatty acids (FAs), respectively. The fraction was
submitted to Florisil column chromatography (WAK Florisil, Wako) to separate the
seven fractions of lipids, according to the method of Carroll (1961). The column
(150 g, 3.2 cm ID 31 cm) was successively eluted with 1.5 liters each of hexane
(fraction 1, 0.04 g), hexane–ether [ 95:5 (fraction 2, 1.0 g), 85:15 (fraction 3), 75:25
(fraction 4), and 50:50 (fraction 5, 1.2 g)], 2% methanol in ether (fraction 6, 1.6 g),
and 4% acetic acid in ether (fraction 7, 3.2 g). Fractions 3 and 4 were combined,
because they yielded spots with the same R f value (0.65) in TLC. This combined
fraction (named fraction I, 3.2 g,) and fraction 7 were separately active.
Extraction of Triacylglycerols and Fatty Acids from Chinese Green Beans.
Because we used previously infected Chinese green beans, it is possible that the
source of the TGs and FAs were derived from green beans. To investigate this possi-
bility, 200 g of crushed fresh green beans were extracted with acetone (200 ml 2)
for 24 hr. After evaporation of the solvent, the residue was partitioned with ether
and water. The ether layer (0.47 g) was submitted to SiO2 column chromatography
(5 g, 10 mm ID 14 cm) with stepwise elution of hexane, hexane–ether (90:10,
50:50 v/v), ether, and methanol (50 ml each). TGs and FAs were eluted in 10%
ether in hexane (390.7 mg) and analyzed with Florisil column chromatography
(5 g, 10 mm ID 14 cm). The column was successively eluted with 50 ml each of
hexane, hexane–ether (95:5, 85:15, 75:25, and 50:50 v/v), 2% methanol in ether,
and 4% acetic acid in ether.
Surface Hexane Washing from Previously Infected Green Beans. Approxi-
mately 40 g of green beans were placed into each of five 9-cm plastic Petri dishes,
and ca. 150 adult host weevils were added to each dish. After eight days, all of
the adult weevils had died and were removed with a sieve, but the hatched larvae
had invaded the beans. The surfaces of these beans were washed with hexane for

HOST-RECOGNIZING KAIROMONES FOR A. calandrae
1213
30 sec, and the TGs and FAs were identified with TLC. The same total volume
(200 g) of fresh green beans that had not been invaded by weevils were extracted
with hexane for 30 sec as a control.
Synthesis of Triacylglycerols. TGs were synthesized by the following two
methods. For method A, commercially obtained 1-monoacylglycerols or 1,3-
diacylglycerols (Wako) were treated overnight at 60 C with the desired acyl chlo-
rides in the presence of pyridine. The crude TGs were purified by SiO2 column
chromatography and eluted with 10% ether in hexane. The following nine TGs
were prepared (acronyms and yields are given in parentheses): 1-oleoyl-2,3-
dipalmitoylglycerol (1O2P3P, 26.4 mg, 94.1%), 1,3-dioleoyl-2-palmitoylglycerol
(1O2P3O, 36.7 mg, 89.6%), 1,2-dioleoyl-3-palmitoylglycerol (1O2O3P, 22.1 mg,
76.5%), 1,2-dilinoleoyl-3-palmitoylglycerol (1L2L3P, 10.0 mg, 78.1%), 2-
linoleoyl-1-oleoyl-3-palmitoylglycerol (1O2L3P, 4.6 mg, 59.6%), 1,3-dioleoyl-2-
stearoylglycerol (1O2S3O, 35.3 mg, 50.0%), 2-linoleoyl-1-oleoyl-3-
stearoylglycerol (1O2L3S, 10.4 mg, 73.2%), 2-linoleoyl-1,3-dioleoylglycerol
(1O2L3O, 6.9 mg, 65.0%), and 1,2-dilinoleoyl-3-oleoylglycerol (1L2L3O, 4.4 mg,
55.2%).
For method B, three TGs were prepared according to Golding et al. (1993).
2,2-Dimethyl-1,3-dioxolane-4-methanol (100 l, 0.8 mmol, Wako) was dissolved
in 1.5 ml pyridine, to which excess acyl chlorides were added. This was stirred
overnight at 60 C. The crude monoacylesters of 2,2-dimethyl-1,3-dioxolane-4-
methanol were treated by SiO2 column chromatography and eluted with 10%
ether in hexane to yield pure esters. The acetonides of the esters were cleaved
by treating with trifluoroacetic acid in 80% THF aqueous solution and stirring at
0 C in (Leblanc et al., 1986). The crude 1-monoacylglycerols were purified by
SiO2 column chromatography and then reacted with acyl chlorides (two molar
equivalents) as described in A: 1-palmitoleoyl-2,3-dipalmitoylglycerol (1P’2P3P,
6.1 mg, 13.4%), 1-linoleoyl-2,3-dipalmitoylglycerol (1L2P3P, 30.3 mg, 17.4%),
and 3-oleyl-1,2-distearoylglycerol (1S2S3O, 55.9 mg, 6.2%).
Some acyl chlorides were prepared by the reaction of the desired acids with
oxalyl chloride. For example, palmitic acid (20 mg, 0.08 mmol) was dissolved in
dry benzene (1 ml), to which equimolar oxalyl chloride was added, and the mixture
was refluxed for 1 hr. After benzene and oxalyl chloride were removed, palmitoyl
chloride (21.7 mg) was obtained. Palmitoleoyl chloride and stearoyl chloride were
prepared in the same way.
RESULTS
Chemical Analysis of Active Fraction 7. Kairomonal activity was found only
2
1
from fraction 7 (active at 10 BE) and Fraction I (active at 10 BE). No syn-
ergistic effect was observed between the fractions because their mixture did not

1214
ONODERA, MATSUYAMA, SUZUKI, AND FUJII
TABLE 1. KAIROMONAL ACTIVITY OF FRACTIONS
Total stinging behaviors (N)a
Dose
Sample
(mg BE)
Sample
Control
P
Activity
2
Acetone extract
Ether layer
10
10
10
10
10
10
42
42
51
36
33
27
18
24
8
12
28
47
<0.025
NS
<0.005
NS
NS
NS
+
3
2
+
3
2
Aqueous layer
3
Fractions separated by SiO2 column chromatography
1
H fraction
10% E/H fraction
10
10
10
10
10
10
23
27
29
12
15
12
21
1
11
14
14
15
NS
<0.025
NS
NS
NS
1
2
1
1
1
+
50% E/H fraction
E fraction
MeOH fraction
NS
Fractions separated by Florisil column chromatography
Fraction 1
100
10
35
16
13
12
36
20
7
29
4
13
4
4
8
3
4
12
20
9
NS
NS
NS
NS
<0.025
NS
NS
NS
NS
NS
<0.05
<0.025
NS
<0.025
NS
1
Fraction 2
100
10
1
1
2
Fraction 3 + fraction 4 (fraction I)
Fraction 5
10
+
10
100
10
1
3
5
Fraction 6
100
10
1
1
2
3
2
3
13
26
34
19
36
18
Fraction 7
10
10
10
10
+
+
5
31
12
15
Fraction I + fraction 7
+
10
a Total number of stinging behaviors for each bean during 15 min (N = 15). BE = bean equivalent;
NS: not significant; H: hexane; E: ether; MeOH: methanol; AcOH: acetic acid. +: active; : inactive.
enhance the activity (Table 1). In view of their TLC (hexane–ether 7:3, v/v) and
Florisil column chromatographic behaviors, the chemicals were considered to be
FAs ( R f = 0.13, tailing) and TGs (R f = 0.65), respectively.
Fraction 7 was treated with diazomethane to yield methyl esters, which were
analyzed by GC and GC-MS and compared with the authentic samples. Ten peaks
were detected and identified (Table 2). The fraction 7 (3.2 g) obtained from 1.04 kg
green beans, with emerged weevil removed, had a 10 2 BE (active dose) of 30.9 ng
FAs. Thus, the absolute amount of each acid in 10 2 BE of fraction 7 was calculated
on the basis of the GC peaks (Table 2). The most abundant acid was palmitic acid
(40.8%). Kairomonal activity of authentic acids or their mixtures was assayed

HOST-RECOGNIZING KAIROMONES FOR A. calandrae
1215
TABLE 2. FATTY ACID COMPOSITION OF ACTIVE FRACTION
Composition (%) Absolute amount (ng)a
Fatty acid
Lauric acid
0.1
0.8
0.1
40.8
0.7
0.3
0.03
0.25
0.03
12.61
0.22
0.09
2.07
10.20
3.92
1.48
Myristic acid
Pentadecanoic acid
Palmitic acid
Palmitoleic acid
Heptadecanoic acid
Stearic acid
Oleic acid
Linoleic acid
Linolenic acid
6.7
33.0
12.7
4.8
2
a Absolute amount in active dose (10 mg BE, 30.9 ng). BE = bean
equivalent.
3
at doses of 100–10 BE (Table 3). Individual acids, or a mixture of five major
acids, were inactive, but the complete mixture (10 acids) was active at a dose of
2
10 BE (natural ratio), whose activity was equal to those of the acetone extract
and fraction 7.
TABLE 3. KAIROMONAL ACTIVITY OF AUTHENTIC ACIDS
Stinging behaviors (N)a
Dose
Sample
(mg BE)
Sample
Control
P
Activity
1
Mixture of 10 acidsb
10
10
10
100
10
10
10
10
10
10
10
10
10
10
10
10
46
25
18
6
16
4
44
20
16
9
22
26
28
14
14
24
10
9
14
5
<0.025
<0.05
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
+
+
2
3
Mixture of 5 acidsc
1
10
15
25
11
11
12
11
29
22
13
11
16
2
1
Lauric acid
1
Myristic acid
Pentadecanoic acid
Palmitic acid
Palmitoleic acid
Heptadecanoic acid
Stearic acid
Oleic acid
Linoleic acid
Linolenic acid
1
1
1
1
1
1
1
1
a Total no. of stinging behaviors for each sample during 15 min (N = 15). NS: not significant; +: active;
: inactive. BE = bean equivalent.
b A mixture of 10 acids (natural ratio).
c A mixture of palmitic, stearic, oleic, linoleic, and linolenic acids (natural ratio).

1216
ONODERA, MATSUYAMA, SUZUKI, AND FUJII
Chemical Analysis of Active Fraction I. Another active fraction (Fraction I)
was comprised of TGs. Its IR spectrum showed the following frequencies: 2940,
2860, 1750 (ester C O), 1470, and 1160 cm 1. No OH stretching frequency
was observed. In TLC, fraction I gave a spot with R f = 0.64 (trioleoylglycerol,
R f = 0.64) (hexane–ether 7:3, v/v).
Fraction I showed the activity at a dose of 10 1 BE, which contained 309 ng
of TGs. An aliquot of fraction I was hydrolyzed with KOH in ethanol–water
overnight at room temperature. The acidic fraction was analyzed by GC and
GC-MS, after methylation with diazomethane. The GC profile was very simi-
lar to that of the methyl esters of fraction 7. The aqueous layer, upon evaporation
of ethanol and water after neutralization, was heated with potassium hydrogen sul-
fate (KHSO4) to confirm the presence of glycerol, using the acrolein test (Hayashi,
1991). The characteristic pungent odor of acrolein confirmed the presence of
glycerol.
The MS spectrum of the TGs (fraction I) was measured with the direct inlet
method, and the following fragments were detected; m/z 57 (base peak), 237, 239,
261, 263, 265, 267 (fatty acid residues), 336, 338, 340, 549, 551, 571, 573, 575,
579, 601, 603, and 605 (residues leaving FAs from TGs) (Murata and Takahashi,
1973).
Direct GC analysis of the TGs involved in fraction I was conducted using an
MP65HT capillary column. The column temperature was 300–360 C programmed
at 5 C/min. A total of 13 peaks were observed from the gas chromatogram
(Figure 1), which were separated on the basis of total number of carbons that
constituted the FAs of the TGs. They were C48 (2.5%), C50 (33.1%), C52 (50.0%),
and C54 (14.4%), compared with those of the synthetic TGs (Table 4). The major
components of the TGs were PPO (24.2%) and POO (26.3%). In this column,
TGs with the same total number of carbons as those of FAs, but with different
carbon positions, could not be distinguished. For example, 1,2-dipalmitoyl-3-
oleoylglycerol (1P2P3O), and 1,3-dipalmitoyl-2-oleoylglycerol (1P2O3P) yielded
one peak.
TGs and FAs from Chinese Green Beans. Florisil column chromatography
separated 153 mg TGs and 124.1 mg FAs from 5 and 15% ether in hexane and
4% acetic acid in ether fractions, respectively. These fractions were obtained from
200 g (ca. 60 mg/bean) of crushed green beans and were calculated as 45.9
TGs and 37.3 g FAs per bean, or 765 g TGs and 620.5 g FAs per g beans.
g
TGs and FAs in Surface Hexane Washings of Green Beans from Which
Weevils Were Removed. The hexane washing from the beans oviposited by weevils
produced 72.5 mg or 38.5 g/bean of hydrocarbons (the main component), TGs
and FAs, which were identified with TLC (hexane–ether 7:3, v/v). Conversely,
the hexane washing from the fresh bean control (no weevil oviposition) produced
almost no residue (0.3 mg).

HOST-RECOGNIZING KAIROMONES FOR A. calandrae
1217
FIG. 1. Gas chromatogram of the active triacylglycerols (fraction I). Condition: MP65HT
capillary column; 300–365 C (10 min) at 5 C/min.
DISCUSSION
The host-recognizing kairomones for the parasitic wasp, A. calandrae, were
demonstrated to be comprised of fatty acids (FAs) and triacylglycerols (TGs).
Both fractions separately elicited kairomonal activity, but no synergistic effect was
observed. The activity of the FAs (fraction 7) was more than 10 2 mg BE (30.9 ng),
1
whereas for the TGs (Fraction I) it was more than 10 mg BE (309 ng). Thus,
fraction 7 was ten times more active than Fraction I. Fraction 7 was composed of 10
carboxylic acids, of which palmitic acid was main component (40.8%). Individual
acids or a mixture of five major acids were inactive, but the complete acid mixture
was active. This suggests that minor acids were required for activity.
The FAs constituting the TGs in fraction I were characterized, although
their attaching positions to glycerol were not determined. The carbon numbers

1218
ONODERA, MATSUYAMA, SUZUKI, AND FUJII
TABLE 4. TOTAL CARBON NUMBERS OF FATTY ACIDS AND THEIR
CONSTITUENTS, AND COMPOSITION OF EACH TRIACYLGLYCEROL
INVOLVED IN FRACTION I
GC peak
no.a
Total carbon no. of
acid constituents
Acid constituent Composition (%)
1
C48
PPP’
2.5
2.5
Subtotal
2
3
C50
PPO
PPL
24.2
8.9
Subtotal
33.1
4
5
6
7
C52
PSO
POO
POL
PLL
10.6
26.3
10.1
3.0
Subtotal
50.0
8
9
10
11
12
13
C54
SSO
SOO
OOO
SOL
OOL
OLL
1.5
2.7
4.2
0.9
3.5
1.6
Subtotal
14.4
a GC peak no. corresponds that of Figure 1. P: palmitic; P’: palmitoleic;
S: stearic; O: oleic; L: linoleic acids.
constituting the FA component of the TGs in this fraction ranged from C48 to C54,
in which C52 (50.0%) was the major components.
Azuki bean weevils secrete an oviposition-marking pheromone to prevent
multiple oviposition for a bean, which consists of FAs, hydrocarbons, and TGs
(Oshima et al., 1973; Sakai et al., 1986; Honda and Ohsawa, 1990) in a ratio of
ca. 1:4.7:2. Of these three constituents, the FA fraction is composed of palmitic
(21.6%), stearic (11.5%), oleic (45.5%), linoleic (10.7%), and linolenic (4.1%)
acids and 7.4% other acids; the TG fraction comprises a mixture of C48 (1.0%), C50
(18.9%), C52 (47.7%), C54 (31.7%), and C56 (0.4%) (Honda and Ohsawa, 1990).
In the present study, the active kairomones for A. calandrae were also comprised
of FAs and TGs and are almost the same constituents as for those of the host
pheromone. Although TGs and FAs were components of the control fresh green
beans (765 and 620.5 g/bean, respectively), the major fractions of the separated
kairomones are thought to be insect-derived, because the amount of TGs and FAs
obtained from previously infected green beans was calculated to be 3120 g/g

HOST-RECOGNIZING KAIROMONES FOR A. calandrae
1219
beans, i.e., ca. 4–5 times more than those obtained from the control beans. In
addition, hexane washing of surface green beans yielded almost no residue. Thus,
the major part of the kairomones must be derived from the host pheromone.
We previously reported that other pteromalid wasp, Dinarmus basalis, uti-
lizes the saturated hydrocarbons (and diacylglycerols) of the oviposition-marking
pheromone of azuki bean weevils as a host-recognizing kairomone (Kumazaki
et al., 2000). Thus, the two pteromalid wasps selectively utilize different con-
stituents of the oviposition-marking pheromone of azuki bean weevils as host-
recognizing kairomones. For D. basalis, TGs as well as diacylglycerols were
active. Moreover, both saturated hydrocarbons and diacylglycerols (triacylglyc-
erols) were required for activity in this wasp, whereas in A. calandrae, FAs and
TGs were separately active, and no synergistic effect was observed between them.
D. basalis mainly parasitizes azuki bean weevils, whereas A. calandrae also at-
tacks other bruchid species. The narrower host range of D. basalis compared to A.
calandrae might be because both constituents of its host pheromone are necessary
to search for, or recognize, its host larvae. It would be interesting to know whether
FAs and TGs can independently promote stinging behavior.
REFERENCES
ARBOGAST, R. T. and MULLEN, M. A. 1990. Interaction of maize weevil (Coleoptera: Curculionidae)
and parasitoid Anisopteromalus calandrae (Hymenoptera: Pteromalidae) in a small bulk of stored
corn. J. Econ. Entomol. 83:2462–2468.
CARROLL, K. K. 1961. Separation of lipid classes by chromatography on florisil. J. Lipid Res. 2:135–
141.
CLINE, L. D., PRESS, J. W., and FLAHERTY, B. R. 1985. Suppression of the rice weevil, Sitophilus
oryzae (Coleoptera: Curculionidae), inside and outside of burlap, woven polypropylene, and
cotton bags by the parasitic wasp, Anisopteromarus calandrae (Hymenoptera: Pteromalidae). J.
Econ. Entomol. 78:835–838.
GHANI, M. A. and SWEETMAN, H. L. 1955. Ecological studies on the granary weevil parasite, Aplas-
tomorpha calandrae (Howard). Biologia (Lahore) 1:115–139.
GOLDING, B. T., GRIFFIN, A. L., and ROBINSON, D. H. 1993. Synthesis of a “reverse ester” analogue
of 1,2-sn-diglycerides from (S)-1,2-di-O-isopropylidene-glycerol; efficient, stereospecific nucle-
ophilic displacement via a triflate at glycerol C-2. Tetrahedron Lett. 34:6459–6462.
HAYASHI, A. 1991. Chapter 9. Handling of Lipids, pp. 149–170, in T. Goto, T. Shiba, and T. Matsuura
(eds.). Introduction to Experiments for Organic Chemistry. Kagakudojin, Kyoto (in Japanese).
HONDA, H. and OHSAWA, K. 1990. Chemical ecology for stored product insects. J. Pestic. Sci. 15:263–
270. (in Japanese with an English abstract).
KUMAZAKI, M., MATSUYAMA, S., SUZUKI, T., KUWAHARA, Y., and FUJII, K. 2000. Parasitic wasp
Dinarmus basalis, utilizes oviposition-marking pheromone of host azuki bean weevils as host-
recognizing kairomone. J. Chem. Ecol. 26:2677–2695.
LEBLANC, Y., FITZSIMMONS, B. J., ADAMS, J., PEREZ, F., and ROKACH, J. 1986. The total synthesis of
12-HETE and 12, 20-DiHETE. J. Org. Chem. 51:789–793.
MATSUMOTO, H., TEBAYASHI, S., KUWAHARA, Y., MATSUYAMA, S., SUZUKI, T., and FUJII, K. 1994.
Identification of taxifolin present in the azuki bean as an oviposition stimulant of the azuki bean
weevil. J. Pestic. Sci. 19:181–186.

1220
ONODERA, MATSUYAMA, SUZUKI, AND FUJII
MURATA, T. and TAKAHASHI, S. 1973. Analysis of triglycerides mixtures by gas chromatography-mass
spectrometry. Anal. Chem. 45:1816–1823.
OHSHIMA, K., HONDA, H., and YAMAMOTO, I. 1973. Isolation of an oviposition marker from azuki
bean weevil, Callosobruchus chinensis (L.). Agric. Biol. Chem. 37:2679–2680.
OKAMOTO, K. 1971. The synchronization of the life cycles between Callosobruchus chinensis (L) and
its parasite Anisopteromalus carandrae (Howard). Jpn. J. Ecol. 20:233–237 (in Japanese with an
English synopsis).
OKAMOTO, K. 1972. The synchronization of the life cycles between Callosobruchus chinensis (L) and
its parasite Anisopteromalus calandrae (Howard) II. The relationship between the development
of the parasite and the developmental stage of the host. Jpn. J. Ecol. 22:238–244 (in Japanese
with an English synopsis).
PRESS, J. W., CLINE, L. D., and FLAHERTY, B. R. 1984. Suppression of residual population of the rice
weevil, Sitophilus oryzae, by the parasitic wasp, Anisopteromalus calandrae. J. Ga. Entomol. Soc.
19:110–113.
SAKAI, A., HONDA, H., and YAMAMOTO, I. 1986. Oviposition marking pheromone of two bean weevils,
Callosobruchus chinensis and Callosobruchus maculatus. J. Pestic. Sci. 11:163–168.
SMITH, L. 1992. Effect of temperature on life history characteristics of Anisopteromalus calandrae
(Hymenoptera: Pteromalidae) parasitizing maize weevil larvae in corn kernels. Environ. Entomol.
21:877–887.
SOKAL, R. R. and ROHLF, F. J. 1973. Introduction to Biostatistics. Kyoritsu Publishing, Tokyo, 449 pp.
(K. Fujii Japanese translation, 1983).
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.
TEBAYASHI, S., MATSUYAMA, S., SUZUKI, T., KUWAHARA, Y., NEMOTO, T., and FUJII, K. 1995.
Quercimeritrin: The third oviposition stimulant of the azuki bean weevil from the host azuki
bean. J. Pestic. Sci. 20:299–305.
UENO, T., KUWAHARA, Y., FUJII, K., TAPER, M. L., TOQUENAGA, Y., and SUZUKI, T. 1990. D-Katechin:
An oviposition stimulant of azuki bean weevil Callosobruchus chinensis in the host azuki bean.
J. Pestic. Sci. 15:573–578.
UTIDA, S. and NAGASAWA, S. 1949. On the developmental period and that of adult life of Neocatolaccus
mamezophagus Ishii et Nagasawa, a parasite wasp of the azuki bean weevil. Kontyu 17:1–21 (in
Japanese).
VAN DEN ASSEM, J., PUTTERS, F. A., and PRINTS, T. C. 1984. Host quality effects on sex ratio of the
parasitic wasp Anisopteromalus calandrae (Chalcidoidea, Pteromalidae). Neth. J. Zool. 34:33–62.
WILLIAMS, R. N. and FLOYD, E. H. 1971. Effect of two parasites, Anisopteromalus calandrae and
Choetospila elegans, upon populations of the maize weevil under laboratory and natural condi-
tions. J. Econ. Entomol. 64:1406–1408.