nating from 100-/zg stimulus loads of the 29 cotton compounds. ..... DICKENS, J.C., SMITH, J.W., and LIGHT, D.M. 1992. .... John Wiley & Sons, New York.
Journal of Chemical Ecology, Vol. 18, No. 10, 1992
ANTENNAL OLFACTORY RESPONSIVENESS OF Microplitis croceipes (HYMENOPTERA: BRACONIDAE) TO COTTON PLANT VOLATILES
YONGSHENG
LI, 1 JOSEPH
WILLIAM
W.M.
C . D I C K E N S , 2'* a n d STEINER 3
tDepartment of Entomology University of Missouri Columbia, Missouri 65211 2USDA/ARS, Boll Weevil Research Unit Mississippi State, Mississippi 39762 3USDA/ARS, Biological Control of Insects Research Laboratory Columbia, Missouri 65205 (Received December 9, 1991; accepted May 29, 1992) Abstract--Antennal olfactory responses of the parasitoid, Microplitb croceipes (Cresson) (Hymenoptera: Braconidae), to 29 cotton volatile compounds were measured by electroantennogram (EAG) techniques. No significant sexual differences were found in EAGs of males and females to volatiles emanating from 100-/zg stimulus loads of the 29 cotton compounds. Green leaf volatiles (saturated and monounsaturated six-carbon alcohols, aldehydes, and their acetate derivatives), heptanal, and the benzene derivatives, benzaldehyde and acetophenone, elicited the largest EAGs. Monoterpenes elicited moderate EAGs with 3-ocimene being the most effective monoterpene tested. Among the sesquiterpenes tested, 3-bisabolol was the most effective stimulus at the 100-#g dose. Dose-response curves constructed from EAGs of females revealed a low threshold for (Z)-3-hexenyl acetate, a compound previously shown to be an effective attractant in wind-tunnel bioassays. Comparison of relative volatilities of the various odorants indicated differential selectivity and sensitivity of M. croceipes antennal receptors to them. The roles of cotton plant volatiles in host habitat location of M. croceipes are discussed. Key Words--Microplitis croceipes, Hymenoptera, Braconidae, chemoreceptors, cotton, electroantennogram, host habitat location, plant volatileg, parasitoid, EAG. *To whom correspondence should be addressed.
1761 0098-0331/92/1000-1761506.50/0 9 1992PlenumPublishingCorporation
1762
LI ET AL. INTRODUCTION
Microptitis croceipes (Cresson) (Hymenoptera: Braconidae) is a predominant parasitoid of two major cotton pests: the cotton bollworm [Helicoverpa zea (Boddie) (Lepidoptera: Noctuidae)] and the tobacco budworm [Heliothis virescence (F.)] in the United States (Powell and King, 1984). As an important biological control agent, M. croceipes has been studied extensively in recent years (Drost et al., 1986; Eller et al., 1988; Hopper and King, 1984; Lewis et al., 1988; Powell and King, 1984; Whitman and Eller, 1990). The success of the parasitic wasp in suppressing the pest population depends on its ability to locate hosts in a complex olfactory and visual environment. Understanding the mechanisms governing host-searching behavior by the wasp is critical to successful implementation of biological control programs. Host-finding behavior in parasitoids is a hierarchy of behavioral events, which include host habitat location and host location (Vinson, 1981, 1984). Chemical cues are known to play a central role in each level of these behaviors. Electroantennograms (EAGs) to several plant volatiles by two parasitoids, Campoletis sonorensis (Ichneumonidae) (Baehrecke et al., 1989) and Microplitis demolitor (Ramachandran and Norris, 1991), have been previously reported. Numerous volatiles have been identified from the cotton plant (Chang et al., 1988; Hedin, 1976; Thompson et al., 1971). Some of these compounds play important roles in mediating host-searching behaviors in Campoletis sonorensis (Elzen et al., 1983, 1984; Baehrecke et al., 1989). Behavioral responses of M. croceipes to volatiles from the plant-host complex have been studied by several workers (Drost et al., 1986; Eller et al., 1988; Whitman, 1988). Drost et al. (1986) showed that M. croceipes responded in a wind tunnel to odors by increasing the frequency of sustained and oriented flight. M. croceipes also showed a dose-dependent response to volatiles from its planthost complex in an olfactometer (Eller et al., 1988). M. croceipes was also shown to orient in a flight tunnel to green leaf volatiles (Whitman, 1988; Whitman and Eller, 1990). No information is available on the nature of neurophysiological responses of M. croceipes to volatiles emanating from its plant-host complex. The purpose of this study was to investigate neurophysiological responses of M. croceipes to cotton volatiles by means of electroantennograms (EAGs). Our results provide a basis for our future studies of single olfactory receptor neurons on the antenna. METHODS AND MATERIALS
Insects. Microplitis croceipes used in this study were from a colony maintained at the USDA, ARS, Biological Control of Insects Research Lab, Columbia, Missouri. Pupae of the wasp were shipped overnight to the behavioral
ANTENNAL RESPONSIVENESS
1763
physiology lab at the USDA, ARS, Boll Weevil Research Unit, Mississippi State, Mississippi, where they were kept in Petri dishes at room temperature (25 _+ 1 ~ until emergence. Newly emerged wasps were sexed and maintained in Petri dishes at room temperature and a light cycle of 16 hr light and 8 hr darkness. Adult wasps were provided water and honey until testing three to six days after emergence. Odorous Stimuli. Compounds tested, their purities, and supply sources are listed in Table 1. These compounds were chosen based primarily on their presence in the cotton plant and/or in the green leaf volatile complex (Dickens, 1984). Compounds were diluted in spectrometric grade hexane, and tested for purity using a gas chromatograph (Tracor Instruments model 540, Austin, Texas). Serial dilutions of 0.001-100/zg//zl were prepared for eight of the 29 compounds tested. All test solutions were kept at - 4 ~ before and after each experiment. Solutions were allowed to reach room temperature before testing. The odor delivery system and stimulation technique were the same as that described by Dickens (1984). Stimulus compounds diluted in hexane were delivered as 1-/zl or 10-/~1 samples placed on No. 1 filter paper (7 x 16 mm) inserted into glass cartridges (80 mm long x 5 mm ID) and oriented toward the preparation from ca. 1.5 cm. Filtered and dried hydrocarbon-free air carried odor molecules evaporating from the filter paper over the preparation. Stimulus duration was 1.0 sec; air flow was 1 m/sec as measured by a thermistor. The atmosphere around the preparation was continuously exhausted. Electrophysiological Recording Technique. Electroantennogram (=EAG) techniques used in this study were a modification of previous techniques and are described in detail elsewhere (Schneider, 1957; Dickens and Payne, 1977; Dickens, 1984). Ag-AgC1 glass capillary microelectrodes were used. Following prepuncture with a sharpened tungsten needle, the recording electrode was inserted into the distal segment of the antenna. In the same manner, the indifferent electrode was inserted into the scape. The signal was amplified 10 x by a Grass P-16 AC/DC microelectrode amplifier and viewed on a Tektronix 5111A storage oscilloscope. EAGs were recorded on graph paper by an Omniscribe recorder (Houston Instruments, Houston, Texas) for subsequent analyses and storage. Experimental Procedure. Two series of experiments were conducted. In the first series of experiments, responses of the antennal receptors were measured by recording EAGs to 100-/zg stimulus loads of each test odorant. In the second series of experiments, eight compounds were chosen based on results of the first series of experiments, and their function in host location of other parasitic wasps. Serial dilutions of each odorant were presented from the lowest to the highest dose. Because of the variation in volatility of the test compounds, only relative comparisons can be made between the stimuli. In the first experimental series, EAGs were recorded from five wasps of
1764
LI ET AL.
TABLE 1. SOURCE AND PURITY OF VOLATILE COMPOUNDS USED IN L A G STUDIES OF
M. croceipes Compound Aliphatic alcohols Propan- 1-ol Butan- 1-ol Pentan- 1-ol Hexan- 1-ol (E)-2-Hexen-l-01 (Z)-3-Hexen-l-o! Heptan- 1-ol Octan- t -ol Nonan- 1-ol Decan- 1-ol Aliphatic aldehydes Hexanal (E)-2-Hexenal Heptanal Monoterpene hydrocarbons (-)-a-Pinene (-)-3-Pinene 3-Ocimene (-)-Limonene Myrcene Oxygenated monoterpenes Geraniol Nerol Linalool Sesquiterpenes ~-BisaboloI 3-Bisabo[ol Caryophyltene oxide 3-Caryophyllene Miscellaneous compounds (Z)-3-Hexenyl acetate Acetophenone Benzaldehyde Decane Undecane
Chemical purity (%)
Source of supply a
> 99 > 99 > 99 98 97 98 98 > 99 97 > 99
A A B C C C C C C C
99 99 95
C C C
98 80-90 60 97 85
C C D C C
> 90 > 65 99 92 83 97 90.2 97 99 > 98 > 99 99
E E C F G C F H C C C C
aA, Fisher Scientific Co., Pittsburgh, Pennsylvania; B, J.T. Baker Chem. Co., Phillipsburg, New Jersey; C, Aldrich Chem. Co., Milwaukee, Wisconsin; D. Alfred Bader Lib. Rare Chem., Milwaukee, Wisconsin; E, Pfaltz & Bauer, Inc., Stamford, Connecticut; F, ICN-K&K Laboratories, Inc., Plainview, New York; G, P.A. Hedin, USDA, ARS, Mississippi State, Mississippi; H, Sigma Chemical Co., St. Louis, Missouri.
1765
ANTENNAL RESPONSIVENESS
each sex for each test compound. Only female wasps were tested for the doseresponse studies for each of eight chosen compounds. In both experiments, control stimulations (1 #1 or 10 #1 of hexane solvent on filter paper in a cartridge) were tested three to five times in each preparation. Hexan-l-ol (100 #g on filter paper) served as a standard and was tested following every two experimental stimuli. Response to the solvent control, if any, was deducted from all responses. Responses to the standard were used to normalize all responses so that the responses from different preparations could be compared. At least 3 rain were allowed between each stimulation for recovery of the EAG. EAGs to test compounds were evaluated by measuring the maximum amplitude of depolarization elicited by a stimulus, then subtracting the amplitude of the response to the preceding control. Miltivolt responses to all stimuli were converted to percentage values of the mean responses to the two nearest standards (Dickens, 1984). Threshold was defined as the stimulus load at which the mean EAG minus the standard error was greater than the mean EAG to the lowest stimulus load tested plus the standard error. Saturation was considered to be the stimulus load at which the mean EAG plus the standard error was greater than or equal to the mean EAG minus the standard error for the next highest dose tested. Statistical Analyses. For the first series of experiments, EAGs to 100-/zg doses of the various odorants were compared for significant differences using a general linear models procedure (SAS Institute, Cary, North Carolina). Doseresponse curves were evaluated by analysis of variance. Where differences were noted, comparisons were made using Duncan's new multiple-range test.
RESULTS
Mean responses of the male [ - 3 . 5 3 mV (SE = +0.14 mV)] and female [ - 3 . 8 1 mV (SE = _+0.13 mV)] M. croceipes to the 1-hexanol standard (100 ~g) were not significantly different (P < 0.05). Selectivity. Although no significant differences were found in responses of male and female wasps to the various odorants, EAGs to individual compounds did differ (P < 0.05) (Table 2). The most effective odorants were heptanal, benzenoids (i.e., benzaldehyde and acetophenone), and green leaf volatiles (GLVs). Responses of both sexes to primary alcohols with three to 10 carbons were maximal for the six-carbon alcohol (hexan-l-ol, a GLV), and declined as the carbon chain length increased or decreased (Figure IA). All GLVs tested elicited strong responses in both male and female wasps, but EAGs for individual GLVs did not differ significantly (P < 0.05) (Figure 1B). Monoterpenes were less effective than GLVs and elicited moderate responses. Among the oxygenated monoterpenes tested, linalool elicited significantly higher responses
1766
LIET AL.
TABLE 2. MEAN L A G s OF FIVE MALE AND FIVE FEMALE M. c r o c e i p e s TO 100-/zg STIMULUS LOAD OF SELECTED COTTON VOLATILES AND OTHER COMPOUNDS a
Compound
Mean LAG b
lk c
Heptanal Benzaldehyde Acetophenone (E)-2-Hexenal Heptan-l-ol (Z)-3-Hexen-l-ol (E)-2-Hexen-l-ol Hexanal Octan- 1-ol Pentan-l-ol /3-Bisabolol Decan-l-ol /3-Ocimene Decane Linalool Nonan-l-ol Undecane Butan-l-ol (-)-/3-Pinene Myrcene Geraniol (-)-Limonene Nerol /3-Caryophyllene (-)-a-Pinene Propan- 1-ol ot-Bisabolol Caryophyllene oxide
101.2 A 101.1 A 100.8 A 97.0 A 94.8 A 86.0 A 85.6 AB 84,6 ABC 71.2 BCD 67.0 CDE 62.3 DEF 57.2 DEFG 56.3 DEFG 55.8 DEFG 51.6 DEFGH 48.5 EFGHI 47.2 EFGHI 44.9 FGHI 44.0 FGHI 41.9 GHI 40.3 GHI 37.5 GHIJ 33.7 HIJ 32.8 HIJ 30.3 IJ 18,9 J 18.6 J 18.0 J
1040 1172 1270 965 [1080] 990 995 945 [ 1180] 880 1795 [1380] 1210 1000 [1280] [1100] [780] 1105 1115 1425 1168 1410 1582 1080 [680] 1765
a LAGs from males and females were combined since chemical * sex interaction was not significant. Numbers followed by different letters are significantly different (P < 0.05; Duncan's multiplerange test). bEAGs are represented as percent response to the standard = hexan-l-ol (100 ttg). CKov~ts' indices on OV-17 column from Hedin et al. (1975). Values in brackets were extrapolated.
than either geraniol or nerol. /3-Ocimene was the most effective monoterpene hydrocarbon with decreasing responses elicited by the remaining compounds of the same class: (-)-13-pinene > myrcene > (-)-limonene > (-)-o~-pinene. With the exception of /3-bisabolol, which elicited a moderate response, the sesquiterpenes were ineffective. Both of the alkanes tested, decane and undecane, elicited moderate responses that did not differ significantly. Sensitivity. Dose-response curves were constructed from LAGs of female
1767
ANTENNAL RESPONSIVENESS
A*
% EAG relative to standard* 120 100
8O 6O
40 20 0
3
4
$
6
7
8
9
10
N u m b e r of carbon atoms
B. % EAG relative to standard* 120 100
80
60 40 20 0 E-2-6:OH 6:AL Green leaf volatile
Z-3-6:AC
FIc. 1. Mean EAGs of M. croceipes (N = 10, five males and five females) to: (A) stimulus load of 100/zg of saturated primary alcohols with chain length from three to ten carbons; (B) stimulus load of 100 ~g of green leaf volatile alcohols [hexan-l-ol = 6: OH, (E)-2-hexen-l-ol = E 2 - 6 : OH, (Z)-3-hexen-l-ol = Z 3 - 6 : OH), aldehydes (hexanal = 6:A1, (E)-2-hexenal = E 2 - 6 : A 1 ] , and acetate [(Z)-3-hexenyl acetate = Z 3 - 6 : Ac, N = 5 females for this compound]. See Table 2 for statistical analyses.
1768
LI~T AL.
wasps to eight volatiles distributed among four chemical classes: monoterpenes, sesquiterpenes, green leaf volatiles, and benzene derivatives (Figure 2). Thresholds for compounds relative to those in the same class were: monoterpenes--/3ocimene (0.01 tzg) > linalool (10 ~g) > (-)-/3-pinene (100 t~g); sesquiterpenes--/3-bisabolol (1/~g) = /3-caryophyllene (1 tzg); green leaf volatiles--(Z)3-hexenyl acetate (0.01 #g) > (E)-2-hexenal (1/zg). Benzaldehyde had a threshold of 1.0 /zg. Only /3-caryophyllene clearly reached saturation at a stimulus load of 10/zg.
DISCUSSION
Characteristics of Electroantennograms. No significant sexual differences were found in EAGs to 100-#g stimulus loads among the 29 compounds tested. Similarity in antennal responses of male and female insects to plant odors was previously noted for several phytophagous pests (Fein et al., 1982; Dickens, 1984; Wellso et al., 1984; Light et al., 1988; Hansson et al., 1989). The explanation for such similarity was that males and females live in the same habitat and may share the same chemical cues to locate host plants on which to feed and mate. In contrast to phytophagous insects, female parasitic wasps use plant volatiles to locate host insects, while male wasps probably share some of the same chemical cues, possibly in conjunction with a sex pheromone (Powell and King, 1984), to locate females for mating. Differences in behavioral responses of male and female wasps to green leaf volatiles (Whitman, 1988; Whitman and Eller, 1990) may arise from sexual differences in higher-order processing of incoming peripheral olfactory information. Both the differential EAGs recorded to 100-tzg stimulus loads of the various odorants (Table 2) and the differing dose-response curves to selected odorants recorded from females (Figure 2) indicate differential selectivity and sensitivity of the antennal olfactory receptor system of M. croceipes for host habitat odors. Clearly the number of molecules emanating from the filter paper in the glass odor cartridge, and ultimately reaching the preparation, differs for the various odorants tested due to differing volatilities of the compounds and binding characteristics of the compounds to the filter paper. Thus, what is being compared is stimulus loading and not the number of molecules reaching the preparation. However, the observed differences in EAGs cannot be ascribed solely to differences in volatility of the various compounds. This point is illustrated by comparing Kovhts' retention indices (a measure of relative volatility) (Kovhts, 1961) for the various odorants with observed thresholds of neurons responsive to them (Table 2). Kovhts' indices previously calculated for the compounds, and threshold stimulus loads (/~g) were: /3-ocimene = 1210, 0.01; /3-bisabolol = 1795, 1.0;/3-caryophyllene = 1582, 1.0; (E)-2-hexenal = 965, 1.0; benzaldehyde =
1769
ANTENNAL RESPONSIVENESS
Ae
Be
%EAG relative to standard 120 100
% EAG relative to standard 1201
-
1 O0 ]
80-
so 1 a+
6040 -
a
~ 40
200 ~ , , , , 0.001 0.01 0.1 LO 10 100 Stimulus load (/~g on filter paper) (-)-/3-plnene ~ # - o e i m e n e
a ~-
~t 20
+
b
0 0.001 0.01 0.1 1.0 10 100 Stimulus load ( # g on filter paper)
linalool
J ~-bisabolol
~
