Electrophysiological and behavioural responses of female Isoceras sibirica (Lepidoptera: Cossidae) to volatiles produced by the plant, Asparagus officinalis

In herbivorous insects plant volatiles have an important role in locating mates and oviposition sites. The aim of this study was to test the antennal and behavioural responses of females of Isoceras sibirica Alpheraky (Lepidoptera: Cossidae) to the volatiles produced by Asparagus offi cinalis L. Electroantennographs (EAG) revealed that the antennae of I. sibirica respond in a dose dependent way to hexanal, limonene, 2-ethyl-2-hexenal, linalool and α-terpineol. In fl ight tunnel experiments, 2-ethyl-2-hexenal, α-terpineol, hexanal, ρ-cymene and geraniol were signifi cantly more attractive to females of I. sibirica than other chemicals. These fi ndings indicate that host volatiles are important for host recognition in I. sibirica. * Corresponding author; e-mail: zhangjintong@126.com. INTRODUCTION Plant volatiles play a signifi cant role in mate and host recognition in phytophagous insects, especially herbivores with narrow host ranges, because they must fi nd a single species of host plant surrounded by a range of non-hosts (Tahvanainen & Root, 1972; Reed & Landolt, 2002; Ansebo et al., 2004; Hern & Dorn, 2004; Natale et al., 2004; Tasin et al., 2005, 2006a, b, 2007). That host volatiles are important in host-plant recognition in polyphagous insects is documented for many species. Eleven chemicals in the shoots of riverbank grape (Vitis riparia Michx.) elicit female grape berry moth (Paralobesia viteana descriptor) to fl y upwind fl ight in fl ight-tunnel tests ( Cha et al., 2008a, b). Volatiles collected from grapevine Vitis vinifera elicit strong antennal responses in females of Lobesia botrana (Denis & Schiffermüller) and in wind tunnel tests, the main compound (E)-β-caryophyllene elicit a strong upwind fl ight (Tasin et al., 2005, 2006a). In addition, synthetic grape volatiles attract mated females of Lobesia Botrana in the fi eld (Anfora et al., 2009). Volatiles from Aquilaria sinensis (Lour.) Gilg (Thymelaeaceae) leaves attract Heortia vitessoides Moore (Lepidoptera: Crambidae), and odour blends of young leaves play an important role in H. vitessoides host plant recognition ( Qiao et al., 2012). Floral odours elicit signifi cant EAG responses from female Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae) (Burguiere et al., 2001). Therefore, EAG and behavioural Eur. J. Entomol. 114: 101–105, 2017 doi: 10.14411/eje.2017.014


INTRODUCTION
Plant volatiles play a signifi cant role in mate and host recognition in phytophagous insects, especially herbivores with narrow host ranges, because they must fi nd a single species of host plant surrounded by a range of non-hosts (Tahvanainen & Root, 1972;Reed & Landolt, 2002;Ansebo et al., 2004;Hern & Dorn, 2004;Natale et al., 2004;Tasin et al., 2005Tasin et al., , 2006aTasin et al., , b, 2007)).That host volatiles are important in host-plant recognition in polyphagous insects is documented for many species.Eleven chemicals in the shoots of riverbank grape (Vitis riparia Michx.)elicit female grape berry moth (Paralobesia viteana descriptor) to fl y upwind fl ight in fl ight-tunnel tests ( Cha et al., 2008a, b).Volatiles collected from grapevine Vitis vinifera elicit strong antennal responses in females of Lobesia botrana (Denis & Schiffermüller) and in wind tunnel tests, the main compound (E)-β-caryophyllene elicit a strong upwind fl ight (Tasin et al., 2005(Tasin et al., , 2006a)).In addition, synthetic grape volatiles attract mated females of Lobesia Botrana in the fi eld (Anfora et al., 2009).Volatiles from Aquilaria sinensis (Lour.)Gilg (Thymelaeaceae) leaves attract Heortia vitessoides Moore (Lepidoptera: Crambidae), and odour blends of young leaves play an important role in H. vitessoides host plant recognition ( Qiao et al., 2012).Floral odours elicit signifi cant EAG responses from female Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae) (Burguiere et al., 2001).Therefore, EAG and behavioural bulb (light intensity approximately 0.5 lux).During the experiments, the temperature and relative humidity in the fl ight tunnel were 25 ± 1° and 75% ± 10, respectively, and the photoperiod 14L : 10D.All experiments were done during dusk, as the circadian fl ight activity of this species peaks in the fourth hour of the scotophase (Liu et al., 2013).Four-day-old females (mating peaks in 3-day-old moths, Liu et al., 2013) from a mating cage were used for fl ight-tunnel bioassays.We used rubber septa (Baoji Guangren Biotechnology Co., Shanxi, China) loaded with chemicals as stimuli.Ten ng of the stimulus was diluted in 1 μl of hexane and placed in the cavity of each septum.Preliminary fl ight tunnel experiments revealed that the moth responds to this concentration.
At the beginning of the scotophase all female moths were transferred into the climatic chamber housing the wind tunnel.The females were tested for 15 min, using batches of 10-15 females per test stimulus.For each moth, we recorded whether it left the 15 × 10 cm cylindrical screen cage and made an upwind fl ight (more than 50 cm of tight, zigzag fl ight to within 10 cm of the target) as well as whether it landed on or made contact with the target.For data analysis, we categorized each moth based on the most complete behaviour the moth displayed within a period of 8-min.Thus, the behavioural responses of moths were categorized as "no upwind fl ight" (no directed fl ight toward the target), "upwind fl ight" or "landing".Two to four batches of females were tested per day, and each odour source was tested on at least four different days.In total, each chemical was tested using four to six batches of 10-15 females.The respon ses of female I. sibirica to different treatments were compared by fi tting a generalise d linear model (GENMOD) with upwind fl ight or landing as dependent variables and different chemicals as fi xed independent variables by using binomial distribution with logit link function and maximum likelihood estimation (Proc Glimmix, SAS Institute, 2006).Test effects were compared by using contrast statements and mean % response, and standard errors were estimated using lsmeans statement with diff option.

Electroantennogram responses
Electroantennogram responses of female I. sibiric a to different chemicals are shown in Fig. 1.Female moths showed highest EAG responses to 2-ethyl-2-hexenal and the values were higher than the controls to hexanal.Linalool, hexanal and α-terpineol were in the second class and there was no difference among them.Limonene, p-cymene, ocimene and nonanal did not differ in their activity, which was greater than 57% of that of hexanal.The activity recorded for the other chemicals was lower than 30% of hexanol, except for geraniol (43% of hexanal).
The responses of female I. sibirica to these chemicals were all dose dependent.For every compound, responses to the maximum dose of 10 mg•mL -1 were signifi cantly different from those to other doses (Fig. 2).

Behavioural response of I. sibirica females to plant volatiles
Females responded differently to different single chemicals in terms of upwind fl ight (P < 0.001) and landing (P < 0.001) (Fig. 3).Females responded in terms of upwind fl ight most strongly to 2-ethyl-2 -hexenal (P = 0.99) compared to the control, but this was not statistically different from their response to α-terpineol (P = 0.85) a nd hexanal (P = 0.12).Attractiveness of ρ-cymene to females was gence, adult males and females were provided ad lib with a 10% sucrose solution.

Chemicals
Chemicals were selected on the basis of the results of GC-MS (Sun et al., 2002) and the bioactive effect the chemicals on moths.The source and level of purity of the chemicals are shown in Table 1.

Electroantennograph detection
The EAG was recorded using Tang et al.'s (2012) method.Excised antennae, which were prepared by cutting both distal and basic segments, were each suspended between two glass capillary Ag-AgCl electrodes fi lled with a 0.2M KCl solution.The antenna was positioned 1 cm from a glass tube (0.5 cm inner diameter, 10 cm long) that directed test chemicals into a charcoal-fi ltered humidifi ed air stream (95% RH) at 1 m s −1 on to each antenna.A CS-05 stimulus controller (Syntech) continuously passed humidifi ed air over the antennae at 1 l/min.Antennal signals were passed through a high impedance amplifi er (CS-05 model, Syntech, the Netherlands) and the EAG responses were initially measured in millivolts (peak height of depolarization) and then converted to normalized responses using the Syntech EAG 2000 program (Syntech, The Netherlands).
The test chemicals were dissolved in paraffi n oil at a concentration of 10 mg •mL -1 , and concentrations of each of the chemicals of 0.01, 0.1, 1 and 10 mg•mL -1 were used in the tests.Ten μL of each chemical solution was placed on a fi lter paper strip (5 × 60 mm) and the solvent allowed to evaporate for 30 s before the strip was placed inside a glass Pasteur pipette cartridge (14.5 cm long) used to dispense the test volatiles.Filter papers treated with 5 μL of paraffi n oil and hexenal was used as the control and s tandard chemical, respectively.The chemical stimuli were tested randomly, with a stimulus duration time of 0.5 s, followed by a 2 min purge using humidifi ed and purifi ed air to ensure that the antennal receptors recovered.Response to the solvent control was subtracted from all of the normalized responses and the normalized EAG responses were expressed as a percentage of the response to the standard chemical.Each chemical was tested using 15 individual female antennae, and each antenna tested three times.The EAG response data for different chemicals and different concentrations were tested using a one-way ANOVA and compared using Du ncan's multiple range test (P < 0.05) in the program SPSS 18.0 (SPSS Inc., Chicago, IL, USA).

Flight-tunnel assay
The behavioural responses of individual females of I. sibiraca to the eleven chemicals was tested in a poly (methyl methacrylate) fl ight tunnel (195 × 60 × 60 cm) as described in Cha et al. (2008a).Light was provided by a single red incandescent light signifi cantly lower than to hexanal (P < 0.001), but not statistically different from the response to geraniol (P = 0.71), which was signifi cantly greater than to nonanal (P < 0.001).Limonene was signifi cantly less attractive to females than nonanal (P = 0.05), but not statistically different from the response to ocimene (P = 0.15) and linalool (P = 0.11).Attractiveness of α-pinene and β-pinene to females was not statistically signifi cant.The percentage of females that landed on the geraniol target was signifi cantly higher than on the 2-ethyl-2-hexenal target (P < 0.001), but there were no statistical differences between the landings recorded on 2-ethyl-2-hexenal and hexanal (P = 0.12), hexanal and α-terpineol (P = 0.08) and α-terpineol and ρ-cymene targets (P = 0.05), and no landings were recorded on the targets with other chemicals.
Among the chemicals tested, although 2-ethyl-2-hexenal, hexanal and α-terpineol attracted more females than other chemicals, the upwind fl ight responses were only 43.2% and 42.4% respectively.The low responses recorded indicate that in addition to specifi c plant odours a particular ratio of different chemicals might be more attractive.Visser (1986) and Bruce et al. (2005) argue that the ratio of the chemicals in plant volatiles provide the specifi c recognition cues.This has been shown for females of L. botrana, where the ratio of 3 synthetic grapevine volatiles is important for host recognition in this species (Tasin et al., 2006b).Furthermore, in the behaviour test, only a few females arrived at the source of the stimulus.A possible explanation is that, in nature, background odour from plants surrounding the host will interfere with the recognition process, in addition fl ight tunnel and biological factors, such as wind speed, diel periodicity and mating status need to be considered.Cha et al. (2008a) report that the behaviour of females depended on wind speed and mated females were more likely to land on the source than unmated females.
Our fi nding that females of I. sibirica respond to host plant volatiles, indicate it is possible to develop synthetic host plant lures and these volatiles will be useful in future studies on the processing of olfactory information.

Fig. 1 .
Fig. 1.Mean relative electroantennogram (EAG) response (expressed as a % of that recorded in the control plus standard deviations, see text) of females of I. sibiraca to 11 chemicals.The test of each chemical was done using the antennae of 15 females and the response of each antenna was recorded three times.Bars followed by different letters differ signifi cantly based on an ANOVA followed by Duncan's new multiple range test (P < 0.05).

Fig. 2 .
Fig.2.Mean relative electroantennogram response (expressed as a % of that recorded in the control plus standard deviations, see text) of females of I. sibiraca to 11 chemicals at four different doses (0.01, 0.1, 1 and 10 mg•mL −1 ).The test of each chemical was done using the antennae of 15 females and the response of each antenna was recorded three times.

Fig. 3 .
Fig. 3. Flight tunnel response (%) of females I. sibiraca to the different chemicals listed below the X axis (N = 636).Different letters (capital letters for upwind fl ight and small letters for landing) on bars indicate signifi cant differences in the responses (P < 0.05).

Table 1 .
Purity and source of the chemicals used.