Antennal and Behavioral Responses of Drosophila suzukii to Volatiles from a Non-Crop Host, Osyris wightiana

Simple Summary The spotted-wing drosophila Drosophila suzukii, native to Southeast Asia, has become a new threat for cultivated and wild soft-skinned fruits in both their native and invaded areas. The attractiveness of volatiles from fresh ripening and ripe fruits of cultivated crops, such as raspberry, blueberry, strawberry, and cherry, has been extensively studied. To date, however, little attention has been paid to the non-crop host of D. suzukii. In this study, we collected headspace volatiles from ripe fruits of a wild non-crop host, Osyris wightiana, and identified seven compounds attractive to D. suzukii from the fruit volatiles. A blend of the seven active compounds captured a significant amount of female and male D. suzukii in field, suggesting that the blend can be used to develop an attractant for management and monitoring of D. suzukii. Abstract Drosophila suzukii (Diptera: Drosophilidae) infests a variety of commercial fruits, including cherries and other soft-skinned fruits. After the cropping season of most cultivated crop hosts, it heavily infests the fruit of a wild host-plant, Osyris wightiana in southwest China. Here, we employ gas chromatography-electroantennographic detection (GC-EAD) together with behavioral bioassays and a trapping experiment to identify volatile semiochemicals emitted by O. wightiana that are involved in D. suzukii attraction. GC-EAD recordings of D. suzukii antenna showed responses to 13 compounds, including α-pinene, 3-methylbutyl acetate, 2-hexanol, E-β-ocimene, Z-3-hexenol, β-caryophyllene, α-humulene, and six unidentified compounds. The flies were attracted by seven individual EAD-active compounds at low doses (0.01 and 0.1 μg), but were repelled at high doses (10 and 100 μg). In a similar manner, a blend of seven EAD-active compounds at low doses (0.1 and 1 μg) was attractive to female flies, but became repulsive at high doses (10 μg). The low dose of the blend was as attractive as the fruit volatiles, although both were less attractive than the fruits. The blend attracted both female and male D. suzukii and other Drosophila flies. The percentage of D. suzukii out of all flies captured by the blend was significantly greater than that captured by the control. These results indicate that the EAD-active volatile compounds emitted by fruits of O. wightiana play an important role in D. suzukii attraction, and have the potential to be used for management of D. suzukii.

Wild growing crop species, such as blueberry and blackberry, are preferred hosts of D. suzukii [6][7][8]. Additionally, D. suzukii is known to attack a wide variety of wild/ornamental non-crop plants [9]. More than 150 non-crop plants belonging to 27 families have been determined to be suitable hosts of D. suzukii, providing a rich food resource throughout the year [6,[10][11][12][13].
Osyris wightiana Wall (Santalaceae) is an evergreen shrub or short tree, 2 to 5 m in height in southwest China. It grows on rocky slopes and sparsely distributes in hilly land. Its fruit is similar in appearance, size, and texture to the common cherry (Cerasus yunnanensis (Franch.) Yu et Li, Rosaceae), but has a bitter taste. The fruiting season of O. wightiana is in late September to December and cultivation of crop fruits generally ends in September. Therefore, D. suzukii may migrate from cultivated crops to neighboring O. wightiana bushes and continue infestation before overwintering as adults in reproductive diapause. The heavy infestation of the fruits of O. wightiana between years increases the number of overwintering reproductive flies. Up to 80% of ripe fruits can be infested by D. suzukii, and usually 35 larvae are present in a fruit (0.6 g on average) (unpublished data). Osyris wightiana appears to act as an alternative host source for D. suzukii when crop hosts are not available, which may allow for an early-season rapid increase in the fly population in crop fields.
Like most insects, D. suzukii rely on olfactory detection of volatile compounds released from host plants to search for suitable feeding and oviposition sites [14,15]. For obvious reasons, much attention has been paid to identification of behaviorally attractive volatile compounds from crop hosts for D. suzukii [14,[16][17][18], whereas no emphasis has been placed on non-crop hosts. To identify potential attractive compounds from non-crop host for development of a species-specific lure to manage this pest, we investigated the attraction of mated female D. suzukii to volatiles from ripe fruits of O. wightiana. Specifically, the main objectives of this study were to: (1) assess the behavioral response of mated female D. suzukii to volatiles from ripe fruits of O. wightiana; (2) identify volatile compounds that can be detected by mated female D. suzukii; and (3) evaluate the attractiveness of individual antennally active compounds and their mixture in a series of laboratory twochoice bioassays and field trapping experiments.

Insects
The colony of D. suzukii used in this study was obtained from infested fruits of bayberries, Myrica rubra (Lour.) Sieb. et Zucc. in Yunnan Province, China in May 2016. The flies were reared on peeled bananas in an incubator chamber at 24 ± 1 • C, 65 ± 5% R.H., and 14:10 h L:D.
Newly-enclosed flies were kept in a cage (30 × 30 × 30 cm) with access to food and water, and were allowed to mate for 3 days. Adult flies were starved for approximately 24 h prior to experiments. Only five-day-old mated females were used in behavioral and electrophysiological experiments.

Collection of Fruit Volatiles
Newly collected O. wightiana fruits at ripe stage (about 0.6 kg) from Changchong Mountain, Kunming, Yunnan Province were enclosed in a glass vessel (6 cm diam., 20 cm high). Air entering the vessel was drawn through a filter filled with activated charcoal. Air was pulled out a vacuum pump (QC-1S; Beijing Municipal Institute of Labor Protection, Beijing, China) from the vessel at 400 mL/min through an adsorbent tube (5 mm Insects 2021, 12, 166 3 of 10 internal diam.; 500 mg Porapak-Q, 80/100 mesh, Supelco, Bellefonte, PA, USA) for 3 h. All connections were made with Teflon tubing. Volatiles were eluted from the Porapak-Q trap with 3 mL of hexane (98%, Merck, KGaA, Darmstadt, Germany). The volatile samples from 4 collections were combined and concentrated to 500 µL under a mild N 2 stream for chemical analysis, electrophysiological experiments, and behavior experiments. A second set of 4 collections were conducted as described above for quantitative analysis by GC-FID. Nonyl acetate [19] was added to volatile samples of each collection as the internal standard for subsequent quantification.

Coupled Gas Chromatography-Electroantennographic Detection (GC-EAD)
The O. wightiana fruit volatile sample (2 µL) was injected in a Shimadzu GC-2010plus equipped with a DB-Wax column (30 m × 0.25 mm inner diameter × 0.25 µm film thickness; Agilent) and interfaced with the EAG apparatus. Helium (2.0 mL/min) was used as the carrier gas. The oven temperature was held at 40 • C for 2 min and programmed at 5 • C/min to 120 • C, then 15 • C/min to 240 • C and held for 4 min. The column effluent was split at a 1:1 ratio, with one part to a heated line into a humidified airstream (400 mL/min) which was directed at the antenna preparation, and the other to the Flame Ionization Detector (FID) of GC. Electroantennogram (EAG) recordings were made using Ag-AgCl glass microelectrodes filled with a Ephrussi-Beadle Ringer solution. The intact female fly was immobilized by cotton inside a pipette tip, while its head stretched out. The recording electrode was inserted into the distal region of the terminal antennal segment, while the reference electrode was positioned into the base of the antenna. The signals generated by the EAD and FID were passed through a Syntech IDAC-2 high-impedance amplifier and analyzed with Syntech GC-EAD software. FID peaks that elicited EAD responses for at least three runs were considered electrophysiologically-active and marked for identification by coupled gas chromatography-mass spectrometry (GC-MS).

Chemical Analyses
The volatile sample was analyzed with an Agilent 7890A GC coupled to a 5975C mass selective detector equipped with a polar DB-Wax column. Helium (1.0 mL/min) was used as carrier gas. The oven temperature was maintained at 40 • C for 4 min, and then increased to 60 • C at 1 • C/min, to 75 • C at 3 • C/min, to 180 • C at 5 • C/min, and subsequently to 240 • C at 10 • C/min, with a final 4-min holding time. The injector and transfer line temperatures were set at 230 • C and 250 • C, respectively. Mass spectra were obtained by using the electron impact (70 eV). The chemicals that were consistently EAD-active were tentatively identified by comparing their mass spectra with those of the NIST08 database. Their chemical identities were further confirmed by (1) comparing chromatographic retention times through co-injection with those of corresponding authentic standards, and (2) calculating Kovats Indices by comparing the retention times of the compounds of interest with those of the C7-C23 normal alkanes under the same GC-MS conditions. GC-FID analyses were conducted using a Shimadzu GC-2010plus equipped with a DB-Wax capillary column in splitless mode. The oven temperature was programmed in accordance with GC-EAD recordings. Eighty µL of 0.1 µg/µL nonyl acetate was added to volatile samples of each collection (3 mL) as an internal standard for chemical quantification by comparing their peak areas.

Y-Tube Olfactometer Bioassays
The behavioral responses of female D. suzukii to host fruits, volatile extract of host fruits and synthetic compounds were investigated in a Y-tube olfactometer (stem and arms: 15 cm and 10 cm long, respectively, 2 cm internal diam., and 60 • Y angle). Two arms of the olfactometer were connected to two bulbs with an open end (ϕ = 2 cm). The open end of the arm slightly approaches into the bulb, which can partially prevent flies coming back to the main stem. The whole Y-tube olfactometer was housed inside a plastic box (43 × 30 × 18 cm) which was covered with an opaque black cloth to avoid visual distraction of the flies. Air entering the arms was pumped (atmospheric sampler) out the stem of the olfactometer at 200 mL/min. To test the response of mated D. suzukii females to host fruits, 25 intact fresh O. wightiana fruits were enclosed in one bulb as the other bulb was left empty (control). Ten females at a time were introduced at the entrance of the main stem. The number of flies that entered into the side arms was counted after 15 min. The flies that remained in the main stem after this time period were recorded as 'no choice'. The olfactometer set-up was cleaned with ethyl alcohol and the glass parts were baked at 200 • C after every trial. There were 10 replicates for each treatment, and all flies were used only once.
The responses of mated D. suzukii females to fruit volatile sample, individual EADactive compounds and their mixture at natural ratio (α-pinene:3-methylbutyl acetate:2hexanol:β-ocimene:Z-3-hexenol:β-caryophyllene:α-humulene = 38:127:21:31:17:8:5) were conducted in the same manner. One glass bulb holding a filter paper disk (ϕ = 1.5 cm) loaded with 10 µL of the sample solution served as treatment while the other glass bulb holding a filter paper disk of the same size received 10 µL of hexane and served as control. All the synthetic compounds were prepared in hexane and diluted to 10, 1, 0.1, 0.01, and 0.001 µg/µL while the mixture was diluted with hexane to 1, 0.1, and 0.01 µg/µL. wightiana plants were about 2-meters tall. One hundred µL of the mixture of the 7 EAD-active compounds, at natural ratio mentioned above (0.01 µg/µL as Blend 1, and 0.1 µg/µL as Blend 2), or hexane (control) were loaded into an aboral bell-shaped rubber (length = 1.5 cm, Pherobio Technology Co., Ltd., Beijing, China). The lure was placed in a hole at the center of a blue sticky trap (25 cm, Tianxia Wuchong Pest Control Co., Ltd., Nanjing, China). All the traps were hung on O. wightiana tree branches 1~1.5 m above the ground, and 20 m apart.

Field-Trapping Experiments
A randomized complete block design was used for the field experiment. There were 9 replicate blocks, having two treatment traps and a control trap in each block. The numbers of female and male D. suzukii and other flies captured on each trap were counted at the end of the field trapping experiment (12 days in total).

Dual-Choice Olfactometer Bioassays
Significantly more D. suzukii were attracted to fruits and the volatile sample of fruits of O. wightiana than the control (Figure 2).  Table 1.

Dual-Choice Olfactometer Bioassays
Significantly more D. suzukii were attracted to fruits and the volatile sample of fruits of O. wightiana than the control (Figure 2).
In the same manner, the blend of seven EAD-active compounds (0.1 and 1 μg) was significantly attractive to tested females at lower doses, but showed apparent repellency

Field-Trapping Experiment
The number of female and male D. suzukii captured by traps baited with the volatile blend was significantly higher than by control traps (Levene test: df 1 = 2, df 2 = 24, p = 0.246; F = 74.864, p < 0.0001). The percentages of D. suzukii out of all flies captured by the blend were significantly higher than that captured by hexane control traps (Levene test: df 1 = 2, df 2 = 24, p = 0.326; F = 111.654, p < 0.0001). Both blends and hexane control traps captured more males than females of D. suzukii (Figure 3). About half of the non-target flies captured were D. melanogaster. The number of female and male D. suzukii captured by traps baited with the volatile blend was significantly higher than by control traps (Levene test: df1 = 2, df2 = 24, p = 0.246; F = 74.864, p < 0.0001). The percentages of D. suzukii out of all flies captured by the blend were significantly higher than that captured by hexane control traps (Levene test: df1 = 2, df2 = 24, p = 0.326; F = 111.654, p < 0.0001). Both blends and hexane control traps captured more males than females of D. suzukii (Figure 3). About half of the non-target flies captured were D. melanogaster.

Discussion
Several studies have reported on the olfactory responses of D. suzukii to volatile semiochemicals from intact fruits of cultivated host plants [14,[16][17][18]. Drosophila suzukii appears to respond significantly to these volatiles. Here we reported the identification of active volatiles from O. wightiana, a preferred wild host plant of D. suzukii. A non-crop plant species was tested for the first time.
The percentage of responding flies and the attractive effect of the seven EAD-active compound blend at lower doses to D. suzukii were similar to those of the fruit volatile extract of O. wightiana, further indicating that the seven EAD-active compounds are key components from O. wightiana fruits responsible for attracting D. suzukii. The percentage

Discussion
Several studies have reported on the olfactory responses of D. suzukii to volatile semiochemicals from intact fruits of cultivated host plants [14,[16][17][18]. Drosophila suzukii appears to respond significantly to these volatiles. Here we reported the identification of active volatiles from O. wightiana, a preferred wild host plant of D. suzukii. A non-crop plant species was tested for the first time.
The percentage of responding flies and the attractive effect of the seven EAD-active compound blend at lower doses to D. suzukii were similar to those of the fruit volatile extract of O. wightiana, further indicating that the seven EAD-active compounds are key components from O. wightiana fruits responsible for attracting D. suzukii. The percentage of active components in the total amount of the sample analyzed by GC was 3.42% (Table 1), while over 95% of volatile components were not detected by the antennae of D. suzukii in our GC-EAD recordings. These results suggest that although fruits of O. wightiana emit a variety of volatile compounds that signal their identity and physiological state, only a small portion of these volatile compounds can be detected and used by D. suzukii for host location. These green leaf volatiles and species-specific odors may convey both general and specific information on fruits of O. wightiana. In this study, however, banana was used to rear D. suzukii. Laval experience on banana may affect olfactory preferences of adult D. suzukii to fruits of O. wightiana. Innate olfactory response in D. suzukii reared on artificial diet to volatiles of O. wightiana fruits awaits further investigation.
The seven individual EAD-active compounds and their blend displayed dose-dependent olfactory behavior in D. suzukii. The overall response pattern followed the same trend with at-traction to low doses, indifference to intermediate dose and repulsion to high doses ( Figure 2). This dose-dependent behavioral response pattern could be a common feature of olfactory perception in Drosophila [28]. For instance, six attractive odorants from ripen fruits, ethyl acetate, acetoin, butyl butyrate, ethyl hexanoate, ethyl 3-hydroxybutyrate, and phenylacetonitrile were attractive to D. melanogaster at lower concentrations, but aversive to them at higher concentrations [29]. Caged greenhouse experiments showed that traps baited with a lower dose (5.5 µL) of ethyl acetate captured more D. suzukii than that with higher doses (55 µL and 550 µL) [30]. In a recent report, we demonstrated that four EAD-active compounds released from Chinese bayberry, methyl (E)-3-hexenoate, methyl (E)-2-hexenoate, ethyl (E)-2-hexenoate and α-humulene were attractive to mated female D. suzukii at lower doses (0.01 and 0.1 µg), but showed repellency at higher doses (10 and 100 µg) [16]. The mechanism underlying the behavioral switch in response to higher concentrations of odorant in D. melanogaster has been well studied. The activation of two glomeruli, DM1 and VA2, was found to be responsible for the attraction of apple cider vinegar at low concentrations to D. melanogaster. A higher concentration of vinegar is less attractive to flies and excites an additional glomerulus, DM5 that independently mediates aversion. The activation of the aversive glomerulus DM5 may counterbalance the activation of the above two attractive glomeruli. The activation of DM1 or DM5 by any odor should be sufficient for concentration-dependent behavioral switch in D. melanogaster [28]. It needs to be tested whether D. suzukii has a similar brain structure or nervous system to D. melanogaster.
Field capture of both female and male D. suzukii confirmed the attraction of the seven EAD-active compound blend. Interestingly, the blend and control (hexane) captured significantly more males than females, same as sugar-vinegar-water solution for annual monitoring of D. suzukii (unpublished data). A reasonable interpretation for this difference was that the population of D. suzukii in Yunnan province was male-biased during the season of the field trapping experiment. Similarly, significantly more D. suzukii males were caught by the bayberry volatile blend than females in our previous study [16]. The seasonal sex bias in D. suzukii flies have been reported in previous studies from other regions. Rossi-Stacconi et al. [31] observed that male D. suzukii were more abundant than females in both spring and autumn in an Italian mountain region. Significantly higher numbers of male D. suzukii were captured in summer and autumn in the same area [2]. A strong male bias was observed in late autumn [7]. Furthermore, the percentage of D. suzukii captured by the blend was significantly higher than that captured by the control (hexane), indicating that the blend of volatiles derived from a wild host plant was selective in attracting D. suzukii. Moreover, the six unidentified EAD-active compounds could also contribute to attraction of O. wightiana fruits to D. suzukii. The identification of these EAD-active volatiles might allow the development of more selective and powerful attractant lures.

Conclusions
Field margins have been considered to play an important ecological role in providing alternative host resources, overwintering habitats, and refuge areas for insects [32,33]. High infestation rate (unpublished data) in O. wightiana shrubs suggested that O. wightiana can act as an alternative host source for D. suzukii during the post-harvest season of their cultivated host plants in Yunnan province because of its capability for long-distance dispersal across large geographic landscapes [34,35]. As a wild host plant, O. wightiana surrounding crop fields may play an important role in maintaining off-season D. suzukii populations [36]. Management tools such as host volatile compound-based attractants [37], food-based lures and traps [1], protein bait [38], or synthetic attractants [39] applied to field margins may reduce D. suzukii populations, especially if implemented in area-wide programs. Our study demonstrates that the fruit odors become less attractive and eventually repellent as their intensity is increased, suggesting that control of appropriate releasing rate of volatile mixtures is important for future development of lures used for management of D. suzukii in the field. Further investigations should test the effect of late season management of D. suzukii in O. wightiana habitats on its population size on the next-season crop cultivars. Institutional Review Board Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available on request from the corresponding authors.