The Role of Green Gram Plant Volatile Blends in the Behavior of Arctiid Moth, Spilosoma obliqua

This study investigated effects of volatile blends released from undamaged (UD), insect-damaged [ID, plants fed by larvae of Spilosoma obliqua Walker (Lepidoptera: Arctiidae)] and mechanically-damaged (MD) plants of three green gram cultivars [PDM 54, Pusa Baisakhi and Samrat] including synthetic blends on the behavior of conspecific adult moths in Y-tube olfactometer bioassays. Females showed attraction towards volatile blends of UD, ID and MD plants of these green gram cultivars against the control solvent (CH2Cl2). The components of volatile blends in UD plants of three green gram cultivars are not similar, but no any difference was found among three cultivars in term of the attractive effect on the insect moths when volatile blends from UD plants of these three cultivars were tested against one another. Females were more attracted towards volatile blends of ID plants of a particular cultivar compared to UD plants of the same cultivar. Total amount of volatiles was higher in ID plants than UD plants. Some herbivore-induced plant volatiles – (Z)-3-hexenal, 1-hexanol, (Z)-3-hexenyl acetate, 2-octanol and ocimene were attractive to the insect moths. Females were attracted towards three synthetic blends resembling amounts present in natural volatile blends of ID plants of these three cultivars in Y-tube olfactometer and wind tunnel bioassays, suggesting that involvement of host-specific chemical cues in long-range host location by S. obliqua females. If attraction of adult S. obliqua to these synthetic volatile blends is upheld by field trials then these blends may find practical application in detection and monitoring of S. obliqua populations.


Introduction
Green gram [Vigna radiata (L.) Wilczek] (Fabaceae), commonly known as mung bean, is one of the most important pulse crops in India. This leguminous crop is thought to be originated from India, and is grown as a fallow crop after cultivation of rice in India. Asian countries (India, Bangladesh, Korea, Pakistan, China, Japan, Sri Lanka, Indonesia, Philippines, Burma and Thailand) are the main producers of this pulse crop (Gurung et al. 2020;Singh and Sehgal 1992). Now, this pulse crop is also grown in America and Africa (Anonymous 2020). At present, approximately 70% of this pulse crop is contributed by India (Anonymous 2020). The cooked green gram seeds are valuable sources of proteins, carbohydrates, manganese, magnesium, phosphorous, iron, copper, potassium, zinc, and vitamins B1, B2, B3, B5, B6 and B9. Green gram seeds are also consumed as sprouts. The sprouting green gram seeds had higher level of protein, calcium, iron and zinc content as well as lower level of antinutrients such as total phenols, flavonoids and tannins (Oghbaei and Prakash 2017). Attacks by various insect pests such as Lamprosema indica (F.) (Lepidoptera: Pyralidae), Spodoptera exigua (Hubner) (Lepidoptera: Noctuidae), Anticarsia irrorata (F.) (Lepidoptera: Noctuidae), Spodoptera litura (F.) (Lepidoptera: Noctuidae), Agris convolvuli (L.) (Lepidoptera: Sphingidae), Acherontia styx (Westwood) (Lepidoptera: Sphingidae), Spilosoma obliqua Walker (Lepidoptera: Arctiidae), etc. associated with green gram plants cause a reduction in pulse yield (Swaminathan et al. 2012).
The polyphagous insect pest S. obliqua, commonly known as Bihar hairy caterpillar, is of oriental origin and known to cause serious damage to various crops of agricultural importance (Gurung et al. 2020) such as cotton, cauliflower, cabbage, broccoli, mulberry, castor, broad bean (Gurung et al. 2020), jute (Selvaraj et al. 2015), black gram, cowpea (Haribhai 2015), soybean (Chaudhary 2009), sesame (Biswas 2006), sunflower (Varatharajan et al. 1998), groundnut and brinjal (Edde 2021). The insect has been recorded to feed on 126 plant species, which belongs to ca. 24 plant families (Gurung et al. 2020;Singh and Varatharajan 1999). This insect is also a serious pest of green gram plant in India as the larvae of this insect pest feed on the leaves and flowers of green gram, and reduces yield of this pulse crop (Mobarak et al. 2020a). Females of this insect lay eggs on the underside of leaves, and the emerged first instars consume green and soft tissues of the leaves behind the veins, while fourth to sixth instars are the most damaging as they defoliate the whole plant. Nineteen to 20 days are required to complete the larval development of S. obliqua on green gram leaves, and 8-9 days are required to complete pupation, while newly emerged adults (males and females) live for 4-5 days (Mobarak et al. 2020a). Synthetic insecticides (Indoxacarb, Imidacloprid, Thiamethoxam, Lambda cyhalothrin, Triazophos, etc.) are applied to control S. obliqua population (Mohapatra and Gupta 2018), but intensive applications of these substances have problems such as insecticide resistance, harmful effects in environment and reduction of natural enemies of the insect pest. Thus, the primary objective of the present study was to identify the volatile organic compounds (VOCs) from green gram plants causing attraction of adult S. obliqua with the aim of developing a tool for its management.
Female moths use diverse sensory cues in host plant location and acceptance (Calatayud et al. 2014). Before touching host plants, olfaction plays an important role in host recognition and, thus, plant volatile blends serve as potent cues to attract lepidopteran insects (Rojas et al. 2000;Calatayud et al. 2014). However, VOCs emitting from green gram plants causing attraction of adult S. obliqua have not been investigated. Several reports revealed that Y-tube olfactometer is suitable to study the behavioral responses of lepidopteran insects (Calatayud et al. 2014;Li et al. 2012;Mobarak et al. 2020bMobarak et al. , 2022Najar-Rodriguez et al. 2010). In addition, wind tunnels provide more space to the flying insects to observe orientation and landing of the insects (Calatayud et al. 2014). So, we have used Y-tube olfactometer and wind tunnels to observe the behavioral responses of S. obliqua. The survival of lepidopteran offsprings is dependent on the selection of suitable host plants by females as the first instars are sluggish and have no choice to locate alternative host plants (Renwick 1989;Renwick and Chew 1994). After hatching, the first instars of Lepidoptera feed on the leaf tissue of the selected host plant and results in the changes of total amounts of VOCs compared to undamaged plants, and some novel herbivore-induced plant volatiles (HIPVs) are emitted by insect-damaged plants (Sun et al. 2021). Therefore, it is interesting to observe whether some HIPVs are released by S. obliqua larvae-damaged green gram plants and their role in the conspecific female moths. Composition of VOCs varies between insect-damaged and mechanicallydamaged plants due to the metabolic effects of oral secretions by the insect herbivores on the insect-damaged plants Piesik et al. 2010;Portillo-Estrada et al. 2015, 2021. Therefore, it is of substantial interest to monitor whether there are changes in the composition of VOCs in volatile blends emitted by mechanically-damaged green gram plants compared to volatile blends emitted by insectdamaged green gram plants.
Plant cultivars can be categorized depending on the VOCs, which reveal a noteworthy correlation between attraction of insect herbivore and mixture of volatile emission (Darshanee et al. 2017;McDaniel et al. 2016). In the present study, three green gram cultivars (cv. PDM 54, Pusa Baisakhi and Samrat) were taken because these three green gram cultivars are commonly cultivated in West Bengal, India due to high yielding potential as genetic make-up of these cultivars is suitable in present conditions. So, we first studied whether volatile blends emitting from undamaged plants of these three green gram cultivars are causing similar behavioral responses to S. obliqua. We also studied the behavioral responses of S. obliqua towards volatile blends emitted from insectdamaged and mechanically-damaged plants of these three green gram cultivars. To compare the VOC profiles among differently treated plants of these green gram cultivars, we have identified and quantified the VOCs of these cultivars by gas chromatography-mass spectrometry (GC-MS) and gas chromatography-flame ionization detection (GC-FID) analyses, respectively. We evaluated the role of individual synthetic VOCs and synthetic blends resembling amounts present in differently treated green gram cultivars in the insect responses to determine whether synthetic blends could act as olfactory cues to attract S. obliqua. The current study shows how biologically active components of volatile blends from differently treated green gram cultivars act as cues to attract S. obliqua, which helps to realize the mechanisms underlying chemically mediated interactions between volatile blends of differently treated green gram cultivars and the insect pest.

Methods and Materials
Plants Seeds of three green gram cultivars (PDM 54, Pusa Baisakhi and Samrat, hereafter these three cultivars will be expressed as PDM, PUSA and SAM, respectively) were germinated on moistened filter papers. Sufficient amounts of soil were collected from the field of Crop Research Farm (CRF), The University of Burdwan (23°16′ N, 87°54′ E), West Bengal, India and sterilised in an autoclave at 121 °C for 30 min. Each seed with cotyledon was sown in an earthen pot containing ca. 1500 cm 3 of sterilised soil, and grown 1 3 under natural photoperiod of 13:11 L/D at 30-35 °C during July -September, 2020. The whole plant with the pot was covered with fine mesh nylon net (180 cm height, 90 cm diameter) to avoid insect attack and unintentional infection. Plants (2-3-wks-old) of each green gram cultivar with a height of ca. 20-25 cm and 10 fully developed leaves were used for the experiments.
Insects Sixth instars of S. obliqua were collected from jute plants adjoining to the fields of CRF, The University of Burdwan. Larvae were fed on castor plants in an environmental chamber at 27 ± 2 °C, 65 ± 10% relative humidity (r. h.) and a photoperiod of 12:12 L/D. Emerged adults were kept on castor leaves as females could lay eggs and F1 eggs were collected. The emerged larvae of S. obliqua were also fed on castor leaves. Second generation newly emerged 4 th instars of S. obliqua were placed on the plants of each green gram cultivar for insect feeding damage treatments. Newly emerged males and females (males were identified by bipectinate antenna, while females were identified by serrated antenna) were paired, and mating occurred between four and six hours of emergence. Second generation females (1-2 days old) were used for olfactometer and wind tunnel bioassays. Larvae were not reared on green gram leaves as insects might be habituated to the volatile blends of green gram plants, which might cause a bias of S. obliqua females towards volatile blends of green gram plants in bioassays.

Plant Volatile Blend Collection
Undamaged (UD) plants of three green gram cultivars (PDM, PUSA and SAM) were used for collection of volatile blends. For insect damage (ID) treatments, a group of five newly emerged 4 th instars of S. obliqua had been continuously fed for 12 h on the plants of PDM, PUSA and SAM (hereafter referred as ID PDM, ID PUSA and ID SAM, respectively). Volatile blends were collected from ID plants just after feeding by 4 th instars. For the treatment of mechanical damage (MD), 10 leaves of a plant were wounded twice with hole punches (each hole 0.5 cm), and volatile blends were collected without delay after wounding. An experimental plant was positioned in a closed glass dome (4 L) with Teflon base having a small opening for passing the stem of plant ) (Supplementary Fig. S1). After placing each plant in the closed glass dome, sterilized non-absorbent cotton was encircled in the stem of an experimental plant to avoid damage in the stem by the Teflon base. Five experimental plants of each treatment were used to collect volatile blends for 4 h between 9 am and 1 pm, and a fraction of volatile blends was applied to identify and quantify the VOCs while remaining fractions were used for olfactory bioassays. For olfactory bioassays, volatile blends were collected from 33 UD plants of each green gram cultivar, 82 plants of ID PDM, 75 plants each of ID PUSA and ID SAM, and 33 MD plants of each green gram cultivar over 4 h between 9 am and 1 pm. Charcoal-filtered air was propelled (4 L min −1 ) into the top of the closed glass dome and pulled (2 L min −1 ) through four volatile collector traps (150 mm long × 5 mm o.d.). Each trap comprised of 80 mg HayeSep Q (80-100 mesh) as adsorbent, and four volatile collector traps were put into four side sampling ports round about the base of closed glass dome.
To elute volatile blends from the adsorbent of an experimental plant, HayeSep Q from four volatile collector traps were amalgamated and 600 μl methylene chloride were applied. The collected volatile blends in methylene chloride were concentrated to 100 μl through a gentle nitrogen flow. A fraction (75 μl) was employed for olfactory bioassays, and the rest (25 μl) was applied for identification and quantification of VOCs through GC-MS and GC-FID, respectively (the volatile blends collected from five experimental plants of each treatment). For olfactory bioassays, an aliquot of 25 μl (equivalent to volatile blends emitted by a plant during 1 h) was applied to Whatman No. 41 filter paper (1 cm 2 ).

Olfactometer Bioassays
We performed olfactory bioassays of S. obliqua females using a glass Y-tube olfactometer (each arm 15 cm, radius of each arm 4 cm and 45° Y angle) ( Supplementary Fig. S2). The stem of the olfactometer was attached with a porous glass vial (4 cm radius × 7 cm long) (as the air can flow through the olfactometer). Each arm of the olfactometer was connected with an adapter attached with a glass vial (4 cm radius × 4 cm long), which contained a piece (1 cm 2 ) of filter paper (Whatman No. 41) loaded with 25 μl volatile blends (VOCs collected from variously treated experimental plants or individual synthetic compounds or synthetic blends) and other glass vial contained a filter paper of the same size loaded with 25 μl control solvent (methylene chloride), and a female was placed in the stem of the olfactometer. Charcoal-filtered air was passed through each arm of the Y-tube at 125 ml min −1 . Teflon tubes were used to connect between different parts of the olfactometer set-up.
Behavioral responses of S. obliqua females (1-2 days old) using Y-tube olfactometer were conducted in the laboratory between 16.30 and 20.30 h at 27 ± 2 °C, 70 ± 5% r. h. and a light intensity of 150 lx. Females did not show bias towards the control solvent (methylene chloride) in preliminary olfactory bioassays. The choice of a female was recorded as positive response when the female reached the end of the odor-loaded arm of the Y-tube, while a negative response was recorded when the female moved towards the solvent-loaded arm. Behavioral response was recorded as no choice when a female was in the stem of the Y-tube for 3 min ). A female was tested once and it was discarded from bioassays. Responses of S. obliqua females towards one volatile blend sample were finished when 60 adult females had responded. After testing five insects, the filter papers with either volatile blend sample or the control solvent were changed and the Y-tube was rotated 180° to keep away from directional effects. Further after testing five insects, the Y-tube olfactometer set-up was thoroughly washed with petroleum ether followed by acetone and dried in a hot-air oven at 50 °C for 1 h. Teflon tubes were cleaned by 70% ethanol two times and dried at 27 ± 2 °C for 1 h after testing five insects.
Behavioral Responses of S. obliqua Towards Natural Volatile Blends in Olfactometer Bioassays Responses of unmated females or gravid females or males towards volatile blends of UD plants of each green gram cultivar (PDM or PUSA or SAM) were tested against the control solvent [methylene chloride (CH 2 Cl 2 )] to observe whether there were differences in the responses of unmated females, gravid females and males towards natural volatile blends of UD plants of each green gram cultivar.
Bioassays using unmated females towards volatile blends emitted by differently treated plants (UD, ID and MD) of three green gram cultivars (PDM, PUSA and SAM) were performed in a variety of combinations (Supplementary  Table S1).

Volatile Blend Analysis
Identification of collected volatile blends from five separate experimental plants of each treatment were carried out through a Perkin-Elmer Clarus 690 coupled to a Mass selective detector fitted with a HP-5 capillary column (Agilent; Palo Alto, CA, USA; length: 30 m × 0.32 mm × 0.25-μm film thickness). The oven temperature program was held at 50 °C for 3 min then programmed at 3.8 °C min −1 to 240 °C and finally held for 5 min. Carrier gas was helium (1 ml min −1 ), and sample (1 μl) was injected with a split ratio of 1:5 . The MS parameters were 250 °C at the interface, ionization energy 70 eV and scan speed 5 scans/ sec. The identities of VOCs were confirmed by comparing the diagnostic ions (NIST 2014 library) including GC retention times with those of relevant authentic synthetic compounds.
For quantification of VOCs through GC-FID, nonyl acetate was incorporated as internal standard (IS) at 1 μg μl −1 in each sample. Analysis of volatile blends were carried out by a Techcomp GC (Em Macau, Rua De Pequim, Nos. 202A-246, Centro Financeiro F7, Hong Kong) model 7900 attached with a HP-5 capillary column and a flame ionization detector, which was performed at the same temperature conditions mentioned for GC-MS analysis. Nitrogen was carrier gas (18 ml/min), and 1 μl sample injected with a split ratio of 1:5. Temperature of the injector port was 280 °C. In addition, straight-chain alkanes from C7 to C28 were injected in this equipment at the same temperature programming. Retention time was applied to compute the Kovats retention indices (for temperature-programming) according to van Den Dool and Kratz (1963).

Synthetic Compounds
HayeSep Q (80-100 mesh) was purchased from Sigma Aldrich, Germany. All synthetic compounds were purchased from Sigma Aldrich, Germany (Supplementary Table S2).

Behavioral Responses of S. obliqua Females Towards Individual Synthetic Compounds and Synthetic Blends in Olfactometer Bioassays
Responses of females towards individual synthetic compounds or complete synthetic blends resembling amounts emitted by UD, ID and MD plants of each green gram cultivar (μg/4 h) were dissolved in 100 μl CH 2 Cl 2 , and 25 μl of this odor (volatile blends emitted by plants during 1 h) were tested against the control solvent (25 μl) to observe the role of individual synthetic compounds in S. obliqua (Supplementary Table S3 and S4). In addition, responses of S. obliqua towards synthetic blends (consisting of those individual synthetic compounds to which S. obliqua showed responses or attraction) (25 μl volatile odor) resembling amounts emitted by UD, ID and MD plants of each green gram cultivar (volatile blends emitted by plants during 1 h) vs. the control solvent (25 μl) were conducted (Supplementary Table S3 and S5).
Responses of females towards certain synthetic blends resembling natural volatile blends of ID plants of each green gram cultivar were tested against certain synthetic blends resembling natural volatile blends of the other green gram cultivars (μg in 25 μl CH 2 Cl 2 ) (Supplementary Table S3).

Behavioral Responses of S. obliqua Females Towards Certain Synthetic Blends in Wind Tunnel Bioassays
The wind tunnel comprised of a plexiglass observation chamber (120 × 30 × 30 cm) with attachments for regulating the speed, filtering, channeling and expelling the air. Charcoalfiltered air was passed through the chamber at 0.20 m s −1 with air temperature 27 ± 1 °C and 70 ± 5% r. h. The sides and floor of the chamber were covered with black paper to insulate insects from external visual stimuli in the chamber. A plexiglass platform 15 cm above the floor of the chamber was used for release of the test insect, whereas a plexiglass platform 3 cm above the floor of the chamber was used for the attachment of filter papers. A piece (1 cm 2 ) of filter paper (Whatman no. 41) was fixed on a plexiglass stick (4 mm diameter × 2 cm high), and the stick was inserted into a rubber stopper (2.5 cm diameter × 1 cm high). Filter papers were placed centrally in inverted direction 30 cm from the upwind end on the floor of the observation chamber and the test insect was released 90 cm from the upwind end. All tests with females were carried out between 16.30 and 20.30 h. Each female was introduced into the wind tunnel, observed for 5 min for a given treatment and then recaptured. A female was tested once and it was discarded from bioassays. After testing every five insects, the chamber and release platform were cleaned with deionised water and the chamber was ventilated for 15 min before next bioassays. The target positions of filter papers with volatile cues were changed after each replicate to avoid positional bias in two-choice test. Each filter paper with volatile blends was replaced by a new one after testing five insects. A total of 50 insects (each replicate five insects) were tested in each experiment excluding the number of females that did not respond. The responses of the female S. obliqua were carried out by the following treatments: Condition 1: In no-choice assay, certain synthetic blend (25 μl of PDM blend 7 or PUSA blend 5 or SAM blend 5) resembling each green gram cultivar (equivalent to volatile blends released by each green gram cultivar during 1 h that elicited highest attraction to the insect in Y-tube olfactometer bioassays) was applied on the centre of each filter paper. Condition 2: In two-choice test, two filter papers were placed with a gap of 5 cm in between them at the centre of the observation chamber. The two-choice test was carried out in the following combinations: PDM blend 7 vs. PUSA blend 5 or SAM blend 5, and PUSA blend 5 vs. SAM blend 5.

Statistical Analyses
The data recorded as total amounts of VOCs, and amounts of individual VOCs from UD, ID and MD plants of three green gram cultivars were log (x + 1) transformed before performing statistical analyses. The distribution pattern of the data was normal as indicated by Kolmogorov-Smirnov tests, and Levene's test for homogeneity of variance demonstrated that the data were also homogeneous with respect to treatments (except those individual compounds which are not recorded in all treatments) using SPSS software (version 25.0). Discriminant function analysis (DA) was conducted using different treatments of three green gram cultivars -UD plants (UD PDM, UD PUSA and UD SAM), ID plants (ID PDM, ID PUSA and ID SAM) and MD plants (MD PDM, MD PUSA and MD SAM) as response variables against the VOCs recorded as explanatory variables to substantiate the differences among green gram cultivars in the context of VOC bouquets using XLSTAT software (Zar 1999). Data obtained on the behavioral responses of S. obliqua towards VOCs were analysed assuming the null hypothesis that the possibility of scores for the test sample(s) or the control solvent is equivalent to 50%, i.e., Chi-square analysis (H 0 : P = 50%) . Insects that remained in the stem of olfactometer were not used for the analysis. Data recorded on the responses of S. obliqua in wind tunnel bioassay experiments were subjected to t-test (Rojas and Wyatt 1999;Karmakar et al. 2016).
Females showed attraction towards volatile blends of UD, ID and MD plants of each green gram cultivar compared to the control solvent (Fig. 1). Females did not show preference towards volatile blends from UD plants of a particular green gram cultivar when tested against one another (Supplementary Table S7  vs. the control solvent (CH 2 Cl 2 ) in Y-tube olfactometer bioassays. Each test was performed with at least 60 responding females. Nonresponders were other than 60 responding females Identification and Quantification of VOCs The DA elucidated that first two factors F1 and F2 expressed more than 97% of the variation, and the biplot represented the spatial direction of response variables, UD PDM, UD PUSA, UD SAM, ID PDM, ID PUSA, ID SAM, MD PDM, MD PUSA and MD SAM against the eight extracted factors (Fig. 2). As indicated through the Fisher's distance, plants are significantly different in the context of VOC bouquet (Fig. 2, Supplementary Table S8). The eigen values and canonical correlations specify > 99% of variation being illustrated by three axes (Supplementary Table S8).
The total amounts of VOCs were higher in ID or MD plants of each green gram cultivar compared to UD plants of the same cultivar (Table 1). There were no significant differences in the total amounts of VOCs among three cultivars in the context of UD or ID or MD treatments.
The volatile profile of UD PDM, UD PUSA and UD SAM comprised of 24, 22 and 22 compounds with 79%, 86% and 86% of the compounds shared among UD plants of these cultivars, respectively (Table 1, Supplementary Fig. S3). In ID PDM, ID PUSA and ID SAM, 35, 33 and 30 compounds were detected, demonstrating 86%, 91% and 100% of the compounds shared among ID plants of these cultivars, respectively (Table 1, Supplementary Fig. S3). Thirty, 26 and 26 VOCs were identified in volatile blends of MD PDM, MD PUSA and MD SAM, respectively, indicating 77%, 88% and 88% of the compounds shared among MD plants of these cultivars, respectively (Table 1,

Behavioral Responses of Adult S. obliqua Females Towards Individual Synthetic Compounds and Synthetic Blends
Females showed attraction towards complete synthetic blends resembling natural volatile blends of UD, ID and MD plants of each green gram cultivar (PDM, PUSA and SAM) compared to the control solvent (Table 2, Supplementary Tables S9, S10 and S11).
Females showed responses towards seven individual synthetic compounds (benzaldehyde, myrcene, benzyl alcohol, acetophenone, linalool, decanal and geraniol) resembling amounts present in volatile blends of UD PDM in comparison to the control solvent, and females were attracted towards a synthetic blend of these seven compounds compared to the control solvent (Supplementary Table S9). Females demonstrated responses towards five individual synthetic compounds (hexanal, benzaldehyde, myrcene, benzyl alcohol and acetophenone) resembling amounts present in volatile blends of UD PUSA in comparison to the control solvent, and females were attracted towards a synthetic blend of these five compounds in comparison to the control solvent (Supplementary Table S9). Females displayed responses towards four individual synthetic compounds (hexanal, benzaldehyde, myrcene and benzyl alcohol) resembling amounts present in volatile blends of UD SAM in comparison to the control solvent, and females showed attraction towards a synthetic blend of these four compounds compared to the control solvent (Supplementary Table S9).  Females showed responses towards 12 individual compounds [(Z)-3-hexenal, hexanal, 1-hexanol, benzaldehyde, myrcene, (Z)-3-hexenyl acetate, 2-octanol, benzyl alcohol, ocimene, acetophenone, decanal and geraniol] resembling amounts present in volatile blends of ID PDM in comparison to the control solvent, and females were attracted towards a synthetic blend of these 12 compounds (PDM blend 12) ( Table 2, Supplementary Table S10). Females were attracted towards seven individual compounds -(Z)-3-hexenal, 1-hexanol, benzaldehyde, (Z)-3-hexenyl acetate, 2-octanol, ocimene and acetophenone or a synthetic blend of these seven compounds (PDM blend 7) compared to the control solvent ( Table 2, Supplementary Table S10).
Females demonstrated responses towards eight individual compounds [(Z)-3-hexenal, hexanal, benzaldehyde, benzyl alcohol, ocimene, acetophenone, linalool and decanal] resembling amounts present in MD PUSA compared to the control solvent, while a synthetic blend of these eight compounds attracted the female moths (Supplementary Table S11). The insect showed attraction towards four individual compounds -hexanal, benzaldehyde, linalool and Within the rows means followed by different letters are significantly different (P < 0.05) by Tukey test decanal or a synthetic blend of these four compounds compared to the control solvent (Supplementary Table S11).
Females exhibited responses towards seven individual compounds [(Z)-3-hexenal, hexanal, (Z)-3-hexen-1-ol, benzyl alcohol, acetophenone, linalool and decanal] resembling amounts present in MD SAM compared to the control solvent, while a synthetic blend of these seven compounds elicited attraction of the females (Supplementary Table S11). The insects showed attraction towards two individual compounds -(Z)-3-hexen-1-ol and decanal or a synthetic blend of these two compounds compared to the control solvent (Supplementary Table S11).
Females did not differentiate between natural volatile blends released by ID PDM plants and PDM blend 12 or PDM blend 7 (Table 3, Supplementary Table S12). Females did not discriminate between volatile blends emitted by ID PUSA plants and PUSA blend 10 or PUSA blend 5 (Table 3,  Supplementary Table S12). They also did not discriminate between volatile blends emitted by ID SAM plants and SAM blend 10 or SAM blend 5 (Table 3, Supplementary  Table S12).
Females did not distinguish between PDM blend 7 and PUSA blend 5 or SAM blend 5 (Fig. 3). Females also did not distinguish between PUSA blend 5 and SAM blend 5 (Fig. 3).

Discussion
In natural conditions, plants produce volatile blends which consists of monoterpenes, sesquiterpenes, aldehydes and green leaf volatiles (GLVs), whose chemical composition can strongly depend on cultivars or varieties of an individual plant species (Darshanee et al. 2017;McDaniel et al. 2016).
In this study, acetophenone and linalool were present in volatile blends of UD PDM and UD PUSA, whereas decanal and methyl jasmonate were present in UD PDM and UD SAM. Geraniol was present in volatile blends of UD PDM, whereas nerolidol was detected in volatile blends of UD PUSA and UD SAM. Therefore, our findings support the hypothesis that the VOC profiles differ within cultivars of a same plant species (Avellaneda et al. 2021;McDaniel et al. 2016;Mitra et al. 2021). The insect showed responses to some compounds resembling amounts present in undamaged plants (benzaldehyde, myrcene and benzyl alcohol -common among three cultivars, acetophenone -common in PDM and PUSA, hexanal -common in PUSA and SAM, and linalool, decanal and geraniol -specific in PDM), suggesting the attractions of insects are dependent on the key compounds as well as their amounts present in volatile blends Mitra et al. 2021).
The composition and emission of VOCs from insectdamaged plants are influenced by the nature of insect feeding damage (Lin et al. 2016;Rodriguez-Saona et al. 2009). Among insect-damaged treatments of three green gram cultivars, hexanal, limonene oxide and 1-nonanol were detected in volatile blends of ID PDM and ID PUSA plants, whereas 1-hexanol and 2-octanol were specific in volatile blends of ID PDM plants, suggesting qualitative variations among VOCs in three green gram cultivars are due to nature of feeding damage by the larvae of S. obliqua. Eight compounds [(Z)-3-hexenal, (Z)-3-hexen-1-ol, 1-heptanol, (Z)-3-hexenyl acetate, ocimene, 1,3-diethyl benzene, farnesene and 1-heptadecanol] were specific in volatile blends of insect-damaged plants of three green gram cultivars compared to volatile blends of undamaged plants of these green gram cultivars, supporting the hypothesis that insect feeding causes qualitative and quantitative changes in VOCs among volatile blends of insect-damaged plants compared to volatile blends of undamaged plants Koner et al. 2022;Mitra et al. 2020). However, four herbivore-induced plant volatiles (HIPVs) -1-heptanol, 1,3-diethyl benzene, farnesene and 1-heptadecanol present in volatile blends of these green gram cultivars did not elicit response in the females of S. 39 μg (Z)-3-hexenal, 6.29 μg benzaldehyde, 8.74 μg (Z)-3-hexenyl acetate, 6.17 μg ocimene and 3.69 μg acetophenone] were tested against each other in Y-tube olfactometer bioassays. Each test was performed with at least 60 responding females. Non-responders were other than 60 responding females obliqua. Two HIPVs (1-hexanol and 2-octanol) specific in volatile blends of ID PDM compared to volatile blends of UD PDM elicited attraction of the insect pest. The above observations suggest the hypothesis that feeding by an insect herbivore on different cultivars of a plant species influences production of different VOCs (Mitra et al. 2021).
Among mechanical damage treatments of three green gram cultivars, (Z)-3-hexen-1-ol and 1-nonanol were detected in volatile blends of MD PDM and MD SAM plants, whereas benzaldehyde and ocimene were detected in volatile blends of MD PDM and MD PUSA plants. In addition, 1-hexanol, 2-octanol and limonene oxide were specific in volatile blends of MD PDM plants. These observations suggest that there were qualitative variations in VOCs among mechanically-damaged plants of three green gram cultivars. Here, nonane and myrcene were detected in volatile blends of UD and ID plants of three green gram cultivars but not in volatile blends of MD plants of these cultivars.
Two compounds [(Z)-3-hexenyl acetate and geraniol] were recorded in volatile blends of ID plants of three green gram cultivars but not in volatile blends of MD plants of these cultivars. There were also qualitative variations in individual VOCs emitted by MD plants of three green gram cultivars compared to VOCs present in ID plants of these cultivars such as hexanal was not detected in volatile blends of ID SAM plants. Five compounds (1-hexanol, benzaldehyde, 2-octanol, ocimene and acetophenone) were responsible for attraction of S. obliqua towards volatile blends of MD PDM, whereas four compounds (hexanal, benzaldehyde, linalool and decanal) were responsible for attraction of the insect towards volatile blends of MD PUSA. Two compounds [(Z)-3-hexen-1-ol and decanal] caused attraction of female moths towards volatile blends of MD SAM. Therefore, our findings support the hypothesis that HIPVs differ qualitatively as well as quantitatively from the volatile blends of mechanicallydamaged plants Koner et al. 2022;Malik et al. 2016;Mitra et al. 2017;Piesik et al. 2010). It is in agreement with the idea that the composition of VOCs differs between plants after insect-damage and mechanicaldamage due to the metabolic effects of oral discharges by the insect herbivore on the insect-damaged plants (Holopainen and Gershenzon 2010;Portillo-Estrada et al. 2021).
HIPVs attract herbivore enemies of the host plant (McCormick et al. 2012;Qiao et al. 2018;Rodriguez-Saona et al. 2020). In the current study, (Z)-3-hexenal, (Z)-3-hexenyl acetate and ocimene, which were specific in volatile blends of ID plants of three green gram cultivars compared to volatile blends of UD plants of these cultivars, attracted females of S. obliqua. 1-Hexanol and 2-octanol specific in volatile blends of ID PDM compared to other two ID green gram cultivars induced attraction of S. obliqua. Except for UD SAM, where acetophenone was not detected, increased emissions of benzaldehyde and acetophenone from ID plants compared to UD plants caused attraction of S. obliqua towards volatile blends. Therefore, these findings suggest that HIPVs not only assist predators and parasitoids to locate the insect-herbivore, but also assist conspecific adults to further attract the host plant (El-Sayed et al. 2016;Guo and Wang 2019). Several studies demonstrated that caterpillar-induced plant volatiles attract conspecificadults in nature such as feeding by the larvae of the light brown apple moth, Epiphyas postvittana (Walker) on apple seedlings elicited more attraction of the adult moth on the insect-damaged apple seedlings compared to undamaged plants (El-Sayed et al. 2016), and attack by the larvae of the cabbage moth, Mamestra brassicae (Linnaeus) on cabbage plants resulted in more attraction of the conspecific adult moths compared to undamaged plants (Rojas 1999). Caterpillars of Spodoptera littoralis (Boisduval) fed on cotton plants resulted in preference of conspecific female moths to oviposit on insect-damaged plants compared to undamaged

Data Availability
The authors confirm that the data supporting the findings of this study are available within the article and its supplementary material.
Code Availability Not applicable.

Conflicts of Interest/Competing Interests
The authors declare that they have no conflict of interest/competing interests.
Ethics Approval Not applicable.

Consent for Publication
All authors agreed to submission of final manuscript.