Athetis dissimilis strain was obtained from Shandong Agricultural University in 2017 and was reared in an artificial climate chamber with a temperature of 26°C ± 2°C, a photoperiod of 14 h light and 10 h dark, and a relative humidity of 80% ± 5%. Larvae were fed an artificial diet (including soybean flour, wheat germ flour, bran, yeast powder, vitamin C, sorbic acid, methylparaben, and agar). A 10% honey-water solution was used to supplement the nutrients for the adults. After the mature larvae pupated, distinguished the male from the female, and placed the male and female pupae respectively in the mating cages. Three-day-old unmated female and male adults were used for the EAG and fumigation experiments.

According to previous studies on the identification of volatile compounds in maize and wheat (Lan et al., 2022; D'Isita et al., 2024), 61 compounds were selected for the determination of the EAG reaction (Supplementary Table S1). Liquid paraffin was purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. All compounds to be analyzed were diluted with liquid paraffin to a concentration of 1 mol/L and stored at −20°C for future use.

The EAG reaction of A. dissimilis with volatile compounds in the host plants was assessed using the method described by Li et al., 2022. Adult antennae were excised from the base using ophthalmic tweezers and the tip of the antennal flagellate segments were removed with a sharp blade under an anatomical mirror. The two ends of the antennae were affixed respectively to positive and negative electrodes with conductive glue. The potential signals from the antennae were amplified by an Syntech IDA232 device (Synchronous Technology Co., LTD., Shenzhen, China) and connected to a host computer, where the amplitudes of the antennal potential response were recorded using EagPro V2.0.2 software. A strip of filter paper measuring 3 cm length and 0.5 cm in width was cut and positioned the inner wall of the 200 µL pipette tip. Subsequently, 20 µL of the prepared liquid was applied to the center of the concave strip. To prevent contamination of the odor tube by chemical compounds, the strip must be at least 0.5 cm away from the open of 200 µL pipette tip. The distance from the antenna to the opening of the odor stimulation tube was set to 1 cm. Both the continuous and pulsed stimulated gases for humidification were regulated using a Syntech CS-55 gas generator (Synchronous Technology Co., LTD., Shenzhen, China) with a continuous gas flow rate of 450 mL/min and a pulsed gas flow rate of 120 mL/min. Each compound was stimulated for 0.5 s, with a 60-s interval between stimuli. 6-7 compounds were detected by each antenna and presented in random order. Each adult was tested with only one antenna, and ten antennae were used for each compound. The measurements of liquid paraffin were compared before and after testing for each compound (EAG_CK1 and EAG_CK2). The relative value of the EAG reaction was calculated using the following formula: Relative value = absolute value of treated EAG reaction − EAG CK 1 + EAG CK 2 / 2

EAG responses concentration-dependent test was conducted using 17 volatile compounds that elicited heightened EAG response in male and female adults. Each compound was successively diluted with liquid paraffin at a 10-fold decreasing rate, resulting in six concentration gradients: 1, 10⁻¹, 10⁻², 10⁻³, and 10⁻⁴ mol/L. Different concentrations of the same compound were measured, from low to high concentrations. Liquid paraffin served as a control both before and after stimulation with each compound concentration. The concentration-response of a single antenna was measured for each compound and the concentration gradient of each compound was assessed using ten antennae.

Based on the results of the EAG experiment and the Y-tube test (Supplementary Figures S1, S2), two host plant volatiles benzyl acetate and β-caryophyllene, were selected for chemical fumigation tests to assess the expression of the AdisOBP genes. 30 male and 30 female adults who had not mated for 3 days after emergence were selected, starved for 24 h and respectively placed in plexiglass cages (35 × 35 × 35 cm). A culture dish was positioned at the center of each cage, with a rubber septum inside the dish, onto which active ingredients 10 mg plant volatiles were applied. Liquid paraffin fumigation and non-fumigation were used as a control. Each group was treated with 4 biological replicates. After 24 h of exposure to this environment, the 60 male and 60 female antennae were respectively removed, immediately placed in liquid nitrogen, and stored at −80°C until further use.

Total RNA was extracted using an RNAiso Plus kit (TaKaRa, Dalian, China). First-strand cDNA was synthesized using PrimeScript RT reagent and gDNA Eraser (TaKaRa, Dalian, China), according to the manufacturer’s instructions. Real-time quantitative PCR was conducted on a Bio-Rad instrument (Bio-Rad Laboratories, Inc., United States) using SYBR® Premix Ex Taq II (TaKaRa, Dalian, China). All AdisOBPs and qPCR primer sequences (Supplementary Table S2) were sourced from a previous study ^[19]. GAPDH served as an internal reference gene to account for variations between samples. Each amplification reaction was carried out in a 20 μL reaction mixture under the following conditions: denaturation at 95°C for 3 min, followed by 40 cycles of 95°C for 10 s and 58°C for 30 s. No template controls were included to detect potential contamination. Three biological replicates were analyzed, and the relative expressions of the AdisOBPs were quantified using the −2^−ΔΔCT method (Livak and Schmittgen, 2001). Expression was calculated relative to levels of AdisOBPs in the female and male antennae of non-fumigation, which were arbitrarily set to 1.

Blastp was used to search for protein spatial structure templates in the Protein Data Bank. The Swiss-Model online service (https://swissmodel.expasy.org) was used to enhance the simulation of the three-dimensional structure of the protein (Waterhouse et al., 2018). Molecular docking of AdisOBPs with odor compounds was conducted using the CB-Dock2 online platform (https://cadd.labshare.cn/cb-dock2/index.php) (Liu J.et al., 2022). LigPlot Software (v. 2.2.8) was used for the 2D interaction plotting (Laskowski and Swindells, 2011). Three-dimensional structure illustrations were created using PyMOL software (v. 2.5.0, open source).

SPSS software (version 23.0) was used for statistical analysis of the experiment data. One-way analysis of variance, followed by the Student-Newman-Keuls (S-N-K) multiple comparison, was used to assess the relative value of EAG responses to various plant volatiles, different dosage gradients of the same plant volatiles, and the relative expression levels of AdisOBPs after fumigation with different plant volatiles. An independent samples t-test (equal variance hypothesis) was conducted to analyze the significant differences in the EAG responses values between male and female insects exposed to the same plant volatiles.

Figure 1 illustrates the EAG responses of adult insects to 61 host plant volatiles. The relative EAG responses for cis-3-hexen-1-ol and trans-2-hexen-1-ol in male adults were the highest, measuring 1.30 ± 0.10 mV and 1.27 ± 0.18 mV, respectively, significantly surpassing those of other similar compounds. In contrast, the EAG responses of female adults to alcohols were relatively weak, with the strongest responses observed for cis-2-hexen-1-ol and trans-2-hexen-1-ol, both at 0.44 ± 0.04 mV. There were significant differences in the responses of male and female adults to the same alcohol substances, such as the EAG responses of male adults to trans-2-hexen-1-ol (t = 4.445, df = 8, p = 0.011), cis-3-hexen-1-ol (t = 8.880, df = 8, p = 0.001), octadecanol (t = 21.112, df = 8, p < 0.0001), geraniol (t = 5.516, df = 8, p = 0.005), 1-heptanol (t = 15.123, df = 8, p < 0.0001), hexyl alcohol (t = 6.914, df = 8, p = 0.002), 1-nonanol (t = 9.784, df = 8, p = 0.001), and trans-nerolidol (t = 2.814, df = 8, p = 0.048) were significantly stronger than that of female adults (Figure 1A). Male adults exhibited the strongest EAG responses to benzyl acetate, cis-3-hexenyl butyrate, and methyl benzoate, with values of 0.61 ± 0.02 mV, 0.58 ± 0.05 mV, and 0.48 ± 0.03 mV, respectively. Except for cis-3-hexenyl butyrate (t = 12.237, df = 8, p < 0.0001), benzyl acetate (t = 3.903, df = 8, p = 0.017), and methyl benzoate (t = 7.616, df = 8, p = 0.002), there were no significant differences in the EAG responses of male and female adults to the other six ester compounds (Figure 1B). The EAG response to trans-2-hexen-1-al (1.35 ± 0.02 mV) was the strongest among male adults, followed by (E, E)-2,4-hexadienal (1.01 ± 0.06 mV), octanal (0.69 ± 0.04 mV), and hexanal (0.67 ± 0.04 mV). Female adults exhibited a significantly higher EAG response to nonanal (t = −4.245, df = 8, p = 0.013) than male adults, whereas their responses to other volatiles were either significantly lower or not significantly different from those of male adults, such as the EAG responses of male adults to trans-2-hexenal (t = 45.152, df = 8, p < 0.0001), (E,E)-2,4-hexadienal (t = 11.499, df = 8, p < 0.0001), hexanal (t = 4.019, df = 8, p = 0.014), and octanal (t = 4.880, df = 8, p = 0.008) were significantly stronger than that of female adults (Figure 1C). The EAG response of female adults to undecane (0.24 ± 0.02 mV) (t = −10.372, df = 8, p < 0.0001) was the highest and was significantly greater than that of male adults. Except for undecane (t = −10.372, df = 8, p < 0.0001) and dodecane (t = −3.725, df = 8, p = 0.020), there were no significant difference in the EAG responses of male and female adults to other alkanes. (Figure 1D). The EAG value for limonene (0.44 ± 0.01 mV) was the highest among male adults, followed by β-caryophyllene (0.35 ± 0.05 mV) and styrene (0.34 ± 0.03 mV). The EAG responses of male adults to limonene (t = 9.610, df = 8, p = 0.001), styrene (t = 7.255, df = 8, p = 0.002), β-caryophyllene (t = 5.755, df = 8, p = 0.005), and β-pinene (t = 5.563, df = 8, p = 0.005) were significantly stronger than that of female adults (Figure 1E). The EAG response of male adults to acetylacetone (0.72 ± 0.07 mV) was the highest, followed by 2-heptanone (0.50 ± 0.04 mV) and 2-nonanone (0.44 ± 0.02 mV). The EAG responses of male adults to all measured ketone compounds were significantly greater than those of female adults, such as the EAG responses of male adults to acetylacetone (t = 9.333, df = 8, p = 0.001), 2-heptanone (t = 5.361, df = 8, p = 0.006), β-lonone (t = 4.520, df = 8, p = 0.011), cis-jasmone (t = 6.561, df = 8, p = 0.003), camphor (t = 6.058, df = 8, p = 0.004), and 2-nonanone (t = 4.329, df = 8, p = 0.012) were significantly stronger than that of female adults (Figure 1F). The EAG response of male adults to benzothiazole (0.64 ± 0.01 mV) (t = 25.604, df = 8, p < 0.0001) was the highest and significantly exceeded that of female adults (Figure 1G). Secondly, there was indole (0.178 ± 0.005 mV), and the response of male adults to it (t = 10.366, df = 8, p < 0.0001) is also significantly stronger than that of female adults. Overall, the EAG response of male adults demonstrated greater sensitivity to odors than female adults across all tested compounds in A. dissimilis.

From the 61 host plant volatilities, 17 compounds were selected to elicit strong EAG responses in both female and male adults, and the EAG responses of the adults to varying concentrations of these compounds were measured. As illustrated in Figure 2, increasing the compound concentration from 10⁻⁴ to 10⁻³ mol/L did not result in a significant change in the relative EAG response values of female and male adults to the various compounds. At a concentration of 10⁻² mol/L, the EAG response of male adults to cis-3-hexenyl butyrate, hexyl alcohol, 2-nonanone, and β-caryophyllene increased significantly (Figures 2B,E,I,K). In contrast, female adults exhibited significant increases in EAG responses to cis-3-hexenyl butyrate, cis-2-hexen-1-ol, 2-heptanone, trans-2-hexen-1-al, octanal, nonanal and hexanal (Figures 2B,F,H,M,N,P,Q). The EAG response values for male adults to cis-3-hexenyl butyrate, cis-jasmone, β-caryophyllene and benzaldehyde (Figures 2B,J,K,Q), and for female adults to 1-nonanol and trans-2-hexen-1-al reached their maximum values at a concentration of 10⁻¹ mol/L (Figures 2G,M). When the concentration was increased to 1 mol/L, the EAG response of both male and female adults to most plant volatiles peaked. Such as male adults responded to benzyl acetate, geraniol, 1-hydroxyheptane, hexyl alcohol, 1-nonanol, 2-heptanone, 2-nonanone, undecane, trans-2-hexen-1-al, octanal, nonanal, and hexanal (Figures 2A,C–E,G–I,L–N,P,Q), whereas female adults responded to cis-3-hexenyl butyrate, geraniol, 1-hydroxyheptane, hexyl alcohol, cis-2-hexen-1-ol, 2-heptanone, 2-nonanone, cis-jasmone, undecane, octanal, nonanal and hexanal (Figures 2B–F,H–J,L,N,P,Q).

The expression levels of AdisOBP11 were the highest in the female antennae, followed by AdisOBP24, AdisOBP9, AdisOBP21, and AdisOBP4 (Figure 3). However, among the male antennae, AdisOBP21 and AdisOBP24 exhibited the highest gene expression, followed by AdisOBP9, AdisOBP11, AdisOBP7, AdisOBP4, AdisOBP50, AdisOBP20, AdisOBP13, and AdisOBP10 in non-fumigation test (Figure 4). After fumigation, the expression of AdisOBPs in both female and male antennae showed varying degrees of increase compared with non-fumigation. The expression levels of AdisOBP9 and AdisOBP11 in the female antennae were significantly higher than those in the paraffin control after fumigation with β-caryophyllene. Conversely, the expression of AdisOBP11, AdisOBP21, and AdisOBP24 in the female antennae after fumigation with benzyl acetate was significantly lower than that in the paraffin control (Figure 3). After fumigation with by β-caryophyllene, the expression of AdisOBP9, AdisOBP11, AdisOBP24, and AdisOBP21 in the male antennae increased significantly compared with that in the paraffin control. After fumigation with benzyl acetate, the expression levels of AdisOBP24, AdisOBP21, AdisOBP9, AdisOBP11, AdisOBP7, AdisOBP4, and AdisOBP50 in the male antennae were significantly higher than those in the paraffin control (Figure 4). After odor fumigation, the number of genes upregulated by AdisOBP in male antennae was greater than that in female antennae.

Based on the results of the fumigation test, we selected AdisOBP9, AdisOBP11, AdisOBP21, and AdisOBP24, which exhibited significantly increased odor-induced expression, for homology modeling and molecular docking studies. Homologous modeling of AdisOBP9, AdisOBP11, AdisOBP21, and AdisOBP24 was respectively compared with the sequences of Helicoverpa armigera OBP3, Pieris rapae OBP10, Agrotis ypsilon OBP2, and A. lepigone OBP24. Sequence identities with the template were 76.19, 72.95, 72.80, and 100%, respectively. The stereochemical properties of the protein model were evaluated using the ramachandran plot. 93.10% of the amino acid residues in AdisOBP9 were located in the optimal region, 97.50% in AdisOBP11, 99.19% in AdisOBP21, and 99.20% in AdisOBP24 (Supplementary Figure S3). The model indicated that the four AdisOBP proteins share a similar structure with other insect OBPs and are characterized by six α-helical components (Figure 5). Odor ligands, including β-caryophyllene and benzyl acetate, were downloaded from the ZINC database, energy-minimized, and subsequently docked with AdisOBP9, AdisOBP11, AdisOBP21, and AdisOBP24. Molecular docking results suggest that potential hydrophobic and electrostatic interactions play significant roles in binding (Figure 6). The predicted free binding energies of AdisOBP9, AdisOBP11, AdisOBP21, and AdisOBP24 with β-caryophyllene were – 5.9, −7.6, −7.4, and −7.3 kJ/mol, respectively. The predicted free binding energies of the recombinant proteins AdisOBP9, AdisOBP11, AdisOBP21, and AdisOBP24 with benzyl acetate were – 4.3, −5.9, −5.7, and −5.7 kJ/mol, respectively (Supplementary Table S3). The Thr84 residue of AdisOBP11 was involved in the formation of hydrogen bonds with benzyl acetate. Hydrophobic interactions were observed between all compounds and AdisOBPs (Figure 6).