Synthetic analogues of the volatiles identified from P. vulgaris were purchased from Shanghai Macklin Biochemical Co., Ltd., and Aladdin Industrial Corporation, China. A three-arm olfactometer (Brand: YMM3-150), a GC-EAG measurement system, and a QP 2010 SE GC-MS system were purchased from Shanghai Yuming Instrument Co., Ltd., Syntech Limited, and Shimadzu Corporation, respectively. The adsorption column consisted of a glass tube (15 cm long and 0.7 cm in diameter) with an airflow outlet and inlet equipped with 200 mg SuperQ adsorbent, purchased from Altec, United States.

About 15 P. vulgaris pods were placed inside a glass jar (35 × 20 cm – height and diameter) provided with an airflow inlet and outlet at the top. An adsorption column was attached to the outlet exit, while an activated carbon filter (air flow rate: 600 L/min) was attached prior to the inlet. After collecting air volatiles for 8 h, the adsorption column was removed to be eluted with GC-MS grade hexane.

Two 4.5 cm long glass capillaries were filled with 0.9% sodium chloride saline and inserted with two 4.5 cm long clean silver wires as electrodes. One glass tube was used as the measuring pole and the other one served as the reference pole.

Antennae of adult bugs were excised at the base using a scalpel under a dissection microscope and chipped at their tip. One antenna was attached to the reference pole by the surface tension of the saline solution and connected to the measuring pole under a stereo microscope. When the baseline readings of the EAD remained stable and the GC oven was ready. 4 μl of plant volatile solution was injected into the GC port. Obtained data were recorded and analyzed with GC-EAD 2014 V1.2.5 software.

The procedure for collecting volatiles was as previously stated. Gas chromatograph settings included an Agilent DB-wax chromatographic column (30 m. 0.25 mm; 0.25 µm, Agilent Technologies). The injector temperature was set to 225°C, carrier gas was helium, and injections were made on splitless mode. The oven temperature was programmed to three steps: 1) 40°C hold for 5 min; 2) increase at the rate of 5°C/min to 110°C, hold for 1 min; 3) increased at the rate of 10°C/min to 240°C and hold for 10 min. The mass spectrometer ion source temperature was set to 300°C, and the scanning speed used at 2500, with a 0.3 s interval. Obtained mass spectra were compared with external synthetic standards using Labsolution software to confirm the chemical identity of the main volatiles recovered from P. vulgaris pods.

Volatiles eliciting significantly stronger EAG responses were selected for attractiveness field tests using synthetic analogues. The field tests were conducted from 28 August to 3 September 2020 in a cotton field measuring about 4,000 m² in Ezhou, Hubei, China. Respective synthetic compounds were diluted to 150 mg/ml with ddH₂O, and 1 ml solution was added to individual micro centrifuge tubes sealed with plastic film, minutely punctured to prevent quick evaporation. The tubes were taped to the middle of a yellow board (RAL 1016 Sulfur yellow according to the international standard colour metric scale). Glue was spread on the outer face of the yellow board to capture the bugs. Each compound was tested three times using different yellow boards. Three boards were presented parallel as controls containing no attractants, using ddH₂O as a control. Treatment and control groups were randomly distributed, distanced from each other by at least 20 m, where each board was held suspended at about 20 cm above a cotton plant. After 7 days, the numbers and species of mirid bugs captured on the yellow boards were counted.

Volatiles presenting a positive attractiveness response in field experiments were further tested by a two-way competition experimental design in laboratory conditions. Prior to the tests, male and female adults of A. suturalis and A. lucorum were starved for 4 h and submitted to olfactory behavior assays using a three-arm olfactometer (Figure 1). The arms were ventilated for 5 min before the experiment, with the airflow rate set to 2 L/min. The synthetic volatiles were diluted to 1.5 mg/ml and 15 mg/ml in ddH₂O and added 30 μl anhydrous ethanol to assist in complete solubilization. One of the olfactometer arms was blocked with absorbent cotton, and the other two were connected to glass bottles containing either 50 μl volatile solution or ddH₂O with 30 μl anhydrous ethanol as a control group. Ten adult bugs were placed in the response chamber from an opening at the top. The test was repeated nine times for each species, totaling 90 male and female adults assayed for each species. All behavioral response assays were done in the dark, being recorded for 30 min once insects were introduced. Any insects moving beyond 50% of an arm’s length were considered to have shown a positive attractiveness response to the tested volatile. In setting up each replicate, the olfactometer parts were washed with ethanol, and the direction of odor sources were randomized.

The selected response rate and selection coefficient were calculated as follows (Zhou et al., 2009): S e l e c t e d   r e s p o n s e   r a t e   = N o .   a d u l t s   s e l e c t e d   t r e a t m e n t   N o . a d u l t s   s e l e c t e d   t r e a t m e n t + N o . a d u l t s   s e l e c t e d   c o n t r o l   × 100 S e l e c t i o n   c o e f i c i a n t   = N u m b e r   o f   a d u l t s   s e l e c t e d   t r e a t m e n t − N u m b e r   o f   a d u l t s   s e l e c t e d   c o n t r o l N u m b e r   a d u l t s   s e l e c t e d   t r e a t m e n t + N u m b e r   o f   a d u l t s   s e l e c t e d   c o n t r o l

To test the attractiveness effects of tetradecane analogues, seven analogues, including aldehydes, aldehydes, alcohols and acids, were employed for two-way competition experiments of female adults as described above. The analogues were 7-tetradecene, 1-tetradecene, tetradecanal, 2-tetradecanol, 1-tetradecanol, 2-tetradecenoic acid and tetradecanoic acid.

Fruits of Vitis vinifera, Zea mays, Pyrus bretschneideri, Malus pumila and seedlings of Pisum sativum (100 g each) were placed in individual glass jars to collect volatiles, as previously described. Five different plant parts of cotton, including buds, bolls, seedling leaves, bolls leaves, and bud leaves, were obtained from Xinzhou in 2019 and stored in afrigerator at −80°C as ground powders. Fifteen grams of each different plant part sample were likewise used to collect volatiles.

Volatile compounds were identified by GCMS, but using a Rtx-5 MS chromatographic column (30 m × 0.25 mm x 0.25 µm), injection port temperature was 250°C, helium as the carrier gas, injections done in splitless mode. The oven temperature program was set to three steps: 1) 40°C for 1 min; 2) increasing at the rate of 4°C/min to 130°C hold for 5 min; 3) increasing at the rate of 10°C/min to 250°C, final hold for 5 min. The mass spectrometer ion source temperature was set to 250°C; the scanning speed was 2500 with a 0.3 s interval. Runs were repeated three times, and the relative content of tetradecane in each sample was based on obtained peak areas in respective chromatograms.

Further 15–20 g of Vitis vinifera, Zea mays fruits, and 5-7 cotton leaves were employed to test for attractiveness to mirid bugs, but these tests were designed as ‘strictly olfactory’ assays and normal assays using female adults. In the case of the ‘strictly olfactory’ assays, three different plants were presented as host options inside separate glass bottles connected to three arms of the olfactory response chamber and the odor of host plants was blown into the response chamber from glass bottles by filtered air so the mirid bugs could smell them but could not reach the host plant through the hose that connected the response chamber and the glass bottles. Hence, they were unable to access them visually or physically. Evaluation methods were as described above.

In the case of normal assays, the three samples of different plant parts were offered as host options directly inside the different arms so that mirid bugs could evaluate hosts by smell, touch and gustation cues because in this case there is no barrier between the host plants and pests. Twenty adult A. suturalis and A. lucorum were placed into the olfactory response chamber from an opening at the top centre in each replicate and left overnight between 17:00–8:00. Lights were switched off to maintain dark conditions. Each assay was repeated three times, where the positions of host options were changed and randomized. Response rates were calculated as described previously.

Plants presenting the highest and lowest tetradecane contents were selected for this experiment. The plant with the lowest tetradecane content was supplemented with added synthetic tetradecane to quantify any relative attractiveness changes using the three-arm olfactometer described above. In this experiment, one treatment group and two control groups were set up and 15–20 g of the host plants were used in each group. The treatment group contained host plants with the lowest tetradecane content, and the plants were artificially supplemented with 50 μl of tetradecane solution. Control group 1 included host plants containing the highest tetradecane contents, and Control group 2 included host plants containing the lowest tetradecane content but no artificial tetradecane supplement. Three different concentrations of tetradecane supplementation were tested: undiluted solution, 15 mg/ml and 1.5 mg/ml. Treatment and two control groups were thus allocated to the three glass bottles connected to the olfactometer to be assayed as described above in ‘strictly olfactory’ experiments.

The Least Significant Difference (LSD) test was employed to check for differences in the numbers of mirid bugs trapped by boards presented in the field, in the content of tetradecane in each examined plant bouquet, and in the relative attractiveness of different plants to hosts. Olfactory responses to volatiles and analogues were analyzed using the chi-square test. All analyses were conducted using SPSS software, while data of GC-EAD were analyzed with GC-EAD 2014 V1.2.5 software.

The results of GC-EAD indicated three compounds from P. vulgaris pods inducing antennal responses in males of A. suturalis, and females of A. suturalis and A. lucorum. Their respective retention times were 11.39 min, 17.70 min and 22.61 min (Figure 2), corresponding to 2-propyl-1-pentanol, tetradecane and dodecanal (Table 1), respectively. Interestingly, 2-propyl-1-pentanol did not induce any EAG response in A. lucorum males.

Tetradecane, 2-propyl-1-pentanol, and dodecanal were obtained as synthetic compounds for field tests as baits on yellow board traps. As shown in Figure 3, tetradecane yielded high attractiveness: within 7 days, the average amount of captured mirid bugs was 30.33 ± 2.19, significantly more than with other attractants (p < 0.01) and controls (p < 0.05). In fact, the number of captured mirid bugs trapped by 2-propyl-1-pentanol and dodecanal was significantly lower than in controls (p < 0.05).

In field experiments, tetradecane appeared as the most attractive volatile to bugs; nevertheless, many uncontrollable factors may have influenced these findings, necessitating confirmation in a laboratory setting. In two-choice olfactory tests, volatile concentrations of 1.5 mg/ml and 15 mg/ml were used. Results showed that tetradecane significantly increased female mirid bugs’ attractiveness, with selected response rates above 60%. Tetradecane was most effective in attracting females of both bug species at a concentration of 1.5 mg/ml. While tetradecane was less attractive to males, it showed no significant olfactory response at the concentration of 1.5 mg/ml (Table 2).

Seven synthetic derivatives from various chemical groups, including alcohols, aldehydes, acids and olefins, were chosen for the two-way competition experiment to assess the relative attractiveness of tetradecane analogues to A. lucorum and A. suturalis. As tetradecane proved comparatively less attractive to males, we tested the olfactory response of females only. Overall, tetradecane derivatives proved most attractive to A. lucorum than to A. suturalis, especially at the lowest concentrations. Four analogues showed stronger attractiveness effects on A. lucorum (p < 0.001) compared to controls at the concentration of 1.5 mg/ml. Among them, tetradecanal was the most attractive to A. lucorum, eliciting a higher selected response rate than tetradecane. Only tetradecanoic acid yielded a higher attraction (p < 0.001) of A. suturalis at the concentration of 15 mg/ml (Table 3).

GC-MS analysis showed tetradecane is present as a volatile in the other ten plant species commonly exploited by mirid bugs and their relative amounts are significantly different (p < 0.05). The concentration of tetradecane was the highest in V. vinifera grapes, reaching 1.07 ± 0.06%, followed by P. sativum seedlings and Z. mays fruits (p < 0.05). Other species had statistically similar, lower amounts, including five different plant parts of G. herbaceum (p < 0.01) (Figure 4).

Therefore, fruits of V. vinifera and Z. mays and seedling leaves of G. herbaceum were selected to test for attractiveness to A. suturalis and A. lucorum adults as representative of significantly different tetradecane concentrations. Since tetradecane proved most attractive to females, experiments were designed with females only. Selected response rates of mirid bugs to three host plants in a “strictly olfactory” experiment were correlated with the tetradecane concentration of their volatiles. The selected response rate of A. lucorum to V. vinifera was the highest and significantly higher than that of G. herbaceum (p < 0.01), reaching 44.44 ± 0.80%. While for A. suturalis, there were no significant differences between the selected response rates for V. vinifera and Z. mays, reaching 37.98 ± 2.00% and 38.04 ± 0.54%, respectively. The selected response rate of G. herbaceum was significantly lower than that of the other two host plants (p < 0.01), reaching 23.97 ± 1.60%.

Interestingly, the selected response rates in normal assays (i.e., where bugs can select hosts using multiplecues) were irrelevant with relative contents of tetradecane in the volatiles from three host plants: A. lucorum preferred Zea mays fruits, while A. suturalis preferred G. herbaceum leaves, with respective selected response rates of 56.33 ± 3.30% and 57.92 ± 5.61% (Figure 5).

As shown in Figure 6, supplementation with the undiluted solution of tetradecane significantly increased the olfactory selected response rates of both A. lucorum and A. suturalis to G. herbaceum leaves (46.41 ± 3.15% and 45.31 ± 3.86%, respectively). This response was significantly higher (p < 0.01) than the pertinent control group and more elevated than V. vinifera fruits. On the other hand, when tetradecane was added at concentrations of either 15 mg/ml or 1.5 mg/ml to G. herbaceum leaves, response rates did not change significantly (p > 0.05).