Control of Bemisia tabaci in Greenhouse Tomato using Systemic Insecticide-treated Trap Plant

Trap plant selection

Trap plant preference tests were conducted as follows: The 8 selected solanaceous plants were divided into two groups of four plants each (4 choice test). Tobacco, black nightshade, petunia, and Goji berry constituted group 1 while chili pepper, eggplant, capsicum, and jimsonweed were tested in group 2. The leaves of each plant (from one month old plants) were wrapped with wet cotton and the four plants of the group were placed in an acryl round (0.22 m diameter × 0.17 m) container at a spacing of 10 cm apart (n=10) (Fig. 1A). The container was self-designed and built in our laboratory. Fifty insects were then released into each container (n=10). The number of whiteflies settled on each leaf was counted after 48 hrs of releasing for comparing host preference among 4 plants. The final choice test was conducted with the most preferred plant along with tomato following the same procedure. Tobacco (the most preferred trap plant) and tomato plants of similar size were placed in an acryl round container and 50 whiteflies were released into the container.

Fig. 1.

The design of four choice test. A. Cotton-wrapped leaves were placed in an acryl round container; B. Whiteflies on chili pepper leaf.

Selection of systemic insecticides and mortality of B. tabaci in tobacco trap plant

Six systemic insecticides selected based on the Xerces Society’s database (2021) were tested to determine the appropriate insecticide for use with a trap plant (Table 1). Each insecticide (100 mg/kg) was prepared by diluting it in distilled water. The diluted insecticides were subsequently added to 0.2 g of instant bed soil (Seoul Bio Inc., Korea) in each pot (0.08 m × 0.1 m) of the targeted plant. Fifty adults were released into the cage and the dead and live insects were counted at 3 d after the pesticide treatment. Mortality rates were compared among the 6 tested insecticides, and the best performing insecticide was selected for further experimentation. Finally, the selected insecticide was tested on both tomato and tobacco plant following the same procedure as described above. Fifty adults were released in cage and monitored.

Table 1.

Six systemic insecticides used in this study

Dinotefuran analysis in tobacco leaves

Analytical grade dinotefuran (98.5% purity) was provided by Dr. Ehrenstöfer (London, UK). Extraction packets containing 4 g MgSO₄, 1 g sodium chloride, 1 g sodium citrate, 0.5 g disodium citrate sesquihydrate and dispersive solid phase extraction (d-SPE) containing (150 mg MgSO₄/25 mg primary secondary amine (PSA) and (150 mg MgSO₄/25 mg PSA/25 mg graphitized carbon black (GCB)) were obtained from Agilent (Agilent Technologies, USA).

A liquid chromatograph (Shimazu LC) equipped with tandem mass spectrometry (MS/MS) (Shimazu-8060) was operated for the analysis. The analyte was separated on a non-polar column (Agilent Eclipse Plus C18 2.1 mm i.d. × 100 mm L, 1.8 μm particle size, Waters, MA, USA) using mobile phase A (0.1% formic acid, 5mM ammonium formate in DW) and mobile phase B (0.1% formic acid, 5 mM ammonium formate in MeOH). Injection volume was 5 μL and flow rate was 0.2 mL. In the multiple reaction monitoring mode, qualitative ion and confirmation ion (m/z) were 203.15 > 114.15 and 203.15 > 87.00 in positive electrospray ionization mode, respectively. The MS conditions were as follows: ion spray voltage of 5,500 V, nebulizer gas of 50 psi, curtain gas of 25 psi, drying gas of 50 psi, collision gas of 10 psi, and drying gas temperature of 500^oC.

Two or three tobacco leaves from upper, middle, and lower parts of each plant were collected (n=3) 2, 4, 6, 9, 12 hrs, 1, 2, 3, 5, 7, 9, 14, 21, 28, 35, and 42 d after treatment. Each sample was pulverized with dry ice using a blender. The target compound in tobacco leaf samples was extracted using the modified QuEChERS (quick, easy, cheap, effective, rugged, and safe) method (Chamkasem et al., 2013). Final sample preparation for the instrumental analysis is displayed in Fig. 2. At two fortification levels of 0.01 and 0.1 mg/kg, recovery rates (%) with relative standard deviation (RSD, %) were achieved by plotting matrix-matched standards for the analysis of LC-MS/MS. For quantitative analysis, matrix-matched standards were prepared by mixing solvent standard solution with blank matrix extracts (1:9) at the levels of 1, 2.5, 5, 10, 25, 50, and 100 ng/mL. Limit of quantitation (LOQ) and linearity were evaluated by plotting each matrix-matched calibration curve. LOQ was < 0.01 mg/kg and the correlation coefficients (r²) of matrix-matched standards were > 0.99.

Fig. 2.

Analytical method flow chart. MeCN, acetonitrile; dSPE, dispersive solid phase extraction; LC-MS/MS, liquid tandem mass spectrometry.

Insecticide treatment and mortality test

Tobacco seeds were planted and grown for about 45 d in greenhouse conditions for systemic insecticide treatment. Dinotefuran WG (20%) (200 mg/kg) was diluted in distilled water and was subsequently added to the bed soil (1.4 kg) in each pot (0.25 m × 0.25 m) of tobacco plant. Two or three leaves collected from upper, middle, and lower parts of each plant were collected at 2, 4, 6, 9, and 12 hrs and 1, 2, 3, 6, 9, and 35 d after the treatment (n=3). Each leaf was cut into a disk of about 3cm in diameter before being transferred into a petri dish. A portion of each trimmed leaf was kept in a deep freezer (-80^oC) for further analysis of dinotefuran residue. Thirty adult B. tabaci were introduced into the petri dish. The number of dead and live B. tabaci were checked at 3 d after inoculation (n=3).

Greenhouse experimentation

Tomato seedlings were planted at two lines in four separate greenhouses (Control, A, B, and C) located in the National Institute of Agricultural Sciences, Wanju, Jollabuk do on the 16^th of May in 2023 (Fig. 3). On 23^rd of May, 10 L of dinotefuran (200 mg/kg) was prepared and then, 1 L was added to soil (20 kg) of each tobacco plant. One week after, the same amount of dinotefuran was applied to each tobacco plant. No tobacco plant was placed in the control greenhouse (18 m × 8 m). In greenhouse A and B, two dinotefuran-treated potted tobacco plants were placed at 10 m distance between the two lines. In greenhouse C, 3 tobacco plants were instead used but the spacing distance was reduced to 5 m apart. A total of 100 whiteflies were released into each greenhouse and three density yellow traps (Greenagrotech Inc., Gyeongsan, Korea) were installed at 3m distance. The number of insects in each yellow trap was recorded on a weekly basis for 8 weeks from one week after the introduction of tobacco plants.

Fig. 3.

Field experimentation. Control greenhouse where no tobacco plants were placed. A & B. 2 tobacco plants placed at 10 m distance; C. 3 tobacco plants placed at 5 m distance in furrow between two tomato planted lines.

Statistical analysis

Variance analysis using Duncan’s Multiple Rage Test (DMRT) and Tukey’s test were performed for the determination of statistical differences among samples in the preference test and the mean mortalities of B. tabaci in systemic insecticide treatment using SPSS v.13 software (IBM Analytics, Armonk, NY).

Selection of a trap plant

Out of the 8 solanaceous plants, tobacco was the most attractive trap plant, and was selected for further experimentation. Figure 4 shows the preference test results. In the first group test, tobacco (36%) and black nightshade (35%) were preferred to petunia and Goji berry (Fig. A). Eggplant (41%) and belly pepper (29%) were the most preferred plants in the second group (Fig. 4B). Through the results of two preference tests, four candidate plants were selected. In the final test, tobacco (40%) and eggplant (32%) were the most preferred plants for B. tabaci (Fig. 4C). The DMRT test showed that they were not significantly different, which is consistent with the previous study showing that B. tabaci preferred eggplants, tobacco, and cucumber to cotton and tomato (Tian et al., 2020). However, tobacco was selected as a trap plant in this study because its leaves have a large surface area and therefore can attract and harbor more whiteflies, and the plant is easy to be managed as a trap. This has already been demonstrated in previous studies. For instance, eggplant was used as a trap plant due to its easy to manage attributes compared to cucumber even though eggplant (82.3%) and cucumber (82.5%) yielded similar attraction rates (Choi et al., 2016). It was also hypothesized that B. tabaci adults were more sensitive to visual cues than to odor and the pest preferred the eggplant and cucumber to horseweed. In the final test, the attraction rates of the tobacco plant to B. tabaci adults were 65%, compared to that of tomato (35%) (Fig. 4D), indicating that tobacco has high a potential as a trap plant to lure B. tabaci adults in tomato greenhouse cultivation.

Fig. 4.

Choice test results of B. tabaci. A. attraction rates (%) of tobacco, black nightshade, petunia, and Goji berry; B. attraction rates of eggplant, bell pepper, chili pepper, and jimsonweed; C. attraction rates (%) of the four candidate plants selected, D. attraction rates of tobacco and tomato. The different letters (lowercase) in the attraction rates indicate significant differences in the 95% DMRT test.

Selection of systemic insecticides and their efficacy against B. tabaci in tobacco trap plant

Dinotefuran was finally selected as the best compound for systemic treatment in the tobacco plant and the mortality test results are shown in Fig. 5. Out of six systemic insecticides, cyantraniliprole EC (96.7% ± 1.2%), dinotefuran WG (97.3% ± 4.6%), and sulfoxaflor SC (98.7%± 1.2%) showed high mortalities in our study. The DMRT tests showed that they are not significantly different. Insect mortalities recorded in treatments with emamectin benzoate, imidacloprid, and flonicamide were 28.0%± 6.9%, 52% ± 16.0%, and 67.3% ± 4.2%, respectively. Even though imidacloprid is known to be effective for control of B. tabaci, it showed a low mortality of 52% ± 16%, which is consistent with several other published literature (Kumar et al., 2016).

Fig. 5.

Mortality (%) of B. tabaci on tobacco plants treated with systemic insecticides. The different letters on bars indicate significant differences in the 95% DMRT test.

Additional experiment was conducted to check whether the three best performing insecticides (cyantraniliprole EC, dinotefuran WG, and sulfoxaflor SC) can effectively control B. tabaci in the presence of tomato plants (Fig. 6). Dinotefuran exhibited less or non-disruptive tendency on insect behavior and highest mortality rate of 66% was recorded in tobacco leaves treated with dinotefuran. Mortality rates in treatments with cyantraniliprole EC and sulfoxaflor SC were 58% and 48%, respectively. These results indicated that treatment of tobacco plants with these systemic insecticides have a limited potential to reduce the density of B. tabaci in the tobacco greenhouse cultivation.

Fig. 6.

B. tabaci distribution between tomato and insecticide-treated tobacco. Distribution rate (%) of B. tabaci found on tomato plants (yellow color) and the insecticide-treated tobacco plant (orange color) at 3 d after being released into the cage. Green color indicates the percentage of dead B. tabaci.

In the presence of tomato plants, the mortalities of B. tabaci were reduced compared with no tomato plants. These results may be due to the feeding behavior disorder or repelling effect of these insecticides. Previous studies have shown that cyantraniliprole has a negative effect (repellent effect) on the feeding behavior of B. tabaci (Kwon et al., 2008), and dinotefuran plays a repellent role against green peach aphid (Seo et al., 2009). In the case of dinotefuran, 67.3% (66.0% of mortality and 1.3% of live individuals) of B. tabaci preferred dinotefuran-treated tobacco plants, indicating that the insecticide did not greatly influence the preference of B. tabaci. Compared with the control (8%), dinotefuran-treated tobacco plants gave higher mortality of B. tabaci. Previous study also indicated that dinotefuran, as the third generation of neonicotinoids is one of the most powerful insecticides against B. tabaci (Qu et al., 2017). It has also been used as a systemic insecticide in eggplant trap-plants for B. tabaci (Choi et al., 2016).

Analytical method optimization and validation

This study applied the modified QuEChERS method to analyze dinotefuran in tobacco leaves because it is very simple but robust. Recovery rates from both clean-up methods were not significantly different (Table 2). At two spiking levels, the recoveries of Clean-up 1 ranged from 93.2% to 93.7% and the recoveries of Clean-up 2 ranged from 87.6% to 92.8%, demonstrating that the analytical method was effective. Relative standard deviation for both methods was less than 7.0%, indicating both methods are accurate. However, d-SPE containing GCB (Clean-up 1) with MgSO₄ and PSA was selected for real sample analysis in this study. GCB in the d-SPE clean-up procedure is known to remove high contents of chlorophyll derived from green vegetables (Kim et al., 2019; Lee et al., 2017). MgSO₄ and PSA are common sorbents for adsorbing strong acids or polyacidic materials mostly derived from sample matrices in the QuEChERS clean-up. However, it has been reported that GCB retains planar pesticides (Kim et al., 2019) but in this study, the recoveries indicated that no dinotefuran was retained by GCB in the clean-up procedure. It can be hypothesized that the use of GCB can reduce the chromatographic problems that interfere with sensitive analysis or escalate the maintenance cost of the instrument by removing the tobacco-derived chlorophyll.

Table 2.

Recovery rates obtained from two clean-up procedures

Dinotefuran concentrations in tobacco leaves and mortality of B. tabaci

The dinotefuran concentrations in tobacco leaf location (upper, middle, and lower) were obtained after its application to the soil in greenhouse conditions (Fig. 7). The results showed that the dinotefuran concentrations in upper leaves were generally higher than those in middle and lower leaves (Fig. 5), Within 12 hours, the dinotefuran concentration reached 4 mg/kg on the upper leaves. The peak concentration was recorded at 20 days after treatment. After 20 days, the insecticide concentration of upper side leaves slowly decreased but steadily increased in middle and lower leaves. Dinotefuran has relative stability in soil, and its half-life has been reported to be 50 to 100 days in soil (Liu et al., 2021; Morrissey et al., 2015). This data demonstrated that the insecticide quickly moves through the plants, and can persist in the leaves for more than 3 months. Therefore, dinotefuran-treated plants can potentially sustain a control effect on B. tabaci populations over an extended period.

Fig. 7.

The concentrations of dinotefuran in tobacco leaves by leaf location (lower leaves (a), middle (b), and upper leaves (c)).

The mortalities of B. tabaci recovered from tobacco leaves were 73-94% at 12 hrs and 72-100% at 1, 2, 3, 6, 9 and 35 d after treatment (Fig. 8). Slightly higher mortality was recorded in the upper leaves. Additionally, there was a steady increase in mortality rates over time. Mortality steadily increased from 12-20% at 2 hrs to 24 hrs after the introduction of dinotefuran-treated tobacco plants in greenhouse. This data indicated that B. tabaci can be controlled at one day after dinotefuran-treated tobacco plants are placed in the greenhouse.

Fig. 8.

The mortality of B. tabaci in tobacco leaves treated with dinotefuran (lower leaves (a), middle (b), and upper leaves (c)).

Greenhouse experimentation

The results (Fig. 9) showed that the insecticide-treated tobacco plants significantly controlled B. tabaci populations in the greenhouse. The number of B. tabaci in the control greenhouse skyrocketed eight weeks after the release of the initial B. tabaci population (Fig. 9). Compared with that of control greenhouse (100%), the population of B. tabaci in greenhouses A and B increased by 53% and 63% and the population in greenhouse C increased by 22%. The results in greenhouse C highlight the importance of selecting the appropriate spacing of trap plants to maximize efficiency in attracting and controlling the targeted pest (placed at 5 m distance). Correct spacing can be effective in controlling B. tabaci populations in greenhouse. The control effect of dinotefuran lasted for more than two months in this study. This research demonstrated that tobacco plants can play an effective role in controlling the rapid increase of B. tabaci in the tomato greenhouse condition.

Fig. 9.

The number of B. tabaci per trap. (A) two tobacco plants placed in the greenhouse located nearby control greenhouse, (B) two tobacco plants, and (C) three tobacco plants.

It also proved that the method using dinotefuran-treated tobacco plant can be a robust, reliable, and cost-effective strategy in reduction of the B. tabaci population in tomato greenhouse cultivation. High concentration of dinotefuran was shown to persist in the tobacco leaves for 35 d, and B. tabaci population was controlled for more than 2 months with dinotefuran treated tobacco plants in the tomato greenhouses. However, considering the possibility of insecticide resistance development in the targeted pest, it is advisable to develop a treatment regime including different systemic pesticides such as cyantraniliprole EC or sulfoxaflor SC, which are equally effective as demonstrated in the current study during the screening phase. Tobacco plants have many advantages such as easy management and can be grown for more than 3 months, and thus, new tobacco plants treated with different kinds of insecticides can easily be placed in the same greenhouse.

Disclosure of potential conflicts of interest

The authors declare that they have no conflict of interest.