Insect collection and rearing
B. dorsalis larvae were collected from rot fruits in carambola (Averrhoa carambola) and mango orchards in Haidian Campus of Hainan University, Hainan Province. The larvae were group-cultured in a mesh cage (1.0 × 1.0 × 1.0 m) under laboratory conditions (27 ± 1 ℃, L:D = 16:8 h, RH = 60–75%). Adult flies were reared on a jelly-like food consisting of 90 g sugar, 30 g yeast, 15 g honey, 5 g agar, and 1000 mL water, while eggs were collected using an egg-collecting device made from a 10-mL centrifuge tube. The tube wall had about 50–60 holes (d = 1 mm) for ovipositor insertion, and 2–3 mL of fresh orange juice were added at the bottom to stimulate oviposition. Eggs laid in the tube were collected and cultured on a maize-based artificial diet containing 500 g of corn flour, 500 g of banana, 2 g of sodium benzoate, 100 g of yeast, 100 g of sucrose, 100 g of paper towel, 4 mL of hydrochloric acid, and 800 mL of water. The larvae hatched from eggs were consistently raised on this diet until they reached the pupal stage. Additionally, S. invicta workers were collected directly from carambola orchards and used in the experiments without undergoing any additional manipulation or treatment.
The repellent effects of S. invicta semiochemicals on the visiting behavior of B. dorsalis
Five different sources of semiochemical solutions were prepared: S. invicta poop: Fecal matter were collected from 20 confined worker ants after 12 hours of confinement in a 10 mL centrifuge tube; the collected fecal matter was then dissolved in 1 mL of sterile water. S. invicta saliva: obtained by cutting bite points from 100 ants stimulated with a wooden stick and dissolved in 1 mL sterile water. S. invicta footprints: Twenty starved worker ants were confined in 10 mL centrifuge tubes for 12 hours; before cleaning the tubes with water, any visible fecal matter was removed; the tubes were then cleaned using 1 mL of sterile water, and the resulting clean water was collected. S. invicta residues: Twenty worker ants were confined in a 10 mL centrifuge tube for 12 hours, the tubes were then cleaned using 1 mL of sterile water, and the resulting clean water was collected. Control: 1 mL sterile water without any S. invicta semiochemicals.
A mesh cage (20 × 20 × 20 cm) was utilized and a cup (diameter = 2.5 cm, height = 1.5 cm) was placed inside. A total of 1 mL fresh orange juice, mixed with 200 uL of one of the aforementioned solutions, was added to the cup. One B. dorsalis, starved for 12 hours, was introduced into the cage and allowed to remain for 5 minutes. The observation period ended when the tested B. dorsalis either visited the orange juice or did not visit within the allotted 5-minute period, at which point it was removed and replaced with a new one until 10 B. dorsalis had been tested. Three growth stages were used, namely 1–5 days-old, 6–8 days-old, and 10–15 days-old (oviposited) females. Each semiochemical treatment or growth stage was repeated 20 times, and the number of visited B. dorsalis was recorded.
The repellent effects of S. invicta semiochemicals on B. dorsalis oviposition behavior
An egg collection device (as described previously) containing 2 mL fresh orange juice was utilized. Ant poop, saliva, footprints, and residual materials were prepared as outlined previously. Using a brush, 500 uL of each solution was evenly spread on the internal wall of the egg collection device. The device was then placed in a mesh cage (20 × 20 × 20 cm) and 10 female B. dorsalis (10–15 days old) were introduced into the cage for egg laying. The experiment was conducted between 2:00 pm and 6:30 pm, during which time eggs in the device were collected and counted. Each treatment was performed with 20 replicates.
Chemical composition of S. invicta footprints
The previous experiments revealed that only S. invicta footprints had a significant repellent effect on the feeding and oviposition behavior of female B. dorsalis. Therefore, we conducted a GC-MS analysis to investigate the chemical composition of S. invicta footprints. One milliliter of S. invicta footprints was obtained as previously described, but hexane was used as the solvent instead of water. The resulting solution was concentrated into 200 uL under a gentle nitrogen stream and stored at − 20°C. For the control treatment, hexane was used to wash a clear centrifuge tube, and the washing water was collected and concentrated into 200 uL. Each treatment had 3 replicates.
Two hundred microliters of each sample were added to a 20 mL headspace bottle, and 10 µL of 2-Octanol (10 mg/L stock in dH2O) was added as an internal standard. All samples were analyzed using GC-MS with a SPME cycle of PAL rail system. The incubation temperature was 60°C, the preheat time was 15 min, the incubation time was 30 min, and the desorption time was 4 min. The analysis utilized an Agilent 7890 gas chromatograph system coupled with a 5977B mass spectrometer. The system used a DB-Wax, injected in Splitless Mode, and helium as the carrier gas. The front inlet purge flow was 3 mL min− 1, and the gas flow rate through the column was 1 mL min− 1. The initial temperature was kept at 40°C for 4 min, and then raised to 245°C at a rate of 5°C min− 1, and kept for 5 min. The injection, transfer line, ion source, and quad temperatures were 250, 250, 230, and 150°C, respectively. The energy was − 70 eV in electron impact mode, and the mass spectrometry data were acquired in scan mode with the m/z range of 20–400, solvent delay of 0 min. Chroma TOF 4.3X software of LECO Corporation and Nist database were used for raw peaks exacting, the data baselines filtering and calibration of the baseline, peak alignment, deconvolution analysis, peak identification, integration, and spectrum match of the peak area. The target compounds were identified by comparing the GC and fragmentation patterns between the footprints and control sample.
Biological assessment of S. invicta footprint compounds on the behavior of B. dorsalis
We identified nine compounds, 7 of which were purchased from Hainan Hifly Industrial Co. Ltd, including undecane, d-limonene, dodecane, acetic acid, dodecanoic acid, tetradecanoic acid, and hexadecanoic acid. Because 2,6,10-trimethyltridecane and 4,6-dimethyl-dodecane are difficult to purchase, so these chemicals were not included in our experiments. Each compound was individually mixed with sterile water to form a single emulsion (undecane, d-limonene, dodecane, dodecanoic acid, tetradecanoic acid, or hexadecanoic acid emulsion) or solution (acetic acid solution), with the concentration based on its relative abundance in the footprint extracts shown in Table 1. Additionally, the 7 compounds were mixed together to create a mixture emulsion with the concentration indicated in Table 1. In the feeding behavior assay, 1 mL of fresh orange juice was mixed with 200 uL of each emulsion/solution in a cup, and the number of B. dorsalis visits was counted using the methods described earlier. In the oviposition behavior assay, the emulsion/solution was applied to the internal wall of a tube as described above, and the number of eggs laid by B. dorsalis in each device was counted. Fifteen replicates were conducted for both feeding and oviposition behavior assays.
Table 1
Identification of the chemical compositions present in S. invicta footprints using GC-MS analysis
Peak number (see Fig. 3) | Name | Relative abundance (%) |
1 | Undecane | 4.87 |
2 | D-Limonene | 4.18 |
3 | Dodecane | 2.24 |
4 | 2,6,10-Trimethyl-Tridecane | 0.66 |
5 | 4,6-dimethyl-Dodecane | 1.54 |
6 | Acetic acid | 4.69 |
7 | Dodecanoic acid | 11.74 |
8 | Tetradecanoic acid | 1.24 |
9 | Hexadecanoic acid | 0.79 |
Field evaluation of S. invicta footprint compounds for B. dorsalis control
A mixture emulsion with seven compounds (see above) was foliage sprayed at three fruit orchards: mango (18°27'19.8"N, 109°42'30.5"E), guava (19°46'11.2"N, 110°37'35.7"E), and wax apple (19°54'52.1"N, 110°33'58.5"E), located in Baoting, Haikou and Qionghai city, Hainan province. These orchards had abstained from pesticide use for roughly 45 days prior to these experiments. The experiments were conducted between April 12th to May 20th, 2023 in mango and wax apple orchards, and between May 12th to June 2nd, 2023 in guava orchards. Each orchard area was over 2 acres, containing more than 200 trees. Before the fruits matured, the B. dorsalis population density was monitored using methyl eugenol traps, ensuring that the population was abundant during the experiment period. Each orchard was divided into 12 plots, with an area of about 200 m2 (roughly 20–25 trees per plot), and a distance of more than 10 m between plots.
Each fruit species was treated with either a footprint emulsion spray or a water spray (as a control) for 2 and 7 days. The five-point sampling method was used in each plot, with each sample point comprising 1–2 trees and 50–150 fruits. To prevent damage from B. dorsalis, the fruits were covered with paper bags before foliage spraying.
For each fruit species, there were 12 plots in total, with 6 plots exposed to the footprint emulsion spray and the other 6 plots treated with water spray. Fruits from 3 treated plots and 3 control plots were examined for damage 2 days after spraying. Fruits in the remaining 3 plots (3 for treated plots and 3 for control plots) were exposed to the treatments for an additional 5 days. Finally, all the fruits were checked for damage rate between 5–7 days after being collected.
Data analysis
A generalized linear model (GLM) with a Poisson distribution was used to compare the number of visits by B. dorsalis, considering S. invicta semiochemicals and growth stages (B. dorsalis) as predictors. Kruskal-Wallis tests were used to analyze the number of eggs laid when B. dorsalis was exposed to S. invicta semiochemicals. The chemical composition of S. invicta footprints and control samples was compared using PERMANOVA. The impact of S. invicta footprint compounds on the number of visits and eggs laid by B. dorsalis was assessed with Kruskal-Wallis tests. The fruit damage rate was analyzed using a binomial distributed GLM, taking into account treatments (spraying of footprint compounds or water), host types (mango, guava, and wax apple), and time intervals (2 and 7 days) as predictors. Multiple comparisons were conducted using Tukey’s tests. The data analyses were performed using R 4.2.2 software (R Core Team 2022).