The experimental procedure was performed according to the description of Xiang et al. (2019), with some modifications. The screening operation procedure is shown in Figure 1. According to the bioassay results of the F₀ generation, the 3rd-instar nymphs of susceptible L. striatellus were treated with the LC₅₀ of triflumezopyrim. Then, the surviving nymphs were collected, and the F₁ generation was obtained after spawning. The toxicity of triflumezopyrim was assessed in the F₁ generation, and then, the survived 3rd-instar F₁ nymphs were screened with the newer LC₅₀ value. After five rounds of drug screening from the F₀ generation, survival of the F₄ generation was achieved. Then, 150 surviving nymphs were selected from the F₄ generation and transferred into flat-bottomed test tubes (diameter × height: 20.0 mm × 145.0 mm) containing fresh rice seedlings. The first 100 nymphs were numbered and used for formal experiments, and the last 50 nymphs were used as substitutes. After emergence, each pair of male and female adults was transferred one by one to a flat-bottomed test tube containing two fresh rice seedlings. If a pair did not mate, they were provided with substitute partners. Third-instar F₀ nymphs were used for the control treatment. The rice seedlings were replaced every day until the adults died, and the fecundity and longevity of the pair were recorded. The life tables of the F₄ generation were made at the same time as those of the F₀ generation.

Third-instar nymphs of the F₀ and F₅ generations were analyzed for the activity of three detoxifying metabolic enzymes. Carboxylesterase (CarE) activity was determined according to the method described by Asperen (1962). Twenty nymphs (3rd instar) were transferred into a centrifuge tube and stored in liquid nitrogen as quickly as possible, homogenized in 2 mL phosphate buffer (0.04 mol/L phosphate buffer, pH 7.0) ice-bath slurry, and centrifuged at 4°C and 10,000 × g for 15 min. Then, the supernatant was transferred into a clean Eppendorf tube as the crude enzyme solution. A mixture of 0.45 mL of phosphate buffer (0.04 mol/L, pH 7.0), 1.8 mL of 3 × 10^–4 mol/L α-NA solution (containing 3 × 10^–4 mol/L physostigmine), and 50 μL of diluted enzyme liquid was added to each tube. After thorough mixing, the reaction was performed for 15 min in a constant-temperature bath at 30°C, and then 0.9 mL of staining solution (0.2 g of fast blue-B salt in 20 mL of distilled water plus 50 mL of 5% sodium dodecyl sulfate) was added. The absorbance values were recorded at 600 nm after 5 min in a UV 2000 Spectrophotometer [Unic (Shang Hai) Instruments Incorporated, Shanghai, China].

Glutathione S-transferase (GST) activity was determined by using 1-chloro-2,4-dinitrobenzene (CDNB) as a substrate, according to the method of Wang et al. (2018), with minor revisions. Thirty nymphs were placed into a centrifuge tube and stored in liquid nitrogen as quickly as possible. Then, 2 mL of phosphate buffer (0.1 mol/L phosphate buffer containing 1.0 mmol/L EDTA, pH 6.5) was added, and the mixture was centrifuged at 4°C and 10,000 × g for 10 min. Then, the supernatant was used as the enzyme solution. The total volume of the reaction system was 2.7 mL, including 2,470 μL of phosphate buffer (0.1 mol/L, pH 6.5), 90 μL of CDNB (15 mmol/L), 50 μL of the enzyme source, and 90 μL of reduced GSH (30 mmol/L). After rapidly shaking the mixture, the absorbance values were monitored within 2 min with a spectrophotometer at 340 nm.

Cytochrome P450 (P450) activity was assayed using the method of Rose et al. (1995) with some modification. One hundred fifty (3rd-instar) nymphs were homogenized on ice with 2 mL of phosphoric acid buffer (0.1 mol/L, pH 7.6, containing 20% glycerol, 0.1 mmol/L EDTA, 0.1 mmol/L DTT, and 0.4 mmol/L phenylmethylsulfonyl fluoride). The homogenate was centrifuged at 4°C and 10,000 × g for 10 min to obtain the supernatant, which was stored at low temperature as the enzyme solution. The OD value was determined at 405 nm by adding 100 μL of 4-nitroanisole (2 × 10^–3 mol/L) and 90 mL of enzyme solution to the enzyme labeling plate, incubating for 3 min at 30°C, and then adding 10 mL of NADPH (9.6 × 10^–3 mol/L). The OD values were recorded after every 20 s for 2 min, and a standard curve of p-nitrophenol was prepared in advance. The specific activity of P450s was finally calculated as nanomoles of p-nitrophenol per minute per milligram of protein [nmol/(min mg pro)].

All treatments were performed with three samples (tubes) as biological repetitions, and each enzyme sample was replicated three times as technical repetitions. The total protein content of the enzyme solution was determined by the Bradford method using bovine albumin as a standard (Bradford, 1976). The activities of CarEs, GSTs, and P450s were analyzed using unpaired Student t-tests, and the significance level of the results was set at P < 0.05.

Females of the F₀ and F₅ generations were collected on the second day after emergence, and triplicates were performed for each treatment. Twenty insects of each sample type were frozen with liquid nitrogen and stored at −80°C. The Vg and VgR contents were detected using the double-antibody sandwich method according to the enzyme-linked immunosorbent assay kit of Alpha Enzyme Biotechnology Co., Ltd. (Shanghai, China). The contents of Vg and VgR were analyzed using unpaired Student t-tests, and the significance level of the results was set at P < 0.05.

Females of the F₀ and F₅ generations were collected on the second day after emergence. Total RNA was extracted by using TRIzol reagent (Shanghai Yubo Biological Technology Co. Ltd., Shanghai, China) following the manufacturer’s specifications, and cDNA was synthesized from the total RNA using TransScript cDNA Synthesis SuperMix (Novoprotein Scientific Inc., Shanghai, China) for qPCR. TransStart TopTaq (Novoprotein Scientific Inc.) was used to conduct the reverse transcriptase-quantitative polymerase chain reactions (RT-qPCRs). Primers were designed as shown in Table 1, with the EF-1Fq gene as an internal reference gene (Jia et al., 2015). The RT-qPCRs for each treatment were replicated three times. The relative expressions of EcR (Jia et al., 2015), Vg, and VgR were represented by relative quantification values calculated by the math formula method. The relative expression of the EcR, Vg, and VgR genes was analyzed using unpaired Student t-tests, and the significance level of the results was set at P < 0.05.

To understand the changes in the susceptibility of L. striatellus under the screening pressure of triflumezopyrim, we used the rice seedling dipping method to evaluate the toxicity of triflumezopyrim and found that the LC₅₀ values of triflumezopyrim on 3rd-instar nymphs of the F₀ and F₄ generations were 0.443 ± 0.171 μg/mL and 5.682 ± 1.498 μg/mL, respectively, with a resistance ratio of 12.84-fold (Table 2).

The fecundity, APOP, TPOP, and longevity of the F₅ generation are displayed in Table 5. The mean fecundity of the F₅ generation (141 eggs) was significantly lower than that of the F₀ generation (186 eggs) (P < 0.05), whereas the APOP, TPOP, and adult longevity were not significantly different between the two strains (P > 0.05). The proportions of males to females in the F₅ and F₀ generations were 1:0.85 and 0.89:1, respectively.

The sublethal effects of triflumezopyrim on population dynamics were calculated with a bootstrapping procedure based on a life table (Table 6). The intrinsic rate of increase (r), finite rate of increase (λ), and net reproductive rate in the F₅ generation (0.12, 1.13, and 57.87, respectively) were all lower than those in the F₀ generation (0.14, 1.15, and 87.69, respectively) (P < 0.05). However, there was no difference in the average generation time (T) between these two strains (P > 0.05).

The s_xj values in the F₀ and F₅ generations indicated that their growth curves had substantial overlap and complex growth relationships among individuals. The number of nymphs that could complete development in the two strains was not significantly different (83 and 77%, respectively). In addition, the specific age survival rate of male adults in the F₀ generation was higher than that of female adults, whereas a different pattern was observed in the F₅ generation, and the s_xj values of male adults were similar to those of female adults (Figure 2).

According to Figure 3, the e_xj values of the F₀ and F₅ generations generally decreased with age x and phase j. For the F₀ and F₅ generations, the highest e_xj values all appeared at the egg stage, at 41.41 and 38.95 days, respectively. Overall, the e_xj values of the F₅ generation were lower than those of the F₀ generation, whereas the e_xj values of the female adults were higher than those of the male adults. Among all the adults, the e_xj values of females in the F₀ generation were the highest.

The l_x results indicated that the l_x curves of the F₀ and F₅ generations generally first increased and then decreased. The highest peak fertility in the F₀ and F₅ generations was on the 28th and 27th days, with the highest average yield of 11.0 and 17.1 hatched eggs, respectively (P < 0.05). The female f_x curve of the F₅ generation suggested a lower and less variable value than that of the F₀ generation. Similarly, the curves of m_x and l_xm_x showed a declining trend in the F₅ generation (Figure 4).

The v_xj values in the F₅ generation were lower than those in the F₀ generation for the female adult stage and total v_xj. The v_xj values of all strains gradually increased from the egg stage, and that in the F₀ generation reached its maximum (124.65 eggs) on the 23rd day, especially for female adult reproduction, which showed a peak value of 86.11 eggs on the 27th day. However, the v_xj values in the F₅ generation reached their highest peaks (111.45 eggs) on the 25th day, and those of female adults reached their maximum (67.06 eggs) on the 27th day (Figure 5).