Transcriptomic Analysis Reveals the Detoxification Mechanism of Chilo suppressalis in Response to the Novel Pesticide Cyproflanilide

Chilo suppressalis is one of the most damaging rice pests in China’s rice-growing regions. Chemical pesticides are the primary method for pest control; the excessive use of insecticides has resulted in pesticide resistance. C. suppressalis is highly susceptible to cyproflanilide, a novel pesticide with high efficacy. However, the acute toxicity and detoxification mechanisms remain unclear. We carried out a bioassay experiment with C. suppressalis larvae and found that the LD10, LD30 and LD50 of cyproflanilide for 3rd instar larvae was 1.7 ng/per larvae, 6.62 ng/per larvae and 16.92 ng/per larvae, respectively. Moreover, our field trial results showed that cyproflanilide had a 91.24% control efficiency against C. suppressalis. We investigated the effect of cyproflanilide (LD30) treatment on the transcriptome profiles of C. suppressalis larvae and found that 483 genes were up-regulated and 305 genes were down-regulated in response to cyproflanilide exposure, with significantly higher CYP4G90 and CYP4AU10 expression in the treatment group. The RNA interference knockdown of CYP4G90 and CYP4AU10 increased mortality by 20% and 18%, respectively, compared to the control. Our results indicate that cyproflanilide has effective insecticidal toxicological activity, and that the CYP4G90 and CYP4AU10 genes are involved in detoxification metabolism. These findings provide an insight into the toxicological basis of cyproflanilide and the means to develop efficient resistance management tools for C. suppressalis.


of 13
Cyproflanilide is a novel meta-diamide insecticide with toxicological activity against lepidopteran, coleoptera and thysanoptera pest species. It was classified into Group 30 "GABA-Gated Chloride Channel Allosteric Modulators" by the Insecticide Resistance Action Committee (IRAC). Several studies have identified metabolic and target resistance in C. suppressalis as resistance mechanisms to diamide insecticides [1,7,[9][10][11][12]. Huang et al. (2020) reported that resistance to diamide insecticides was associated with a mutation of the ryanodine receptor (RyRs) at the I4578M, Y4667D/C, G4915E and Y4891F sites in C. suppressalis [8]. Furthermore, the overexpression of CYP genes (CYP6CV5, CYP9A68, CYP321F3, CYP324A12, UGT40AL1 and UGT33AG3) were found to be involved in the metabolic resistance of C. suppressalis to chlorantraniliprole. [1,10] Cyproflanilide has been identified as playing an important role in pest resistance management because it binds to different insect receptors and induces toxicity via different modes of action than other diamide insecticides. However, the underlying mechanisms of C. suppressalis resistance remain unknown.
The ability of insects to metabolize and detoxify pesticides is thought to play a significant role in pesticide resistance. The detoxification enzyme genes involved in xenobiotic metabolism, degradation and detoxification belong to three distinct pathways: phase I (hydrolysis and oxidation-reduction), phase II (conjugation) and phase III (transport) [13,14]. Several important enzymes are involved in the metabolism of xenobiotics, including cytochrome P450 monooxygenase (P450s), esterase (EST, including carboxylesterase and phosphoesterase), glutathione S-transferase (GST), UDP-glycosyltransferase (UGT) and ATP-binding cassette transporter (ABC) [15,16]. P450 belongs to the heme family of proteins involved in the metabolism of endogenous and exogenous compounds and the regulation of insect growth and development [15,17]. It is worth noting that the CYP gene plays a central role in insect xenobiotic metabolism, hydrolysis and oxidation-reduction and regulates a variety of endogenous signaling molecules [16]. However, we do not know whether CYP genes are involved in the metabolism of cyproflanilide.
To evaluate the control efficiency of cyproflanilide on C. suppressalis, we carried out a laboratory bioassay and field trial. Our results revealed that cyproflanilide has effective toxicological insecticidal activity against C. suppressalis. Furthermore, in order to find the detoxification metabolism genes that were mainly involved in cyproflanilide, a combined transcriptome RNA-seq and RNA interference approach identified two P450 genes, CYP4G90 and CYP4AU10, that are involved in cyproflanilide metabolism in C. suppressalis. Our findings contribute to a better understanding of cyproflanilide metabolism in C. suppressalis and provide insight into the development of pest and resistance management strategies for this species.

Bioassay Results
Bioassay experiments showed that cyproflanilide exhibited potent pesticide activity against C. suppressalis. The probit analysis showed that LD 10 , LD 30 and LD 50 were 1.421, 5.261 and 13.022 ng/per larva at 24 h ( Table 1). The third instar larvae were then treated with LD 10 , LD 30 and LD 50 cyproflanilide to test the expression of detoxification genes.

Field Experiment Results
After 21 days of spraying, compared to 200 g/L chlorantraniliprole (SC), the most effective pesticide for controlling C. suppressalis was 20% cyproflanilide (SC). The control efficiency on C. suppressalis of 16.2, 32.4, 48.6 and 64.8 g/hectare of 20% cyproflanilide (SC) were 77.38%, 84.60%, 87.26% and 91.24%, respectively (Table 2). There was a significant difference in control efficiency among low, medium and high doses (F = 74.941, p = 0.0001). However, there was no significant difference in control efficiency between 48.6 and 64.8 g/hectare of 20% cyproflanilide (SC) ( Table 2). The dead heart rate was significantly reduced when rice was treated with different doses of 20% cyproflanilide (SC) for 21 days compared to 200 g/L chlorantraniliprole (SC). The six individual cDNA libraries constructed from the contrast and cyproflanilide treatments were sequenced on Illumina Nova Seq 6000. Clean reads for subsequent analysis were obtained after filtration of the original data, a sequence error rate check and a GC content distribution check. A total of 36.58 GB of clean data was obtained from the RNA sequences of six samples. The GC content of the sequence data ranged from 46.29% to 46.83%, and the Q20 and Q30 ratios were > 97% and 93%, respectively, indicating good quality (Table S1).
Pearson's correlation coefficient was used to analyze the relationship between the control and treatment groups. The square of coefficient was > 0.8 ( Figure S1), indicating that the within-group biological replication was good. Gene expression levels (FPKM values) of each sample were used to calculate the within and between group correlation coefficients which were then plotted into a heat map (Figure 1a).

GO and KEGG Analysis of DEGs
GO pathway enrichment analysis was performed to investigate the function of DEGs in C. suppressalis third instar larvae exposed to cyproflanilide. Regarding the 19 significant GO terms, 393 DEGs were enriched, of which 339 were up-regulated and 54 were downregulated. The 19 significantly enriched GO terms were divided into three categories: molecular functions, biological processes and cellular components. The most abundant genes in the molecular function terms category were "structural molecular activity" and "structural constituent of cuticle". For the biological processes category, the most abundant of genes were "carbohydrate derivative metabolic process" and "drug metabolic process". The "extracellular region" had the most genes in the cellular component category. Downregulated genes were enriched in DNA replication. Additionally, "drug metabolic process", "serine hydrolase activity" and "oxidoreductase activity" may also be related to cyproflanilide exposure (Figure 1b and S4).
The results of the enrichment analysis revealed that "DNA replication" was significantly enriched in 101 KEGG pathways ( Figure S2). The abundant DEGs found in the KEGG database mainly included categories such as "pentose and glucuronate interconversions", "ubiquinone and other terpenoid-quinone biosynthesis", "drug metabolismother enzymes", "fatty acid biosynthesis" and "base excision repair". It is worth noting that "drug metabolism-other enzymes", "drug metabolism-cytochrome P450" and "metabolism of xenobiotics by cytochrome P450" may also be involved in the metabolism of

GO and KEGG Analysis of DEGs
GO pathway enrichment analysis was performed to investigate the function of DEGs in C. suppressalis third instar larvae exposed to cyproflanilide. Regarding the 19 significant GO terms, 393 DEGs were enriched, of which 339 were up-regulated and 54 were down-regulated. The 19 significantly enriched GO terms were divided into three categories: molecular functions, biological processes and cellular components. The most abundant genes in the molecular function terms category were "structural molecular activity" and "structural constituent of cuticle". For the biological processes category, the most abundant of genes were "carbohydrate derivative metabolic process" and "drug metabolic process". The "extracellular region" had the most genes in the cellular component category. Down-regulated genes were enriched in DNA replication. Additionally, "drug metabolic process", "serine hydrolase activity" and "oxidoreductase activity" may also be related to cyproflanilide exposure (Figure 1b and Figure S4).
The results of the enrichment analysis revealed that "DNA replication" was significantly enriched in 101 KEGG pathways ( Figure S2). The abundant DEGs found in the KEGG database mainly included categories such as "pentose and glucuronate interconversions", "ubiquinone and other terpenoid-quinone biosynthesis", "drug metabolism-other enzymes", "fatty acid biosynthesis" and "base excision repair". It is worth noting that "drug metabolism-other enzymes", "drug metabolism-cytochrome P450" and "metabolism of xenobiotics by cytochrome P450" may also be involved in the metabolism of cyproflanilide. We found that the detoxification metabolism pathway was significantly enriched, indicating that these pathway genes were involved in detoxification metabolism to cyproflanilide.

Transcriptome Profiling Reveals Detoxification Genes Associated with Cyproflanilide
Using a log2 fold change cutoff of ≥ 1 and ≤ − 1, and corrected p value of < 0.05, a total of 788 genes were identified as differentially expressed between the two treatments, of which 483 genes were up-regulated and 305 down-regulated ( Figure S3). Finally, the analysis of DEGs identified 28 detoxification metabolism genes, including 3 glutathione S-transferase, 4 glucosyltransferase, 11 cytochrome oxidoreductase, 5 carboxylesterase and 5 ATP-binding cassette transporters. qRT-PCR was performed on the 12 metabolic detoxification genes to check if there were any differences in the transcriptome data. The qRT-PCR results show a correlation between the metabolic detoxification gene expression level and transcriptome data.

Analysis Detoxification Gene Expression Level by qRT-PCR
We used qRT-PCR to see if the 12 detoxifications genes differed with different pesticide concentrations, and over time.

Transcriptome Profiling Reveals Detoxification Genes Associated with Cyproflanilide
Using a log2 fold change cutoff of ≥ 1 and ≤ − 1, and corrected p value of < 0.05, a tota of 788 genes were identified as differentially expressed between the two treatments, o which 483 genes were up-regulated and 305 down-regulated ( Figure S3). Finally, the anal ysis of DEGs identified 28 detoxification metabolism genes, including 3 glutathione S transferase, 4 glucosyltransferase, 11 cytochrome oxidoreductase, 5 carboxylesterase and 5 ATP-binding cassette transporters. qRT-PCR was performed on the 12 metabolic detox ification genes to check if there were any differences in the transcriptome data. The qRT PCR results show a correlation between the metabolic detoxification gene expression leve and transcriptome data.

Analysis Detoxification Gene Expression Level by qRT-PCR
We used qRT-PCR to see if the 12 detoxifications genes differed with different pesti cide concentrations, and over time.  (Figures 2f and 3b). UGT33AG1 was significantly different only under LD30 stress UGT33AG3, ABCC6 and EST5 had significantly different levels under LD10 stress, while ABCC6 was also significantly different under LD30 stress (Figure 2d-f). Interestingly, the results of the gene expression level showed that CYP4G90 and CYP4AU10 have stable up regulation expression levels at different concentrations and across time (Figure 3a,b).

Sensitivity to Pesticides after Gene Silencing
The high expressions levels of CYP4G90 and CYP4AU10 at different cypr concentrations and exposure times suggest their likely involvement in cyproflan tabolism. Hence, RNAi was used to verify the functions of CYP4G90 and CYP4A suppressalis. After 24 hours and 48 hours of feeding on a diet containing dsRNA for CYP4G90 and CYP4AU10, the mRNA level decreased by 65%, 46% and 44%, ure 4a,b), respectively. When larvae were treated with LD30 cyproflanilide for the knockdown of CYP4G90 and CYP4AU10 significantly increased the mortalit third instar larvae compared to the control groups (fed dsEGFP, Figure 4c). After the knockdown of CYP4G90 during LD30 cyproflanilide treatment resulted in a s increase in the mortality of third instar larvae compared to the control groups (F In contrast, the knockdown of CYP4AU10 had no significant effect on the larvae of LD30 cyproflanilide treatment in 24 hours (Figure 4d). Collectively, these result strate that a decrease in CYP4G90 and CYP4AU10 mRNA levels can result in an in the mortality of larvae exposed to cyproflanilide, further confirming the inv of CYP4G90 and CYP4AU10 in the metabolism of cyproflanilide.

Sensitivity to Pesticides after Gene Silencing
The high expressions levels of CYP4G90 and CYP4AU10 at different cyproflanilide concentrations and exposure times suggest their likely involvement in cyproflanilide metabolism. Hence, RNAi was used to verify the functions of CYP4G90 and CYP4AU10 in C. suppressalis. After 24 h and 48 h of feeding on a diet containing dsRNA specific for CYP4G90 and CYP4AU10, the mRNA level decreased by 65%, 46% and 44%, 42% (Figure 4a,b), respectively. When larvae were treated with LD 30 cyproflanilide for 12 h, the knockdown of CYP4G90 and CYP4AU10 significantly increased the mortality rates of third instar larvae compared to the control groups (fed dsEGFP, Figure 4c). After 24 h, the knockdown of CYP4G90 during LD 30 cyproflanilide treatment resulted in a significant increase in the mortality of third instar larvae compared to the control groups (Figure 4d). In contrast, the knockdown of CYP4AU10 had no significant effect on the larvae mortality of LD 30 cyproflanilide treatment in 24 h (Figure 4d). Collectively, these results demonstrate that a decrease in CYP4G90 and CYP4AU10 mRNA levels can result in an increase in the mortality of larvae exposed to cyproflanilide, further confirming the involvement of CYP4G90 and CYP4AU10 in the metabolism of cyproflanilide.

Discussion
Due to their high efficiency, chemical pesticides such as monosultap, triazophos avermectin and chlorantraniliprole are currently the most effective method of controlling large-scale outbreaks of C. suppressalis [6,7]. Chlorantraniliprole sales in China have in creased significantly due to the increasing number of C. suppressalis outbreaks, resulting in a significant increase in resistance level [18,19]. The new insecticide cyproflanilide is highly effective against C. suppressalis larvae. Over time, lethal doses of the insecticide in the field will gradually reduce to sublethal doses, which will continuously threaten larva growth and development.
Sublethal doses of insecticides have been reported to affect development duration pupal weight and reproduction of pests. For example, chlorpyrifos, etofenprox and phos met prolonged larval development time, and chlorantraniliprole and flubendiamide in creased larval and pre-pupal development times and decreased larval weight [20,21]. Dur ing physiological activity, carbohydrate and lipid metabolism, and energy metabolism are the primary source of energy and substrate for organisms [22]. In this study, 7 out of 12 DEGs involved in carbohydrate metabolism were down-regulated, indicating that it may influence C. suppressalis development. Interestingly, under chlorantranilamide treatment these pathway genes were also altered which affected the development time of C. suppres salis larvae [22,23]. Lipids play an important role in insect reproduction [24][25][26], and stud ies have demonstrated that changes in lipid content affect silkworm reproduction [27]. In our study, 31 DEGs were associated with glycerolipid metabolism, sphingolipid metabo lism, steroid biosynthesis, fatty acid elongation, glycerophosholipid metabolism, fatty acid degradation, fatty acid biosynthesis, fatty acid metabolism and unsaturated fatty acid biosynthesis.
In addition to the effect of exogenous substances on insect reproduction and devel opment, the detoxification and metabolism of exogenous substances by pests have been extensively studied [28][29][30][31]. In this study, 28 DEGs were primarily enriched in the "drug

Discussion
Due to their high efficiency, chemical pesticides such as monosultap, triazophos, avermectin and chlorantraniliprole are currently the most effective method of controlling large-scale outbreaks of C. suppressalis [6,7]. Chlorantraniliprole sales in China have increased significantly due to the increasing number of C. suppressalis outbreaks, resulting in a significant increase in resistance level [18,19]. The new insecticide cyproflanilide is highly effective against C. suppressalis larvae. Over time, lethal doses of the insecticide in the field will gradually reduce to sublethal doses, which will continuously threaten larval growth and development.
Sublethal doses of insecticides have been reported to affect development duration, pupal weight and reproduction of pests. For example, chlorpyrifos, etofenprox and phosmet prolonged larval development time, and chlorantraniliprole and flubendiamide increased larval and pre-pupal development times and decreased larval weight [20,21]. During physiological activity, carbohydrate and lipid metabolism, and energy metabolism are the primary source of energy and substrate for organisms [22]. In this study, 7 out of 12 DEGs involved in carbohydrate metabolism were down-regulated, indicating that it may influence C. suppressalis development. Interestingly, under chlorantranilamide treatment, these pathway genes were also altered which affected the development time of C. suppressalis larvae [22,23]. Lipids play an important role in insect reproduction [24][25][26], and studies have demonstrated that changes in lipid content affect silkworm reproduction [27]. In our study, 31 DEGs were associated with glycerolipid metabolism, sphingolipid metabolism, steroid biosynthesis, fatty acid elongation, glycerophosholipid metabolism, fatty acid degradation, fatty acid biosynthesis, fatty acid metabolism and unsaturated fatty acid biosynthesis.
In addition to the effect of exogenous substances on insect reproduction and development, the detoxification and metabolism of exogenous substances by pests have been extensively studied [28][29][30][31]. In this study, 28 DEGs were primarily enriched in the "drug metabolism-other enzymes", "glutathione metabolism" and "Metabolism of xenobiotics by cytochrome P450" pathways. P450 genes, in particular, have received increasing attention due to their broad range of functions [32]. For example, the overexpression of CYP4G19 in beta-cypermethrin-resistant strains of Blattella germanica was positively correlated with a high level of cuticular hydrocarbons, while the knockdown of CYP4G19 expression resulted in a decrease in cuticular hydrocarbons and reduced insecticide tolerance in resistant strains [33]. Furthermore, previous studies have reported a link between the function of the CYP4 family of genes and the insect cuticle [17,[33][34][35]. In this study, we demonstrated that CYP4G90 and CYP4AU10 are involved in cyproflanilide metabolism, and we found a number of DEGs related to the insect cuticle. Therefore, C. suppressalis may develop cuticular resistance to cyproflanilide. Transcriptome analysis of C. suppressalis treated with chlorantraniliprole revealed that CYP4G90 is also a DEG. However, there was no evidence to suggest that CYP4G90 was involved in chlorantraniliprole metabolism [22]. A significant increase in UGT33AG3 expression was found in C. suppressalis strains resistant to chlorantraniliprole [10]. Our results showed a significant difference in UGT33AG3 expression levels following 24 h of cyproflanilide treatment. It is evident that the development of pest resistance is not dependent on a single gene [36][37][38]. Previous research using RNAi demonstrated that CYP6CV5, CYP9A68, CYP321F3 and CYP324A12 are involved in metabolic resistance to chlorantraniliprole [1].
In this study, CYP4G90 and CYP4AU10 had higher expression levels compared with other detoxifying metabolic genes under the cyproflanilide treatment of C. suppressalis third instar larvae. Meanwhile, we found that knockdown of these two genes increased the mortality of C. suppressalis larvae exposed to cyproflanilide. In the reports of the resistant populations of insects to pesticides, the overexpression of P450 genes were related to the metabolic resistance of pesticides. For example, CYP321A6 and CYP332A1 had mediate chlopyrifos resistance in Spodoptera exigua, CYP6BG1 was involved in the resistance of imidacloprid to whiteflies [39,40] and CYP6CM1 and CYP4C62 were proven to have participated resistance to chlorantraniliproe and chlorpyrifos in Plutella xylostella and Nilaparvata lugens, respectively, by RNAi [41,42]. In future, with the wider use of cyproflanilide, the C.suppressalis may develop resistance to it. Therefore, these two genes are the potential target resistance genes of C. suppressalis to cyproflanilide.
In conclusion, the novel insecticide cyproflanilide is highly effective against C. suppressalis, with a control efficiency of 64.8 g/hectare reaching 91.24% efficacy. We also demonstrated the role of CYP4G90 and CYP4AU10 in the metabolic detoxification of cyproflanilide. Our findings show that the novel pesticide cyproflanilide is highly effective against C. suppressalis and we identified potential targets for further pesticide resistance research into cyproflanilide.

Bioassay
The third instar larvae were used for the bioassay in this research. Each treatment was repeated thrice, with 30 larvae per replicate. Cyproflanilide was dissolved in acetone to obtain concentrations of 40 mg/L, 20 mg/L, 10 mg/L, 5 mg/L, 2.5 mg/L, 1.25 mg/L and 0.625 mg/L, with the acetone-treated group as a control. C. suppressalis larvae were treated with 0.5 µL of the insecticide solution drop on the pronotum using a hand micro applicator (Burkard Manufacturing Co Ltd., Rickmansworth, Hertfordshire, England), and mortality was recorded at 24 h. The sublethal dose of cyproflanilide to the C. suppressalis was determined via a log-probit analysis of bioassay data using SPSS 22. Larvae treated with LD 10 , LD 30 and LD 50 were collected at 24 h and 48 h, and were immediately frozen in liquid nitrogen and then stored at −80 • C for RNA extraction.

Evaluting and Verifying Field Efficacy
The field-plot trials were located in Xidu Town (Hengyang County, Hunan Province, China). When the treatment rice plant was applied, most of the pests were second and third instar larvae. The field experiments were conducted to evaluate the control efficacies of cyproflanilide and chlorantraniliprole against C. suppresalis larvae on rice plants. All of the experiments were designed as randomized complete blocks with four replicates of each treatment. The 7.5 × 4 m plots were used for each treatment application. The following four treatments of 20% cyproflanilide (SC) and one treatment of 200 g/L chlorantraniliprole (SC) were evaluated: 16.2 g/hectare, 32.4 g/hectare, 48.6 g/hectare 64.8 g/hectare and 30 g/hectare, with water as a control. After 21 days of pesticide application, we counted the number of rice dead hearts caused by C. suppressalis. The number of rice dead hearts was used to calculate the insecticide control effect. No other pesticides were applied during the experimental period. The average total number of rice plants in the plot was obtained according to the number of rows and columns of rice in the plot and the average tiller number of 50 clusters. The control efficacies of each insecticide-treated group C. suppressalis larvae were calculated using the following Equations (1) and (2): Dead heart rate (%) = (The number of dead rice hearts/Total number of rice plants in the experimental plot) × 100. (1) Control efficiency (%) = (Number of dead rice hearts of control plants − Dead heart rate of pesticide-treated plants/Dead heart rate of control) × 100. (2)

RNA Sequencing and Annotation of Unigenes
Total RNA was used as the input material for the RNA sample preparations. Briefly, mRNA was purified from total RNA using poly-T oligo-attached beads. Fragmentation was carried out using divalent cations under an elevated temperature in the First Stand Synthesis Reacting Buffer (5X). The first strand cDNA was synthesized using random hexamer primer and M-MuLV Reverse Transcriptase; we then used RNaseH to degrade the RNA. Second strand cDNA synthesis was subsequently performed using DNA Polymerase I and dNTP. The remaining overhangs were converted into blunt ends via exonuclease/polymerase activities. After adenylation of the 3 ends of DNA fragments, Adaptor with a hairpin loop structure was ligated to prepare for hybridization. To select cDNA fragments of preferentially 370~420 bp in length, the library fragments were purified with the AMPure XP system (Beckman Coulter, Beverly, USA). After PCR amplification, the PCR product was purified using AMPure XP beads, and the final library was obtained.
The library was initially quantified using a Qubit2.0. Fluorometer, then diluted to 1.5 ng/µL. The insert size of the library was detected using an Agilent 2100 bioanalyzer. After the insert size met expectations, qRT-PCR was used to accurately quantify the effective concentration of the library (the effective concentration of the library is higher than that of 2 nM) to confirm the quality of the library. The different libraries were then pooled according to the effective concentration and the target amount of data of the machine before being sequenced using the Illumina NovaSeq 6000. The reference genome and gene model annotation files were downloaded from a genome website directly: http://v2.Insect-genome.com/api/Download/.-01_data-01_speciesChilosuppressalis-Chilo_suppressalis.genome.fa (accessed on 11 August 2021). The index of the reference genome was built using Hisat2 (v2.0.5) and paired-end clean reads were aligned to the reference genome using Hisat2 (v2.0.5). Transcripts with an adjusted p-value < 0.05 found by DESeq were assigned as differentially expressed.

Quantitative Real Time PCR
First-stand cDNA templates were synthesized using the PrimeScriptTM RT reagent kit with gDNA Eraser (Taraka, Dalian, China). The specific primers for qRT-PCR were designed using online website NCBI: https://www.ncbi.nlm.nih.gov/tools/primer-blast/ (accessed on 27 December 2021) with the EF-1 house-keeping gene used as the internal gene (Table S2) [43]. The qRT-PCR reaction was performed using Hieff ® qPCR SYBR Green Master Mix (Yeasen, Shanghai, China) following the manufacturer's instructions. The 10 µL PCR reaction volume contained 5 µL SYBR Green Master Mix, 2.2 µL diethylpyrocarbonatetreated water (DEPC), 2 µL diluted cDNA template with a concentration of 100 ng/µL, and 0.4 µL of each primer. The qRT-PCR program was as follows: 95 • C for 30 s, 40 cycles of 95 • C for 5 s and 60 • C for 30 s. The relative gene expression levels were represented using the 2 −∆∆CT method.

RNAi Experiment
Primer for RNA interference (RNAi) was designed using online website siDirect: http://sidirect2.rnai.jp/ (accessed on 27 March 2022) and Vazyme: https://crm.vazyme. com/cetool/en-us/singlefra-gment.html (accessed on 27 March 2022). The templates of double-stranded RNA (dsRNA) synthesis were obtained by the real-time polymerase chain reaction (RT-PCR) using specific primers (Table S3). These two genes, a 528 bp fragment of CYP4G90 and a 553 bp fragment of CYP4AU10, were amplified and subcloned into the pET-2p expression vector using the ClonExpress II One Step Cloning Kit (Vazyme, Nanjing, China). The recombinant vectors of CYP4G90 and CYP4AU10 were transformed into HT115competent cells (Shanghai Weidi Biotechnology Co., Ltd., Shanghai, China) for dsRNA expression. Individual colonies were inoculated and grown until the cultures reached an OD 600 of 0.8. Isopropyl-β-D-Thiogalactoside (IPTG) (Beijing coolaber Technology Co., Ltd., Beijing, China) was added to the cultures to produce a final concentration of 0.1 mM; the culture was then incubated at 37 • C for approximately 4 h. The expression of dsRNA was verified by 1% agarose gel. The induced cultures were centrifuged at 8000× g for 5 min before being resuspended in one-tenth of the original culture volume of 0.05 M phosphate-buffered saline (PBS). The resuspended bacterial solution was then used for oral RNAi. The third instar larvae were starved for 2 h, while Zizania latifolia (Griseb) was soaked in the resuspended bacterial solution and then dried for 30 min. After feeding on dsRNA for 24 h, larvae were treated with of 0.5 µL of the LD 30 cyproflanilide solution. Each treatment was replicated five times with 30 third instar larvae per replication and the mortality was recorded every 12 h until the end of the 24 h period.

Data Processing and Statistical Analysis
We used GraphPad Prism8 and IBM SPSS Statistics Version 22 software for data analyses. T tests (nonparametric tests) were used to compare means between treatments and their respective controls. The results are given as means ± standard error. p < 0.05 was considered a significant difference, and p < 0.01 was considered a highly significant difference.