Identification of the Flavone-Inducible Counter-Defense Genes and Their cis-Elements in Helicoverpa armigera

Flavone is widely found in plants and plays an important role in plant defense against pests. Many pests, such as Helicoverpa armigera, use flavone as a cue to upregulate counter-defense genes for detoxification of flavone. Yet the spectrum of the flavone-inducible genes and their linked cis-regulatory elements remains unclear. In this study, 48 differentially expressed genes (DEGs) were found by RNA-seq. These DEGs were mainly concentrated in the retinol metabolism and drug metabolism-cytochrome P450 pathways. Further in silico analysis of the promoter regions of 24 upregulated genes predicted two motifs through MEME and five previously characterized cis-elements including CRE, TRE, EcRE, XRE-AhR and ARE. Functional analysis of the two predicted motifs and two different versions of ARE (named ARE1 and ARE2) in the promoter region of the flavone-inducible carboxylesterase gene CCE001j verified that the two motifs and ARE2 are not responsible for flavone induction of H. armigera counter-defense genes, whereas ARE1 is a new xenobiotic response element to flavone (XRE-Fla) and plays a decisive role in flavone induction of CCE001j. This study is of great significance for further understanding the antagonistic interaction between plants and herbivorous insects.


Introduction
Chemical pesticides have been the most effective tools for control of agricultural pests, but overuse or misuse of them may lead to environmental pollution and pesticide residues in water, soil, and crops, which eventually harm human health [1]. In addition, the abuse of pesticides generally causes the evolution of resistance by pests [2]; the harmful effects of pesticides on the environment can endanger the sustainability of ecosystems [3]. Therefore, it is very important to find eco-friendly biopesticides with target ability, degradability, and low toxicity. Secondary metabolites such as alkaloids, flavonoids, terpenoids, and sterols extracted from plants are important resources [4]. These biopesticides usually degrade quickly and therefore do not last long in the ecosystem [5]. Studies on the effects of these compounds on insect pests provide new opportunities for the development of new biological insecticides.
During the long evolutionary process, plants have developed defense mechanisms against insect feeding through morphological, biochemical, and molecular regulation [6]. can affect larval midgut transcriptome but has no noticeable impacts on the growth of the larvae. Relative to the control larvae fed on normal diets, the larvae fed on any of the three flavone-containing diets excreted a lower amount of feces ( Figure 1A). The higher the flavone concentration, the lower the fecal volume ( Figure 1A). In terms of body weight, a measurement of larval growth, no significant differences were found between the control larvae and the larvae fed on 0.01% or 0.1% flavone diets ( Figure 1B). However, the body weight of the larvae fed on 1% flavone diets was significantly reduced ( Figure 1B), suggesting that 0.1% flavone is probably the suitable dose for induction of the larval transcriptome. A bioassay of newly-molted 6th instar larvae with 0.1% flavone diets further confirmed that 0.1% flavone is the suitable induction dose because no significant differences in morphology and body weight were detected between the experimental group and the control group ( Figure 1C,D).

Selection of a Flavone Dose for Induction of the Larval Midgut Transcriptome of H. armigera
A bioassay of 5th instar larvae of the H. armigera Baiyun strain with diets incorporated with 0.01%, 0.1%, or 1% flavone was conducted to select a suitable induction dose that can affect larval midgut transcriptome but has no noticeable impacts on the growth of the larvae. Relative to the control larvae fed on normal diets, the larvae fed on any of the three flavone-containing diets excreted a lower amount of feces ( Figure 1A). The higher the flavone concentration, the lower the fecal volume ( Figure 1A). In terms of body weight, a measurement of larval growth, no significant differences were found between the control larvae and the larvae fed on 0.01% or 0.1% flavone diets ( Figure 1B). However, the body weight of the larvae fed on 1% flavone diets was significantly reduced ( Figure 1B), suggesting that 0.1% flavone is probably the suitable dose for induction of the larval transcriptome. A bioassay of newly-molted 6th instar larvae with 0.1% flavone diets further confirmed that 0.1% flavone is the suitable induction dose because no significant differences in morphology and body weight were detected between the experimental group and the control group ( Figure 1C,D).  Fecal quantity of the 5th instar larvae fed with 0.01%, 0.1%, and 1% flavone diets for 3 days. (B) Body weight gain of the 5th instar larvae fed with 0.01%, 0.1%, and 1% flavone diets for 3 days. (C) Photos of the 6th instar larvae before (0 h) and 48 h post feeding with 0.1% flavone diets. (D) Body weight gain of the 6th instar larvae fed with 0.1% flavone diets for 48 h. Bars with different letters such as a, b, c, d (p < 0.05, one-way ANOVA followed by Tukey's HSD test) are significantly different. NS means no significantly different (p > 0.05, independent t test).

Analysis of Flavone-Elicited Differentially Expressed Genes
Statistical analysis of the FPKM (reads per kilo base of exon model per million mapped reads) value of each gene expressed in the midgut of the control and flavone-treated larvae using DESeq2 uncovered 48 genes meeting the criteria (False discovery rate (FDR) < 0.05 and |log 2 fold change (FC)| > 1) for differentially expressed genes (DEGs) (Figure 2A). Among the 48 DGEs, 38 and 10 genes were up-and downregulated ( Figure 2B), respectively, by 0.1% flavone. Two-way hierarchical clustering analysis divided the six larval midgut RNA samples into control and flavone clades and the 48 DGEs into up-and downregulated clades ( Figure 2C), confirming the RNA-seq data were reliable.

Identification of Conserved Motifs in the Promoter Regions of Flavone-Upregulated Genes
We used the MEME program to screen the promoter regions (up to 3 kb upstream of the start codon) of 24 flavone-upregulated genes (see their names in Figure 5) for the presence of conserved motifs that may act as xenobiotic response elements to flavone (XRE-Fla). The 24 genes were selected from the 38 flavone-upregulated genes ( Figure 2C) using Onetailed t test at p < 0.05. (C) Two-tailed t test at p < 0.05. Bar pairs with one (p < 0.05, independent t test) and two asterisks (p < 0.01, independent t test) are significantly and extremely different, respectively.

Identification of Conserved Motifs in the Promoter Regions of Flavone-Upregulated Genes
We used the MEME program to screen the promoter regions (up to 3 kb upstream of the start codon) of 24 flavone-upregulated genes (see their names in Figure 5) for the presence of conserved motifs that may act as xenobiotic response elements to flavone (XRE-Fla). The 24 genes were selected from the 38 flavone-upregulated genes ( Figure 2C) using the following two criteria: (1) FPKM value greater than 5; and (2) significantly higher relative expression in the flavone-treated larvae (two-tailed t test at p < 0.05). One 29 bp GC-rich motif called Motif 1 and one 21 bp TA-rich motif called Motif 2 were detected from the promoter regions of these genes ( Figure 5). Motif 1 was present in the promoter regions of 21 of the 24 genes, whereas the promoter regions of all the 24 genes had Motif 2 ( Figure 5).

Identification of Known Elements in the Promoter Regions of Flavone-Upregulated Genes
We also screened the promoter regions of the 24 flavone-upregulated genes for the presence of previously characterized cis-acting elements including XRE (Xenobiotic response element), TRE (TPA response element), CRE (cAMP-response element), EcRE (Ecdysone receptor element), and XRE-AhR (Xenobiotic response element-Aryl hydrocarbon receptor nuclear) that may mediate the flavone induction of these genes. Among the five known elements, ARE, EcRE, TRE, CRE and XRE-AhR were detected in the promoter regions of 13, 6, 4, 3, and 1 flavone-upregulated genes, respectively ( Figure 6). Given its high prevalence, ARE is likely to be one of the elements that respond to flavone.

Identification of Known Elements in the Promoter Regions of Flavone-Upregulated Genes
We also screened the promoter regions of the 24 flavone-upregulated genes for the presence of previously characterized cis-acting elements including XRE (Xenobiotic response element), TRE (TPA response element), CRE (cAMP-response element), EcRE (Ecdysone receptor element), and XRE-AhR (Xenobiotic response element-Aryl hydrocarbon receptor nuclear) that may mediate the flavone induction of these genes. Among the five known elements, ARE, EcRE, TRE, CRE and XRE-AhR were detected in the promoter regions of 13, 6, 4, 3, and 1 flavone-upregulated genes, respectively ( Figure 6). Given its high prevalence, ARE is likely to be one of the elements that respond to flavone.
Meanwhile, using the pGL3-CCE001j promoter as the template, the luciferase activity of four recombinant plasmids obtained by deletion del-Motif1, del-Motif2, del-ARE1, and del-ARE2 was determined. Only the basal activity of del-Motif1 was significantly higher than that of the pGL3-CCE001j promoter, which increased from 11.90 ± 0.43 to 16.49 ± 2.30. The activity of del-ARE1 decreased significantly from 26.70 ± 2.41 to 13.22 ± 1.74 (p = 0.003). The deletion of the ARE1 element caused the CCE001j promoter to be no longer induced by flavones ( Figure 7C).
Meanwhile, using the pGL3-CCE001j promoter as the template, the luciferase activity of four recombinant plasmids obtained by deletion del-Motif1, del-Motif2, del-ARE1, and del-ARE2 was determined. Only the basal activity of del-Motif1 was significantly higher than that of the pGL3-CCE001j promoter, which increased from 11.90 ± 0.43 to 16.49 ± 2.30. The activity of del-ARE1 decreased significantly from 26.70 ± 2.41 to 13.22 ± 1.74 (p = 0.003). The deletion of the ARE1 element caused the CCE001j promoter to be no longer induced by flavones ( Figure 7C).  The start codon ATG is numbered with +1 and marked in red, the upstream sequence is preceded by a "−" and the downstream sequence by a "+". The predicted Motif1, Motif2, ARE1 and ARE2 are underlined. (B) The Renilla/reference firefly luciferase ratio of 6 upregulated gene promoters before and after flavone induction. (C) The Renilla/reference firefly luciferase ratio of Motif1, Motif2, ARE1, and ARE2 deleted recombinant plasmids was calculated. The structure of the internal deletion is named "del-" and the name of the deletion sequence. Bars with different letters such as a, b (A, B or A', B') (p < 0.05, one-way ANOVA followed by Tukey's HSD test) are significantly different. Bar pairs with two and three asterisks mean p < 0.01 and p < 0.001, respectively. NS means no significantly different (p > 0.05, independent t test).

Discussion
Studying the response of H. armigera larvae to flavone will broaden our understanding of the effects of flavone on herbivorous insects and help us lay the foundation for future pest control [31]. In lepidopteron insects, midgut tissue plays an important role in feeding and toxin metabolism, which means it is the main tissue affected by the flavone's toxin [32]. In this study, transcriptomics were used to analyze gene expression in the midgut of H. armigera treated with flavone, and a flavone-induced gene change pattern was obtained. Based on these differentially expressed genes, GO analysis and KEGG analyses were performed. Finally, cis-regulatory elements in promoters of upregulated gene expression were predicted and the functions of novel and classic elements were verified in the CCE001j promoter.
In this study, three flavone concentration gradients were set, which were 0.01%, 0.1% and 1%. After feeding, the fecal volume of H. armigera larvae decreased significantly with The start codon ATG is numbered with +1 and marked in red, the upstream sequence is preceded by a "−" and the downstream sequence by a "+". The predicted Motif1, Motif2, ARE1 and ARE2 are underlined. (B) The Renilla/reference firefly luciferase ratio of 6 upregulated gene promoters before and after flavone induction. (C) The Renilla/reference firefly luciferase ratio of Motif1, Motif2, ARE1, and ARE2 deleted recombinant plasmids was calculated. The structure of the internal deletion is named "del-" and the name of the deletion sequence. Bars with different letters such as a, b (A, B or A', B') (p < 0.05, one-way ANOVA followed by Tukey's HSD test) are significantly different. Bar pairs with two and three asterisks mean p < 0.01 and p < 0.001, respectively. NS means no significantly different (p > 0.05, independent t test).

Discussion
Studying the response of H. armigera larvae to flavone will broaden our understanding of the effects of flavone on herbivorous insects and help us lay the foundation for future pest control [31]. In lepidopteron insects, midgut tissue plays an important role in feeding and toxin metabolism, which means it is the main tissue affected by the flavone's toxin [32]. In this study, transcriptomics were used to analyze gene expression in the midgut of H. armigera treated with flavone, and a flavone-induced gene change pattern was obtained. Based on these differentially expressed genes, GO analysis and KEGG analyses were performed. Finally, cis-regulatory elements in promoters of upregulated gene expression were predicted and the functions of novel and classic elements were verified in the CCE001j promoter.
In this study, three flavone concentration gradients were set, which were 0.01%, 0.1% and 1%. After feeding, the fecal volume of H. armigera larvae decreased significantly with the increase in the concentration of flavone, indicating that a high concentration of flavone can inhibit feeding of H. armigera. The body weight of larvae treated with 0.01% and 0.1% flavone decreased slightly, but not significantly. After treatment with 1% flavone concentration, body weight decreased significantly. This was consistent with the reported effects of flavone on Spodoptera litura [10], indicating that the low flavone concentration did not affect insect body weight in a short period of time, which might be because of the role of detoxification metabolism genes in insects. A 0.1% concentration was commonly used in previous studies of flavone-induced gene expression [30]. Therefore, in this study, 0.1% flavone concentration was selected to treat the 6th larva of H. armigera, enabling the larvae to fully mobilize the detoxification gene to metabolize flavone and take midgut tissue for transcriptome sequencing.
RNA-seq revealed a total of 48 differentially expressed genes including 38 upregulated gens and 10 downregulated genes. Among the upregulated genes were P450s, carboxylesterase, UGTs, and glutathione transferase; these genes have been reported to be involved in plant toxin metabolism [31,33,34]. The upregulated DEGs were mainly enriched in drug metabolism-cytochrome P450, metabolism of xenobiotics by cytochrome P450, and drug metabolism-other enzymes. These results indicated that detoxificationrelated pathways play an important role in the interaction between H. armigera and flavone. Only 10 downregulated genes were detected, and future research should also focus on these genes, particularly on these genes' downregulated mechanisms and their downregulated functions, especially in antifeeding and insect development.
As the FPKM values of some upregulated genes were relatively low, and the expression trends of some genes were inconsistent with the transcriptome results, 24 promoters of upregulated genes were selected for cis-regulatory element prediction. Two motifs were predicted by the online site MEME and some cis-regulatory elements previously reported to be associated with toxins were also predicted. Six promoters of upregulated genes were cloned. Except for the CCE001f promoter, the other five promoters significantly increased luciferin activity after induction by flavones, but CCE001b, CYP6AE19, UGT40F2, and the basic and induced activity of ALDH1A1L were low, so the CCE001j promoter was selected to predict the function of the element.
Four potential cis-acting elements Motif1, Motif2, ARE1 and ARE2 in the CCE001j promoter were verified by deletion of the elements. The results indicated that the two motifs obtained by MEME prediction were not induced by flavones, but internal deletion of Motif1 significantly enhanced basal promoter activity of CCE001j, and Motif2 may plays a role in other aspects, which needs further study.
In conclusion, our results suggest that high doses of flavone have an antifeeding effect on H. armigera, and cis-acting elements play an important role in activating detoxification metabolic pathways in insects. In particular, the ARE element in the CCE001j promoter determines whether the gene is induced by flavone. The results of this study will help us to understand the reaction process of flavonoids in H. armigera and help us to clarify the action mode of flavone.

Conclusions
Transcriptome sequencing was performed for the first time in the midgut tissues of H. armigera treated with flavone. In terms of a bioassay, we proved that low doses of flavone did not affect the feeding and growth of H. armigera, while high doses of flavone hindered larval feeding and caused the slow growth of H. armigera. In terms of transcriptome sequencing, we reported for the first time differentially expressed genes in the midgut tissues of H. armigera after feeding on flavone, and these genes were mainly enriched in detoxification metabolic pathways. Cis-acting elements in the promoter regions of some upregulated genes were predicted. Finally, the response of some components to flavone was verified in the CCE001j gene promoter. In addition, it was confirmed that element ARE1 in the CCE001j promoter determines whether the promoter is induced by flavone. These results indicated that the expression levels of detoxification metabolism genes in midgut tissues were upregulated after feeding on flavone, and cis-acting elements in these genes played an important role in this process, such as the ARE1 element in CCE001j promoter, which determined whether the promoter was induced by flavone.

Insects
The H. armigera strain used in this study was collected from tobacco fields in Xuchang City (Henan Province, China) in 2018. The larvae were reared on a wheat germ-containing artificial diet [35] at 26 ± 1 • C, 60 ± 10% relative humidity (RH), and a photoperiod of 16:8 h L/D. The larvae were separated into plastic cups individually. Adult moths were placed in plastic mating cages (5 pairs/cage) with 10% honey solution. The cage was covered with a piece of cheese cloth sheet for collecting eggs.

Flavone Feeding Bioassay
Flavone (CAS number: 525-82-6, purity 98%) was purchased from Shanghai Aladdin Biochemical Technology Co. (Shanghai, China). Three different final concentrations (0.01%, 0.1% and 1%, w/w) of flavone were incorporated into a corn flour/soybean flour-based diet. Briefly, when the diet was cooled to about 47 • C, 400 µL of the corresponding concentration of flavone dissolved in dimethyl sulfoxide (DMSO) was added to 40 mL of diet, vortexed to mix well, and then evenly poured into 27 plastic cups. After the diet was cooled and solidified, H. armigera 5th instar larvae of similar size were transferred to the 27 plastic cups (1 larva per cup). Three days later, the body weight and feces of each larva were weighed and counted. There were 3 biological replicates of 9 larvae each per concentration and DMSO control.

RNA Extraction and Library Construction and Illumina RNA-Seq
Newly molted sixth instar of H. armigera were individually transferred to cups (1 larva/cup) with control or 0.1% flavone diets. There were 3 biological replicates in both the experimental group (0.1% flavone diets) and the control group (diets with equal volume of the solvent DMSO), with 6 larvae in each replicate. After 48 h of treatment, midgut tissue was dissected, and a quick-frozen tube stored at −80 • C.
Total RNA of the above six larval midgut samples were extracted according to the steps using RNA extraction kit (Beibei Biological, Zhengzhou). RNA purity, integrity and concentration were determined by Nanodrop 2000 spectrophotometer (Wilmington, DE) cDNA libraries, and then sequenced on Illumina HiSeq 2500 platform by Gene Denovo Biotechnology Co. (Guangzhou, China).

DEGs Analysis
The expression levels were displayed as original reads count and FPKM. The original reads count represented the number of reads contained in the transcript, but being affected by sequencing amount and gene length, the original reads count was not conducive to the comparison of differential genes between samples. To ensure the accuracy of the subsequent analysis, we first corrected the sequencing depth and then corrected the length of genes or transcripts. After obtaining the FPKM value of genes, gene differential expression analysis was conducted. The input data for the gene differential expression analysis were reads count data obtained from gene expression level analysis, which was analyzed by DESeq2 software [36]. The analysis was mainly divided into three parts: Read count was first standardized. Then the hypothesis testing probability (p-value) was calculated according to the model. Finally, multiple hypothesis testing was performed to obtain the FDR value (error detection rate). Genes with a FDR< 0.05 and | log2FC | > 1 were considered as differentially expressed genes (DEGs).

Expression Validation of DEGs by qRT-PCR
Real-time quantitative fluorescence PCR (qRT-PCR) was used to verify the DEGs identified from sequencing analysis. All primers were designed using Primer 5.0 software (Premier Biosoft International, Palo Alto, CA, USA) and DNAMAN (Lynnon Biosoft, San Ramon, CA, USA) according to strict principles (primers are listed in Table S1). The qRT-PCR reactions were performed following the manufacturer's manual for the UltraSYBR Mixture (low ROX) (CWBIO, Beijing, China) on an ABI QuantStudio 5 (manufacture information). Each reaction (20 µL final volumes) contained 1.0 µL cDNA, 10 µL UltraSYBR Mixture (low ROX), 0.5 µL forward and reverse primers, and 8 µL ddH 2 O. PCR conditions were as follows: 10 min at 95 • C, followed by 40 cycles of 10 s at 95 • C, 30 s at 60 • C, 32 S at 72 • C and 10 min at 72 • C EF and Actin (ACT) were used as internal controls. Three biological samples were repeated with three techniques repeats each. The relative expression levels of each gene were calculated by the 2 −∆∆CT method [37]. The 2 −∆∆CT method was used to process the data, and the processed data were plotted by GraphPad Prism software. Primer sequences and amplification efficiency are shown in Table S1.

Conservative Motif Analysis of Flavone-Induced Upregulated Genes in H. armigera
The purpose of MEME (Multiple EM for Motif Elicitation) is to allow users to discover signals (called 'motifs') in DNA or protein sequences. The user of MEME inputs a set of sequences believed to share some (unknown) sequence signal(s). For example, some or all of a set of promoters from co-expressed and/or orthologous genes may contain binding sites (the 'signal') for the same transcription factor. We used the online version of MEME (Multiple EM for Motif Elicitation) (http://meme-suite.org/tools/meme, accessed on 5 November 2021) to predict transcription factor binding sites in upregulated gene promoters. To put it simply, we started by uploading a FASTA file containing the sequence of 24 upregulated gene promoters into MEME. Finally, we changed some parameters, changed the Motif width parameter from 6 to 30, and set the number of pre-find motifs to 20. At the end of the task, we created MEME.xml and MAST.xml files. We visualized the MEME/MAST xml results with TBtools (bioinformatics analysis software). In addition, ALLGEN and JASPAR software were used to predict known gene toxin-related response elements in promoters.

Construction of Wildtype and Deletion Promoter-pGL3 Constructs
Promoters of six genes (CCE001j, CCE001b, CCE001f, CYP9AE19, UGT40F2, and ALDH1A1L) were amplified from genomic DNA. The Mlu1 restriction site was designed on the forward primers (BglII was designed for the forward primers of CCE001j), and the XhoI restriction site was designed for all reverse primers (HindIII was designed for the reverse primers of CCE001j). The PCR products were analyzed by electrophoresis on 1.5% agarose gel; the target bands were recovered from the gel and double-digested by Mlu1 and XhoI (BglII and HindIII for CCE001j). The recombinant plasmid was obtained by attaching the double-digested products to firefly luciferase reporter vector pGL3-basic (XhoI and MluI double-digested) (double-digested products of CCE001j were attached to BglII and HindIII digested pGL3-basic).

Transient Transfection and Dual Luciferase Assay
One hundred uL of H. zea fatbody cells were seeded onto wells of 96-well plates (5 × 10 5 cells/well). After 30 min, the seeded cells were transiently co-transfected with either of the 6 wildtype (pGL3-CCE001j promoter, pGL3-CCE001b promoter, pGL3-CCE001f promoter, pGL3-CYP6AE19 promoter, pGL3-UGT40F2 promoter, pGL3-ALDH1A1L promoter) and 4 deletion promoter-pGL3 constructs (del-Motif 1, del-Motif 2, del-ARE1 and del-ARE2) (0.1 µg/well) and the internal renilla luciferase control reporter plasmid PHRL-TK (Promega, 0.01 µg/well). There were three technical replicates in both the experimental group and the control group. Six hours after transfection, the final concentration of 18.5 µM flavone (induced) or equivalent volume of methanol (control) was added. After 48 h, the cells were harvested and the luciferase activity of kidney and firefly was measured using the dual luciferase reporting assay system (Promega, Madison, WI, USA) on a TD-20/20 single-tube luminescence instrument designed by Turner (Turner Biosystems, Sunnyvale, CA, USA). The relative firefly luciferase activity normalized against the renilla luciferase activity was calculated as an indicator of the basal or flavone-inducible promoter activity of each construct. At least 3 repeated measurements were conducted for each independent transfection.