MAPK-Activated Transcription Factor PxJun Suppresses PxABCB1 Expression and Confers Resistance to Bacillus thuringiensis Cry1Ac Toxin in Plutella xylostella (L.)

ABSTRACT

Deciphering the molecular mechanisms underlying insect resistance to Cry toxins produced by Bacillus thuringiensis (Bt) is pivotal for the sustainable utilization of Bt biopesticides and transgenic Bt crops. Previously, we identified that mitogen-activated protein kinase (MAPK)-mediated reduced expression of the PxABCB1 gene is associated with Bt Cry1Ac resistance in the diamondback moth, Plutella xylostella (L.). However, the underlying transcriptional regulation mechanism remains enigmatic. Here, the PxABCB1 promoter in Cry1Ac-susceptible and Cry1Ac-resistant P. xylostella strains was cloned and analyzed and found to contain a putative Jun binding site (JBS). A dual-luciferase reporter assay and yeast one-hybrid assay demonstrated that the transcription factor PxJun repressed PxABCB1 expression by interacting with this JBS. The expression levels of PxJun were increased in the midguts of all resistant strains compared to the susceptible strain. Silencing of PxJun expression significantly elevated PxABCB1 expression and Cry1Ac susceptibility in the resistant NIL-R strain, and silencing of PxMAP4K4 expression decreased PxJun expression and also increased PxABCB1 expression. These results indicate that MAPK-activated PxJun suppresses PxABCB1 expression to confer Cry1Ac resistance in P. xylostella, deepening our understanding of the transcriptional regulation of midgut Cry receptor genes and the molecular basis of insect resistance to Bt Cry toxins.

IMPORTANCE The transcriptional regulation mechanisms underlying reduced expression of Bt toxin receptor genes in Bt-resistant insects remain elusive. This study unveils that a transcription factor PxJun activated by the MAPK signaling pathway represses PxABCB1 expression and confers Cry1Ac resistance in P. xylostella. Our results provide new insights into the transcriptional regulation mechanisms of midgut Cry receptor genes and deepen our understanding of the molecular basis of insect resistance to Bt Cry toxins. To our knowledge, this study identified the first transcription factor that can be involved in the transcriptional regulation mechanisms of midgut Cry receptor genes in Bt-resistant insects.

INTRODUCTION

The diverse insecticidal crystal proteins generated by Bacillus thuringiensis (Bt) can specifically kill various lepidopteran, coleopteran and dipteran insect pests without causing harm to nontarget organisms or the environment; as such, Bt biopesticides and genetically modified Bt crops have been developed and widely used for pest control (1–3). Unfortunately, different insect pests in the field have developed high-level resistance to Bt sprays or Bt crops due to the resulting strong selection pressure, gravely threatening the sustainable application of Bt products (4, 5). Therefore, understanding the molecular mechanisms underlying insect resistance to Bt Cry toxins is crucial for the long-term utilization of these valuable Bt biotechnologies (6, 7).

As currently understood, the mode of action of Bt Cry toxins involves multiple steps that occur in the larval midgut, and the binding of Bt Cry toxins to functional receptors in the midgut is a key step in this complex toxicity process (8–10). The identified midgut receptors of Bt Cry toxins include cadherin (CAD), alkaline phosphatase (ALP), aminopeptidase N (APN), and ATP-binding cassette (ABC) transporters (i.e., ABCAs, ABCBs, ABCCs, and ABCGs) (11, 12). The downregulation of midgut Cry receptor genes to reduce toxin-receptor interactions is one of the principal drivers of Bt resistance evolution in diverse insects (13, 14). However, little is known about the transcriptional regulation mechanisms of these midgut Cry receptor genes.

ABCB1, also known as P-glycoprotein (PGP) or MDR1, is one of the best-studied ABC transporters. ABCB1 can serve as a pump to efflux toxic xenobiotics out of cells using ATP-driven energy, and upregulation of the ABCB1 gene has been reported to be involved in multidrug resistance (MDR) in mammals (15) and resistance to chemical and biological pesticides in insects (16–19). Moreover, studies in recent years have found that ABCB1 can also serve as a functional midgut receptor of Bt Cry3 toxins, and its gene mutation is associated with high-level resistance to Bt Cry3 toxins in Diabrotica virgifera virgifera and Chrysomela tremula (20–22). Meanwhile, we found that downregulation of ABCB1 is directly associated with Cry1Ac resistance in P. xylostella (23, 24). Although research on the transcriptional regulatory mechanisms of ABCB1 expression has made great progress in mammals (25, 26), little is known about this phenomenon in insects.

Recently, our studies have shown that the insect hormone-activated mitogen-activated protein kinase (MAPK) signaling pathway can act as a common switch to trans-regulate the differential expression of multiple midgut genes, including genes encoding ALP, APNs (APN1, APN3a, APN5, and APN6) and ABC transporters (ABCB1, ABCC1, ABCC2, ABCC3, and ABCG1), thereby conferring Bt Cry1Ac resistance in P. xylostella without fitness costs (14, 23, 24, 27, 28). These researches have also indicated that there is a significant level of redundancy in receptors such that multiple proteins within a given insect can act as functional receptors and that change of individual gene results in incremental increase in resistance. The involvement of the MAPK cascades indicating that there are some MAPK-responsive downstream transcription factors (TFs) that regulate the expression of these midgut genes, including ABCB1 (29). Moreover, some studies have also found that TFs such as FOXA and GATAe can positively regulate the expression of some Bt Cry1Ac receptor genes, including ABCC2, ABCC3, ALP, and CAD, in different insect cell lines (30, 31). Indeed, we have identified a MAPK-activated transcription factor CREB that can increase the expression of a P450 gene CYP6CM1, thereby leading to imidacloprid resistance in Bemisia tabaci (32). Nonetheless, the TFs that regulate the reduction in expression of midgut receptor genes, including ABCB1, in insect resistance to Bt Cry toxins remain largely unknown.

Here, our work has shown that a TF called PxJun can directly bind to the promoter region of PxABCB1 to reduce its expression. Furthermore, midgut PxJun expression was significantly higher in all the Cry1Ac-resistant strains than in the Cry1Ac-susceptible strain. Silencing of PxJun expression upregulated PxABCB1 transcription and enhanced larval susceptibility to Cry1Ac toxin in the Cry1Ac-resistant NIL-R strain, and silencing of PxMAP4K4 expression inhibited PxJun and increased PxABCB1 transcription. These results indicate that PxJun activated by the MAPK signaling pathway participates in Cry1Ac resistance in P. xylostella by inhibiting the expression of the PxABCB1 gene. Our results provide important insights into the transcriptional regulation of Bt toxin receptors, providing a better understanding of the evolution of insect resistance to Bt Cry toxins.

RESULTS

PxABCB1 promoter analyses in susceptible and resistant strains.

Having previously observed that differential expression of PxABCB1 associates with resistance to the Bt Cry1Ac toxin in P. xylostella (23, 24), we sought to compare the promoter region of this gene between a susceptible and a resistant strain in order to investigate possible reasons for this differential expression.

We amplified the 5′-flanking region of the PxABCB1 gene from gDNA samples of the Cry1Ac-susceptible strain DBM1Ac-S and its near-isogenic Cry1Ac-resistant strain NIL-R. Analysis of the PxABCB1 promoter sequence revealed a conserved initiator of transcription site (Inr) motif 5′-TCAGT-3′ located 127 nucleotides (nt) upstream of the start codon (ATG) of the PxABCB1 gene. The adenine (A) of the Inr was designed as the transcriptional start site (TSS) and marked as “+1” (see Fig. S1 in the supplemental material). With respect to promoter identification no typical TATA box was found near the TSS, although a putative CAAT box was predicted 139 nt upstream of the TSS (see Fig. S1). Nucleotide sequence alignment showed that multiple single nucleotide polymorphism sites existed in the PxABCB1 promoters of the two strains and that the PxABCB1 promoter in the NIL-R strain contained a number of deletions, as well as a small insertion, compared to that in the DBM1Ac-S strain (Fig. 1A; see also Fig. S1).

FIG 1.

Transcriptional activity of the PxABCB1 promoter in susceptible and resistant P. xylostella. (A) Diagram of sequence alignment of the 5′-flanking regions of PxABCB1 in susceptible DBM1Ac-S and resistant NIL-R strains. The right-angled arrow denotes the TSS. The numbers with arrows specify the 5′ and 3′ positions of the corresponding nucleotide. The green/white rectangles indicate DNA fragment insertion/deletion (Ins/Del). (B) Detection of PxABCB1 promoter activities in susceptible and resistant P. xylostella. All the fragment constructs are named with “P” as the starting letter, followed by a pair of parentheses that contain two numerals, separated by a slash (/), to specify the 5′ and 3′ positions of the corresponding promoter fragment. Relative luciferase (Luc) activities were detected at 48 h posttransfection in S2 cells. The relative luciferase activity (fold) of different promoter recombinants was calculated based on the value of the pGL4.10 control vector. The values shown are means and corresponding standard errors (SEM) for three independent experiments. The significance of differences was determined by one-way ANOVA with Duncan’s test (P < 0.05). Different letters used to mark bars denote significant differences.

To investigate whether the differences in the 5′-flanking region sequences between DBM1Ac-S and NIL-R strains lead to differences in promoter activity, and thus cause differential expression of the PxABCB1 gene, a transcriptional reporter assay was employed. The full-length and truncated promoters to reflect regions of similarity and dissimilarity between the two strains were cloned into the pGL4.10 vector to drive the expression of the luciferase reporter gene (Fig. 1). The relative luciferase activities of these PxABCB1 promoter recombinant plasmids were detected in S2 cells at 48 h posttransfection. The results indicated that there was no significant difference in activity for any of the PxABCB1 promoter regions (Fig. 1B), suggesting that genomic differences were not responsible for the altered expression of PxABCB1 in the resistant strain. What was noticeable, however, was that the promoter activity was significantly reduced when the section between positions −280 and +125 was used, despite this region containing the putative CAAT box promoter motif (Fig. 1B; see also Fig. S1).

Identification of the critical regulatory regions in the PxABCB1 promoter.

To investigate the loss of promoter activity associated with the −280/+125 region, we created a further range of constructs containing decreasing amounts of the promoter region from the DBM1Ac-S strain (Fig. 2). The data from these constructs confirmed that promoter activity was reduced once sequences upstream of position −460 were removed. Interestingly though, the promoter activity was partially restored when the deletion extended as far as position −72 (Fig. 2). Based on the luciferase activity data, these two important regulatory regions in PxABCB1 promoter from positions −621 to −460 and from positions −112 to −72 likely contain critical positive and negative regulatory elements, respectively, suggesting that they could potentially control PxABCB1 expression through a combination of cis-regulatory elements and trans-acting factors (33).

FIG 2.

Relative luciferase activity analysis of the fragments between positions −765 and +125. Progressive 5′ deletion constructs from positions −765 to +125 were transfected into S2 cells, and luciferase activity was detected. The relative luciferase activity (fold) of different constructs was calculated based on the value of the pGL4.10 vector. The values shown are means and the corresponding SEM. One-way ANOVA followed by Duncan’s test was used for comparison (P < 0.05, n = 3). Different letters used to mark bars denote significant differences.

The transcription factor PxJun represses PxABCB1 promoter activity.

Informatic analyses were undertaken to identify putative TF binding sites. Potential sites for CncC, Vvl, and Ubx were predicted in the positive regulatory region and the binding sites of Jun and Ubx were predicted in the core negative regulatory region (Fig. 3A; see also Fig. S1). To explore whether these proteins were involved in the transcriptional regulation of the PxABCB1 gene, the full-length coding region of each TF was cloned by PCR amplification and subcloned into the expression vector pAc5.1. They were then cotransfected into S2 cells, along with the luciferase reporter plasmid bearing the full PxABCB1 promoter region. The cotransfection assays revealed that PxJun significantly decreased the activity of the PxABCB1 promoter compared to the control, whereas other TFs had little effect on promoter activity (Fig. 3B), suggesting that PxJun might be involved in the negative regulation of PxABCB1 expression through interacting with the Jun binding site (JBS).

FIG 3.

Effects of potential TFs on the activity of the PxABCB1 promoter. (A) TFBS in critical positive and negative regulatory regions (see Fig. S1 for detailed DNA motifs). Different colored ellipses represent different TFBSs. (B) Effects of different predicted TFs on PxABCB1 promoter activity. Every cloned TF was subcloned into the pAc5.1 expression vector to generate a recombinant vector, which was then cotransfected with P(–1122/+125) to determine the luciferase activity. The empty pAc5.1 without PxJun was used as a control. The relative luciferase activity (fold) was calculated based on the value of the empty pAc5.1 vector. The values shown are means and the corresponding SEM. One-way ANOVA followed by Duncan’s test was used for statistical analysis (P < 0.05, n = 3). Different letters used to mark bars denote significant differences.

PxJun interacts with the JBS to repress PxABCB1 promoter activity.

To confirm that PxJun inhibits promoter activity through the predicted JBS between positions −112 and −72 (Fig. 3A), recombinant plasmids bearing PxABCB1 promoters containing a JBS deletion or mutation were constructed and cotransfected with the PxJun plasmid into S2 cells. The luciferase activity of the PxABCB1 promoter containing normal JBS was significantly decreased by PxJun, whereas the activities of the JBS deleted or mutated promoters showed no change compared to the control (Fig. 4A). These results suggested that PxJun repressed PxABCB1 promoter activity mainly through the identified JBS. Subsequently, a yeast one-hybrid (Y1H) assay was carried out to further test the interaction between PxJun and the JBS. The yeast strain containing the PxJun prey vector and normal JBS bait grew normally on SD/–Leu medium supplemented with 500 ng/ml AbA, while the yeast strain containing the PxJun prey vector and mutant JBS-M bait did not grow (Fig. 4B). This result indicated that the PxJun prey plasmid interacted with the JBS but not the JBS-M bait. Together, these results confirmed that PxJun suppressed the promoter activity of the PxABCB1 gene by interacting with the JBS.

FIG 4.

PxJun represses PxABCB1 promoter activity through the Jun binding site (JBS). (A) Effect of PxJun on PxABCB1 promoter activity with either a deleted or a mutated form of the JBS via the dual-luciferase reporter assay. The JBS motif (AAGAAATGAGAGAT, −107 to −94) in P(–1122/+125) was deleted or mutated to GGAGGGCAGAGAGC. PxJun was cotransfected with P(–1122/+125) containing normal JBS (red ellipse), deleted JBS (black ellipse), or mutated JBS (green ellipse). The empty pAc5.1 without PxJun was used as a control. Three biological replicates were performed for all experiments. The values shown are means and the corresponding SEM. One-way ANOVA followed by Duncan’s test was used for statistical analysis (P < 0.05). Different letters used to mark bars denote significant differences. (B) Verification of direct binding of PxJun to the JBS by the yeast one-hybrid (Y1H) assay. Three tandem repeats containing wild-type or mutated JBS sequences were fused to the pABAi vector, which was subsequently integrated into the Y1HGold yeast strain to generate the “bait strain.” A critical aureobasidin A (AbA) concentration of 500 ng/ml was detected to completely repress the growth of the bait strains on SD/–Ura media. PxJun was fused to the pGATD7 vector and then transferred into the “bait strain.” which was grown on SD/–Leu selective media with or without AbA. EV, empty vector; positive control, using pGADT7-p53 + pABAi-p53.

Analysis of the PxJun protein and its possible interaction with PxFos.

The PxJun protein contains two characteristic domains: the Jun domain at the N terminus and the bZIP_Jun domain at the C terminus (see Fig. S2). The bZIP_Jun domain consists of a basic DNA-binding domain for specific DNA recognition and binding, followed by a leucine zipper domain for protein dimerization (see Fig. S3) (29). Both domains are conserved among different insects and mammals (see Fig. S2 and S3). We performed a phylogenetic analysis of PxJun to determine the evolutionary relationships among c-Jun proteins in different species (see Fig. S2). The results showed that the c-Jun proteins are evolutionarily conserved and clearly clustered into groups corresponding to each insect order; as expected, the mammalian c-Jun proteins also grouped into one cluster (see Fig. S2).

Jun-related subfamily members typically interact with themselves or other proteins, and function as dimers in their regulatory role, and Fos subfamily proteins are the most common cofactors (34, 35). Thus, we aimed to further explore whether the PxFos protein is also involved in the regulation of PxABCB1 transcription with PxJun. A recombinant plasmid encoding PxFos with or without PxJun was cotransfected with the PxABCB1 promoter into S2 cells to measure the activity of the luciferase reporter. Transfection of PxFos alone or in combination with PxJun had no significant effect on the activity of the PxABCB1 promoter (Fig. 5), indicating that the PxFos protein did not participate in the regulation of the PxABCB1 gene.

FIG 5.

Effects of PxJun and/or PxFos on the activity of the PxABCB1 promoter. PxJun and/or PxFos were cotransfected with P(–1122/+125), and the luciferase activity was measured. The empty pAc5.1 without TF was used as a control. The relative luciferase activity (fold) was calculated based on the value of the empty pAc5.1 vector. The values shown are the means and the corresponding SEM for three independent experiments. One-way ANOVA followed by Duncan’s test was used for statistical analysis (P < 0.05, n = 3). Different letters used to mark bars denote significant differences.

Increased expression levels of the PxJun gene in resistant strains.

The spatiotemporal transcription profiles of the PxJun gene were monitored by qPCR in different tissues of fourth-instar DBM1Ac-S larvae and different developmental stages. The tissue expression profile showed that the PxJun gene was widely expressed in different tissues with no obvious tissue-specific expression pattern, implying that the PxJun gene plays important roles in a variety of tissues (see Fig. S4). Meanwhile, developmental expression analysis indicated that the PxJun gene was also expressed in different periods with no obvious stage-specific expression pattern (see Fig. S4), suggesting that the PxJun gene is involved in the regulation of growth, development, and reproduction in P. xylostella.

To further investigate the relationship between PxJun and PxABCB1, the transcript levels of PxJun were detected in the midgut tissues of fourth-instar larvae of different P. xylostella strains. As indicated by the qPCR results, the expression levels of the PxJun gene were significantly higher in all four Cry1Ac-resistant strains than in the susceptible DBM1Ac-S strain (Fig. 6). The expression trend of PxJun was negatively correlated with the PxABCB1 expression trend in the midgut tissues of different strains (detected in our previous study) (23), whereas it was positively correlated with Cry1Ac resistance level, supporting the concept that PxJun promotes Cry1Ac resistance by repressing PxABCB1 expression.

FIG 6.

Relative expression levels of the PxJun gene in the midgut tissues of fourth-instar larvae in five P. xylostella strains as detected by qPCR. The RPL32 gene was used as an internal control. The relative expression level (fold) is presented as the ratio to the value of the lowest expression level, observed in the DBM1Ac-S strain. The average relative expression levels and SEM of three independent replicates are presented. Different letters used to mark bars denote significant differences (P < 0.05, Duncan’s test, n = 3).

Silencing PxJun enhances PxABCB1 expression and susceptibility to Cry1Ac toxin.

To validate whether PxJun depresses PxABCB1 expression in P. xylostella, PxJun-specific dsRNA was synthesized and injected into third-instar NIL-R larvae. The transcription levels of PxJun and PxABCB1 were measured by qPCR after 48 h. The expression of PxJun was reduced by approximately 50% in the midguts of double-stranded RNA (dsRNA)-injected larvae compared to controls; in contrast, the transcriptional level of PxABCB1 in their midguts was significantly increased compared to that of controls (Fig. 7A). In addition, toxicity bioassays with 1,000 mg/liter Cry1Ac protoxin were carried out at 48 h postinjection. The mortality of the PxJun-silenced larvae was dramatically increased (from 10 to 35.56%) (Fig. 7B). These data again supported the role of PxJun in negatively controlling the expression of PxABCB1 in vivo, contributing to Cry1Ac resistance in P. xylostella.

FIG 7.

Effect of PxJun gene silencing on PxABCB1 expression and Cry1Ac resistance in resistant NIL-R larvae. (A) Relative expression of PxJun and PxABCB1 at 48 h postinjection with buffer, dsEGFP or dsPxJun. The expression levels of PxJun or PxABCB1 in the control larvae injected with buffer were set as 1. (B) Silencing of PxJun expression decreased the resistance of NIL-R larvae to Cry1Ac protoxin. At 48 h after microinjection with buffer, dsEGFP or dsPxJun, the larvae were fed diets with or without Cry1Ac protoxin (1,000 mg/liter). The percentage of larval mortality was then counted at 72 h posttreatment. The data are presented as mean values ± the SEM for three biologically independent experiments. Different letters in each group indicate statistically significant differences between treatments (P < 0.05; Duncan’s test; n = 3).

The MAPK signaling pathway activates PxJun transcription.

Jun subfamily proteins are typical downstream targets of the MAPK signaling pathway and can be activated at the transcriptional and protein phosphorylation levels in mammals (34, 36, 37). Our previous studies revealed that the insect hormone-activated MAPK signaling pathway regulates the reduced expression of multiple midgut receptors, including PxABCB1 gene in resistant P. xylostella strains (24). Thus, we speculated that the decrease in PxABCB1 expression might be controlled by MAPK-responsive PxJun. To investigate whether PxJun expression leading to the downregulation of PxABCB1 is induced by the MAPK signaling pathway, we silenced PxMAP4K4 expression in resistant NIL-R larvae and then detected the expression levels of PxJun and PxABCB1 at different periods (Fig. 8). The results reflected that after PxMAP4K4 dsRNA injection, the transcript level of PxJun decreased, while the expression level of PxABCB1 increased (Fig. 8). Therefore, PxJun responded to MAPK signaling pathway and repressed the expression of the PxABCB1 gene in Cry1Ac-resistant P. xylostella.

FIG 8.

Effect of PxMAP4K4 gene silencing on the expression levels of PxJun and PxABCB1 in the midgut tissues at different periods. The data are presented as mean values ± the SEM for three biologically independent experiments. Asterisks (*) indicate significant differences among periods for each gene (P < 0.05; Duncan’s test; n = 3).

DISCUSSION

Downregulation of Bt Cry toxin receptor genes in the midgut usually results in high-level Bt resistance in insects; nevertheless, the specific transcriptional regulation mechanisms of these midgut genes remain poorly understood. Changes in gene expression can result from changes in trans-regulatory factors and cis-regulatory elements (TFBS [transcription factor binding sites] in the promoter, enhancer, and silencer) (38). In this study, we revealed that the MAPK-activated TF PxJun regulates the reduced expression of the PxABCB1 gene, thereby mediating Bt Cry1Ac resistance in P. xylostella.

Jun subfamily members in mammals include c-Jun, JunB, and JunD, important stress-responsive TFs that are activated by the MAPK signaling pathway under various physiological and external stimuli and participate in multiple cellular processes, such as cell proliferation, differentiation, apoptosis, and inflammation (34, 39). Intriguingly, whereas there are three Jun paralogs in mammals, only one Jun protein, a homolog of the mammalian c-Jun, has been identified in insects (36, 40). As in mammals, the MAPK signaling pathway in D. melanogaster can activate DJun in response to various stimuli to ensure normal development, immunity and homeostasis (37, 41).

Jun-related subfamily members belong to the basic leucine zipper (bZIP) family of proteins, which typically exert their functions through homodimerization or heterodimerization to form an AP-1 complex (34, 35). Jun proteins in animals participate in essential biological processes, including cell proliferation, differentiation, apoptosis, and immune response (39). PxJun is a homolog of mammalian c-Jun. We demonstrated that PxJun can specifically bind to the JBS and inhibit the promoter activity of the PxABCB1 gene. In mammals, activated c-Jun can act as an activator or repressor to control MDR1 (ABCB1) expression in different cancer cells (25, 42–45). Therefore, c-Jun appears to play a dual regulatory role in the regulation of ABCB1 expression in different types of cells and species, probably depending on its binding partners and on the cellular environment.

In mammals, the AP-1 complexes Jun-Jun and Jun-Fos preferentially bind to 12-O-tetradecanoylphorbol-13-acetate (TPA) response elements [TRE, 5′-TGA(C/G)TCA-3′], while Jun-ATF preferentially binds to the cAMP response element (CRE, 5′-TGACGTCA-3′) (35, 39). Distinct cofactors of Jun have different effects on the DNA binding affinity and function of the dimer, thus greatly expanding the spectrum of regulated genes (34). Our structural analysis of the PxJun protein showed a highly conserved bZIP_Jun domain for DNA binding and protein interaction at the C terminus. Fos is the most common partner of Jun (34, 35). However, we demonstrated that the PxFos protein did not participate in the regulation of the PxABCB1 gene. Thus, further study is needed to identify whether additional TFs interact with PxJun to coregulate PxABCB1 expression in the midgut, which will help us understand the complex functions of AP-1.

Mammalian AP-1 family members are expressed in a cell- and stage-dependent manner during development and under different stimuli, thereby mediating different transcriptional levels of specific target genes (34). In D. melanogaster, DJun expression is observed in all tissues and developmental stages (46). Similarly, we found that the PxJun gene is extensively expressed in multiple tissues and developmental stages of P. xylostella. The temporally and spatially broad expression pattern of c-Jun implies its essential role in maintaining immunity and homeostasis, normal growth and development, and response to various stress factors (35, 39, 47).

Jun subfamily proteins in mammals are activated both at the transcriptional level and via posttranslational modification by the MAPK signaling pathway in response to many physiological and environmental stimuli, affecting the transactivation potential, DNA binding capacity, and stability of AP-1 components (34, 36, 37). Studies in human cells found that c-Jun can be activated by the MAPK signaling pathway, leading to elevated transcript and phosphorylation levels and thus regulating the expression of the MDR1 (ABCB1) gene (43, 44, 48, 49). Silencing of PxMAP4K4 expression downregulated PxJun and upregulated PxABCB1, demonstrating that the expression of the PxABCB1 gene is negatively regulated by the MAPK-responsive PxJun in P. xylostella. Persistent alteration of AP-1 activity and/or ABCB1 expression can induce the oncogenic transformation of cells and tumor formation, as well as the development of chemoresistance in mammals (39). In fact, the expression trend of PxJun in midgut tissues of different P. xylostella strains is similar to that of PxMAP4K4 and opposite to that of PxABCB1 (23, 24, 27). Thus, a model explaining the transcriptional regulation mechanisms of reduced expression of the PxABCB1 gene is shown in Fig. 9. In the Bt-resistant P. xylostella larvae, the activated MAPK signaling pathway can induce PxJun transcription, thereby leading to a reduction in PxABCB1 gene expression and causing Cry1Ac resistance. Therefore, the MAPK/c-Jun signaling cascade might be a common mechanism of transcriptional regulation of the ABCB1 gene. Since the MAPK pathway has been shown to regulate the expression of a number of Cry1Ac receptors within P. xylostella, it is possible that PxJun is directly involved in the downregulation of other receptors and therefore that the increase in mortality observed in Fig. 7B may not be solely attributed to PxABCB1 overexpression. Further research is needed to evaluate whether the MAPK signaling pathway activates PxJun at both transcript and phosphorylation levels and whether PxJun also participates in the regulation of other receptors.

FIG 9.

Proposed model for the transcriptional regulation of reduced PxABCB1 expression by the MAPK-activated TF PxJun. The activated MAPK signaling pathway increases the expression of PxJun, which in turn represses the transcript level of Bt Cry1Ac receptor gene PxABCB1 and enhances larval resistance to Bt Cry1Ac toxin in P. xylostella.

In conclusion, we found that PxJun negatively regulates PxABCB1 expression by interacting with the JBS in the PxABCB1 promoter and that the activated MAPK cascade enhances PxJun expression and thus represses PxABCB1 expression to result in Cry1Ac resistance in P. xylostella. This study provides new insights into the transcriptional regulation mechanisms of midgut receptor genes and lays the foundation for comprehensively understanding the complex molecular mechanisms of insect resistance to Bt Cry toxins.

MATERIALS AND METHODS

Insect strains and cell line.

The five P. xylostella strains used in this study, including one Bt-susceptible and four Bt-resistant strains, were described previously (27, 50, 51). In brief, the DBM1Ac-S strain has been maintained in the laboratory without exposure to any Bt products/toxins or any other insecticides. The four Bt-resistant strains—DBM1Ac-R, NIL-R, SZ-R, and SH-R—exhibit different levels of resistance to Bt Cry1Ac protoxin or B. thuringiensis var. kurstaki (Btk) formulation. Their resistance ratios are approximately 3,500-, 4,000-, 450-, and 1,900-fold, respectively, that of the susceptible DBM1Ac‐S strain. The larvae of all strains were reared on Jing Feng no. 1 cabbage (Brassica oleracea var. capitata) at 25°C, a 65% relative humidity, and a 16:8 h (light-dark) photoperiod. Adults were supplied with a 10% honey/water solution.

The Drosophila melanogaster S2 cell line was transfected for the dual-luciferase reporter assay. S2 cells were cultured in HyClone SFX-Insect medium (GE HyClone, Logan, UT, USA) supplemented with penicillin-streptomycin (Gibco, Rockville, MD) at 27°C.

Toxin preparation and toxicity bioassay.

The Cry1Ac protoxin preparation and subsequent leaf-dip bioassay were performed as described previously (52). Briefly, the Cry1Ac protoxin was isolated and purified from Btk strain HD-73, and its protein concentration was quantified and stored in 50 mM Na₂CO₃ (pH 9.6) for subsequent toxicity bioassays. A 3-day leaf-dip bioassay was conducted using third-instar P. xylostella larvae. Ten individuals in each group were used, and the bioassays were repeated four times for each Cry1Ac concentration. The larval mortality of the control did not exceed 5%.

gDNA extraction, promoter cloning, and TFBS prediction.

Genomic DNA was isolated from fourth-instar P. xylostella larvae using the TIANamp genomic DNA kit (Tiangen, Beijing, China) according to the manufacturer’s instructions. A pair of specific PCR primers to amplify the PxABCB1 promoter was designed based on the 5′-flanking sequence of the PxABCB1 gene in the P. xylostella genome data set of LepBase (http://ensembl.lepbase.org/Plutella_xylostella_pacbiov1/Info/Index). PrimeSTAR Max DNA polymerase (TaKaRa, Dalian, China) was used according to the manufacturer’s protocol for all PCR amplifications in this study. The PCR products were subsequently purified and ligated into pEASY-T1 vectors (TransGen, Beijing, China) for sequencing. TFBS in the PxABCB1 promoter region were in silico predicted via the JASPAR database (http://jaspar.genereg.net) and PROMO virtual laboratory (http://alggen.lsi.upc.es/cgi-bin/promo_v3/promo/promoinit.cgi?dirDB=TF_8.3).

Sample preparation, RNA isolation, and cDNA synthesis.

Samples at different developmental stages (eggs, first- to fourth-instar larvae, prepupae, pupae, and 1-day-old virgin male and female adults) were collected from the susceptible DBM1Ac-S strain, and five tissues (head, midgut, Malpighian tubules, integument, and testis) were dissected from fourth-instar P. xylostella larvae in ice-cold insect Ringer’s solution (130 mM NaCl, 0.5 mM KCl, 0.1 mM CaCl₂).

Total RNA from different samples was extracted with TRIzol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. The first-strand cDNA used for subsequent gene cloning was synthesized using the PrimeScript II first-strand cDNA synthesis kit (TaKaRa) according to the manufacturer’s protocol.

TF cloning and sequence analysis.

The initial open reading frames (ORFs) of Jun, Fos, cap ‘n’ isoform C (CncC), and ultrabithorax (Ubx) genes in P. xylostella were retrieved from the GenBank database (https://www.ncbi.nlm.nih.gov/) under accession numbers XM_011559543, XM_011549928, XM_011570739, and KP245729, respectively. In addition, the ORF of the ventral veins lacking (Vvl) gene in P. xylostella was obtained from the LepBase (http://ensembl.lepbase.org/; Gene ID g11291). All the ORFs of these genes were further corrected in silico with the assistance of our previous P. xylostella midgut transcriptome database (53). Using specific primers for PCR amplification (see Table S1 in the supplemental material), the full-length coding sequences (CDSs) of PxJun (GenBank accession no. MW446637) and four other genes were cloned, and each obtained CDS was translated into amino acid sequences using the ExPASy translate tool (https://web.expasy.org/translate/). The presence of conserved domains in the PxJun protein was analyzed using the Conserved Domain Database (CDD) at NCBI (https://www.ncbi.nlm.nih.gov/cdd/). Multiple sequence alignment was conducted using Clustal Omega (http://www.ebi.ac.uk/Tools/msa/clustalo/) and was further formatted using GeneDoc 2.7 software (http://genedoc.software.informer.com/2.7/).

Phylogenetic analysis.

The Jun protein sequences used for alignment and phylogenetic analysis in insects and representative mammals were obtained from the GenBank database. A phylogenetic tree was built using MEGA 7.0 software (https://www.megasoftware.net/) with the neighbor-joining method, “p-distance” model, and 1,000 bootstrap replicates.

Dual-luciferase assay.

The 5′-flanking sequence of the PxABCB1 gene was truncated into a series of fragments of different sizes. All of the fragments were subcloned into linearized pGL4.10 vector (double digests with BglII and KpnI) using an In-Fusion HD cloning kit (Clontech, Mountain View, CA). The In-Fusion primers used for promoter amplification are listed in Table S1. In addition, full-length promoters with JBS mutation/deletion were obtained by gene synthesis (TsingKe, Beijing, China). The ORFs of TFs were ligated into linearized pAc5.1/V5-His B expression vector (here referred to as “pAc5.1” double digested with KpnI and XhoI) using the In-Fusion HD cloning kit. The In-Fusion primers used for TF amplification are listed in Table S2. The pGL4.73 vector (Promega, Madison, WI) containing a Renilla luciferase gene was used as an internal control.

Vector transfection was performed using Lipofectamine 2000 transfection reagent (Thermo Fisher Scientific, Waltham, MA). S2 cells were seeded at a density of 5 × 10⁵ cells per well in 400 μl of medium without antibiotics in a 24-well plate 6 h before transfection. To detect promoter activity, the pGL4.10-promoter reporter plasmid (600 ng) was cotransfected with pGL4.73 (200 ng), and the empty pGL4.10 vector was used as the control vector. To determine the effect of TF on promoter activity, the pAc5.1-TF expression plasmid (600 ng) and pGL4.10-promoter reporter plasmid (200 ng) were cotransfected with pGL4.73 (100 ng), and the empty pAc5.1 plasmid was used as a control vector. For each transfection reaction, the plasmids and transfection reagent were diluted in 100 μl of medium without antibiotics, incubated at room temperature for 15 min, and then added to each well. After 48 h of transient transfection, the luciferase activity was measured on a GloMax 96 microplate luminometer (Promega) by using the dual-luciferase reporter assay system (Promega) according to the manufacturer’s protocol. The luciferase activity was calculated by normalizing the firefly luciferase level to the Renilla luciferase level. The relative luciferase activity (fold) was calculated by setting the activity of the control to an arbitrary value of 1. Three biological replicates and four technical replicates were conducted for each transfection experiment. The statistical significance of differences between different groups were analyzed by one-way analysis of variance (ANOVA) with Duncan’s test (P < 0.05).

Yeast one-hybrid assay.

A yeast one-hybrid (Y1H) assay was performed using the Matchmaker Gold yeast one-hybrid system (Clontech) according to the manufacturer’s instructions. Three tandem copies of the predicted JBS (5′-AGAAAGAAATGAGAGATACG-3′) or mutant JBS (5′-AGAGGAGGGCAGAGAGCACG-3′) were ligated to linearized pAbAi vector (double digested with XhoI and HindIII) to construct bait plasmids. The bait strains were generated by integrating the pBait-AbAi vector (linearized with BstBI) into the Y1HGold yeast genome and then selected on SD/–Ura medium. After 3 to 5 days at 30°C, colonies were picked and analyzed by colony PCR and sequenced to further identify bait sequence inserts. For each confirmed bait strain, the MIC of aureobasidin A (AbA) to suppress basal expression of the bait construct was determined (<1,000 ng/ml), and this AbA concentration was used to screen the prey-bait interaction. PxJun cDNA was subcloned into linearized pGADT7 vector (double digested with EcoRI and XhoI) to construct a prey plasmid, which was then transformed into the bait strains and selected on SD/–Leu medium with the minimal AbA inhibitory concentration. After 3 to 5 days at 30°C, colonies were picked, analyzed by colony PCR, and sequenced to confirm the prey protein inserts. The confirmed positive interaction colony was rescreened on SD/–Leu medium with AbA to establish individual interaction strains. The positive control was generated by cotransforming the pGADT7-p53 and pAbAi-p53 plasmids into the Y1HGold strain. A negative control was created by cotransforming the empty pGADT7 with the normal pAbAi-JBS plasmids into the Y1HGold strain.

qPCR analysis.

As mentioned previously (14, 23), the transcript levels of the PxJun and PxABCB1 genes were quantified by real-time quantitative PCR (qPCR) using the specific primers listed in Table S3. The qPCR experiment was run on the QuantStudio 3 real-time PCR system (Applied Biosystems, Carlsbad, CA) using FastFire qPCR PreMix (SYBR green) (Tiangen, Beijing, China) according to the manufacturer’s instructions. Three biological replicates and four technical replicates were performed for each experiment. The relative expression levels of target genes were determined using the 2^–ΔΔCT method and normalized to the internal control ribosomal protein L32 (RPL32) gene (GenBank accession no. AB180441). The statistical significance of differences between groups were analyzed by one-way ANOVA with Duncan’s test (P < 0.05).

RNA interference.

RNA interference (RNAi) of PxMAP4K4 and PxJun was carried out to investigate the regulatory relationships among the PxMAP4K4, PxJun, and PxABCB1 genes and to explore whether the PxJun gene is involved in Cry1Ac resistance in P. xylostella. The cDNA fragments of PxMAP4K4, PxJun, or EGFP used as the templates for subsequent dsRNA synthesis were amplified by gene-specific dsRNA primers containing a T7 promoter on the 5′ end (see Table S3). The gene-specific primer set used to produce dsRNA of PxJun was designed for the 5′-terminal gene-specific region and not for the 3′-terminal conserved bZIP_Jun region to avoid potential off-target effects, and we could not detect any specific hits to other Jun genes by BLASTN search of GenBank (https://www.ncbi.nlm.nih.gov/) or the P. xylostella genome database (LepBase: http://ensembl.lepbase.org/Plutella_xylostella_pacbiov1/Info/Index), further confirming the specificity of the selected dsRNA fragment. PxMAP4K4, PxJun, or EGFP dsRNA was synthesized using the T7 RiboMAX Express RNAi system (Promega) as indicated by the manual.

The protocol for RNAi experiments in the early third-instar larvae of the resistant NIL-R strain was conducted by dsRNA microinjection as mentioned earlier (54). Thirty larvae were microinjected with buffer, dsEGFP (300 ng), dsPxJun (300 ng), or dsPxMAP4K4 (300 ng) for each treatment, and three biological repeats were performed. The injected larvae were reared and subjected to qPCR analysis at different times. In addition, toxicity bioassays were carried out at 48 h postinjection. The larvae were fed cabbage containing 1,000 mg/liter Cry1Ac protoxin (the lethal concentration causing 10% mortality [LC₁₀] for NIL-R larvae) for 72 h to calculate larval mortality as described previously (14, 52). The significance of the differences between the treatment and control groups were determined by one-way ANOVA with Duncan’s test (P < 0.05).

Data availability.

The full-length cDNA sequence of the cloned PxJun gene in this study has been deposited in the GenBank database under accession number MW446637.