Transcriptional regulation and overexpression of GST cluster enhances pesticide resistance in the cotton bollworm, Helicoverpa armigera (Lepidoptera: Noctuidae)

Abstract

The rapid evolution of resistance in agricultural pest poses a serious threat to global food security. However, the mechanisms of resistance through metabolic regulation are largely unknown. Here, we found that a GST gene cluster was strongly selected in North China (NTC) population, and it was significantly genetically-linked to lambda-cyhalothrin resistance. Knockout of the GST cluster using CRISPR/Cas9 significantly increased the sensitivity of the knockout strain to lambda-cyhalothrin. Haplotype analysis revealed no non-synonymous mutations or structural variations in the GST cluster, whereas GST_119 and GST_121 were significantly overexpressed in the NTC population. Silencing of GST_119 or co-silencing of GST_119 and GST_121 with RNAi significantly increased larval sensitivity to lambda-cyhalothrin. We also identified additional GATAe transcription factor binding sites in the promoter of NTC_GST_119. Transient expression of GATAe in Hi5 cells activated NTC_GST_119 and Xinjiang (XJ)_GST_119 transcription, but the transcriptional activity of NTC_GST_119 was significantly higher than that of XJ_GST_119. These results demonstrate that variations in the regulatory region result in complex expression changes in the GST cluster, which enhances lambda-cyhalothrin resistance in field-populations. This study deepens our knowledge of the evolutionary mechanism of pest adaptation under environmental stress and provides potential targets for monitoring pest resistance and integrated management.

Fig. 1. Insecticide monitoring results of cotton bollworm in North China and Xinjiang from 2008 to 2019.

PX phoxim, LC lambda-cyhalothrin, EB emamectin benzoate. Data source: National Agro-Tech Extension and Service Center (www.natesc.org.cn). White boxes represent no monitoring data.

Fig. 2. GST clusters under selective sweep in North China population.

a Selection signs on chromosome 23 (2.71–3.01 Mb), including the GST cluster. The putative sweep region was validated by θ_π, θ_w and Tajima’s D test. Positions of hard/soft sweep classification of the GST cluster were inferred by LASSI analyses. The peak point of the T value and its corresponding m value are marked by vertical dashed lines. Selective sweeps were classified as hard when m = 1 (marked by horizontal dashed lines) and soft when m > 1. Gene annotations in the sweep region are indicated at the bottom. b Haplotypes of six GST genes in the GST cluster. The genotype of the samples was categorized into two haplotypes (North China and Xinjiang). The presence of the wild allele and mutation allele is shown in blue and red, respectively. The white color symbolizes missing information. c Local LD heatmap of GST cluster region in Xinjiang, gene models, and local LD heatmap of GST cluster region in North China. The block of the GST cluster is marked with a red arrow.

Fig. 3. Expansion of the GST cluster.

a Maximum likelihood analysis of GST genes using 54 homologous genes. The homologous gene sequences were downloaded from NCBI, including Bombyx mandarina, Bombyx mori, Manduca sexta, Arctia plantaginis, Trichoplusia ni, Spodoptera frugiperda, Spodoptera litura, Spodoptera littoralis, Chrysodeixis includens, Spodoptera exigua, Manduca sexta, Plodia interpunctella, Amyelosis transitella, Plutella xylostella and Helicoverpa zea. Bootstrap support for representative nodes is shown. b Conserved syntenic genomic region containing the GST cluster in H. armigera and the ancestor species B. mori. Genes belonging to the same orthologous groups are connected by lines. Relative gene expression levels of GST genes in different developmental stages and tissues of H. armigera are shown using FPKM values.

Fig. 4. GST cluster genetic linkage with lambda-cyhalothrin resistance.

a Single-pair crosses between XJ and NTC virgin adults produced families of hybrid F₂ offspring. Progeny from F₂ families was exposed to diagnostic concentrations of phoxim (PX, 0.64 μg/cm²), lambda-cyhalothrin (LC, 2.0 μg/cm²), and emamectin benzoate (EB, 0.001 μg/cm²). Ninety-six survivors of each treatment were selected for linkage analysis. b Molecular marker that distinguished between the XJ and NTC populations. c Number of survivors with different genotypes. χ² test, based on the hypothesis of the random assortment ratio rr:rs:ss = 1:2:1. rr homozygote site in NTC, ss homozygote site in XJ, rs heterozygotes. d P values of three insecticides based on the χ² test.

Fig. 5. GST cluster knockout using the CRISPR/Cas9 system.

a Schematic diagram of sgRNA positions (sgRNA1 and sgRNA2), three pairs of detecting primers (KO_F/R, 117 F/R and 120 F/R), sgRNA target sequences and the PAM sequences (red) and chromatogram of the mutation type, a 26 kb deletion. The cutting sites of the Cas9 protein are indicated with red triangles. b Genotyping of individuals for deletion of the GST cluster according to the banding patterns of the PCR products amplified with three primer pairs. HO homozygote, He heterozygote, WT wild type. c Bioassay results of field-resistant NTC strain, GST-KO strain, and susceptible 96 S strain treated by three insecticides lambda-cyhalothrin, phoxim, and emamectin benzoate. The GST-KO strain was derived from the NTC strain by knocking out the GST cluster. Resistance ratios were calculated as the LC₅₀ of NTC/LC₅₀ of 96 S and the LC₅₀ of GST-KO/LC₅₀ of 96 S. Asterisks indicate the resistance ratios that were significantly different between the two strains based on their fiducial limits of LC₅₀ values, which did not overlap. na, no significant change.

Fig. 6. Mutations in the promoter of GST_119 enhance the expression of this GST gene, resulting in resistance in the NTC population.

a Relative expression of each GST gene in the resistant (NTC, n = 3) and susceptible strain (XJ, n = 3) of cotton bollworm, as determined by RT-qPCR. Error bars represent standard deviation. A significant difference in expression between the susceptible and resistant strains is indicated using an asterisk (Student’s t test, **, P < 0.01). na means that no expression was detected. b Effects of injecting dsRNA on the mortality of larvae treated with lambda-cyhalothrin (n = 3). dsGFP (n = 3) was used as a control. Each third-instar larva was injected with 1.5 μg dsRNA. Twenty-four larvae were used as a treatment, with three replicates. The concentration of lambda-cyhalothrin was 100 μg/mL. Larval mortality rates were recorded 3 days after treatment. The P value on the error bar indicates a significant difference analyzed by Student’s t test. Error bars represent standard deviation. c Prediction of transcription factor binding sites in the promoter region of GST_119 from North China population (NTC) and Xinjiang population (XJ). The nucleotides are numbered relative to the translation start site (ATG), with the sequence upstream of it preceded by “-”. The transcription factor binding sites are shown in different colors. GATAe binding sites are represented by vertical lines. (d) Dual-luciferase assay. The pGL3 GST_119 and GST_121 promoter constructs of the NTC or XJ population were transfected with GATAe constructs into sf9 cells. The cells were collected 48 h after transfection, and luciferase activities were measured (n = 3). Error bars represent standard deviation. A significant difference in the activity of luciferase between the NTC and XJ populations is indicated using asterisks (Student’s t test, **, P < 0.01). na, no significant changes.

Fig. 7. Schematic of the promoter variation leading to overexpression of GST genes, resulting in insecticide resistance in a natural field population.

In the naturally resistant population from North China, the GST cluster was positively selected. Mutation of the promoter enhances the expression of the GST gene, leading to insecticide resistance.