Retrotransposon-mediated disruption of a chitin synthase gene confers insect resistance to Bacillus thuringiensis Vip3Aa toxin

Abstract

The vegetative insecticidal protein Vip3Aa from Bacillus thuringiensis (Bt) has been produced by transgenic crops to counter pest resistance to the widely used crystalline (Cry) insecticidal proteins from Bt. To proactively manage pest resistance, there is an urgent need to better understand the genetic basis of resistance to Vip3Aa, which has been largely unknown. We discovered that retrotransposon-mediated alternative splicing of a midgut-specific chitin synthase gene was associated with 5,560-fold resistance to Vip3Aa in a laboratory-selected strain of the fall armyworm, a globally important crop pest. The same mutation in this gene was also detected in a field population. Knockout of this gene via CRISPR/Cas9 caused high levels of resistance to Vip3Aa in fall armyworm and 2 other lepidopteran pests. The insights provided by these results could help to advance monitoring and management of pest resistance to Vip3Aa.

Fig 1. Mapping Vip3Aa resistance in the S. frugiperda genome.

(A) Manhattan plot from BSA showing a ΔSNP index peak in chromosome 1 reflecting a high proportion of alleles associated with resistance to Vip3Aa. (B) Chromosome 1 with ΔSNP index highest from 6.56 to 10.92 Mb (blue shading). The solid line indicates the LOESS regression fitted ΔSNP index value. The dashed red line in (A) and (B) indicates the threshold for the top 1% of values for the ΔSNP index. (C) Fine-scale mapping shows complete linkage with markers at 8.17 and 8.62 Mb. Only the last 4 digits are shown for the 8 markers from 6.56 to 10.92 Mb. The full name for each marker starts with LOC11827. (D) Genes from 8.14 to 8.64 Mb including chitin synthase 2 (SfCHS2). The data underlying this figure can be found in S1 Data and https://doi.org/10.5281/zenodo.11395059. BSA, bulked segregant analysis; SNP, single nucleotide polymorphism.

Fig 2. Retrotransposon insertion associated with resistance to Vip3Aa and reduced abundance of wild-type SfCHS2 transcripts.

(A) Exon 21 to 22 of SfCHS2 in the SS (susceptible) and Sfru_R3 (Vip3Aa-resistant) strains showing the LTR retrotransposon (Yaoer) insertion. (B) Yaoer with LTR regions at the 5′ and 3′ termini and the intervening region encoding AP, RT, ribonuclease H (RNase H), and integrase (INT). (C) SfCHS2 genotype of F₇ Vip3Aa-susceptible (F₇-S) and resistant (F₇-R) larvae determined using gDNA from individual larvae and primers shown in (A). The e21-F/e22-R primers flanking intron 21 produced a strong band indicating wild-type from all 16 F₇-S larvae and no F₇-R larvae. The positive control primers (e21-F/e21-R) in exon 21 produced a strong band in all F₇-S and F₇-R larvae. The molecular weight marker containing DNA with length of 100, 250, 500, 750, 1,000, and 2,000 bp was used for agarose gel electrophoresis analysis. (D) Primers for analyzing relative transcript abundance of SfCHS2 via RT-qPCR. Primers SfCHS2-q-F1/R1 in exons 20 and 21 evaluated total transcript abundance of SfCHS2 in SS versus Sfru_R3 (E) and individuals from F₇-S (F) versus F₇-R (G). Primers SfCHS2-q-F2/R2 flanking intron 21 evaluated wild-type transcript abundance of SfCHS2 in SS versus Sfru_R3 (H) and individuals from F₇-S (I) versus F₇-R (J). Transcript abundance of SfCHS2 in Sfru_R3 and individuals from F₇-S and F₇-R was normalized to the fold value of 2^-ΔCt relative to SS. Bars for SS and Sfu_R3 in (E) and (H) show mean relative transcript abundance ± SEM. Based on t tests, NS in (E) indicates no significant difference between strains in total transcript abundance and ** in (H) indicates SS greater than Sfu_R3 (P < 0.01, see text for details). The data underlying this figure can be found in S2 Data. AP, aspartate protease; LTR, long terminal repeat; RT, reverse transcriptase; SS, susceptible strain.

Fig 3. Knockout of SfCHS2 via CRISPR/Cas9 confers resistance to Vip3Aa in S. frugiperda.

(A) CRISPR/Cas9-mediated double sgRNA system and various types of mutations in G1 larvae identified through sequencing of individual PCR clones. Deleted bases are indicated as red dashes and inserted bases are indicated as red letters. The CRISPR target sites and the number of deleted and inserted bases (+, insertion;–, deletion) are shown. The chromatogram shows the sequence of the mutant isolated from a homozygous knockout larva in G2. (B) Log Vip3Aa concentration–response curves for SS, SfCHS2-KO-A, and the progeny of reciprocal crosses between SfCHS2-KO-A and Sfru_R3. (C) SfCHS2-KO-A did not show strong cross-resistance to Cry1Ab or Cry1Fa. The EC₅₀ did not differ significantly between strains for Cry1Ab (SfCHS2-KO-A: 3.2 [95% CI = 2.0–6.8] versus SS: 1.5 [1.0–2.1]; RR = 2.1) or Cry1Fa (SfCHS2-KO-A: 0.016 [0.013–0.019] versus SS: 0.019 [0.014–0.024], RR = 0.8), n = 168 larvae tested in each of the 4 bioassays. The data underlying this figure can be found in S3 Data and S9 Data. CI, confidence interval; RR, resistance ratio; SS, susceptible strain.