ABCC transporters mediate insect resistance to multiple Bt toxins revealed by bulk segregant analysis

Abstract

Background: Relatively recent evidence indicates that ABCC2 transporters play a main role in the mode of action of Bacillus thuringiensis (Bt) Cry1A-type proteins. Mapping of major Cry1A resistance genes has linked resistance to the ABCC2 locus in Heliothis virescens, Plutella xylostella, Trichoplusia ni and Bombyx mori, and mutations in this gene have been found in three of these Bt-resistant strains.

Results: We have used a colony of Spodoptera exigua (Xen-R) highly resistant to a Bt commercial bioinsecticide to identify regions in the S. exigua genome containing loci for major resistance genes by using bulk segregant analysis (BSA). Results reveal a region containing three genes from the ABCC family (ABBC1, ABBC2 and ABBC3) and a mutation in one of them (ABBC2) as responsible for the resistance of S. exigua to the Bt commercial product and to its key Spodoptera-active ingredients, Cry1Ca. In contrast to all previously described mutations in ABCC2 genes that directly or indirectly affect the extracellular domains of the membrane protein, the ABCC2 mutation found in S. exigua affects an intracellular domain involved in ATP binding. Functional analyses of ABBC2 and ABBC3 support the role of both proteins in the mode of action of Bt toxins in S. exigua. Partial silencing of these genes with dsRNA decreased the susceptibility of wild type larvae to both Cry1Ac and Cry1Ca. In addition, reduction of ABBC2 and ABBC3 expression negatively affected some fitness components and induced up-regulation of arylphorin and repat5, genes that respond to Bt intoxication and that are found constitutively up-regulated in the Xen-R strain.

Conclusions: The current results show the involvement of different members of the ABCC family in the mode of action of B. thuringiensis proteins and expand the role of the ABCC2 transporter in B. thuringiensis resistance beyond the Cry1A family of proteins to include Cry1Ca.

Figure 1

Schematic representation of the BSA index for the SNPs in the S. exigua unigenes with orthologs into the silkworm genome. Each data point (from 0 to 1) is the average index for three consecutive orthologs calculated using overlapping rolling-windows (BSA index for each unigene individually is provided in Additional file 3). Average index >0.60 are painted in red. BSA, bulk segregant analysis; SNPs, single nucleotide polymorphism.

Figure 2

Phylogenetic analysis and ortholog asignment of S. exigua ABCC transporters. Neighbor-joining consensus tree of representative members of ABC transporters from Lepidoptera and Tribolium castaneum (Coleoptera). Nomenclature for each member is based on the name of the species (Bmo: Bombyx mori; Se: Spodoptera exigua; Dpe: Danaus plexippus; Bma: Bombyx mandarina; Hvi: Heliothis virescens; Hsu: Heliothis subflexa; Mse: manduca sexta; Tca: Tribolium castaneum) followed by their accession number from the National Center for Biotechnology Information (NCBI), Silkdb and Manduca Base (Agripest) databases. Sequences were selected based on their homology to the ABCC1-3 transporters from the linkage region in B. mori. Some homologs in T. cataneum were selected as a kind of outgroup. For the sake of clarity bootstrap values are only reported for the branches identifying the main ABCC subgroups. NCBI, National Center for Biotechnology Information.

Figure 3

Localization of the deletion in resistant insects into the ABCC2 gene genomic region (A) schematically represented at the genomic region (B) and on the hypothetical protein (C). Sequences were obtained by PCR amplification of the reported region using genomic DNA from the Xen-R and FRA strains. ‘E’ labels are indicative of the predicted exons in the reported region. The dashed line indicates the deletion in the Xen-R insects. PCR, polymerase chain reaction.

Figure 4

Expression patterns of SeABCC2 or SeABCC3 gene in different developmental stages (A) and in different tissues (B) of the fifth instar larvae in S. exigua. ‘E’, ‘L1 – L5’, ‘P’, and ‘A’ represent egg, larval instars, pupa, and adult, respectively. ‘FB’, ‘HC’, ‘NV’, and ‘SG’ represent fat body, hemocytes, nerve, and salivary gland, respectively. Expression of β-actin confirms the integrity of cDNA preparation. Toxicity comparison of Cry1Ac and Cry1Ca protoxins against third instar larvae of S. exigua(C). Each bioassay was conducted with ten larvae per replicate and three replicates per concentration. Mortality was scored at five days post-treatment. Suppression of SeABCC2 or SeABCC3 gene expression using a specific double-stranded RNA (dsSeABCC2 or dsSeABCC3) in the whole body of S. exigua larvae at different periods (D and E, respectively). SeABCC2 or SeABCC3 expression was analyzed by qRT-PCR. Each treatment was independently replicated three times. Different letters above standard deviation bars indicate significant differences among means at Type I error = 0.05 (LSD test). The effect of suppression of SeABCC2 or SeABCC3 gene expression on susceptibility of the third instar S. exigua to Cry1Ac or Cry1Ca protoxins (F). A viral gene (ORF302)-specific dsRNA (dsCON) was used for a dsRNA control. At 24 hours after dsRNA treatment, the larvae were exposed to Cry1Ac (10 μg/cm²) or Cry1Ca (1 μg/cm²) protoxins treated by the leaf-dipping method. Mortality was assessed at five days after Cry toxin treatment. Each treatment was replicated three times. Each replication used 10 larvae. Different letters above standard deviation bars indicate significant difference among means at Type I error = 0.05 (LSD test). dsRNA, double-stranded RNA; LSD, least significant difference; qRT-PCR, quantitative reverse transcriptase-polymerase chain reaction.

Figure 5

Binding of ¹²⁵I-labeled Cry1Ca to S. exigua BBMV. (A) Homologous competition. ¹²⁵I-labeled Cry1C was incubated with BBMV from susceptible (FRA) or resistant (Xen-R) insects in the presence of increasing concentrations of unlabeled Cry1C toxin. (B) Dissection of total binding into their reversible and irreversible components. Contribution of both components to the binding of ¹²⁵I-labeled Cry1Ca to BBMV from susceptible and resistant insects were compared. T-test (P = 0.7244 for the reversible component) (P = 0.0170 for the irreversible component). BBMV, brush border membrane vesicles.

Figure 6

Effect of suppression of SeABCC2 or SeABCC3 gene expression on pupation (A, B, and C) and adult emergence (D) of S. exigua. The expression of SeABCC2 or SeABCC3 gene was suppressed by feeding dsSeABCC2 or dsSeABCC3 to newly molted second instar larvae. A viral gene (ORF302)-specific dsRNA (dsCON) was used for a dsRNA control. Each treatment was replicated three times. (A) Pupation, (B) pupal weight, (C) pupal duration and (D) adult emergence after dsRNA feeding were measured. Each replicate used 10 larvae. Different letters above standard deviation bars indicate significant differences among means at Type I error = 0.05 (LSD test). dsRNA, double-stranded RNA, LSD, least significant difference.