Transgenic crops that produce Bacillus thuringiensis (Bt) proteins for pest control are grown extensively, but insect adaptation can reduce their effectiveness. Established mode of action models assert that Bt proteins Cry1Ab and Cry1Ac are produced as inactive protoxins that require conversion to a smaller activated form to exert toxicity. However, contrary to this widely accepted paradigm, we report evidence from seven resistant strains of three major crop pests showing that Cry1Ab and Cry1Ac protoxins were generally more potent than the corresponding activated toxins. Moreover, resistance was higher to activated toxins than protoxins in eight of nine cases evaluated in this study. These data and previously reported results support a new model in which protoxins and activated toxins kill insects via different pathways. Recognizing that protoxins can be more potent than activated toxins against resistant insects may help to enhance and sustain the efficacy of transgenic Bt crops.

BACKGROUND: Transgenic crops that make insecticidal proteins from Bacillus thuringiensis (Bt) have revolutionized management of some pests. However, evolution of resistance to Bt toxins by pests diminishes the efficacy of Bt crops. Resistance to crystalline (Cry) Bt toxins has spurred adoption of crops genetically engineered to produce the Bt vegetative insecticidal protein Vip3Aa. Here we used laboratory diet bioassays to evaluate responses to Vip3Aa by pink bollworm (Pectinophora gossypiella), one of the world's most damaging pests of cotton.RESULTS: Against pink bollworm larvae susceptible to Cry toxins, Vip3Aa was less potent than Cry1Ac or Cry2Ab. Conversely, Vip3Aa was more potent than Cry1Ac or Cry2Ab against laboratory strains highly resistant to those Cry toxins. Five Crysusceptible field populations were less susceptible to Vip3Aa than a Cry-susceptible laboratory strain (APHIS-S). Relative to APHIS-S, significant resistance to Vip3Aa did not occur in strains selected in the laboratory for > 700-fold resistance to Cry1Ac or both Cry1Ac and Cry2Ab.CONCLUSIONS: Resistance to Cry1Ac and Cry2Ab did not cause strong cross-resistance to Vip3Aa in pink bollworm, which is consistent with predictions based on the lack of shared midgut receptors between these toxins and previous results from other lepidopterans. Comparison of the Bt toxin concentration in plants relative to the median lethal concentration (LC 50 ) from bioassays may be useful for estimating efficacy. The moderate potency of Vip3Aa against Cry1Ac-and Cry2Ab-resistant and susceptible pink bollworm larvae suggests that Bt cotton producing this toxin together with novel Cry toxins might be useful as one component of integrated pest management.