Structure-function studies on Bacillus thuringiensis Cry toxins active against Aedes aegypti

The Aedes aegypti mosquito, a carrier of various arboviruses, is spreading to new regions due to increased human activity, global travel, and climate change. Controlling the Aedes vector with environmentally safe methods, increased efforts to mitigate climate change, and policies enacted for travel across borders are paramount. Bacillus thuringiensis (Bt) is a gram-positive spore-forming bacterium that produces crystalline inclusions composed of proteins known as delta endotoxins and has been effective in mosquito control, but the demand for biodegradable insecticides requires more potent Bt toxins. Delta endotoxins comprise many multigenic families, one of which is the Cry toxin. Cry2Aa is a subset of the Cry family and has a dual specificity to lepidopterans and dipterans. Its domain I has been implicated in its specificity to Aedes as opposed to other Cry toxins whose specificity has been linked to domains II and III. The primary research question we attempted to answer in this study is to understand if there is something fundamentally different about the mechanism of action of Aedes active toxins that involves domain I in specificity determination. Previous studies have suggested that four amino acids E/RTD were necessary for Cry2Aa’s Aedes activity, and that the RTD were needed for proteolytic cleavage. We investigated the role of the E27 amino acid residue by mutating E27 to a range of other amino acids. We were able to establish that E27 amino acid residue was in fact not an obligatory residue of the quadriad for Aedes activity. The correlation between structure at the N-terminal region comprising the RTD amino acid sequence and Aedes activity was investigated. Multiple sequence alignment and structural comparison of Cry2Aa’s N-terminus with other Cry toxins used in this study showed no clear association between structure and Aedes activity. To test whether the N-terminus would confer Aedes activity to a Cry toxin lacking in the activity, we constructed a hybrid by domain swapping. The lepidopteran specific Cry1Ac has no natural activity against Aedes aegypti. Attempts were made to add Aedes activity to Cry1Ac via a domain I swap with Cry2Aa. The Cry2Aa-1Ac hybrid was characterised and purified, and expression was verified by SDS-PAGE analysis. The 130kDa hybrid protein was soluble in Na2CO3 and, on trypsin activation, produced a 65kDa band. However, no toxicity was observed when the hybrid was tested against Aedes larvae. Results obtained from this study suggest that domain I of Cry2Aa alone is insufficient to confer Aedes activity to a Cry toxin lacking the activity. Site-directed mutagenesis was used to probe the function of surface-exposed amino acid residues in the loop 2 region of Cry2Aa for possible involvement of domain II in Aedes activity. Single alanine substitutions were carried out on the highly conserved 383DRE385 amino acid residue in Cry2Aa’s loop 2 region. D383A, R384A and E385A were all expressed like wild-type Cry2Aa. When a bioassay was conducted to assess their toxicity against Aedes larvae, only the R384A mutant showed a decreased toxicity, an indication of the possible involvement of Cry2Aa’s domain II in Aedes activity. Finally, we attempted to determine whether Cry2Aa binds differently from other Aedes active toxins via competition feeding assays using a disabled insecticidal protein (DIPs). Based on the results obtained from this study, a binding model of Cry2Aa towards the Aedes aeyppti larvae was proposed in which Cry2Aa shares a binding site with Cry1Ca, Cry2Ab, Cry4Aa and Cry11Aa but also has a separate site not shared by these toxins. We speculate that this separate binding site may be associated with the involvement of Cry2Aa domain I in Aedes activity.