In the previous sections, we described the factors that govern the stability of proteins and then outlined a framework for interpreting the energetics of protein toxin con-formational flexibility in structural terms. In thinking of the life cycle of a protein toxin, a couple of points appear to be critical in the framework of conformational changes and their energetics — cellular localization and formation of the toxic entity. Indeed, looking at Table 2.1, these two steps are quite active areas of protein toxin research. In particular, the tools of molecular biology can be brought to bear on identifying specific residue interactions that may play the key role in modulating the conformational changes necessary for forming toxic conformations.
Of the two types of interactions responsible for protein stability discussed above, hydrogen bonds can provide specific interactions that may be turned on or off depending upon environmental conditions. A common theme emerges wherein changes in pH initiate membrane insertion. This can be illustrated by a couple of well-known examples. Both diphtheria toxin T-domain and protective antigen (PA) from Bacillus anthracis have been shown to insert into target membranes in response to a drop in cellular pH. We have recently shown that a conserved hydrogen-bonded interaction between side-chains between helix 5 and helix 6 in the d -endotoxins from Bacillus thuringiensis may serve as a pH-dependent switch that controls membrane insertion.18 This hydrogen bond between a histidine side-chain and a tyrosine side-chain contributes directly to the overall stability of the toxin, as mutation of either residue results in an inactive protein that is highly susceptible to proteases. Mutation also results in significant destabilization of the toxin as judged by chemical denaturation. The conservation of histidine at one of these positions is significant because this side-chain titrates over the physiological pH range. Protonation of the histidine side-chain in response to pH changes will break this hydrogen bond, allowing the more flexible helix 5 to insert into a membrane. In the case of d-endotoxins, the tyrosine side-chain may also titrate at physiological pH ranges since the insect gut pH of lepi-dopteran insects can be as high as 10.5. Deprotonation of this tyrosine hydroxyl will also break the hydrogen bond with histidine causing insertion into the membrane.
These pH-driven conformational changes resulting in membrane insertion are not exclusive to bacterial protein toxins. Another well-characterized example is the pH-driven insertion of viral hemagglutinin, which then results in the fusion of viral membrane to the target cellular membrane. Observing pH-driven conforma-tional changes in viral membrane fusion as well as bacterial protein toxin membrane interactions suggests that protein systems in general may have evolved specific amino acid interactions that take advantage of pH differences resulting in protein activity.
The previous examples highlight the importance of specific hydrogen-bonded interactions in switching toxins from inactive to active states. On the other hand, as discussed in the introductory section above, hydrophobic interactions can provide little specificity that would function like hydrogen bonds. However, the shielding of hydrogen bonds from solvent by hydrophobic interactions can result in a stronger hydrogen bond. Combining these two ideas, we propose that hydrophobic interactions surrounding hydrogen bonds that are important for conformational switching may modulate the strength of the hydrogen bond, thereby exerting indirect control over the specific switching behavior.
In the context of our framework, using the tools of molecular biology and biophysics, regions of the protein that undergo conformational changes observed during the life cycle of the protein can be identified. Once these smaller regions of the protein have been identified, a more detailed dissection of the interactions at the interface of the flexible regions can be performed to identify specific residues that may potentially contribute to both structure and stability of the protein states for that stage of the life cycle. Identification of these specific residue types can then lead to specific hypotheses about the types of factors that can exert some control over the necessary conformational changes for that step in the mode of action, i.e., pH-controlled hydrogen bonds. This approach should be useful in identifying the steps in the life cycle of protein toxins; the conformational changes necessary for the toxic action of the protein; and the specific interactions that control the conditions for the formation of the toxic entity. Painting a picture at that level of detail for a bacterial toxin of interest will allow for both the specific modulation of protein activity as well as the understanding of conditions that will allow for the safe use of that protein in crop protection or as a therapeutic agent.
Was this article helpful?