Designing glycopeptide inhibitors of multivalent pathogen interactions

Multivalent interactions between proteins and carbohydrates are critical in various biological processes and are frequently exploited by pathogens to invade host cells. The AB5 family of toxins produced by a variety of disease-causing bacteria leverages these interactions for cell entry. Perhaps the most well studied of the AB5 toxins is Cholera toxin (CT), produced by the gram-negative bacteria Vibrio cholerae. The pentameric B subunit of CT (CTB) contains two carbohydrate binding sites on each subunit, which interact with carbohydrates on the surface of host cells to initiate cell entry and cause disease. Many multivalent inhibitors have been designed to inhibit CTB-host interactions, but these molecules are often large, complex, and costly to make, rendering them unsuitable candidates for therapeutic development. This thesis proposes a novel CTB inhibitor consisting of a linear peptide spacer bearing GM1 and Lewis-Y ligand moieties for hetero-bivalent binding to CTB's carbohydrate binding sites. The design approach focuses on identifying an optimal peptide spacer sequence by predicting entropic and enthalpic factors that contribute to bivalent binding affinity through multiscale molecular simulations. Predictions made here suggest that a ligand containing a rigid peptide spacer is likely to achieve the highest binding affinity to CTB. However, a flexible spacer with a short linear interaction motif also shows promise in achieving comparable affinity levels. This work lays the foundation for developing a new class of effective and simplified inhibitors targeting multivalent interactions in AB5 toxins, potentially revolutionising therapeutic strategies against a variety of diseases.