The ability to identify the effects of mutations is allowing us to optimise proteins for therapeutic and biotechnological purposes.
Antibodies are becoming increasingly important in therapeutic capacity, due to their ability to bind with high specificity and affinity to an enormous variety of substances. Although many programs have been developed to predict the affinity of a protein-protein interaction, none had been specifically designed for antibodies. Antibodies rely on a unique binding mode, interactions through 6 highly variable loops, which is significantly different from general protein-protein interfaces. We hope to harness these features for modelling an antibody's affinity toward its antigen. Using graph-based signatures we have been able to identify mutations altering an antibodies binding affinity. Furthermore, we have been able to identify antibody escape mutations, which lead to reduced effectiveness of these therapies. We are now further developing this into a validated platform that can be used to guide tomorrow's antibody engineering solutions.
Other specific efforts include mapping and optimisation of antigenic epitopes, and optimisation of therapeutic antibodies to minimise aggregation and poor pharmacokinetics.
Many potential biotherapeutics are restricted in their therapeutic potential due to inherent significant limitations- including thermal and in vivo instability, immunogenicity and rapid plasma clearance. We have been overcoming these limitations through directed protein engineering, using the predictions from the mutational analysis platform to identify optimal stabilising and immune masking mutations, with minimal effect on protein activity. In particular we have been applying this to the optimisation of proteins that could be used as enzyme replacement therapies in rare genetic diseases such as Alkaptonuria and OTC deficiency. We are also trying to modify the binding properties of toxins to improve their therapeutic potential.
Optimising biotechnological processes
Using our mutational analysis pipeline we have been analysing the active sites of important industrial enzymes, such as cellulases in bioethanol production, in order to improve their efficiency, activity, stability and reduce feedback inhibition, major limitations in most biotechnological processes.
Designing peptides and nucleic acids that bind to specific proteins
We are developing novel computational methods for assessing peptide and nucleic acid binding affinity to a protein. This will be used as a basis for the development of a de novo design platform of peptides and nucleic acids to target specific proteins with high affinity and specificity. This has practical applications in the in silico identification of optimal binding motifs and the design of novel therapeutics.