Rational Design and Application of Antibodies for Amylin and Amyloid Beta

Abstract

Misfolded proteins and polypeptides such as amylin (islet amyloid polypeptide, IAPP) and amyloid beta are associated with numerous age-related pathological conditions such as Alzheimer's, Parkinson's, and type II diabetes, but effective treatments for these diseases are currently lacking. Antibodies have great potential as tools for research, diagnostics, and therapeutics of these diseases due to their high specificity and affinity for targets. However, traditional antibody discovery methods often fail to produce antibodies with desirable properties such as epitope selectivity, conformational stability, and solubility. Computational antibody design methods combining epitope targeting, solubility calculation and specificity prediction may help overcome these challenges. In this thesis, we used a computational antibody design called cascade method to successfully increase the potency of a single domain antibody by over 200-fold against Aβ42 aggregation. Utilizing display techniques and chemical modification can further enhance the binding affinity and specificity of these antibodies, making them a more viable treatment option. However, the current monoclonal and single domain antibody scaffolds contain conserved disulfide bonds that can misform, resulting in a loss of activity, formation of oligomers and precipitates, and necessitating strict expression conditions. This presents a challenge in the application of display techniques and chemical modification to these antibody scaffolds. Therefore, we engineered the current single domain antibody and an antibody-like protein scaffold to overcome these difficulties and make them compatible with our computational design binding loops. We then used these engineered antibody scaffolds to generate antibody libraries for scanning different epitopes of amylin with binding affinities ranging from hundreds of nM to low μM. These antibodies allow us to identify IAPP aggregates in serum samples of type II diabetes patients, which may provide a powerful tool for the study and diagnosis of type II diabetes in the future.