A protein surface properties perspective on multimodal chromatography : from fundamental understanding to predictive tools

Abstract

During downstream process development of proteins, high throughput screening is conduct- ed to identify conditions for multimodal steps due to the difficulty of identifying important binding regions and predicting selectivity a priori. This work addressed these challenges by connecting protein surface properties to protein behavior in multimodal systems to increase understanding of ligand-protein interactions. This knowledge informed the creation of predictive models to facilitate in silico process development of multimodal steps. A library of multimodal anion exchange (MM AEX) prototypes was designed to investigate the impact of aromatic ring substituents on selectivity for a diverse set of model proteins. Aliphatic prototypes showed dramatic changes in selectivity with important implications for mAb monomer-aggregate separa- tions. A quantitative structure activity relationship (QSAR) model was developed to predict protein retention in the various MM AEX resins.