Deamidation of asparagine residues is one of the most common post translational modifications (PTMs) that can lead to chemical degradation of therapeutic antibodies and antibody-drug conjugates (ADCs). This non-enzymatic degradation can occur during fermentation and purification of the drug substance, as well as processing of the drug product, and eventual storage with potential for significant impact to potency and stability. Given the importance of controlling deamidations throughout the biopharmaceutical industry, many attempts have been made to develop predictive algorithms for deamidation from protein sequence. To date, these algorithms have had poor predictive power, and experimental deamidation identification is required.
In this study, we investigate the deamidation susceptibility of asparagine residues of Trastuzumab under different conditions. Proteins were placed into one of five buffers at different pH values (11 conditions in total) and then stored at temperatures ranging from 4-50 °C. Deamidation was evaluated by LC-MS/MS after samples were digested with trypsin. Analysis of the data reveals that different deamidation sites show variable responses to increasing pH, resulting in varying deamidation product profiles. The experimental data is complemented with kinetic modeling and molecular dynamics simulations to present hypotheses linking pH effects on hotspot local environment to deamidation rate. This work will provide fundamental knowledge to guide protein design or mitigation approaches for chemical hotspot liabilities.
