Modeling and prediction of protein solubility using the second osmotic virial coefficient

The development of a precipitation or crystallization step requires knowing the solubility of the target protein and its crystallization behavior in aqueous solutions at different pH, temperatures and in the presence of precipitating agents, especially salts. Within this work, a solubility model for proteins based on the second osmotic virial coefficient B22 is developed. For this, a relation between protein solubility and B22 was combined with the extended DLVO model. This solubility model was then used to model and also predict the protein solubility of lysozyme and monoclonal antibody for different salts, salt concentrations, and pH. The modeled as well predicted B22 and protein solubility data of lysozyme in the presence of sodium chloride and sodium p-toluenesulfonate and of a monoclonal antibody in the presence of ammonium sulfate at different pH shows good agreement with experimental data.