Investigation of multi-modal high-salt binding ion-exchange chromatography using quantitative structure-property relationship modeling.

A quantitative structure-property relationship (QSPR) modeling approach was employed to correlate the physicochemical properties and structural components of multi-modal ion-exchange ligands with their ability to bind proteins under high-salt conditions. These ion-exchange ligands contain various substructures, which may contribute to secondary interactions that promote protein binding. A set of molecular descriptors was calculated based on the structures of these cation-exchange ligands. The molecular descriptors used to generate the QSPR models were used to characterize the ligand molecules and QSPR models were generated for predicting the elution conductivity of three test set proteins. The resulting models were able to predict the performance of test ligands not included in the generation of the models and the interpretation of the selected descriptors provided insight into the important physicochemical properties and structural characteristics required for protein binding under high-salt conditions. The results indicate that while the aromatic ring plays an important role in promoting protein binding under high-salt conditions, moieties associated with intermediate hydrophobicity (e.g. aliphatic side chain) or the presence of hydrogen bond donors (e.g. NH and OH) tended to suppress the binding. Further, regions of the ligands with negative partial charge also tended to promote protein binding at high-salt conditions in these multi-modal cation-exchange systems. This work demonstrates that ligand-based QSPR models may have utility for the a priori design of mixed mode chromatographic systems for protein separations.