Evaluation of selectivity in multimodal anion exchange systems: A priori prediction of protein retention and examination of mobile phase modifier effects

Although recent advances in multimodal chromatography have shown significant potential for selective protein purification, there is a need to establish a deeper understanding of the nature of selectivity in these systems. In this work, the adsorption behavior of a library of commercially available proteins with varying physicochemical properties was investigated. Linear gradient experiments were carried out with the multimodal anion exchanger and the results were compared to those obtained with a traditional anion exchange material. Proteins were more strongly retained on the multimodal resin and a different elution pattern was obtained as compared to anion exchange. Quantitative structure–property relationship models using a support vector regression technique were developed and the resulting models were shown to have good predictive abilities in both systems. Molecular descriptors selected during the generation of these models suggested that multiple interaction modes contributed to the stronger retention and different elution patterns observed in the multimodal system. Finally, mobile phase modifiers such as ethylene glycol, urea and arginine were shown to be able to impart unique selectivity trends in multimodal chromatography.