Fundamental questions in optimizing ion-exchange chromatography of proteins using computer-aided process design

The major objectives for preparative protein chromatography are maximal loading and increased flow rate while maintaining defined resolution. Conventionally a series of chromatographic experiments are performed and the optimal conditions are selected according to the separation criteria. Computer-aided process design uses the same strategy, except a group of related experiments are generated by computer simulation. The access to concrete separation parameters for valid simulation necessitates chromatographic experiments. Optimal conditions are determined in the same manner as conducted in the conventional strategy. Beside other parameters, the distribution coefficient (K) determines the performance of a chromatographic purification under overloading conditions. In ion-exchange chromatography the distribution coefficient is strongly influenced by the protein concentration (C) and the salt concentration (I). A strategy to derive the distribution coefficient from chromatographic experiments, such as isocratic runs (pulse response), linear gradients, and frontal analysis, is described and compared to previously published strategies. In ion-exchange chromatography, the number of plates and transfer units change with the salt concentration. The distribution coefficient for salt also changes under various conditions including salt and protein concentration. The number of plates and transfer units also vary with the flow rate. Furthermore criteria such as the multicomponent situation require a more complex mathematical treatment. Several solutions have been validated to circumvent those obstacles.