Productivity for free: Residence time gradients during loading increase dynamic binding capacity and productivity

Abstract

• An algorithm to improve productivity and dynamic binding capacity in protein chromatography. • A controller to optimize the flow rate gradient during loading. • The theoretical concept has been validated by chromatography experiments. • The algorithm allows a productivity increase of 68% in Protein A chromatography without additional expense. Traditional chromatographic purification steps consist of long loading steps at a constant residence time, especially during early stages of the downstream process. The breakthrough behavior of the product is usually well defined. This information can be used to optimize the loading phase by starting the loading phase at a low residence time and gradually increasing the residence time to boost productivity and resin utilization. Process modeling of the loading of lysozyme on a Toyopearl SP-650C stationary phase, a strong cation exchanger, revealed optimal residence time gradients. These gradients result in a faster breakthrough in the time domain, while having the same breakthrough as constant residence time loading, in the volume domain. Modeling also revealed that a delayed start of the gradient further increases productivity. Eventually, several mathematical optimizations were employed to ensure the optimality of the idea. We confirmed the process model results in lab experiments using lysozyme breakthrough in ion-exchange chromatography and monoclonal antibody breakthrough in protein A affinity chromatography. Productivity of the loading step could be increased by 68 % from 40.8 to 68.7 g L -1 h −1 resin for protein A chromatography while retaining the same breakthrough behavior and better manufacturability of unstable proteins. Consequently, the bound fraction of protein is maintained above 99 %. This affords similar productivity gains as multicolumn chromatography while employing an easy-to-implement process model that can be used on conventional systems.