Characterization of the redox environment in protein refolding processes

Abstract

Protein refolding, is often done via an empirical approach rather than a Quality by Design approach. Quality by Design aims to improve process understanding and robustness by identifying critical process parameters and, ideally, linking them to desired critical quality attributes. The redox potential represents an interesting parameter for a possible process analytical technology tool, since the formation of disulfide bonds occurs via a redox reaction and since in vivo folding processes are reported to be influenced by the redox potential. Oxygen is a cost-effective oxidant and, furthermore,it can support the formation of disulfide bonds, so its effect was also investigated. Inorder to avoid the influence of air/water interfaces, bubble free control strategies for the dissolved oxygen were developed. In this work, the control of the redox potential was carried out exclusively via the dissolved oxygen, since no alternative oxidant couldbe found for the galactose oxidase. Consequently, the parameters were not considered independently of each other. With the developed control strategies, several reactor runs were performed under controlled conditions. A decrease in activity was observed with increasing dissolved oxygen as well as increasing redox potential. Only under oxygen-free conditions and a low redox potential, an increase in activity could be measured over several hours. Both parameters could not be mathematically linked to the activity, but further critical process parameters were found, which probably allow a linkage. The dissolved oxygen or the redox potential show an influence on the enzymatic activity and should, in refolding processes, therefore be observed or controlled.Since both parameters could not be considered independently and they correlate with each other, the influence cannot be attributed exactly. Another parameter that turned out to be promising is the free thiol concentration, which is composed of the cysteines of the protein and the cysteamine of the refolding buffer. In the future, this parameter could possibly be determined by a model that takes into account the redox potential,the dissolved oxygen and the amount of reductant used, which could lead to a better understanding of the process.i