Quality by Design strategies in Inclusion Body processing

Abstract

Recombinant protein production in E. coli poses several significant advantages compared to other production hosts, such as short process times and high cost-efficiency. However, due to high expression levels and the lack of post-translational modification machineries, proteins of interest are often produced as insoluble aggregates, more commonly called Inclusion Bodies (IBs). Therefore, several additional unit operations are required during down-stream processing (DSP), namely IB washing, solubilization and refolding. Up to now, these unit operations are developed and optimized using empiric approaches since no universal refolding protocol is available. In recent years, several paradigm changes for production processes of pharmaceuticals took place, changing the focus to Quality by Design (QbD) and Process analytical technology (PAT) (e.g. ICH Q8-12). These strategies require scientific knowledge and process understanding to be generated during process development as well as suitable analytical methods in order to monitor and control the process. While readily implemented for small molecules, strategies are still lacking for biopharmaceuticals and, in particular, for those produced as IBs. The reason for this is twofold: (i) The production of biopharmaceuticals in living organisms presents highly complex systems. Therefore, the product requires analytical methods able to assess very specific properties of one target molecule in a highly complex matrix. (ii) Many of the analytical techniques able to fulfill the requirements listed in (i) are not applicable for the refolding process due to harsh conditions during solubilization and changing Critical Quality Attributes (CQAs) during the DSP. In this work, the development of IB processes for the enzyme horseradish peroxidase (HRP) as model protein was investigated. Screening of suitable refolding conditions was performed as Design of Experiment (DoE) in small-scale approaches. These conditions were then scaled up to the controlled environment of a bench-scale refolding vessel. Different analytical strategies were applied for the entire DSP, including high performance liquid chromatography, Quantum cascade laser-IR-spectroscopy, and enzymatic activity assays. Several of the shown methods were then employed for IB forming proteins from other, different protein families. Overall, the applicability of QbD approaches to specific parts of IB processes were demonstrated. This cumulative thesis contains four first author papers, one patent, and two book chapters. While the definition of CQAs is still an integral part of these strategies, the high structural complexity and conformation-change of IBs through various unit operations makes the identification of CQAs for intermediates challenging. Furthermore, identifying the material attributes and process parameters that can have an effect on product CQAs as well as establishing PAT tools requires significant process understanding. This can, at least partly, be generated through unit-operation spanning multivariate approaches and an iterative approach to process development requiring a “Design of Design of Experiments” concept.