Cell Line and Process Development for Improved Transient Production of a "Difficult-to-Express" Fusion Protein by CHO Cells

Despite the remarkable yield improvements of recombinant proteins produced in mammalian cells, some “difficult-to-express” (DTE) proteins achieve considerably lower production titres. The bottlenecks are exacerbated in the case of transient gene expression (TGE) systems as the host cells are easily overloaded with recombinant genes, hence necessitating intensive cell line and process development. The aims of this thesis are to study the limitations to high TGE yields of DTE proteins, and subsequently investigate strategies to efficiently alleviate the bottlenecks. For this purpose, we used a model DTE Fc-fusion protein (Sp35:Fc; proprietary of Biogen Idec) expressed in Chinese hamster ovary (CHO) cells, as well as secreted alkaline phosphatase and green fluorescent protein for comparisons. Through analyses of intracellular and extracellular Sp35:Fc polypeptides, we found that post-translational mechanisms were limiting in the cells. The saturation of Sp35:Fc expression coincided with the retention of folded proteins in the ER and the increase in disulphide bonded aggregates. Further in silico analysis via a mathematical model enabled identification of the relative importance of specific cellular process on Sp35:Fc productivity (qP). Based on these observations, three strategies aimed at improving the transient production of Sp35:Fc were investigated. The first strategy involved the evaluation of functional performance of clonally derived cell populations to produce Sp35:Fc. We critically assessed the key intrinsic functional traits of the clones, and their impact on Sp35:Fc production. The data indicate that cell lines with the capability to accumulate high biomass while maintaining relatively high specific growth rate (µ) were likely to be high producers for DTE proteins. For the second strategy, we utilised a novel vector system specifically beneficial for DTE proteins by incorporating ER stress response elements (ERSE) into the SV40 vector expressing Sp35:Fc or the UPR transactivator ATF6c. The ERSE-SV40 vectors acted as a synthetic "amplifier/dual activator" circuit, where expressed ATF6c amplified both its own and Sp35:Fc expression via activation of ERSE-SV40, whilst generally transactivating cellular ER capacity via endogenous ERSE. In the third approach, we addressed the hypothesis that specific functional proteins and chemical chaperones could improve ER capacity for Sp35:Fc folding reactions, increase secretion rate and/or relieve host cells from ER stress. We employed cell/process engineering by co-expressing a variety of molecular chaperones or UPR transactivators with Sp35:Fc, as well as a range of chemicals and hypothermic condition. We observed that Sp35:Fc production could be improved via two distinct modes; (i) increase in qP correlated to repression of µ, and (ii) stimulation of µ with general reduction in qP. In this regard, genes and chemicals could work synergistically to provide an optimal solution. Overall, this study illustrates that effective cell line and process development for DTE protein production requires a synergistic combination of vector, cell and process engineering strategies designed to alleviate cellular bottlenecks simultaneously, enabling the host cell to attain both high qP and cell density. Using two clonal variants and combinations of cell and process engineering, the transient Sp35:Fc production yield could be increased by more than six-fold. Rapid, high-throughput predictive mathematical tools would be particularly valuable in assessing the relative/synergistic impact of different engineering interventions.