Reprogramming of Sterol Biosynthesis in Chinese Hamster Ovary Cells for Enhanced Recombinant Protein Production

The biopharmaceutical industry is a multi-billion dollar global market with projections that within the next 5 to 10 years, up to 50% of all drugs in development will be biopharmaceuticals. The majority of these drugs are recombinantly produced protein based therapeutics with the demand for such new therapeutic products continuing to increase, broadening the range of medical conditions that they are being used to treat. Chinese hamster ovary (CHO) cells are the preferred and most commonly

used system for large scale production of recombinant glycoproteins for therapeutic use due to their ability to perform complex human-like post-translational modifications required for biological activity. CHO cells also display robust growth, high productivity and stability, attributes that have enabled them to become widely used. Whilst there have been large increases in the yield and quality of recombinant protein that can be obtained from mammalian expression systems over the years, particularly CHO cells, many recombinant proteins remain difficult to express in CHO cells or any other system. Many of these proteins are highly complex, highly glycosylated, large and unstable proteins. As such, there remains a need to develop CHO cell systems that are able to produce such difficult to express proteins rapidly in higher yield and quality. The endoplasmic reticulum (ER) is a key compartment in the secretory pathway of mammalian cells and during the production of recombinant proteins in CHO cells this has reportedly been a potential bottleneck and site of

perceived cellular stress as a result of the load imposed on the cell by the recombinant protein.

This study has investigated two sterol reductases; Transmembrane 7 Superfamily member 2 (TM7SF2) and 7-dehydrocholesterol reductase (DHCR7). ER resident proteins involved in the biosynthesis of cholesterol that have been reported to influence endoplasmic reticulum (ER) and

nuclear envelope (NE) expansion when over-expressed in mammalian cell lines. The work described here therefore set out to determine whether overexpression of these sterol reductases could enhance the secretory capacity of CHO cells by expansion of the ER and hence lead to increased yields of difficult to express recombinant proteins compared to those cells where these were not overexpressed. The TM7SF2 and DHCR7 genes were cloned into a mammalian expression vector with a hygromycin selection marker to allow for the selection of stable cell lines where the genes had been integrated into the genome. Stable CHO cells overexpressing TM7SF2 and DHCR7 were generated to

assess the impact on cell growth and stable and transient recombinant protein expression, monitored using western blot techniques and microscopy analysis. Control cell lines over-expressing GFP (but not the sterol reductases) and the pcDNA3.1Hygro (empty) vector were also developed. Stable

expression levels of the sterol reductases were determined in heterogeneous pools and then used for limited dilution cloning to isolate clonal cell lines with differing sterol reductase expression which were further investigated. Growth profile assays were undertaken to monitor any toxicity effect of over-expression of TM7SF2 and DHCR7 on CHO cells. Transient expression studies of the recombinant bio therapeutic protein etanercept, a 150 kDa biomolecule and erythropoietin (EPO), a 36 kDa recombinant protein, were carried out to investigate the production capacity of lipid engineered CHO cells with reference to controls. Immunocytolocalisation studies were also undertaken to investigate pools and clones of lipid targets predicted to be ER-localized or associated. A YFP tag was present on the N-terminal of TM7SF2 and DHCR7 molecules and was used to monitor expression by fluorescence in stable CHO cells.

The data from these experiments show that both TM7SF2 and DHCR7 engineered CHO cells had improved cell growth and culture viability over the course of 10-day experimental study compared to the control cells. Western blot studies showed that both TM7SF2 and DHCR7 could be stably over-expressed in CHO cells and the amount of expression varied among the generated clones. Further, the high expressers of TM7SF2 and DHCR7 were identified and investigated for EPO and etanercept production and expression levels of these recombinant biomolecules were observed to be proportional to the levels of TM7SF2 and DHCR7 expressions in CHO cells. More importantly, TM7SF2 and DHCR7 engineered cells showed increased expression levels of these recombinant products when compared with pcDNA3.1Hygro controls as determined by western blot analysis of the amount

of secreted recombinant target protein in the cell culture supernatant. Confocal images also showed clones expressed different levels of YFP signal which related to expression levels observed when western blot techniques were used. In summary, the data presented here shows that the manipulation of cellular circuits such as cholesterol biosynthesis has the potential to enhance cellular growth and recombinant protein yield. By designing new hosts and cellular circuits to reprogramme

the CHO cell ER there is the potential of expanding the secretory capacity and/or subsequent secretory vesicle system.