A high-capacity cell macroencapsulation system supporting the long-term survival of genetically engineered allogeneic cells

The rapid increase in the number of approved therapeutic proteins, including recombinant antibodies, for diseases necessitating chronic treatments raises the question of the overall costs imposed on healthcare systems. It is therefore important to investigate alternative methods for recombinant protein administration. The implantation of genetically engineered cells is an attractive strategy for the chronic long-term delivery of recombinant proteins. Here, we have developed a high-capacity cell encapsulation system for the implantation of allogeneic myoblasts, which survive at high density for at least one year. This flat sheet device is based on permeable polypropylene membranes sealed to a mechanically resistant frame which confine cells seeded in a tailored biomimetic poly(ethylene glycol) (PEG)-based hydrogel matrix. In order to quantitate the number of cells surviving in the device and optimize initial conditions leading to high-density survival, we implant devices containing C2C12 mouse myoblasts expressing a luciferase reporter in the mouse subcutaneous tissue. We show that initial cell load, hydrogel stiffness and permeable membrane porosity are critical parameters to achieve long-term implant survival and efficacy. Optimization of these parameters leads to the survival of encapsulated myogenic cells at high density for several months, with minimal inflammatory response and dense neovascularization in the adjacent host tissue. Therefore, this encapsulation system is an effective platform for the implantation of genetically engineered cells in allogeneic conditions, which could be adapted to the chronic administration of recombinant proteins.

Transient gene expression for rapid protein production

Natalie Muller

Recombinant proteins are important for biomedical research and for the treatment of human disease. Therefore it is necessary to develop reproducible bioprocesses to rapidly produce proteins of adequate quality and quantity. Expression in mammalian cells is preferred if the proteins are to be properly folded and post-translationally modified. For the rapid production of milligram to gram quantities of a protein in mammalian cells, large-scale transient gene expression is an attractive option. The two most important components of this technology are the cell cultivation system and the gene delivery method. For this reason the goal of this thesis was to develop novel cost-efficient cell cultivation systems and gene delivery methods for the large-scale transfection of mammalian cells in chemically defined media free of any animal-derived components including serum. Cultivation of mammalian cells in suspension is essential for large-scale transient gene expression. A superior system for the cultivation of mammalian cells based on the agitation of cells in "square-shaped" glass bottles (square bottles) was developed. It is an inexpensive but efficient means to grow cells to a high density without instrumentation. The growth and viability of cultures of both human embryo kidney (HEK) 293E and Chinese hamster ovary (CHO) DG44 cells in agitated one-liter square bottles exceeded that in spinner flasks, reaching cell densities 1.5 times higher on average. Furthermore, an efficient and reproducible method to transfect cells in agitated square bottles was developed for both HEK 293E and CHO DG44 cells. A novel gene delivery method termed calfection that relies on the addition of DNA and CaCl2 directly to a suspension culture was developed. Calfection has a number of advantages over existing gene delivery methods such as calcium phosphate-DNA coprecipitation in that there are no time-dependent steps. The DNA/CaCl2 solution can be stored for long periods and is filterable. It is suitable for both large-scale transfections in bioreactors and for high-throughput transfections in microtiter plates. One drawback, however, is its dependence on the presence of serum in the culture medium. A second gene delivery method, serum-free transfection with polyethylenimine (polyfection), proved to be highly efficient for the transfection of both HEK 293E and CHO DG44 cells. A reproducible method for polyfection in microtiter plates, 50 ml centrifuge tubes, agitated square and round bottles, spinner flasks, and bioreactors was optimized for these two cell lines. Polyfection proved to be a simple and successful gene transfer method in both instrumented and non-instrumented cultivation systems. Importantly, it could be performed in serum-free chemically defined media. With HEK 293E cells, 80 mg/l of antibody was produced in agitated square bottles and with CHO DG44 cells, more than 2 g of antibody were produced in one week in a 150-liter bioreactor.

EPFL2005

Development of high-titer transient gene expression processes for manufacturing recombinant proteins

Gaurav Backliwal

The market for recombinant therapeutic proteins (including antibodies) is estimated to be greater than $50-billion and has a compounded annual growth rate of around 20%. More than 50% of all approved processes for manufacturing recombinant proteins use mammalian cells as expression hosts due to their ability to carry out complex assembly and processing, human-like glycosylation and secretion of proteins. This percentage is not expected to change and should even increase with time. Classically, these manufacturing processes use stable cell lines for protein expression, wherein the plasmid coding for the protein of interest is stably integrated into the host cell chromosomes (Stable Gene Expression, SGE). An alternative method for expressing recombinant proteins is Transient Gene Expression (TGE). Herein, the expression of the protein of interest takes place from plasmids which are transfected into the cells and are maintained extra-chromosomally. TGE has become a rapid and easy method for obtaining proteins for structural, biochemical or pre-clinical studies, where only small amounts of one or more proteins might be required. As yet, TGE has not been used for manufacturing recombinant proteins for clinical testing or approved clinical use due to the requirement of large amounts of protein of consistent quality. Even though TGE has been scaled-up to the 100-liter scale and is being attempted at the 1000-liter scale, due to low specific productivity, low volumetric yields and the high cost of DNA, TGE processes are not able to economically produce sufficient amounts of proteins for such purposes. Therefore, the goal of this thesis was to improve TGE in order to create high-titer processes, which would be economically more feasible for manufacturing large-amounts of recombinant proteins. This was achieved by: Improving and simplifying gene delivery – by the combination of high cell density at time of transfection and in-situ complex formation (as apposed to a-priori complex formation), we could enhance protein expression levels and gene delivery. This made implementation more robust and easy, with greater choice of media which could be used and cell density which could be maintained. Improving gene expression – by using a combination of inhibitors of histone deacetylases like valproic acid, and growth factors like acidic fibroblast growth factor, we could enhance specific productivity by ∼20-fold. Enhancing the cell densities at which cells could be maintained after transfection – by the combination of cell cycle regulators like human p18 and p21 and treatment with inhibitors of histone deacetylases, we blocked cell growth which allowed cells to be maintained at significantly higher cell densities, allowing us to enhance volumetric productivity by ∼100-fold. Overall these improvements allowed antibody titers in excess of 1 g/l. The gram per liter range of volumetric productivity has previously only been reported for optimized SGE based manufacturing processes. We therefore improved the economic feasibility of TGE for manufacturing recombinant proteins and reduced the difference in manufacturing costs between TGE and SGE, for a given amount of protein, by an estimated 50-fold. With some further development work and study of related issues like protein quality, TGE based processes can be expected to become part of mainstream recombinant protein manufacturing in the future.

EPFL2008

Transient recombinant protein expression in mammalian cells

Sarah Wulhfard

Transient gene expression (TGE) is a rapid method for generating recombinant proteins in mammalian cells, but the volumetric productivities for secreted proteins in transiently transfected CHO DG44 cells are typically more than an order of magnitude lower than the yields achieved with recombinant CHO-derived cell lines. The goals of the thesis are to identify the limitations to higher TGE yields in CHO DG44 cells and to find possible solutions to overcome the problems. Initially an attempt was made to enhance TGE production by increasing the amount of transfected plasmid DNA. However, this approach did not result in increased recombinant protein levels; on the contrary, transfection with an excess of plasmid DNA (> 1.25 μg/ml) had a negative impact on transgene mRNA levels and protein production. Moreover, it was also observed that recombinant protein yield was strongly dependent on the mRNA level. Therefore, three strategies aimed at increasing the amount of transgene mRNA were investigated. For the first approach, transfected cells were exposed to hypothermic conditions during the production phase. It was already known that lower temperatures increase protein production several fold in recombinant CHO DG44-derived cell lines. The second strategy involved the treatment of transfected cells with valproic acid, a histone deacetylase inhibitor that reduces the effects of gene silencing. The third approach aimed to increase transgene mRNA levels by overexpressing transcription factors and growth factors. With the first two strategies recombinant antibody yields of 60-80 mg/L were achieved whereas the untreated control transfections produced only 5-10 mg/L. Combination of the two strategies led to the production of 90 mg/L of antibody. Moreover, in the treated cultures, the steady-state level of transgene mRNA was 3-5 times higher than in the untreated cultures and remained stable up to 6 days post-transfection. Using the third approach, the increase in recombinant protein production was moderate and transgene mRNA amounts were only 2-fold higher in treated samples compared to the control. When specific proteins such as c-fos, c-jun, NF-kB, and acidic fibroblast growth factor (aFGF) were overexpressed the recombinant antibody production was 20 mg/L compared to 5 mg/L for the control transfection. Overexpression of either a transcription factor or a growth factor in combination with treatment with valproic acid allowed the recombinant protein yield to reach 90 mg/L. However, the benefit of the overexpressed factors was minimal compared to the effect of valproic acid alone. In conclusion, it was demonstrated that the level and stability of transgene mRNA are important factors for increasing volumetric yields in transiently transfected CHO DG44 cells. Furthermore, three approaches aimed to increase mRNA amounts were tested. Exposure to hypothermic conditions and treatment with valproic acid were the two best strategies tested. Both are simple, cost-effective, and scalable making transient gene expression in CHO DG44 cells a feasible alternative for rapid production of gram amounts of recombinant protein.

EPFL2009