Development of a Mathematical Model for Evaluating the Dynamics of Normal and Apoptotic Chinese Hamster Ovary Cells Saeideh Naderi and Mukesh Meshram Dept. of Chemical Engineering, University of Waterloo, Waterloo, ON, Canada N2L 3G1 Catherine Wei and Brendan McConkey Dept. of Biology, University of Waterloo, Waterloo, ON, Canada N2L 3G1 Brian Ingalls Dept. of Applied Mathematics, University of Waterloo, Waterloo, ON, Canada N2L 3G1 Hector Budman and Jeno Scharer Dept. of Chemical Engineering, University of Waterloo, Waterloo, ON, Canada N2L 3G1 DOI 10.1002/btpr.647 Published online May 25, 2011 in Wiley Online Library (wileyonlinelibrary.com). A metabolic flux based methodology was developed for modeling the metabolism of a Chinese hamster ovary cell line. The elimination of insignificant fluxes resulted in a simpli- fied metabolic network which was the basis for modeling the significant metabolites. Employ- ing kinetic rate expressions for growing and non-growing subpopulations, a logistic model was developed for cell growth and dynamic models were formulated to describe culture composition and monoclonal antibody (MAb) secretion. The model was validated for a range of nutrient concentrations. Good agreement was obtained between model predictions and experimental data. The ultimate goal of this study is to establish a comprehensive dynamic model which may be used for model-based optimization of the cell culture for MAb produc- tion in both batch and fed-batch systems. V V C 2011 American Institute of Chemical Engineers Biotechnol. Prog., 27: 1197–1205, 2011 Keywords: CHO, metabolic flux analysis, apoptosis, mathematical modeling, logistic equation Introduction In comparison to various microorganisms, mammalian cells are superior platforms for secreting secondary metabo- lites including recombinant proteins (r-proteins) with thera- peutic potential such as humanized monoclonal antibodies (MAbs). Chinese hamster ovary (CHO) cells have been rou- tinely employed as hosts for a range of recombinant glyco- sylated protein production, including MAbs. Their complex metabolism allows them to perform appropriate post-transla- tion modification (such as glycosylation and sialylation) of the r-proteins making these bioproducts suitable for thera- peutic and diagnostic application. 1–3 Significant progress has been made in cell culture technol- ogy through genetic engineering, 4,5 which led to improve- ment of the strains in terms of robustness and high specific protein expression. However, often low productivity and cell death are still major obstacles for efficient and economical production on an industrial scale. In batch culture, which is still a common cell culture sys- tem for a majority of industrial bioprocesses, cells are prone to programmed cell death or apoptosis. 6,7 Apoptosis can be triggered by environmental causes, such as nutrient deple- tion, metabolic byproduct accumulation, changes in pH, or oxygen limitation. Besides apoptosis, another type of pro- grammed cell death called autophagy has been attributed to nutrient exhaustion. 8 It is important to assess the extent of apoptosis and autophagy since they affect glycosylation, thus impacting the quality of MAb. 1,9 Since cells produce r-pro- tein continuously, often even after the cessation of growth, the control and optimization of cell culture conditions can prolong viability, and hence productivity. Fed-batch strategy with interval supplementation of essen- tial nutrients such as glucose and glutamine as the main source of energy has been employed with CHO cell lines, resulting in longer cell viability and higher MAb produc- tion. 9,10 However, formulating an optimal nutrient feeding strategy (i.e., frequency and volume of feeding) requires a detailed understanding of cell metabolism and physiology. Due to the complexity of cell metabolism and the many unknown intracellular factors involved in regulating cellular mechanisms, most applied strategies have been based on trial-and-error approach. 11 Mathematical modeling has been recognized as a rational approach to systematize the experi- mental observations with the aim of identifying the key metabolites, bioreactions, and process parameters. These models can then be used for the optimization and effective control of bioprocess performance. In order to achieve this goal, a model must provide an accurate explanation of intra- cellular reactions involving all metabolites. Correspondence concerning this article should be addressed to H. Budman at hbudman@engmail.uwaterloo.ca. V V C 2011 American Institute of Chemical Engineers 1197