(518g) A Metabolic Flux Based Dynamic Model for Antibody Producing Mammalian Cells

The demand for high-value biopharmaceuticals is on a rapid rise due to their therapeutic and diagnostic applications. Monoclonal antibodies (MABs) are important reagents used in biomedical research, diagnosis and treatment of diseases, affinity production systems, and in vivo imaging. Monoclonal antibodies account for nearly 40% of the sale of biologics in the U.S. and are produced in animal cell bioreactors at a variety of scales. Mammalian cell cultures have been the favored production hosts for MABs, since microbial systems are not able to carry out the complex post-translational and functional modifications of these proteins, such as glycosylation. Chinese Hamster Ovary (CHO) cells and hybridoma cells, which share many similar metabolic characteristics, have been the popular cell types for production of MABs. The large scale production of MABs occurs via in vitro cultivation of mammalian cell lines in bioreactors using techniques similar to those used for microbial cultivation. The simplest approach for producing MABs in vitro is to grow the mammalian cells in batch and fed-batch cultures, since these are easy to operate and scale up, and recover and purify the MABs from the culture medium. However, under batch conditions, mammalian cells exhibit an inefficient glucose and glutamine metabolism, causing increased production of lactate and ammonia, major waste products of metabolism of glucose and glutamine, respectively. Both lactate and ammonia inhibit growth of mammalian cells and deteriorate MAB production. To reduce formation of these toxic by-products, fed-batch operations with intermittent supplementation of principal carbon and nitrogen sources have been performed with these cell lines, resulting in a higher volumetric productivity and extended cell viability. A cost-effective production of MABs requires an understanding of the effects of bioreactor process variables on the physiology of mammalian cells. The development of methods for optimization of cell growth and cell productivity has become a crucial issue to enhance MAB yield in vitro production. Although fed-batch operations have been successfully employed for microbial systems, kinetics of mammalian cell cultures is still under investigation to determine quantitatively as well as qualitatively cost-effective production strategies. Creating these strategies requires understanding of the effects of cell metabolism on the system dynamics. Substantial cell death and sensitivity of MAB productivity to culture conditions are important issues.

Mathematical models have been significant tools for combining cell physiology and engineering to predict the behavior of cell metabolism and optimize culture conditions by identifying process parameters that significantly impact cell growth and target metabolite productivity. The cell metabolism can be described and characterized using a metabolic network that accounts for biotic phase reactions. Metabolic flux analysis is a widely used approach to characterize the state of cellular metabolism and activities of various metabolic pathways. The analysis enables computation of intracellular fluxes from information on experimentally determined metabolite uptake and excretion rates by using the stoichiometry of an identified metabolic network. The analysis enables identification of all significant fluxes and significant metabolites influencing cell metabolism so that it is possible to obtain a set of macro-reactions linking the substrates to the end-products, the so-called elementary flux modes. In this work, a novel dynamic model is developed on the basis of these macro-reactions involving extracellular substrates and products for a mammalian cell culture with a MAB as the target product. The processes accounted for in the model are cell growth and death, uptake of key nutrients, glucose and glutamine, accumulation of lactate and ammonia, and MAB synthesis. In the end, the cell-specific kinetics of growth and death of mammalian cells, utilization of glucose and glutamine, and generation of MAB, ammonia, and lactate are expressed in terms of concentrations of glucose, glutamine, lactate, and ammonia. The model incorporates a structured kinetic representation of MAB synthesis. The model accounts for the impact of glutamine availability and ammonia accumulation on translation rates and stability of mRNAs involved in assembly of MAB. Glutamine provides remarkable metabolic energy for cell growth and protein synthesis and is an important precursor of proteins and peptides, as well as amino sugars, purines and pyrimidines. It has been observed that glutamine depletion in the culture greatly influences the overall antibody synthesis rate and increases the apoptotic death rates of mammalian cells. This is accounted for in the dynamic model. Performance of batch and fed-batch cultures with intermittent or sustained addition of glucose and/or glutamine is simulated. The model predictions are in good agreement with experimental data reported in the literature for Immunoglobulin G (IgG) antibodies.