(58b) Antibody Production in the GS-NS0 System under Normal and Hyperosmotic Culture Conditions: a Combined Modeling and Experimental Study

Large-scale monoclonal antibodies (MAb) production for a wide range of therapeutic and diagnostic applications account for a significant part of the current biotechnology industry. Mammalian cell culture systems are most commonly used for industrial MAb production due to their ability to carry out complex post-translational modifications and protein folding in an authentic manner (Hauser 1997). The Glutamine Synthetase-NS0 (GS-NS0) system, an industrially significant cell line that has been used in recent years in the production of many important therapeutic antibodies including Synagis?, Zenapax® (Lonza Press Release. 1998a, 1998b) and Mylotarg® (Wyeth Pharmaceuticals Inc. 2005), is the chosen cell culture system for this study.

The continual increase in demand for therapeutic MAb products has resulted in predictions of a shortfall in plant manufacturing capacity in the biopharmaceutical industry within the next five years (Ransohoff et al. 2004). As a consequence, a key aim of the industry is the maximization of MAb production from the culture process. At present, trial-and-error cell culture experimentation is frequently carried out to optimize culture parameters and improve antibody production yields. However, this approach tends to require conducting a large number of experiments that are expensive and time-consuming (Jacquez 1998). In view of this limitation, a combined experimental and modeling approach has been taken for this study of the antibody production pathway in GS-NS0 cells. The utilization of a mathematical model can help in the identification of key factors affecting MAb production, thereby reducing the required number of cell culture experiments for product optimization. In this work, a novel hybrid model which describes the antibody production mechanism in GS-NS0 cells is first developed using preliminary experimental data available in literature. The hybrid model consists of unstructured elements describing cell growth, death and basic metabolic processes with a structured element for the description of the antibody production process in GS-NS0 cells. The combination of structured and unstructured elements allows for the detailed description of the antibody production pathway without the high computation cost normally associated with a fully structured model (Sidoli et al. 2005). A global sensitivity analysis carried out on the model has subsequently identified key parameters influencing MAb production rates and additionally provided a description of the variation in importance of different model parameters with respect to culture phases. The results indicate that the specific transcription rates of heavy and light chain mRNA, as well as the translation rates of heavy and light immunoglobulin chains have significant effects on antibody production rates throughout all phases of the cell culture, while the heavy and light chain mRNA half-lives become increasingly important as the culture progresses.

Additionally, due to existing literature reports on its economic viability as a means to enhance antibody production in GS-NS0 cells, hyperosmotic pressure has been chosen as the environmental factor under investigation. A plausible representation of the developed model for MAb production under hyperosmotic culture conditions is first derived using the global sensitivity analysis results. Increased values of approximately 40% for the specific transcription rates and 60% for the initial specific translation rates were required in order to accurately describe antibody production at hyperosmotic culture conditions.

The currently ongoing experimental study of GS-NS0 cells under normal and hyperosmotic conditions is focused specifically on the key parameters that have been identified by the global sensitivity analysis. Northern and western blotting techniques have been utilized to measure intracellular mRNA and polypeptide chains concentrations in order to obtain estimates for the specific transcription and translation rate parameters. The derived parameter estimates will then be used to verify the validity of the developed model under both normal and hyperosmotic cultures conditions and subsequently deduce the means through which hyperosmotic pressure improves antibody production.

References:

Hauser H (1997) Heterologous expression of genes in mammalian cells. In: Hauser H, Wagner R (eds) Mammalian cell biotechnology in protein production. Walter de Gruyter, Berlin, p 3-32.

Jacquez JA (1998) Design of experiments. J. Franklin Inst. 335B:259-279.

Lonza Press Release. (1998a) Lonza's glutamine synthetase expression system used in production of Hoffmann-La Roche's Zenapax®.

Lonza Press Release. (1998b) Lonza Biologics announces that MedImmune Inc.'s Synagis? becomes the second therapeutic monoclonal antibody to be produced using the glutamine synthetase expression system.

Ransohoff TC, Mittendorff II RE, Levine HL (2004) Forecasting Industrywide Biopharmaceutical Manufacturing Capacity Requirements Advances in Large Scale BioManufacturing and Scale-Up Production, October issue.

Sidoli FR, Mantalaris A, Asprey SP (2005) Toward global parametric estimability of a large-scale kinetic single-cell model for mammalian cell cultures. Ind. Eng. Chem. Res. 44:868-878.

Wyeth Pharmaceuticals Inc. (2005) Mylotarg® Label.