Applications of Quasi Real-Time Metabolic Flux Analysis in Mammalian Cell Culture Process Development

Abstract

Metabolic flux analysis offers a means of quantifying changes in cellular physiology and metabolism and thus has the potential to play an important role in accelerating bioprocess development towards optimizing protein productivity. Most implementations of MFA to date have been of an off-line nature while to be most useful for on-line process diagnosis and control, effectively realtime estimation of metabolic fluxes is necessary. We have developed a framework that allows for quasi realtime estimation of metabolic fluxes for mammalian cells in high-density perfusion cultures by using a combination of on-line and rapid off-line measurements of experimental variables. The above approach was used on experimental data obtained from long-term perfusion experiments with CHO cells at ∼20 × 106 cells/mL. Metabolic fluxes were estimated for various steady states over the course of the experiment resulting in useful information on the changes in cellular physiology and metabolism in response to changes in bioreactor conditions. The approach we present allows for robust estimation of metabolic fluxes in a quasi real-time fashion and has the potential to accelerate bioprocess development by reaching desired protein yields with fewer experiments.