Kinetic-model-based pathway optimization with application to reverse glycolysis in mammalian cells

For over 20 years, model-based tools have been used to optimize metabolic pathways in microorganisms for metabolite production. These approaches typically rely on simplified models that do not account for complex regulatory interactions. Mammalian cells, however, require more detailed modeling due to nonlinear kinetics and regulatory feedback. Prior research has shown that standard optimization methods fail to capture these interactions in higher organisms. This gap motivated the development of a new framework that integrates detailed kinetic data. The study addresses a key limitation in applying metabolic optimization to mammalian systems. Existing tools lack the ability to handle multiple physiological scenarios simultaneously. This paper introduces a novel two-stage approach to address these challenges.

The aim of this study is to develop a pathway optimization framework that can be applied to mammalian cells. The framework is designed to incorporate nonlinear kinetic models and regulatory interactions. The specific problem addressed is the lack of tools for multi-scenario optimization in complex cellular systems. The motivation comes from the need to improve metabolic processes in mammalian cell cultures. The study targets reverse glycolysis as a test case for the framework. Reverse glucose flow could reduce the need for glucose feed in later stages of cell culture. The framework is intended to identify key enzymes and optimal interventions. This approach aims to enhance process robustness and metabolic flexibility.

The framework consists of two stages for pathway optimization. Stage 1 involves solving optimization problems to rank enzymes by their contribution to the metabolic objective. This stage identifies enzymes that have the greatest impact on achieving the desired outcome. Stage 2 determines the optimal enzyme adjustments for a given number of interventions. It uses multi-scenario optimization to consider multiple physiological conditions. The model incorporates detailed kinetic and regulatory data. The approach integrates nonlinear equations to capture concentration changes. The framework is applied to reverse glycolysis in cultured mammalian cells. Computational simulations are used to test the effectiveness of the proposed method.

The framework successfully identified enzymes most critical to reverse glycolysis. Computational results showed that the method captures key regulatory interactions. The model identified differences in metabolic requirements for various carbon sources. The approach also revealed commonalities in enzyme adjustments across scenarios. The results demonstrated the framework's ability to handle multiple physiological conditions. The method outperformed standard optimization techniques in capturing nonlinear effects. The simulations showed improved process robustness in late-stage cell culture. The findings suggest that the framework can be applied to other metabolic objectives.

The proposed framework effectively captures regulatory and kinetic interactions in mammalian cells. The results suggest that the two-stage approach improves pathway optimization accuracy. The method identifies key enzymes and optimal interventions for reverse glycolysis. The framework supports multi-scenario optimization for different physiological conditions. The study demonstrates the feasibility of applying kinetic models to mammalian systems. The findings align with the authors' goal of enhancing metabolic process robustness. The framework may be useful for other metabolic engineering applications. The authors propose that this approach could be extended to other cellular processes.