Poly-pathway model, a novel approach to simulate multiple metabolic states by reaction network-based model – Application to amino acid depletion in CHO cell culture

• Published in: Journal of biotechnology Vol. 259; pp. 235 - 247
• Main Authors: Hagrot, Erika, Oddsdóttir, Hildur Æsa, Hosta, Joan Gonzalez, Jacobsen, Elling W., Chotteau, Véronique
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 10.10.2017
• Subjects: Amino acid metabolism; amino acids; Amino Acids - metabolism; Animals; biochemical pathways; bioprocessing; cell culture; Chinese hamster ovary cells; CHO Cells; Computer Simulation; Cricetinae; Cricetulus; culture media; Elementary flux modes; equations; mammals; Metabolic Flux Analysis; Metabolic Networks and Pathways - physiology; metabolism; metabolites; Modeling; Models, Biological; Poly-pathway model; Modeling; Amino acid metabolism; Metabolic flux analysis; Elementary flux modes; Poly-pathway model; Chinese hamster ovary cells
• ISSN: 0168-1656; 1873-4863; 1873-4863
• DOI: 10.1016/j.jbiotec.2017.05.026

•A method for modeling multiple states in one model, named poly-pathway, is proposed.•The macroscopic model is based on elementary flux modes and generalized kinetics.•CHO cells were cultured under varied amino acid availability.•Distinct metabolic states were obtained.•A single model simulates observed variations in growth and metabolic rates. Mammalian cell lines are characterized by a complex and flexible metabolism. A single model that could describe the variations in metabolic behavior triggered by variations in the culture conditions would be a precious tool in bioprocess development. In this paper, we introduce an approach to generate a poly-pathway model and use it to simulate diverse metabolic states triggered in response to removal, reduction or doubling of amino acids in the culture medium of an antibody-producing CHO cell line. Macro-reactions were obtained from a metabolic network via elementary flux mode enumeration and the fluxes were modeled by kinetic equations with saturation and inhibition effects from external medium components. Importantly, one set of kinetic parameters was estimated using experimental data of the multiple metabolic states. A good fit between the model and the data was obtained for the majority of the metabolites and the experimentally observed flux variations. We find that the poly-pathway modeling approach is promising for the simulation of multiple metabolic states.