¹³C Flux Analysis Reveals that Rebalancing Medium Amino Acid Composition can Reduce Ammonia Production while Preserving Central Carbon Metabolism of CHO Cell Cultures

Abstract

C metabolic flux analysis (MFA) provides a rigorous approach to quantify intracellular metabolism of industrial cell lines. In this study, C MFA was used to characterize the metabolic response of Chinese hamster ovary (CHO) cells to a novel medium variant designed to reduce ammonia production. Ammonia inhibits growth and viability of CHO cell cultures, alters glycosylation of recombinant proteins, and enhances product degradation. Ammonia production was reduced by manipulating the amino acid composition of the culture medium; specifically, glutamine, glutamate, asparagine, aspartate, and serine levels were adjusted. Parallel C flux analysis experiments determined that, while ammonia production decreased by roughly 40%, CHO cell metabolic phenotype, growth, viability, and monoclonal antibody (mAb) titer were not significantly altered by the changes in media composition. This study illustrates how C flux analysis can be applied to assess the metabolic effects of media manipulations on mammalian cell cultures. The analysis revealed that adjusting the amino acid composition of CHO cell culture media can effectively reduce ammonia production while preserving fluxes throughout central carbon metabolism.

Perfusion Cell Culture Decreases Process and Product Heterogeneity in a Head‐to‐Head Comparison With Fed‐Batch

“…This was the case for both molecules tested (IgG1 and IgG4 sub‐types). One potential explanation for this phenomenon is that, though the extracellular environment is quite different throughout the course of a fed‐batch reactor, the impact of this variability is partially mitigated by the ability of the cell to self‐regulate in the presence of a variety of external conditions . In addition to the relatively decreased dispersion for the cell metabolism PCA within the fed‐batch data set, the distance between the PCA clusters for fed‐batch and perfusion are also decreased in the cellular metabolism measurements, again pointing to the self‐regulating ability of the cells within the culture.…”

Improving the accuracy of flux balance analysis through the implementation of carbon availability constraints for intracellular reactions

“…Comparison of ccFVA to intracellular flux measurements for central carbon metabolism (Glycolysis, TCA, PPP). (a–g): Comparison between ccFVA (blue, green, gray bars correspond to the data from Ahn & Antoniewicz, ; McAtee Pereira et al, ; Templeton et al, , respectively) and 13 C measurements (orange bars). ccFVA, carbon‐constrained FVA; FVA, flux variability analysis [Color figure can be viewed at wileyonlinelibrary.com]…”

“…A series of studies that report experimentally‐measured values for both extracellular (uptake/secretion) and intracellular fluxes were used to evaluate the performance of ccFBA (Ahn & Antoniewicz, ; McAtee Pereira et al, ; and Templeton et al, ). The curated i CHO1766 model was constrained to reflect each experimental dataset using the reported values for extracellular reactions only and the ability of ccFBA/ccFVA to predict intracellular fluxes was compared with that of traditional FBA/FVA.…”

“…Flux variability analysis. (a–g): Comparison between normal FVA (blue, green, gray bars correspond to the data from Ahn & Antoniewicz, ; McAtee Pereira et al, ; Templeton et al, , respectively) and ccFVA (red bars) for central carbon metabolism reactions (glycolysis, TCA, PPP) in i CHO1766 across seven distinct experimental datasets. The initial arbitrary bounds v LB and v UB were set to 0 and 100 mmol gDCW −1 h −1 for unidirectional and − 100–100 mmol gDCW −1 h −1 for bidirectional reactions.…”

“…The model is available in Supporting Information materials (SuppInfo3_iCHO1766_currated) as a SBML file. Datasets containing uptake, secretion and intracellular flux rates for central carbon metabolism (Glycolysis, Tricarboxylic Acid Cycle (TCA), and Pentose Phosphate Pathway (PPP)) were retrieved from literature (Ahn & Antoniewicz, ; McAtee Pereira, Walther, Hollenbach, & Young, ; Templeton, Xu, Roush, & Chen, ) and used to validate the predictions of the proposed algorithm (Table ). In cases where experimentally‐measured fluxes corresponded to more than one reaction in i CHO1766, average FVA or sample means were used for comparison (Table and SuppInfo4 for an example calculation).…”