An unconventional uptake rate objective function approach enhances applicability of genome-scale models for mammalian cells

Yiqun Chen,Harnish Mukesh Naik,Venkata Gayatri Dhara,Chien-Ting Li,Maciek R Antoniewicz,Michael J Betenbaugh,Brian O Mcconnell

doi:10.1038/s41540-019-0103-6

Yiqun Chen, Harnish Mukesh Naik + Show 5 more

Open Access

https://doi.org/10.1038/s41540-019-0103-6

Copy DOI

Abstract

Constraint-based modeling has been applied to analyze metabolism of numerous organisms via flux balance analysis and genome-scale metabolic models, including mammalian cells such as the Chinese hamster ovary (CHO) cells—the principal cell factory platform for therapeutic protein production. Unfortunately, the application of genome-scale model methodologies using the conventional biomass objective function is challenged by the presence of overly-restrictive constraints, including essential amino acid exchange fluxes that can lead to improper predictions of growth rates and intracellular flux distributions. In this study, these constraints are found to be reliably predicted by an “essential nutrient minimization” approach. After modifying these constraints with the predicted minimal uptake values, a series of unconventional objective functions are applied to minimize each individual non-essential nutrient uptake rate, revealing useful insights about metabolic exchange rates and flows across different cell lines and culture conditions. This unconventional uptake-rate objective functions (UOFs) approach is able to distinguish metabolic differences between three distinct CHO cell lines (CHO-K1, -DG44, and -S) not directly observed using the conventional biomass growth maximization solutions. Further, a comparison of model predictions with experimental data from literature correctly correlates with the specific CHO-DG44-derived cell line used experimentally, and the corresponding dual prices provide fruitful information concerning coupling relationships between nutrients. The UOFs approach is likely to be particularly suited for mammalian cells and other complex organisms which contain multiple distinct essential nutrient inputs, and may offer enhanced applicability for characterizing cell metabolism and physiology as well as media optimization and biomanufacturing control.

Highlights

Constraint-based genome-scale models and flux balance analysis (FBA) have been widely used to investigate metabolic systems of various organisms
We introduce an unorthodox FBA approach based on a set of uptake-rate objective functions (UOFs), which utilizes the measured growth rate as a constraint and independently minimizes the uptake rate of each individual non-essential nutrient
We initially demonstrated how the presence of essential amino acids can exert a negative impact on the FBA of a mammalian cell model due to their values being rate-limiting factors for the conventional biomass objective function (BOF), using a case study containing published CHOK1 specific growth rates, 23 metabolites production and consumption rates for 7 cell culture replicates.[20]

Summary

INTRODUCTION

Constraint-based genome-scale models and flux balance analysis (FBA) have been widely used to investigate metabolic systems of various organisms. Direct estimation of these essential amino acid uptake inputs is possible based on measured growth rates and a “essential nutrient minimization” (ENM) approach which solves for the absolute minimal consumption requirements. These rate estimations can be used to adjust the FBA constraints in order to resolve problematic mass balance constraints. Averaging of glucose inputs had no impact on the growth rate prediction results except for replicate 3, in which the predicted growth rate increases by 13% These results cell line evaluation, metabolic engineering to media optimization agree with the dual prices with respect to BOF as illustrated in and biomanufacturing control

RESULTS

Replicates

DISCUSSION