Answers are grounded in this paper and cite their sources.
Sodium butyrate (NaB) is widely used in Chinese hamster ovary (CHO) cell bioprocessing to enhance recombinant protein productivity, yet the systems-level metabolic consequences of NaB treatment across its pharmacological dose range remain poorly characterised. Here, we performed dose-resolved untargeted liquid chromatography-mass spectrometry (LC-MS)-based metabolomics of CHO-K1 cells treated across eight concentrations of NaB (0–8.0 mM) and identified 469 annotated metabolite features. A ten-stage computational pipeline, comprising probabilistic quotient normalisation, principal component analysis (PCA), differential abundance analysis, Spearman dose-correlation, metabolite set enrichment analysis (MSEA), KEGG topology analysis, Random Forest, Least Absolute Shrinkage and Selection Operator (LASSO) regression, and protein-protein interaction (PPI) network construction, was applied without pathway pre-selection. PCA resolved three concentration-dependent metabolic states: sub-threshold [homeostatic] (0–0.25 mM), NaB-responsive [metabolically reoriented (for growth-arrest and potential increased productivity)] (0.5–2.0 mM), and cytostressed (4.0–8.0 mM). At 2.0 mM, 243 differentially abundant metabolites were identified, with decreased metabolites outnumbering increased ones (152 vs. 91; ratio 1.67:1), indicating broad catabolic suppression and selective activation of (stress regulating) pathways. The most enriched pathways included the Warburg Effect, Tryptophan Metabolism, and Glutamate Metabolism. Dominant metabolic changes comprised decreased abundance of metabolites associated with fatty acid oxidation and amino acid catabolism, alongside concurrent increases in metabolites associated with TCA anaplerosis and the glutathione cycle, patterns consistent with, but not a direct measurement of, altered flux through these pathways. Nine metabolites from tryptophan pathway were coordinately altered, with depletion of indole-derived species and induction of NAD-related metabolites. Machine learning identified Oxoproline, Serine, Heptanoate, Cyclohexanoic Acid, and Indole-3-Lactic Acid as the top candidate dose-responsive biomarkers within this dataset; their predictive performance in independent datasets and their identity have not yet been externally or analytically validated. PPI network analysis identified SIRT1, HDAC1, and IDO1 as the most central regulatory nodes. Together, these findings provide a systems-level metabolic framework for NaB-induced CHO cell physiology and nominate SIRT1, IDO1, CPT1A, and NAMPT as candidate targets for functional validation in recombinant producer cell lines.
No takes yet. Share an insight, caveat, or question.
Singh et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: