Abstract

Genome-scale metabolic models (GEMs) are indispensable for studying cellular metabolism. We present iCHO3K, a community-consensus, manually curated reconstruction of the Chinese hamster metabolic network. Spanning 11,004 reactions linked to 3,597 genes, iCHO3K augments the network with 3,489 protein structures and physicochemical descriptors for >70% of 7,377 metabolites, enabling structure-aware analyses. We applied iCHO3K to contextualize transcriptomics and metabolomics from a fed-batch Chinese hamster ovary (CHO) cell line engineered to abolish lactate secretion. The model indicated reduced glycolytic flux with enhanced tricarboxylic acid (TCA) activity and elevated intracellular NADH and phosphoenolpyruvate (PEP), consistent with experimental measurements. Leveraging iCHO3K’s structural annotations, we evaluated potential off-target binding of NADH and PEP across early glycolytic enzymes and identified a putative allosteric PEP interaction with phosphofructokinase, suggesting a structural mechanism underlying reduced glucose uptake and glycolytic flux. Overall, iCHO3K provides a framework for systematic multi-omics integration, improved flux prediction, and structure-guided mechanistic insight, advancing CHO cell engineering and biomanufacturing. A record of this paper’s transparent peer review process is included in the supplemental information.