Key points are not available for this paper at this time.
Background: Osteoarthritis (OA) pathogenesis involves inflammatory-metabolic crosstalk driving cartilage destruction, yet the mechanisms of potential therapeutics like curcumin remain poorly defined. Methods: We integrated untargeted metabolomics, transcriptomic analysis of four GEO datasets (GSE12021, GSE55235, GSE55457, GSE82107), and three machine learning algorithms (LASSO, Random Forest, XGBoost) to characterize curcumin’s effects on IL-1β-induced human chondrocytes. Results: Metabolomic profiling demonstrated that IL-1β caused significant depletion of TCA cycle intermediates compared to blank controls, including pyruvate (log2FC = −1.34, p 0.75). qRT-PCR validation confirmed that high-dose curcumin significantly downregulated pro-inflammatory genes compared to IL-1β treatment alone: JUN (log2FC = −0.68, p < 0.001), IL6 (log2FC = −1.13, p < 0.001), PTGS2 (log2FC = −0.94, p < 0.001), CCL20 (log2FC = −1.51, p < 0.001), and MMP9 (log2FC = −0.42, p < 0.001). Conversely, curcumin significantly upregulated the NF-κB inhibitor NFKBIA (log2FC = 0.40, p < 0.001), whose expression was initially suppressed by IL-1β (−log2FC = 0.76 vs. blank, p < 0.001). Conclusions: This systems-level analysis suggests curcumin modulates metabolic-inflammatory networks in OA chondrocytes, with NFKBIA as a candidate mediator, offering a mechanistic framework for drug-like molecule development despite curcumin’s own translational limitations.
Wang et al. (Sun,) studied this question.