Identification of multimorbidity hub genes for knee osteoarthritis-atherosclerosis and potential clinical applications

The combination of knee osteoarthritis (KOA) and atherosclerosis (AS) is a common multimorbidity. Epidemiological studies have demonstrated the existence of common risk factors, with metabolic syndrome possibly considered the most critical. In the present study, metabolism‑related clinical information was analyzed and metabolic profiles were assessed in healthy controls, patients with KOA, patients with AS and patients with both conditions using untargeted serum metabolomics assays. Potential KOA‑AS multimorbidity hub genes were identified using transcriptomics datasets from the Gene Expression Omnibus database and were validated using clinical samples and animal experiments. Finally, the clinical applications of the analyzed biomolecules were predicted. The results showed that the caffeine metabolic pathway was markedly associated with KOA‑AS multimorbidity and caffeine interacted with two potential hub genes (<i>EGR1</i> and <i>GSK3B</i>). In the validation experiment using clinical samples, early growth response 1 (Egr1) protein was only associated with AS. In the mouse disease model, Egr1 protein in the serum and cartilage was associated with KOA‑AS multimorbidity, with consistent expression trends. Receiver operating characteristic (ROC) analysis showed three metabolites with an area under the ROC curve of >0.7; drug prediction yielded two drugs that interacted with <i>EGR1</i>. In conclusion, KOA‑AS multimorbidity may be associated with metabolic abnormalities in the early stages and could develop into chronic inflammation in the later stages. Through multi‑omics analysis, three caffeine‑related metabolites with diagnostic value were obtained and <i>EGR1</i> was identified as the key gene for KOA‑AS multimorbidity.