Identification of Potential Biomarkers for Coronary Artery Disease Based on Cuproptosis

Abstract

Identifying peripheral biomarkers is an important noninvasive diagnosis method for coronary artery disease (CAD) which has aroused the strong interest of researchers. Cuproptosis, a newly reported kind of programmed cell death, is closely related to mitochondrial respiration, adenosine triphosphate (ATP) production, and the TCA cycle. Currently, no studies have been published about the effects of cuproptosis-related genes (CRGs) on diagnosing CAD. To screen marker genes for CAD from CRGs, we downloaded the whole blood cell gene expression profile of CAD patients and normal samples, i.e., the GSE20680 dataset, from the GEO database. By differential expression analysis, we obtained 10 differentially expressed CRGs (DE-CRGs), which were associated with copper ion response, immune response, and material metabolism. Based on the 10 DE-CRGs, we furtherly performed LASSO analysis and SVM-RFE analysis and identified 5 DE-CRGs as marker genes, including F5, MT4, RNF7, S100A12, and SORD, which had an excellent diagnostic performance. Moreover, the expression of the marker genes was validated in the GSE20681 and GSE42148 datasets, and consistent results were obtained. In mechanism, we conducted gene set enrichment analyses (GSEA) based on the marker genes, and the results implied that they might participate in the regulation of immune response. Therefore, we calculated the relative contents of 22 kinds of immune cells in CAD and normal samples using the CIBERSORT algorithm, followed by differential analysis and correlation analysis of the immune microenvironment, and found that regulatory T cell (Treg) significantly decreased and was negatively correlated with marker gene S100A12. To further reveal the regulation mechanisms, a lncRNA-miRNA-mRNA ceRNA network based on the marker genes was established. Finally, 13 potential therapeutic drugs targeting 2 marker genes (S100A12 and F5) were identified using the Drug Gene Interaction Database (DGIdb). In summary, our findings indicated that some CRGs may be diagnostic biomarkers and treatment targets for CAD and provided new ideas for further scientific research.