Multi-omics data mining combined with experimental validation reveals ferroptosis- and autophagy-associated hub genes
Xinou Zheng1, Jinling Zheng2, Xuezhuang Li1
1Medical Innovation Research Department, Chinese PLA General Hospital, Beijing, China.
Background:
Atherosclerosis (AS) is a chronic inflammatory vascular disease that can lead to severe cardiovascular events. Ferroptosis and autophagy have been increasingly recognized for their significant roles in AS; however, few clinically translatable hub genes that connect these processes in atherosclerotic lesions have been identified. There is an urgent need for novel diagnostic and therapeutic targets to improve the early detection and intervention of AS.
Methods:
We downloaded AS-related datasets from the Gene Expression Omnibus (GEO) database. Common genes were identified using Limma and Weighted Gene Coexpression Network Analysis (WGCNA). We assessed ferroptosis and autophagy marker expression in AS mouse coronary tissues using immunohistochemistry (IHC) and immunofluorescence (IF). Hub genes were identified by intersecting common genes with known ferroptosis- and autophagy-related gene sets. Gene Set Enrichment Analysis (GSEA), receiver operating characteristic (ROC) curves, and 10-fold cross-validation repeated five times were performed to explore the potential roles and diagnostic capabilities of these hub genes. Immune cell infiltration analysis was performed using CIBERSORT, and Spearman's correlation analysis was subsequently performed to evaluate the associations between the identified hub genes and the relative abundance of immune cells. Finally, hub gene expression was validated in AS mouse coronary tissues using IHC and IF.
Results:
Limma and WGCNA identified 104 common genes. IHC and IF analyses in AS mice confirmed the activation of ferroptosis and inhibition of autophagy in the coronary tissue. Intersecting common genes with pathway-specific genes identified three pivotal hub genes: CALCOCO2, TXNRD1, and SELENBP1. ROC analysis and 10-fold cross-validation repeated five times indicated excellent diagnostic efficacy for these genes. Immune infiltration analysis revealed significant alterations in immune cell populations in AS patients, and correlation analysis further demonstrated that CALCOCO2, TXNRD1, and SELENBP1 were significantly associated with multiple immune cell types, directly linking these hub genes to the atherosclerotic immune microenvironment. Consistent with patient data, IHC/IF validation showed significantly lower expression of the hub genes in AS mouse coronary tissues compared to controls.
Conclusions:
CALCOCO2, TXNRD1, and SELENBP1 represent a novel panel of diagnostic biomarkers uniquely positioned at the intersection of ferroptosis and autophagy in AS. The robust diagnostic performance of these hub genes, as demonstrated by ROC analysis and cross‑validation, was further consolidated by IHC/IF validation showing their consistently downregulated expression in AS mouse models. Furthermore, these hub genes are significantly correlated with multiple immune cell populations, directly linking them to the atherosclerotic immune microenvironment and reinforcing their role as immune modulators in AS pathogenesis. Beyond diagnosis, these hub genes illuminate clinically actionable therapeutic targets: restoring their expression or function may simultaneously correct autophagy dysfunction and suppress ferroptosis, offering a dual-mechanism intervention strategy for atherosclerotic cardiovascular disease. Our findings provide a molecular framework for precision medicine approaches targeting ferroptosis-autophagy crosstalk and immune dysregulation in AS.
