Identification of Mitophagy‐Related Genes With Diagnostic Value in Atherosclerosis Using Bioinformatics Analysis and Experiment Validation

Atherosclerosis (AS) is a chronic inflammatory condition with complex molecular underpinnings, where mitophagy-selective mitochondrial autophagy-plays a critical yet poorly defined role. By integrating bulk and single-cell RNA sequencing data from human atherosclerotic plaques, we analyzed 20 mitophagy-related genes and identified 15 that were dysregulated in AS. Machine learning approaches (Random Forest and SVM-RFE) pinpointed four hub genes-PINK1, TOMM40, TOMM7, and VDAC1-which formed the basis of a diagnostic model with solid predictive performance. Single-cell analysis of over 106,000 cells revealed endothelial cells as mitophagy-active and dominant in AS lesions. Trajectory analysis distinguished disease-associated endothelial subtypes, while CellChat uncovered intensified MIF signaling via CD74-CD44 and CD74-CXCR4 axes in mitophagy-high endothelial cells. SCENIC analysis further identified CEBPD, FOS, and JUN family transcription factors as key regulators. Experimental validation using ox-LDL-treated RAW264.7 macrophages confirmed differential expression of all four hub genes. Collectively, our findings highlight endothelial mitophagy dysregulation and immune crosstalk as central to AS pathogenesis and offer promising diagnostic markers and therapeutic targets.