Identification of novel lipid metabolism-related biomarkers of aortic dissection by integrating single-cell RNA sequencing analysis and machine learning algorithms

Zhechen Li¹Yusong Deng²Fei Xiao²Jiashu Sun²Qixu Zhao²Zetong Zheng¹Gang Li^2*

• ¹Beijing Luhe Hospital Affiliated to Capital Medical University, Beijing, China
• ²Beijing Anzhen Hospital Affiliated to Capital Medical University, Beijing, China

Aortic dissection (AD), a lethal disease with increasing incidence and limited preventive options, is characterized by aortic media degeneration and inflammatory cell infiltration. In recent years, dysregulation of lipids metabolism are increasingly recognized as pathological characteristic of AD. However, the exact molecular regulators and critical genetic determinants involved in this process have not been fully elucidated.This study employed an integrative approach combining single-cell RNA sequencing and machine learning to identify novel lipid metabolism-related biomarkers in aortic dissection. Single-cell RNA sequencing data from aortic dissection and control samples were processed to analyze lipid metabolism activity and identify differentially expressed genes. Machine learning algorithms and protein-protein interaction networks were then used to prioritize biomarkers, which were further validated through bulk RNA-seq analysis and immune infiltration studies. Functional characterization included cell-cell communication analysis and pseudo-time trajectory reconstruction to elucidate the roles of candidate genes in aortic dissection pathogenesis. This multi-modal strategy identified PLIN2 and PLIN3 as key regulators of lipid metabolism in aortic dissection, highlighting their potential as diagnostic markers and therapeutic targets.Our analysis revealed significant up-regulation of lipid metabolism in aortic dissection, with PLIN2 and PLIN3 emerging as central regulators. Single-cell profiling showed these genes were highly expressed in monocytic cells, correlating with enhanced inflammatory signaling (e.g., SPP1, GALECTIN). Machine learning and bulk RNA-seq validation confirmed their diagnostic potential. Pseudo-time analysis linked PLIN2 to early monocyte differentiation, while cell-cell communication studies implicated it in pro-inflammatory crosstalk with smooth muscle cells. Molecular docking screened for potential therapeutic compounds that may target PLIN2, among which ketoconazole was identified. Critically, the upregulation of PLIN2 and its specific expression in macrophages were further confirmed in an Ang II-induced aortic dissection mouse model.These findings suggest that *PLIN2/PLIN3* could be key mediators of metabolic dysregulation and immune activation in aortic dissection.