Identification of cuproptosis-related biomarkers in aortic dissection：new insights from bioinformatic analysis

Abstract

Abstract Objective: Aortic dissection (AD) is a cardiovascular disease with a high mortality rate. And the mechanisms of AD are still poorly understood. Cuproptosis is a novel form of programmed cell death that may contributes to occurrence and development of various cardiovascular disease. Therefore, we intend to explore the potential association between cuproptosis-related genes (CRGs) and AD to provide a new biomarker for the treatment and prognosis of AD. Methods: CRGs were obtained from previous literature. Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Set Enrichment Analysis (GSEA) were used to explore the correlation between AD and CRGs. The RNA-seq dataset GSE153434 was used for screening differentially expressed CRGs (DECRGs) between AD and normal group; LASSO and RF machine learning algorithms were used to identify biomarker CRGs and receiver operating characteristic (ROC) curves were used to assess diagnostic efficacy. PPI network was constructed to reveal the interaction between marker CRGs and core CRGs. Subsequent single-gene GSEA and GSVA were performed to explore the function of biomarker CRGs; The mRNA-miRNA-lncRNA network were built to explore the regulatory relationship based on the marker genes. Potential marker CRGs targeted drugs were obtained from Drug Gene Interaction Database (DGIdb). Finally, single-Cell RNA-Seq dataset GSE213740 was used for verification of marker genes distribution and expression in different cell types of aortic tissue.; The RNA-seq dataset GSE52093 was used as validation set for marker genes. Results: First we found potential correlation between AD and CRGs. Then 10 differentially expressed CRGs were obtained from GSE153434, comprising 6 upregulated genes (TOP1M, SLC7A5, WDR12, MAD2L2, LDLR, and SHMT2) and 4 downregulated genes (FZD8, MPC1, CNN1, and N6AMT1). Subsequently, we used LASSO to identify 7 optimal biomarker DECRGs (TOP1M, WDR12, LDLR, FZD8, MPC1, CNN1, and N6AMT1). Then RF model and ROC curves both indicated diagnostic capabilities of those marker genes. PPI network analysis revealed wide interactions between those marker CRGs and core CRGs. Moreover, GSEA and GSVA of marker genes mainly enriched in pivotal pathways related to AD and cuproptosis. Through a drug-gene interaction exploration, we pinpointed potential drugs targeting LDLR, TOP1MT, FZD8 and N6AMT1. Furthermore, the ceRNA network around the 7 marker genes unveiled their regulatory associations with 94 miRNAs and 292 lncRNAs including miR-27a, let-7b, XIST and PVT1. Using Single-cell RNA-seq data from GSE213740, we corroborated the distribution and expression patterns of these marker genes across diverse cell types in aortic tissue. Lastly validation dataset GSE52093 showed that FZD8, MPC1, CNN1 and N6AMT1 expression were consistent with the GSE153434 dataset. Conclusion: Our study systematically illustrates the potential relationship between cuproptosis and AD. We identified several biomarker genes including CNN1, MPC1 and LDLR, which were involved in various pathways related to AD progression. Our findings may provide new insights in diagnosis and clinical treatment strategies for AD.