Dysregulated autophagy-related genes in abdominal aortic aneurysm: Comprehensive bioinformatics analysis and experimental validation

Abstract

Abstract Background Abdominal aortic aneurysm (AAA) is a serious life-threatening cardiovascular disease that occurs in middle-aged and elderly people. Previous experimental studies have suggested that autophagy may be involved in the pathological process of AAA, but the pathogenesis of autophagy in AAA is unclear. We aim to identify and validate key potential autophagy-related genes involved in AAA through bioinformatics analysis to further elucidate the mechanisms of autophagy dysregulation in AAA. Methods The GSE57691 microarray dataset was downloaded from the Gene Expression Omnibus database (GEO), including 49 AAA samples and 10 normal aortic samples. 232 autophagy-related genes were obtained from the Human Autophagy Database (HADb). The GSE57691 dataset was crossed with the autophagy gene set to screen for differentially expressed autophagy-related genes (DE-ARGs) involved in AAA. In addition, gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed on the DE-ARGs in AAA using R software. Protein-protein interaction (PPI) networks were constructed using the STRING database, significant gene cluster modules were identified using the MCODE Cytoscape plugin, and hub genes in AAA associated DE-ARGs were screened using the CytoHubba Cytospace plugin. Meanwhile, DE-ARGs were calculated using the least absolute shrinkage selection algorithm (LASSO) algorithm. By crossing the LASSO calculation results and hub genes, the final key genes were identified, whose expression levels were further validated in AAA aortic samples by qRT-PCR. Finally, the transcription factor regulatory networks and target drugs of these key genes were predicted by the JASPAR database and DsigDB database, respectively. Results A total of 57 DE-ARGs were identified in aortic samples from normal controls and AAA. GO and KEGG analysis showed that these 57 DE-ARGs involved in AAA were particularly enriched in macroautophagy, PI3K-Akt signaling pathway, AMPK signaling pathway, and apoptosis. PPI results indicated that the 57 DE-ARGs interacted with each other. A total of 6 key genes (ATG5, ATG12, MTOR, BCL2L1, EIF4EBP1, and RPTOR) were identified using CytoHubba and LASSO algorithms. Detection of clinical samples by qRT-PCR indicated that ATG5, ATG12, BCL2L1, EIF4EBP1, and RPTOR expression was consistent with bioinformatic analysis. A regulatory network containing 6 key genes and 30 transcription factors was constructed through the JASPAR database. Finally, four targeted autophagy regulatory drugs, rapamycin, Temsirolimus, Sorafenib, and NVP-BEZ235, were screened by the DsigDB database. Conclusions Bioinformatics analysis identified 57 autophagy-related genes that may be involved in AAA. ATG5, ATG12, BCL2L1, EIF4EBP1and RPTOR may serve as potential drug targets and biomarkers as they regulate autophagy. These results expand the understanding of autophagy dysfunction in AAA and may contribute to the diagnosis and prognosis of AAA.