Abstract
BACKGROUND: Autophagy is an evolutionary conserved important process for the turnover of intracellular substances in eukaryotes and is closely related to the development of atrial fibrillation (AF). The aim of this study is to identify and validate potential autophagy-related genes (ARGs) of AF through bioinformatics analysis and experimental validation. METHODS: We downloaded two data sets from the Gene Expression Omnibus (GEO) database, GSE14975 and GSE31821. After merging the data of the two microarrays, adjusting the batch effect, and integrating the differentially expressed genes (DEGs) with ARGs to obtain differentially expressed autophagy-related genes (DEARGs). Functional and pathway enrichment analyses were carried out based on Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG). Use the STRING database to construct a protein-protein interaction (PPI) network. Finally, mRNA expression levels of DEARGs were validated in right atrial tissue samples from AF patients and non-AF controls by qRT-PCR. RESULTS: Through bioinformatics analysis, we finally identified 11 DEARGs (CDKN1A, CXCR4, DIRAS3, HSP90AB1, ITGA3, PRKCD, TP53INP2, DAPK2, IFNG, PTK6, and TNFSF10) in AF using [log2 (fold change)] > 0.5 and P < 0.05. In the pathway enrichment analysis, the most significantly enriched pathway was the autophagy pathway. The results of validation showed that the expression levels of CXCR4, DAPK2, and TNFSF10 corroborating with our computational findings, and the results were statistically significant (P<0.05). CONCLUSION: Our study demonstrates that these 11 potential crucial ARGs, especially CXCR4, DAPK2, and TNFSF10, may be potential biomarkers and therapeutic targets in AF, which will help the personalized treatment of AF patients.