Diagnostic biomarkers and potential drug targets for coronary artery disease as revealed by systematic analysis of lncRNA characteristics.

BACKGROUND

The expression profile of lncRNAs in coronary artery disease (CAD) patients has not yet been fully explored. Therefore, the current study aimed to investigate lncRNA-based prognostic biomarkers for CAD.

METHODS

The expression profiles of lncRNA and messenger RNA (mRNA) were downloaded from the Gene Expression Omnibus (GEO) database. Differentially expressed lncRNA (DElncRNAs) and DEmRNAs were identified from CAD and normal samples, and weighted gene co-expression network analysis (WGCNA) was conducted. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed to investigate the principal functions of significantly dysregulated genes. The potential drugs of new CAD-specific genes were identified by network distance method. Receiver operating characteristic (ROC) was used to verify the classification performance of genes.

RESULTS

A total of 512 differentially expressed genes (DEGs) and 308 DElncRNAs were identified from GSE113079 dataset to classify CAD samples. Through WGCNA co-expression analysis, 24 co-expression modules were obtained. A total of 187 DElncRNAs and 253 DEGs were determined from 7 modules correlated with CAD. Functional enrichment analysis showed that these DEGs were mainly related to inflammatory and immune-related pathways. Furthermore, 36 regulatory pairs of significantly shared micro RNAs (miRNAs) were identified as dysregulated lncRNA-mRNA (LRM-CAD), which contained 11 lncRNAs and 33 genes. Compared with a single lncRNA or gene, LRM-CAD showed stronger classification performance [average area under the curve (AUC) =0.958]. We screened 3 potential therapeutic drugs, DB09105, DB12371, and DB12612, a by binding drug-target gene interaction network. Molecular docking verified that the S1PR1 gene bound relatively closely to DB12371 and DB12612. The ROC analysis on external data sets showed that S1PR1, AC012640.4, and S1PR1-AC012640.4 could effectively distinguish CAD samples from control samples.

CONCLUSIONS

We provided a transcriptome overview of abnormally expressed lncRNAs in CAD patients and identified novel biomarkers for diagnosing CAD.