Integrated aqueous humor ceRNA and miRNA–TF–mRNA network analysis reveals potential molecular mechanisms governing primary open-angle glaucoma pathogenesis

PURPOSE: To conduct an integrated bioinformatics analysis of extant aqueous humor (AH) gene expression datasets in order to identify key genes and the regulatory mechanism governing primary open-angle glaucoma (POAG) progression. METHODS: Two datasets (GSE101727 and GSE105269) were downloaded from the Gene Expression Omnibus, and the messenger RNAs (mRNAs), microRNAs (miRNAs), and long noncoding RNAs (lncRNAs) were identified between controls and POAG patients. Differentially expressed (DE) mRNAs and DElncRNAs were then subjected to pathway enrichment analyses, after which a protein–protein interaction (PPI) network was generated. This network was then expanded to establish lncRNA–miRNA–mRNA and miRNA–transcription factor (TF)–mRNA networks. RESULTS: The GSE101727 dataset was used to identify 2746 DElncRNAs and 2208 DEmRNAs, while the GSE105269 dataset was used to identify 45 DEmiRNAs. We ultimately constructed a competing endogenous RNA (ceRNA) network incorporating 47 lncRNAs, six miRNAs, and 17 mRNAs. The proteins encoded by these 17 hub mRNAs were found to be significantly enriched for activities that may be linked to POAG pathogenesis. In addition, we generated a miRNA–TF–mRNA regulatory network containing two miRNAs (miR-135a-5p and miR-139-5p), five TFs (TGIF2, TCF3, FOS, and so on), and five mRNAs (SHISA7, ST6GAL2, TXNIP, and so on). CONCLUSION: The SHISA7, ST6GAL2, TXNIP, FOS, and DCBLD2 genes may be viable therapeutic targets for the prevention or treatment of POAG and are regulated by the TFs (TGIF2, HNF1A, TCF3, and FOS).