A Novel Competing Endogenous RNA Network Associated With the Pathogenesis of Graves' Ophthalmopathy

Background: Growing evidence has recently revealed the characteristics of long noncoding (lncRNA)/circular RNA (circRNA)-microRNA (miRNA)-mRNA networks in numerous human diseases. However, a scientific lncRNA/circRNA-miRNA-mRNA network related to Graves' ophthalmopathy (GO) remains lacking. Materials and methods: The expression levels of RNAs in GO patients were measured through high-throughput sequencing technology, and the results were proven by quantitative real-time PCR (qPCR). We constructed a protein-protein interaction (PPI) network using the Search Tool for the Retrieval of Interacting Genes (STRING) database and identified hub genes by the Cytoscape plug-in CytoHubba. Then, the miRNAs related to differentially expressed lncRNAs/circRNAs and mRNAs were predicted through seed sequence matching analysis. Correlation coefficient analysis was performed on the interesting RNAs to construct a novel competing endogenous RNA (ceRNA) network. Results: In total, 361 mRNAs, 355 circRNAs, and 242 lncRNAs were differentially expressed in GO patients compared with control patients, 166 pairs were identified, and ceRNA networks were constructed. The qPCR results showed that 4 mRNAs (THBS2, CHRM3, CXCL1, FPR2) and 2 lncRNAs (LINC01820:13, ENST00000499452) were differentially expressed between the GO patients and control patients. Conclusion: An innovative lncRNA/circRNA-miRNA-mRNA ceRNA network between GO patients and control patients was constructed, and two important ceRNA pathways were identified, the LINC01820:13-hsa-miR-27b-3p-FPR2 ceRNA pathway and the ENST00000499452-hsa-miR-27a-3p-CXCL1 pathway, which probably affect the autoimmune response and inflammation in GO patients.