Identifying the Potential Therapeutic Targets for Atopic Dermatitis Through the Immune Infiltration Analysis and Construction of a ceRNA Network

PURPOSE: This study was meant to analyze immune infiltration and construct a ceRNA network to explore the new therapeutic targets for atopic dermatitis (AD) through bioinformatics way. PATIENTS AND METHODS: We downloaded the AD patients’ RNA expression profile datasets (GSE63741, GSE124700) from the Gene Expression Omnibus (GEO) database, which were analyzed through the GEO2R. We explored the hub genes by the enrichment analysis and the protein–protein interaction (PPI) analysis. Moreover, we estimated immune cell types and their proportions by ImmucellAI. GSE121212 dataset validation was performed to verify the robustness of the hub genes. Then, a ceRNA network was constructed by the miRWalk, miRNet, miRDB, DIANA, TargetScan, and starbase database. Finally, gene expression analysis was performed by using RT-qPCR. RESULTS: In total, we detected 22 differentially expressed genes (DEGs), which contained 8 downregulated genes and 14 upregulated genes. There were 5 hub genes confirmed as key genes through PPI network analysis and the ROC curves. KEGG pathway analysis revealed that they were significantly enriched in the IL-17 signaling pathway and GO analysis showed mainly in the immune cell chemotaxis. The immune infiltration profiles were different between normal controls and AD, and each of the key genes (S100A7, S100A8, S100A9, and LCE3D) was significantly correlated with the main infiltration cell of AD. A lncRNA–miRNA–mRNA ceRNA network containing the key genes was constructed, and NEAT1 and XIST, the core of ceRNA network, were significantly overexpressing verified by RT-qPCR in AD patients. CONCLUSION: Altogether, the key genes and their ceRNA network provided a novel perspective to the immunomodulation of AD, which may be potential and new therapeutic targets for AD.