Bioinformatics and experimental validation of druggable targets in non-alcoholic fatty liver disease

Non-alcoholic fatty liver disease (NAFLD) is a common metabolic disease with high heterogeneity and currently lacks approved targeted therapies. Identifying druggable genes with diagnostic relevance and potential translational value may facilitate further research into precision medicine approaches for NAFLD. This study integrated transcriptome data from three independent GEO datasets (GSE33814, GSE63067, and GSE89632) and used sva and limma to correct for batch effects. Differentially expressed genes (DEGs), druggable genes obtained from DGIdb, and genes from WGCNA modules significantly associated with NAFLD were intersected to identify candidate genes. Candidate genes were systematically evaluated using functional enrichment analysis (GO, KEGG, GSEA), immune infiltration analysis (CIBERSORT), receiver operating characteristic (ROC) analysis, and co-expression networks. Finally, the expression of key genes was verified by immunohistochemistry (IHC) and immunofluorescence (IF). A total of nine key candidate drug targets were identified: FABP4, ADAMTS1, FOS, GPR88, IL1RL1, CD52, JUN, SERPINE1, and THBS1. These genes are primarily involved in metabolism, inflammatory response, extracellular matrix remodeling, and immune regulation. ROC analysis showed that FOS and JUN had high diagnostic accuracy. Furthermore, IHC and IF results demonstrated that CD52 was significantly upregulated in NAFLD tissues, suggesting its potential relevance as a candidate target for further investigation. This study systematically identified key druggable genes for NAFLD through bioinformatics analysis and partially validated selected candidates experimentally. In particular, CD52 was upregulated in NAFLD tissues, suggesting a potential association with NAFLD-related pathological alterations. These findings provide new insights into the molecular pathogenesis of NAFLD and may provide a basis for future target validation studies.