Combining bioinformatics, network pharmacology, and artificial intelligence to predict the mechanism of resveratrol in the treatment of rheumatoid arthritis.

Background: Rheumatoid arthritis (RA) is a chronic autoimmune disorder that causes joint inflammation and destruction, resulting in significant physical and economic burdens. Finding effective and targeted therapy for RA remains a top priority. Resveratrol is a potential candidate with anti-inflammatory and immunomodulatory properties for RA treatment. This study aims to determine the therapeutic targets and signaling pathways of resveratrol in the treatment of RA.

Methods: The GSE205962 dataset downloaded from The Gene Expression Omnibus (GEO) database was used to obtain the differentially expressed genes (DEGs) in blood samples from the patients and the healthy. PharmMapper database and Cytoscape (v3.9.1) were applied to construct the resveratrol pharmacophore target network. Gene functional enrichment analysis, including the Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, was based on the BiNGo plug-in of Cytoscape and David's online tool. The intersection of the target genes of resveratrol and the DEGs were considered potential therapeutic genes (PT-genes). The Protein-Protein Interaction (PPI) network of PT-genes was constructed using the STRING tool, and the key therapeutic genes (KT-genes) were determined using the cytoHubba plug-in based on the Maximal Clique Centrality (MCC) algorithms. Molecular docking validation of resveratrol and therapeutic targets was performed based on the protein structure of KT-genes predicted by AlphaFold.

Results: A total of 2202 DEGs and 47PT-genes were identified. GO analysis showed that the three groups of genes, the DEGs, the resveratrol target genes, and the PT-genes, have similar results for the top-five gene functional enrichment. PT-genes were closely related to the pathways of metabolic pathways, pathways in cancer, proteoglycans in cancer, insulin signaling pathway, and chemokine signaling pathway. The common pathway enriched by KEGG for the DEGs, and the resveratrol target genes was up to 36 %. The nine KT-genes were ABL1, ANXA5, CASP3, HSP90AA1, LCK, MAP2K1, MAPK1, PIK3R1, and RAC1, and the lowest free energy indicating the resveratrol/protein affinity were -8.4, -7.4, -6.4, -6.7, -8.0, -7.9, -7.4, -6.7, and -7.9, respectively.

Conclusion: Nine KT-genes were identified and validated as the most potential therapeutic targets in the treatment of RA with resveratrol, which provide new insights into therapeutic mechanisms and may improve the efficiency of drug development.