Identification of key genes and potential therapeutic agents for infantile atopic dermatitis based on the CMap da-tabase

LI Qing-xian; CHONG Shu-bin; ZENG Bing-bing

doi:10.13820/j.cnki.gdyx.20241732

Abstract：Objective This study employed a multifaceted bioinformatics approach,integrating connectivity map-ping via the Connectivity Map(CMap)database,to elucidate the core pathogenic genes implicated in the etiology of infan-tile atopic dermatitis(AD)and to identify potential pharmacological agents for therapeutic intervention.Methods Le-veraging the GSE107361 microarray dataset,differentially expressed genes(DEGs)were meticulously screened from der-mal biopsies obtained from AD-afflicted infants(n=19)and age-matched healthy controls(n=18).Advanced bioin-formatics methodologies,encompassing pathway enrichment analysis,protein-protein interaction(PPI)network con-struction,and module mining,were employed to pinpoint the core pathogenic genes and validate their functional rele-vance.The CMap database was subsequently interrogated to screen for potential therapeutic compounds.Results Emplo-ying the DESeq2 algorithm with stringent criteria(absolute log2 fold change ≥1.0 and adjusted P-value＜0.05),a total of 1 100 upregulated and 1 808 downregulated DEGs were identified.Functional annotation and pathway enrichment analysis,utilizing a threshold of log10P＜-1.30,revealed that upregulated DEGs were significantly enriched in pathways associated with keratinocyte proliferation,cell cycle regulation,and the interleukin(IL)-4/IL-17/tumor necrosis factor-α(TNF-α)inflammatory cascade.Conversely,downregulated DEGs demonstrated significant associations with extracellular matrix(ECM)composition and skin barrier integrity pathways.To delineate the intricate protein-protein in-teraction landscape,a STRING network(confidence score ≥0.9)was constructed,revealing 13 critical modules.Upreg-ulated key hubs were identified within the PI3k/JAK signaling pathway,chemokine-mediated adaptive immune response,and the FGFR2-PDGF-VEGF axis.Downregulated key hubs predominantly clustered within ECM-related signaling pathways.Finally,DEG profiles were uploaded to the CMap database,applying an enrichment score threshold of-90％,resulting in the identification of the top 10 candidate drugs for infantile AD treatment.These included thioguanosine(-97.88％),trichostatin A(-94.86％),melatonin(-95.64％),and sanguinarine(-93.69％),which primarily target cell cycle arrest,IL-4/IL-17 signaling inhibition,and autophagy activation,respectively.Conclusion This comprehensive bioinformatics investigation has unveiled the core gene expression signature of infantile AD,characterized by IL-4/IL-17/TNF-α inflammatory activation and ECM barrier compromise.Furthermore,the identification of ten candidate drugs,including thioguanosine,possessing the potential to reverse the aberrant AD gene expression profile,pro-vides a robust foundation for subsequent preclinical validation and the development of targeted therapeutic strategies.