Exploring hub genes related to adipocytokines in keloids: a combined analysis integrating single-cell, Mendelian randomization and bulk transcriptome data with experimental verification

Background: Keloids are fibroproliferative skin tumors with excessive collagen deposition and are strongly linked to adipocytokine dysregulation. However, the underlying mechanisms remain unclear. This study aimed to identify potential therapeutic targets and elucidate adipocytokine-related pathological mechanisms underlying keloid formation. Methods: Hub genes were identified using differential expression analysis and machine learning. Mendelian randomization assessed causal relationships. Functional insights were gained through gene set enrichment analysis (GSEA) and drug prediction. Single-cell RNA sequencing (scRNA-seq) identified key cell types, and RT-qPCR validated gene expression. Results: We identified 818 differentially expressed genes, narrowing to seven key genes and two hub genes: PIK3R3 and ANGPTL5. MR indicated PIK3R3 as a causal risk factor for keloids, while ANGPTL5 showed no causal association. GSEA linked PIK3R3 to the TEL pathway and ANGPTL5 to adipocyte differentiation. Drug predictions included harmine for PIK3R3 and silica for ANGPTL5. scRNA-seq highlighted fibroblasts as key cells expressing these genes. RT-qPCR confirmed PIK3R3 upregulation in keloids, though ANGPTL5 results were inconsistent, possibly due to sample limitations. Conclusion: This study, based on the integrated re-analysis of existing publicly available transcriptomic data and combined with clinical sample validation, revealed the potential hub roles of PIK3R3 and ANGPTL5 in keloid pathogenesis. PIK3R3 was validated as a causal hub gene, while ANGPTL5 also showed relevance. The study provides new molecular evidence and mechanistic insights for understanding the adipocytokine-related pathological mechanisms of keloids, and suggests potential therapeutic directions such as harmine for future research.