Critical genes in human photoaged skin identified using weighted gene co-expression network analysis

Abstract

Photoaging is unique to the skin and is accompanied by an increased risk of tumors. To explore the transcriptomic regulatory mechanism of skin photoaging, the epidermis, and dermis of 16 healthy donors (eight exposed and eight non-exposed) were surgically excised and detected using total RNA-Seq. Weighted gene co-expression network analysis (WGCNA) identified the most relevant modules with exposure. The hub genes were identified using correlation, p-value, and enrichment analysis. The critical genes were identified using Support Vector Machine-Recursive Feature Elimination (SVM-RFE) and least absolute shrinkage and selection operator (LASSO) regression, then enriched using single-gene GSEA. A competitive endogenous RNA (ceRNA) network was constructed and validated using qRT-PCR. Compared with non-exposed sites, 430 mRNAs, 168 lncRNAs, and 136 miRNAs were differentially expressed in the exposed skin. WGCNA identified the module MEthistle and 12 intersecting genes from the 71 genes in this module. The enriched pathways were related to muscle. The critical genes were KLHL41, MYBPC2, and ERAP2. Single-gene GSEA identified the Hippo signaling pathway, basal cell carcinoma, cell adhesion molecules, and other pathways. Six miRNAs and 18 lncRNAs related to the critical genes constituted the ceRNA network and were verified using qPCR. The differential expression of KLHL41, MYBPC2, and ERAP2 at the protein level was verified using immunohistochemistry. KLHL41, MYBPC2, and ERAP2 genes are related to skin photoaging. The prediction model based on the three critical genes can indicate photoaging. These critical genes may have a role in skin photoaging by regulating cell growth, intercellular adhesion, and substance metabolism pathways.