Identification of keratinocyte-associated genes for immune characterization and drug response prediction in oral squamous cell carcinoma.

Background: Oral squamous cell carcinoma (OSCC) is one of the most frequent types of head and neck tumor. Keratinocytes play a crucial part in tumor cell growth but their role in OSCC remains unknown. Methods: We obtained single-cell RNA sequencing (scRNA-seq) data and bulk RNA sequencing data of OSCC from the Gene Expression Omnibus (GEO) database and utilized the Seurat package for quality control, downscaling, and clustering of the scRNA-seq data. The CellChat package was utilized to develop a ligand-receptor network of keratinocytes. Subsequently, high-dimensional weighted gene co-expression network analysis (hdWGCNA) and differential expression analysis were employed to identify keratinocyte-related gene modules and obtain hub genes. The predictive value of the hub genes was assessed by constructing a diagnostic model, and the CIBERSORT and ESTIMATE algorithms were utilized to analyze the correlation between immune infiltration and the diagnostic model. Finally, the mRNA expressions of the screened genes were measured, and their effects on the proliferation, migration, and invasion ability of OSCC cells were explored using Results: We identified eight major cellular subpopulations including T cells and keratinocytes. Cellular communication revealed that keratinocytes may have close mutual communication with macrophages, fibroblasts, and endothelial cells. The hdWGCNA screening classified nine keratinocyte-related modules and 50 hub genes were extracted, among them Conclusion: Our study provided novel biomarkers targeting keratinocytes for the treatment of OSCC.