Deciphering the oncogenic role of key genes in HNSC: insights from multi-omics and functional studies.

BACKGROUND: Recent evidence highlighted head and neck squamous cell carcinoma (HNSC) as a significant health concern, demanding precise clinical management. This study aimed to identify molecular biomarkers, delineate their functions, and assess clinical relevance via bioinformatics analysis and in vitro validation. METHODS: Using the GSE58911 microarray dataset, we identified differentially expressed genes (DEGs) meeting the criteria of "|log2 fold change (FC) |≥ 1 and false discovery rate (FDR) < 0.05". Subsequently, the top four hub genes were identified based on the degree method. Expression and functional analyses were conducted to explore their roles in tumorigenesis. RT-qPCR and bisulfite sequencing validated hub gene expression in HNSC cell lines. RESULTS: From the GSE58911 dataset, 1781 differentially expressed genes (DEGs) were identified, with 250 selected for further hub gene analysis. STRING and Cytoscape analyses revealed a protein-protein interaction (PPI) network with 122 nodes and 976 edges. MCODE analysis identified a key module with 28 nodes, and Cytohubba analysis highlighted MAPK3, MLLT4, PLAU, and SERPINE1 as top hub genes. Expression profiling in the TCGA-HNSC dataset showed significantly lower MAPK3 and MLLT4, and higher PLAU and SERPINE1 levels in HNSC samples compared to normal tissues. Validation using additional TCGA cohorts confirmed these findings and associated the dysregulation of these hub genes with poor prognosis. Methylation analyses indicated high levels of MAPK3 and MLLT4 methylation, and low levels of PLAU and SERPINE1 methylation in HNSC tissues. Genetic alterations analysis revealed mutations and amplifications, correlating with poorer overall survival. Subcellular localization studies showed distinct protein localizations, and IHC revealed altered protein expression levels. Immune cell infiltration analysis indicated associations between hub gene expression and immune cell types. Drug prediction analysis suggested potential therapeutic interventions, and functional assays in HNSC cell lines demonstrated that PLAU and SERPINE1 knockdown reduced malignancy traits. CONCLUSION: In conclusion, MAPK3, MLLT4, PLAU, and SERPINE1 emerge as promising biomarkers and therapeutic targets for HNSC.