Identification of p53-target genes in human papillomavirus-associated head and neck cancer by integrative bioinformatics analysis.

Abstract

Head and neck cancer (HNC) is a highly prevalent and heterogeneous malignancy. Although extensive efforts have been made to advance its treatment, the prognosis remained poor with increased mortality. Human papillomaviruses (HPV) have been associated with high risk in HNC. TP53, a tumor suppressor, is the most frequently altered gene in HNC, therefore, investigating its target genes for the identification of novel biomarkers or therapeutic targets in HPV-related HNC progression is highly recommended. Transcriptomic profiles from three independent gene expression omnibus (GEO) datasets, including 44 HPV+ and 70 HPV- HNC patients, were subjected to integrative statistical and Bioinformatics analyses. For the top-selected marker, further in-silico validation in TCGA and GTEx databases and experimental validation in 65 (51 HPV- and 14 HPV+) subjects with histologically confirmed head and neck squamous cell carcinoma (HNSCC) have been performed. A total of 498 differentially expressed genes (DEGs) were identified including 291 up-regulated genes and 207 down-regulated genes in HPV+ compared to HPV- HNSCC patients. Functional annotations and gene set enrichment analysis (GSEA) showed that the up-regulated genes were significantly involved in p53-related pathways. The integrative analysis between the Hub-genes identified in the complex protein-protein network and the top frequent genes resulting from GSEA showed an intriguing correlation with five biomarkers which are EZH2, MDM2, PCNA, STAT5A and TYMS. Importantly, the MDM2 gene showed the highest gene expression difference between HPV+ and HPV- HNSCC (Average log2FC = 1.89). Further in-silico validation in a large HNSCC cohort from TCGA and GTEx databases confirmed the over-expression of MDM2 in HPV+ compared to HPV- HNSCC patients (p = 2.39E-05). IHC scoring showed that MDM2 protein expression was significantly higher in HPV+ compared to HPV- HNSCC patients (p = 0.031). Our findings showed evidence that over-expression of MDM2, proto-oncogene, may affect the occurrence and proliferation of HPV-associated HNSCC by disturbing the p53-target genes and consequently the p53-related pathways.