ABSTRACT
BACKGROUND: A growing body of research has revealed the connection of metabolism reprogramming and tumor progression, yet how metabolism reprogramming affects inter-patient heterogeneity and prognosis in head and neck squamous cell carcinoma (HNSCC) still requires further explorations. METHODS: A cellular hierarchy framework based on metabolic properties discrepancy, METArisk, was introduced to re-analyze the cellular composition from bulk transcriptomes of 486 patients through deconvolution utilizing single-cell reference profiles from 25 primary and 8 metastatic HNSCC sample integration of previous studies. Machine learning methods were used to identify the correlations between metabolism-related biomarkers and prognosis. The functions of the genes screened out in tumor progression, metastasis and chemotherapy resistance were validated in vitro by cellular functional experiments and in vivo by xenograft tumor mouse model. RESULTS: Incorporating the cellular hierarchy composition and clinical properties, the METArisk phenotype divided multi-patient cohort into two classes, wherein poor prognosis of METArisk-high subgroup was associated with a particular cluster of malignant cells with significant activity of metabolism reprogramming enriched in metastatic single-cell samples. Subsequent analysis targeted for phenotype differences between the METArisk subgroups identified PYGL as a key metabolism-related biomarker that enhances malignancy and chemotherapy resistance by GSH/ROS/p53 pathway, leading to poor prognosis of HNSCC. CONCLUSION: PYGL was identified as a metabolism-related oncogenic biomarker that promotes HNSCC progression, metastasis and chemotherapy resistance though GSH/ROS/p53 pathway. Our study revealed the cellular hierarchy composition of HNSCC from the cell metabolism reprogramming perspective and may provide new inspirations and therapeutic targets for HNSCC in the future.
Subject(s)
Head and Neck Neoplasms , Humans , Animals , Mice , Squamous Cell Carcinoma of Head and Neck/drug therapy , Squamous Cell Carcinoma of Head and Neck/genetics , Head and Neck Neoplasms/drug therapy , Head and Neck Neoplasms/genetics , Reactive Oxygen Species , Tumor Suppressor Protein p53/genetics , Prognosis , Gene Expression Regulation, NeoplasticABSTRACT
Head and neck squamous cell carcinoma (HNSCC) is one of the most aggressive malignancies that have a poor prognosis. Necroptosis has been demonstrated in recent years to be a form of inflammatory cell death occurring in multicellular organism, which plays complex roles in cancer. However, the expression of necroptosis-related miRNAs and genes in HNSCC and their correlations with prognosis remain unclear. In this study, R software was used to screen differentially expressed miRNAs downloaded from The Cancer Genome Atlas. A prognostic model containing six necroptosis-related miRNAs (miR-141-3p, miR-148a-3p, miR-331-3p, miR-543, miR-425-5p, and miR-7-5p) was generated, whose risk score was validated as an independent prognostic factor for HNSCC. Target genes of the key miRNAs were obtained from TargetScan, miRDB, and miRTarBase, and 193 genes in the intersection of the three databases were defined as consensus genes. Kyoto Encyclopedia of Genes and Genomes and Gene Ontology analyses indicated that the composition of the tumor microenvironment as well as specific pathways may be closely related to necroptosis in HNSCC. Nine key genes were also obtained by the MCODE and cytoHubba plug-ins of Cytoscape: PIK3CD, NRAS, PTK2, IRS2, IRS1, PARP1, KLF4, SMAD2, and DNMT1. A prognostic model formed by the key gene was also established, which can efficiently predict the overall survival of HNSCC patients. In conclusion, necroptosis-related miRNAs and genes play important roles in tumor development and metastasis and can be used to predict the prognosis of HNSCC.
