Abstract
Gliomas are the most prevalent malignant primary brain tumors with poor outcome, and four different molecular subtypes (Mesenchymal, Proneural, Neural, and Classical) are popularly applied in scientific researches and clinics of gliomas. Public databases contain an abundant genome-wide resource to explore the potential biomarker and molecular mechanisms using the informatics analysis. The aim of this study was to discover the potential biomarker and investigate its effect in gliomas. Weighted gene co-expression network analysis (WGCNA) was used to construct the co-expression modules and explore the biomarker among the dataset CGGA mRNAseq_693 carrying 693 glioma samples. Functional annotations, ROC, correlation, survival, univariate, and multivariate Cox regression analyses were implemented to investigate the functional effect in gliomas, and molecular experiments in vitro were performed to study the biological effect on glioma pathogenesis. The brown module was found to be strongly related to WHO grade of gliomas, and KEGG pathway analysis demonstrated that TNFRSF1A was enriched in MAPK signaling pathway and TNF signaling pathway. Overexpressed TNFRSF1A was strongly related to clinical features such as WHO grade, and functioned as an independent poor prognostic predictor of glioma patients. Notably, TNFRSF1A was preferentially upregulated in the Mesenchymal subtype gliomas (Mesenchymal-associated). Knockdown of TNFRSF1A inhibited proliferation and migration of glioma cell lines in vitro. Our findings provide a further understanding of the progression of gliomas, and Mesenchymal-associated TNFRSF1A might be a promising target of diagnosis, therapy, and prognosis of gliomas.