Bioinformatics prediction and experimental verification identify B4GALT1 as a diagnostic biomarker of glioblastoma

Glycolysis plays a key role in the progression of glioblastoma (GBM). Gene expression and clinical information were obtained from The Cancer Genome Atlas, Genotype-Tissue Expression, the Chinese Glioma Genome Atlas database, and five Gene Expression Omnibus datasets. Twenty-one prognosis-related glycolytic differentially expressed genes were identified. The glycolysis-related gene signature associated with poor prognosis was constructed using six upregulated glycolytic genes, including beta-1, 4-galactosyltransferase1 (B4GALT1), and beta-glucuronidase (GUSB). The main immune cell types in GBM samples were M2 macrophages and resting memory CD4+ T cells, and a high-risk score was associated with high neutrophil infiltration. Clinical GBM tissues exhibited higher B4GALT1 and GUSB expression than normal tissues. B4GALT1 expression positively correlated with multiple immune checkpoint genes, and the knockdown of B4GALT1 inhibited the proliferation of GBM cells in vitro and in vivo. Knockdown of B4GALT1 reduced lactic acid levels and increased those of tumor necrosis factor-α and interferon-γ. Accordingly, a prognostic model based on six glycolytic genes could be used to predict poor prognosis and high immune infiltration in patients with GBM. Thus, B4GALT1 may serve as a promising prognostic biomarker for GBM therapy.