Glioma is a primary tumor derived from central nervous system glial cells. RNA binding motif protein 25 (RBM25) has been implicated in glioma progression, yet its underlying molecular mechanism remains incompletely understood. In this study, glioma-related datasets were retrieved from the Proteomic Data Commons (PDC) and Gene Expression Omnibus (GEO) databases. Core glioma-associated targets were screened using machine learning algorithms. Protein expression was assessed by Western blot, while migration, and invasion, cell cycle and apoptosis were analyzed by Transwell assay and flow cytometry. Chromatin immunoprecipitation (ChIP) and dual-luciferase reporter assays were used to validate the interaction between transcription factors and genes. Proteomic analysis revealed distinct differences between glioma and control groups, with differentially expressed proteins mainly enriched in mitochondrial energy metabolism, synaptic function, and neurodegenerative disease pathways. Transcriptomic profiles also exhibited significant alterations, with RBM25 and HSPA8 showing consistent changes at both protein and RNA levels. Multiple machine learning models identified RBM25, NKTR, MAP2K6, TRAPPC3, and HSPA8 as key genes associated with glioma, with RBM25 and NKTR showing strong relevance in model-based feature selection. RBM25 was upregulated in glioma and linked to neuronal signaling pathways, whereas HSPA8 and TRAPPC3 were downregulated. RBM25 knockdown suppressed glioma cell proliferation, migration, and invasion, and induced cell cycle arrest and apoptosis. In conclusion, RBM25 contributes to glioma malignancy and may serve as a potential biomarker and therapeutic target for glioma.
Xu et al. (Fri,) studied this question.