The HMGB1-RAGE Axis Drives the Proneural-to-Mesenchymal Transition and Aggressiveness in Glioblastoma

Glioblastoma (GBM) remains the most lethal primary brain tumor, owing to profound intratumoral heterogeneity and the limited efficacy of standard treatments. The mesenchymal (MES) molecular subtype is particularly aggressive, exhibiting heightened invasiveness, therapy resistance, and dismal patient survival compared with the proneural (PN) subtype. Emerging evidence implicates the High Mobility Group Box 1 (HMGB1) protein and its cognate receptor, the Receptor for Advanced Glycation End Products (RAGE), as drivers of malignant progression, yet their contribution to the PN-to-MES transition is incompletely defined. We integrated transcriptomic analyses of TCGA-GBM and TCGA-LGG cohorts with immunohistochemistry on in-house patient specimens. Functional studies in patient-derived and established GBM cell lines included migration and invasion assays, tumorsphere formation assays, shRNA knockdowns, and Seahorse XF metabolic profiling to interrogate the HMGB1-RAGE axis. HMGB1 and RAGE expression was markedly elevated in MES GBM tissues and cell lines. Importantly, higher HMGB1 expression correlated with shortened overall survival (<i>p</i> < 0.009). HMGB1 silencing curtailed cell motility and downregulated core epithelial-to-mesenchymal transition markers (<i>N</i>-cadherin, Snail). RAGE knockdown diminished tumorsphere formation efficiency and reduced transcription of stemness genes (OCT4), underscoring its role in sustaining tumor-initiating capacity. Metabolically, HMGB1/RAGE activation boosted both mitochondrial respiration and glycolysis, conferring the bioenergetic flexibility characteristic of MES GBM. The HMGB1-RAGE signaling axis orchestrates mesenchymal identity, invasiveness, stem cell-like properties, and metabolic reprogramming in GBM. Targeting this pathway may disrupt the PN-to-MES transition, mitigate therapeutic resistance, and ultimately improve outcomes for glioblastoma patients.