Mechanotransduction in glioma stem cell fate determination: from niche mechanics to therapeutic vulnerability and state plasticity.

Glioblastoma (GBM), the most aggressive primary brain malignancy, harbors glioma stem cells (GSCs) that drive tumor initiation, therapeutic resistance, and recurrence. While the biochemical regulation of GSCs through Notch, Wnt, and Hedgehog pathways is well established, accumulating evidence reveals that the physical and mechanical properties of the tumor microenvironment-including matrix stiffness, solid stress, viscoelasticity, and topographical architecture-constitute a fundamental yet underappreciated layer of GSC regulation. This review provides a comprehensive synthesis of how mechanotransduction pathways regulate GSC fate decisions, including self-renewal, differentiation, migration, and drug resistance. We systematically characterize the mechanical properties of distinct GSC niches, delineate the key mechanosensors and signaling cascades-including integrins/FAK, CD44-hyaluronic acid, Piezo1 channels, and YAP/TAZ-Hippo signaling-that connect extracellular mechanics to intracellular fate decisions. Crucially, aligning with recent single-cell transcriptomic paradigms, we conceptualize GSCs not as a static cellular hierarchy, but as highly plastic populations transitioning among dynamic states (e.g., neural progenitor-like, mesenchymal-like). We examine how mechanical cues serve as profound biophysical drivers of this state plasticity, alongside the emerging roles of nuclear mechanotransduction and mechanical memory. We critically evaluate the discrepancy between 2D and 3D mechanical responses, consolidate the ongoing debate regarding GBM tissue stiffness, and assess the therapeutic potential of targeting mechanotransduction to eliminate GSCs. By integrating insights from mechanobiology, stem cell biology, neuro-oncology, and bioengineering, we propose the "mechano-stemness axis" as a conceptual framework positioning mechanical forces as critical modulators of GSC dynamic plasticity, and identify critical knowledge gaps requiring direct validation in patient-derived models.