Stochastic Nanoscale Biophysical Cues as a Basis for the Induction of Glioblastoma-Like Transcriptional Programs in Astrocytes.

Although direct biological factors underlying the progression of Glioblastoma (GBM), an aggressive form of brain cancer, have been extensively studied, emerging evidence suggests that indirect biological triggers, such as traumatic brain injury (TBI), may also have a role. Since proteoglycans, secreted by reactive astrocytes and astroglial cells contribute to biophysical characteristics (stochastic topography, stiffness) of the brain, we postulated a role for stochastic nanoroughness in the induction of glioma following brain trauma. Using a model system to emulate such physical cues that manifest following traumatic injury, we demonstrate that human cortical astrocytes undergo spontaneous organization into spheroids in response to nanoroughness and retain the spheroid phenotype even upon withdrawal of the physical cues. Furthermore, spheroids serve as aggregation foci for naïve astrocytes, express activated MMP2, and disseminate upon implantation in the mouse brain. RNA-seq analysis revealed that astrocytes within spheroids differentially express genes, including p53, ADAMTS proteases, and NOTCH3, and adopt a transcriptional program enriched for GBM proneural signatures, with reactome analysis pointing toward astrocytes with GBM-associated transcriptional traits. Moreover, nanoroughness mediates a cross-talk between cancer cells and astrocytes through induced senescence. These findings implicate a role for stochastic biophysical cues in driving a potential malignant transformation of astrocytes.