Mechanosensitive expression of the mesenchymal subtype marker connective tissue growth factor in glioblastoma.

Mechanical forces created by the extracellular environment regulate biochemical signals that modulate the inter-related cellular phenotypes of morphology, proliferation, and migration. A stiff microenvironment induces glioblastoma (GBM) cells to develop prominent actin stress fibres, take on a spread morphology and adopt trapezoid shapes, when cultured in 2D, which are phenotypes characteristic of a mesenchymal cell program. The mesenchymal subtype is the most aggressive among the molecular GBM subtypes. Recurrent GBM have been reported to transition to mesenchymal. We therefore sought to test the hypothesis that stiffer microenvironments-such as those found in different brain anatomical structures and induced following treatment-contribute to the expression of markers characterising the mesenchymal subtype. We cultured primary patient-derived cell lines that reflect the three common GBM subtypes (mesenchymal, proneural and classical) on polyacrylamide (PA) hydrogels with controlled stiffnesses spanning the healthy and pathological tissue range. We then assessed the canonical mesenchymal markers Connective Tissue Growth Factor (CTGF) and yes-associated protein (YAP)/transcriptional co-activator with PDZ-binding motif (TAZ) expression, via immunofluorescence. Replating techniques and drug-mediated manipulation of the actin cytoskeleton were utilised to ascertain the response of the cells to differing mechanical environments. We demonstrate that CTGF is induced rapidly following adhesion to a rigid substrate and is independent of actin filament formation. Collectively, our data suggest that microenvironmental rigidity can stimulate expression of mesenchymal-associated molecules in GBM.

Differential response of patient-derived primary glioblastoma cells to environmental stiffness.

The ability of cancer cells to sense external mechanical forces has emerged as a significant factor in the promotion of cancer invasion. Currently there are conflicting reports in the literature with regard to whether glioblastoma (GBM) brain cancer cell migration and invasion is rigidity-sensitive. In order to address this question we have compared the rigidity-response of primary patient-derived GBM lines. Cells were plated on polyacrylamide gels of defined rigidity that reflect the diversity of the brain tissue mechanical environment, and cell morphology and migration were analysed by time-lapse microscopy. Invasiveness was assessed in multicellular spheroids embedded in 3D matrigel cultures. Our data reveal a range of rigidity-dependent responses between the patient-derived cell lines, from reduced migration on the most compliant tissue stiffness to those that are insensitive to substrate rigidity and are equally migratory irrespective of the underlying substrate stiffness. Notably, the rigidity-insensitive GBM cells show the greatest invasive capacity in soft 3D matrigel cultures. Collectively our data confirm both rigidity-dependent and independent behaviour in primary GBM patient-derived cells.