Graphene oxide suppresses the growth and malignancy of glioblastoma stem cell-like spheroids via epigenetic mechanisms

Background Glioblastoma stem-like cells (GSCs) are hypothesized to contribute to self-renewal and therapeutic resistance in glioblastoma multiforme (GBM) tumors. Constituting only a small percentage of cancer cells, GSCs possess stem-like, tumor-initiating properties and display resistance to irradiation and chemotherapy. Thus, novel approaches that can be used to suppress GSCs are urgently needed. A new carbon material-graphene oxide (GO), has been reported to show potential for use in tumor therapy. However, the exact effect of GO on GSCs and the inherent mechanism underlying its action are not clear. In this study, we aimed to investigate the usefulness of GO to inhibit the growth and promote the differentiation of GSCs, so as to suppress the malignancy of GBM. Methods In vitro effects of GO on sphere-forming ability, cell proliferation and differentiation were evaluated in U87, U251 GSCs and primary GSCs. The changes in cell cycle and the level of epigenetic modification H3K27me3 were examined. GO was also tested in vivo against U87 GSCs in mouse subcutaneous xenograft models by evaluating tumor growth and histological features. Results We cultured GSCs to explore the effect of GO and the underlying mechanism of its action. We found, for the first time, that GO triggers the inhibition of cell proliferation and induces apoptotic cell death in GSCs. Moreover, GO could promote the differentiation of GSCs by decreasing the expression of stem cell markers (SOX2 and CD133) and increasing the expression of differentiation-related markers (GFAP and beta-III tubulin). Mechanistically, we found that GO had a striking effect on GSCs by inducing cell cycle arrest and epigenetic regulation. GO decreased H3K27me3 levels, which are regulated by EZH2 and associated with transcriptional silencing, in the promoters of the differentiation-related genes GFAP and beta-III tubulin, thereby enhancing GSC differentiation. In addition, compared with untreated GSCs, GO-treated GSCs that were injected into nude mice exhibited decreased tumor growth in vivo. Conclusion These results suggested that GO could promote differentiation and reduce malignancy in GSCs via an unanticipated epigenetic mechanism, which further demonstrated that GO is a potent anti-GBM agent that could be useful for future clinical applications.