3D brain vascular niche model captures invasive behavior and gene signatures of glioblastoma

Glioblastoma (GBM) is a highly lethal brain cancer with no effective treatments; understanding how GBM cells respond to tumor microenvironment remain challenging as conventional cell cultures lack cytoarchitecture while in vivo models present complexity all at once. Developing a culture system to bridge the gap is thus crucial. Here, we employed a multicellular approach using human glia and vascular cells to optimize a 3-dimensional (3D) brain vascular niche model that enabled not only long-term culture of patient derived GBM cells but also recapitulation of key features of GBM heterogeneity and gene reprogramming in accordance with invasion behaviors and vascular association. Remarkably, vascular contact of GBM cells induced genes concerning neuronal, synaptic, and glia interaction, as well as immune suppression. This gene signature also displayed regional enrichment particularly in leading edge and microvascular proliferation zones and predicted worse prognosis for GBM patients. Furthermore, gene variance analysis uncovered histone demethylation and xylosyltransfererase activity as the main themes for the top induced genes in vivo, signifying the importance of chromatin remodeling and posttranslational modification proteoglygan such as CSPG in GBM adaption. Finally, our model also demonstrated a capacity to maintain GBM quiescence state and a protective niche against chemotherapy. In summary, our 3D model recapitulated key features of GBM heterogeneity and unveiled a previously unappreciated influence of brain glia-vascular unit for transcriptional adaption for neural/synaptic interaction and immunosuppression. The brain vascular niche model may serve as an effective bridge between 2D cultures and in vivo models for GBM modeling and therapeutic development. To investigate how GBM cells respond to different tumor microenvironment, we establisehd various in vitro and in vivo models with patient-derived GBM cells. Then, we performed gene expression profiling analysis using data obtained from RNA-seq of 2 different patient-derived cell lines cultured in 4 different conditions.