Endothelial-cell induced radioresistance in glioblastoma

Abstract

Glioblastoma (GB) is the most frequent malignant primary brain tumour in adults. Despite multimodal therapy prognosis of this locally invasive tumour remains poor. The adaption of the cancer stem cell hypothesis has resulted in a postulated glioblastoma stem cell (GCS) population. This subpopulation is characterized by extended capabilities of self-renewal, tumour initiation and therapy resistance. GCS are located in close proximity to tumour vessels. This special microenvironment of GCS has been defined as perivascular stem cell niche. Endothelial co-culture of glioblastoma cells has been shown to enrich the GCS population in in vitro experiments. The present medical doctoral thesis seeks to establish an in vitro GB/endothelial co-culture model to analyze the effect of endothelial co-culture on stem cell properties of GB cells. Upregulation of stem cell markers, clonogenic survival, radioresistance and migration potential were tested. Therefore, two GB cell-lines (U87MG Katushka, T98G) were co-cultured with two different human endothelial cell-lines (HUVEC, hCMEC/D3) in direct and transfilter co-culture techniques. Radioresistance and clonogenic survival were investigated by flow cytometry and colony formation assays. Migration potential was analyzed through immunofluorescence and RT-PCR of its signalling pathway molecules SDF1/CXCR4, patch-clamp recording of its effector ion channels and on a functional level by in vitro real-time migration assays. Expression of stem cell markers, corresponding ion channels and effector molecules of invasive potential was analyzed with RT-PCR. In the established in vitro co-culture model, endothelial cells stimulate migration, illustrated by significantly increased chemotaxis of GB cells. This stimulation might be induced by observed endothelial- and auto-secreted SDF-1 and mediated by increased BK K+ channel activity. In addition, co-cultured GB cells show a significant upregulation of matrix metalloproteinases, indicating increased invasive potential. In this co-culture model, no effect on stem cell markers, clonogenicity or radioresistance of the GB cells, derived from colony formation assays and flow cytometry, was observed. The increased in vitro migration potential of GB cells by endothelial co-culture is presented for the first time. Our data are congruent with recent literature regarding elevated matrix metalloproteinases levels, increased SDF-1 expression and unaltered levels of most stem cell markers. In contrast, upregulation of stem cell markers, as well as increased clonogenic survival of GB cells through endothelial co-culture or conditioned medium reported by other groups could not be reproduced in the present experiments, given significant differences in the experimental setting. Considering the inherent limitations of an in vitro, two-dimensional co-culture model, the presented data provide evidence for an interaction between endothelial cells and glioblastoma cells in the form of stimulated migration and probably also invasion and therefore contribute to the understanding of the perivascular stem cell niches. This may provide target structures for new treatment approaches for GB.