Functional role of tumor-associated glial cells (TAGs) in brain cancer progression

Several studies have shown that host cells in the tumor bed undergo genetic reprogamming. The resulting changes may affect and often promote cancer cell growth. Currently however, little is known about the role of glial cells in central nervous system malignancies. We have isolated host cells from brain tumors, so-called glial cells and analysed changes in their gene expression profiles. Notably, tumor-associated glial cells (TAGs) overexpress fetal transcription factors, including POU3f2 and SOX2. Furthermore, TAGs proliferate and form spheres in stem cell medium in contrast to glial cells from normal mouse brains. They also release growth factors that may enhance tumor cell growth. We found that overexpression of POU3f2 increased proliferation in human astrocytes. In human brain tumor biopsies we found that POU3f2 expression correlated with tumor grade, being highest in the most aggressive tumors. In cell culture assays we have found that human astrocytes can increase both proliferation and migration of brain cancer (glioma) cells. Currently we are working to establish two mice strains with conditional knock-out of SOX2 and POU3f2 to study their roles in reprogramming tumor-associated glial cells during brain tumor growth.

Malignant gliomas are lethal cancers which infiltrate the surrounding brain extensively, making radical surgery impossible. Adjuvant treatments also have a limited effect, and new therapies are urgently needed. A major advance came with the recognition th at cancer-associated fibroblasts actively foster the growth of malignant cells in several tumor types. Glial cells in the central nervous system may have similar functions in brain tumors, and tumor-associated glial cells (TAGs) are phenotypically altered , frequently adopting a reactive phenotype. We therefore hypothesize that TAGs are major regulators of brain tumor progression, and will combine in vitro functional studies with animal experiments to characterize the TAG cell compartment. This will then g uide the selection of stromal targets for therapeutic validation. We have established fluorescence activated cell sorting protocols for isolating stromal cells from xenograft gliomas established in GFP-scid mice. These cells express predominantly immature and glial cell markers (TAGs), and proliferate in vitro. Co-implantation of TAGs with glioma cells in mice accentuates growth significantly, compared to implanting glioma cells only or together with normal glial cells. Microarray analysis has revealed ge ne clusters differentially expressed between TAGs and normal glial cells. Collectively, our data suggest that TAGs are reprogrammed to a mesenchymal, proliferative state that promotes angiogenesis and invasion through the release of angiogenic and proinva sive factors. Future experiments include functional studies to identify candidate signaling pathways mediating the TAG phenotype, and subsequent selection of TAG specific targets for therapeutic validation. Thus, the work outlined in this project presents an integrated approach for a detailed molecular characterization of TAGs, functional studies into the mechanisms that mediate their tumor-promoting effects, and validation of TAG markers as therapeutic targets