GNS_cells · Biochemical and functional studies of lineage-specific transcription factor complexes in glioblastoma-derived stem cells

Biochemical and functional studies of lineage-specific transcription factor complexes in glioblastoma-derived stem cells

Glioblastoma (GBM) is the most aggressive form of adult brain tumour. GBM tumours display a high level of cellular heterogeneity and are comprised of a variety of both immature and more differentiated cell types. A specific population of cells isolated from brain tumours show neural stem (NS) cells-like characteristics - they can self-renew to produce new cells and they can differentiate into other cell types. Translplantatio of these cells from one species to another can iniciate tumour formation. The major goal of this project was to identify critical factors that drive Glioblastoma derived neural stem (GNS) cell self-renewal and identify their molecular partners. GNS cells can be maintained long term in adherent culture, retaining stem or early progenitor markers, multi-lineage differentiation potential and tumourigenicity (Pollard et al., 2009). This provides a useful model system to study the molecular controls of self-renewal and differentiation in cancer stem cells. Our broad hypothesis is that lineage-specific or 'developmental' transcription factors (TFs) controlling protein expression act as the major integrators downstream of the common mis-regulated cell cycle and 'cancer' pathways in glioblastoma. Understanding their function and biochemical interactions is therefore vital. Previous studies from the Pollard laboratory have included profiling of Gliobastoma cells compared to control cells by measuring mRNA levels - an indirect measurement of protein expression. These data reveal consistent and significant high mRNA levels of a gene called FOXG1. FOXG1 is a forkhead family transcription factor and essential for the development of the forebrain, protecting early progenitor cells from premature arrest and differentiation. Interestingly, using the 'Repository of molecular brain neoplasia data', we were able to find a correlation between levels of FoxG1 and survival for GBM. Furthermore, work from others has recently demonstrated that FOXG1 may co-operate with just one other transcription factor (SOX2) in reprogramming somatic cells to an NS cell like state. The parallels between experimental reprogramming of somatic cells to NS cells and the mechanisms of tumourigenesis are intriguing and suggest FOXG1 may be important in establishing the GBM stem cell state. Due to the lack of highly specific and potent monoclonal antibodies directed against FOXG1, biochemical analyses such as FOXG1 complex purification and characterisation have not been studied extensively. In collabouration with the EMBL in Germany we have therefore generated and validated two high quality monoclonal antibodies for FOXG1. They are highly specific and efficient for detection and enrichment of both human and mouse FOXG1 and will be a highly valuable tool for the scientific community. In addition to the GNS and control cell lines available in the Pollard laboratory, we have generated additional GNS and control NS cell lines from adult brain tumour and fetal fore-, mid- and hindbrain, respectively. These tissue samples have been collected in collabouration with Dr Kathreena Kurian (Frenchay Hospital, Bristol) and Dr Diana Gerelli (Institute of Child Health, University College London). Consistent with the previous mRNA analysis it has been confirmed by Western blotting and immunocytochemistry that FOXG1 protein is indeed overexpressed in all tested GNS cells lines throughout all the different GBM subtypes compared to the control NS cells. Strikingly, we could show that in vitro differentiation of GNS cells into oligodendrocytes leads to downregulation of FoxG1. Similarly, we were able to demonstrate a reduction of protein levels following BMP treatment of mouse NS cells. These results suggest that FoxG1 has a critical role in driving GNS cell self-renewal. The major goal of this project was to identify the molecular partners of FOXG1. We have co-immunoprecipitated FoxG1 from three GNS cell lines. Furthermore, we have also collected and processed primary human foetal forebrain and hindbrain as positive and negative controls respectively. Mass spectrometry was performed by our collaborator Dr Jeroen Krijsveld (EMBL, Germany) using Orbitrap. The results from these experiments are highly encouraging. We found a significant enrichment of FOXG1 and potential interaction partners in all expected samples. To prioritise candidates we restricted our analysis to the set of proteins that were detected at higher levels in all GNS cell lines and foetal forebrain with lack of signal in negative controls. The analysis resulted in around 20 candidate partners. Several of these are now being verified by co-immunoprecipitation and Western blotting with endogenous and tagged overexpressed proteins.

Data: CORDIS, © European Union