Glioblastoma is the most common primary malignant brain tumour in the adult population. Despite multimodality treatment with surgery, radiotherapy and chemotherapy, outcomes are very poor, with less than 15% of patients alive after two years. Increasing evidence suggests that glioblastoma stem cells (GSCs) are likely to play an important role in the biology of this disease and are involved in treatment resistance and tumour recurrence following standard therapy. My thesis aims to address two main aspects of this research area: 1) optimization of methods to evaluate treatment responses of GSCs and their differentiated counterparts (non-GSCs), with a particular focus on a tissue culture model that resembles more closely the tumoral niche; 2) characterization of cell division and centrosome cycle of GSCs, investigating possible differences between these cells and non-GSCs, that would allow the identification of targets for new therapeutic strategies against glioblastomas. In the first part of my project, I optimized a clonogenic survival assay, to compare sensitivity of GSCs and non-GSCs to various treatments, and I developed the use of a 3-dimentional tissue culture system, that allows analysis of features and radiation responses of these two subpopulations in the presence of specific microenvironmental factors from the tumoral niche. In the second part, I show that GSCs display mitotic spindle abnormalities more frequently than non-GSCs and that they have distinctive features with regards to the centrosome cycle. I also demonstrate that GSCs are more sensitive than non-GSCs to subtle changes in Aurora kinase A activity, which result in a rapid increase in polyploidy and subsequently in senescence, with a consistent reduction in clonogenic survival. Based on these findings, I propose that kinases involved in the centrosome cycle need to be explored as a novel strategy to target GSCs effectively and improve outcomes of glioblastoma patients.

Glioblastoma multiforme (GBM) is the most common and most aggressive type of primary brain tumour. It is currently treated by a mixture of ionising radiation and Temozolomide (TMZ) chemotherapy, however virtually all patients experience disease recurrence and 75% die within two years of diagnosis. Tumours which express elevated levels of the DNA repair protein O6-methylguanine DNA methyltransferase (MGMT) have a particularly poor prognosis, suggesting that levels of MGMT and the inefficacy of treatment are linked. MGMT is a “suicide” repair protein that binds irreversibly to a DNA adduct (such as those caused by TMZ) and is destroyed by the proteasome once repair has taken place. Therefore the ability of the cell to repair DNA damage relies on the rate at which it can resynthesise MGMT. Previous research has shown that reducing MGMT levels via promoter silencing increases the effectiveness of treatment, however this causes toxicity in bone marrow stem cells and is therefore unable to be used as a possible treatment option. My preliminary data suggests that inhibition of mTOR signalling reduces the steady state levels of MGMT without affecting mRNA levels, potentially making them more sensitive to TMZ treatment. I am therefore using inhibitors of the mTOR pathway and associated proteins, which have been selected via a novel bioinformatic technique, to ascertain how these affect MGMT protein levels and to determine whether a mixture of these inhibitors with DNA damaging agents could be used to increase the efficacy of TMZ treatment.