Identification of compounds that target glioma initiating cells

Abstract: Glioblastoma multiforme is a common form of brain tumor that leads to debilitating effects despite the current regiment of treatment, which includes surgery, chemotherapy and radiotherapy. With the discovery of glioma-initiating cells (GICs) that exists within the bulk tumor, some light has been shed on this disease. GICs have been shown to be resistant to chemotherapy and radiotherapy, therefore providing a plausible explanation for the high propensity for recurrence in patients. However, effective treatments are still unavailable, and there is an urgency to discover drugs that can eradicate this group of cells. The overall aim of this thesis is to identify drugs from small molecule screens that can kill GICs, as well as to understand the possible causes and mechanisms for drug sensitivity. In Paper I, we have identified a new small molecule, Vacquinol-1, that reduced the viability and growth of GICs, both in vitro and in vivo, and had little effects on normal cells such as fibroblasts and embryonic stem cells. Cell death was caused by an unconventional nonapoptotic manner, where the cells showed massive accumulation of vacuoles that formed through macropinocytosis. This led to the impairment of cell function followed by cell death. Furthermore, treatment with Vacquinol-1 also exhibited excellent improvement in survival of glioma xenografted mice. Using an shRNA screen, mitogen-activated protein kinase kinase 4 (MKK4), which is involved in stress response, was determined to play a role in the unique type of cell death. In Paper II, we have identified that GICs are particularly sensitive to perturbations in calcium (Ca2+) homeostasis, with the relative degree of sensitivity being linked to their degree of stemness. The two compounds employed in this study, Ca2+ ionophore A23187 and the sarco/endoplasmic reticulum Ca2+-ATPase (SERCA) pump inhibitor Thapsigargin, are wellknown compounds that elicit cell death. However, the compounds affected viability to a different degree in the various GIC lines, with the line that is more similar to neural stem cells being more sensitive to perturbations in Ca2+ homeostasis. This sensitivity was correlated to expression levels of different Ca2+ related proteins such as GRIA1 and S100A6, as well as to Nestin (NES). In Paper III, we have repositioned Niguldipine, an old anti-hypertensive drug, as a potential compound for targeting GICs. Through an ion channel drug screen, we have also observed that GICs exhibit sensitivity to Ca2+ modulators, suggesting once again that Ca2+ homeostasis is critical to cell viability in GICs. Niguldipine, which is also a Ca2+ channel inhibitor, also showed a selection for GICs as compared to normal cells such as fibroblasts and neural stem cells. At the effective dose, the compound showed no effects on cardiac rhythmicity, and administration of the drug resulted in a significant improvement in the survival of glioma xenografted mice. In conclusion, we have identified a few small molecules, both old and new, that can reduce the proliferation and viability of GICs without affecting that of normal cells. One of the mechanisms underlying selectivity is that GICs show a greater degree of sensitivity to disturbances in Ca2+ homeostasis, therefore suggesting that Ca2+ modulators should be screened for their potential in cancer therapy.