Gliomas are not responsive to chemotherapy because of tumor cell resistance, by which changes at the molecular level affect important cellular processes, such as cell cycle kinetics and apoptosis. A number of markers of resistance to chemotherapy in glioma models have been discovered; however, the mechanistic relationship between the expression of these markers and failure of chemotherapy remains to be elucidated. Morever, explaining resistance based on changes in a single genetic or molecular entity is inadequate, given the complexity of the system. To understand resistance to chemotherapy, identification of biological systems, and their interactions, is of great promise.

We have focused much of our attention to date on characterizing the response of glioma cell lines to chemotherapeutic drugs, carmustine and etoposide, at the level of cell cycle and apoptosis. In particular, we have developed a mathematical model of cell cycle kinetics that is able to reproduce the dynamics and dose response of cells to the two chemotherapeutic agents based on two parameters relating to cell cycle arrest and entry into apoptosis. We have shown that the model can be used to extract mechanistic information regarding the relative influence of these two processes upon tumor cells simply from pharamacological dose response curves, from which mechanism is not obtained using traditional analyses. The model suggests that carmustine elicits its effect by inducing apoptosis, while etoposide induces cell cycle arrest as a primary mode of action. Our current work seeks to apply this methodology to track the changes in tumor cells as they acquire resistance to chemotherapy.

We have generated glioma cell lines resistant to etoposide and carmustine by incremental stepwise exposure to sublethal doses of each drug. We have compared the effect of drug exposure to chemotherapy on important cellular processes, such as cell cycle distributions and apoptosis, between resistant and parent cell lines. We found that resistant cells have altered cell cycle distributions compared to parent cell lines at drug free conditions; moreover, changes in cell cycle distribution upon drug exposure are more drastic in parent cell lines compared to resistant cell lines. To deepen our understanding of resistant phenotype, we aim to use molecular profiling using microarray technology, which allows for global monitoring of changes in gene expression profiles. Microarrays exploring genes implicated in critical processes, such as cell cycle arrest and apoptosis, will be investigated in order to elucidate pathways implicated in resistance. We plan to construct regulatory networks from gene expression data in order to characterize the process of resistance development from a systems viewpoint.