While extirpative surgery is curative for localized clear cell renal cell carcinoma (ccRCC), many patients develop recurrences or present with metastatic disease. Several aspects of ccRCC biology have been investigated, but these have been done in cell lines, which are known to poorly represent the tumour. Since cell lines are amenable to a wide array of experimental testing, the studies presented here demonstrate a novel method to generate ccRCC cell lines from primary tumours, which increases the rate of primary tumour cell line generation four fold. Additionally, ccRCC cells do not grow in serum-free media, which has been shown to be beneficial in other cancers. Therefore, we interrogated the effect of exogenous growth factors to optimize our serum-free media growth conditions, among which TGFb1 appeared to elicit the largest mitogenic effect. Once optimized, these findings will provide a valuable tool for understanding ccRCC tumour cell biology and identifying therapeutic targets.

Glioblastoma Multiforme (GBM) is the most common and lethal of human primary central nervous system tumours, with a median survival of 14-16 months despite surgery, radiation and chemotherapy. A reason for this dismal prognosis is insufficient understanding of the ontogeny of GBMs, which are highly heterogeneous at a pathological level. This pathological diversity, between and within GBMs as well as varying grades of gliomas, is not fully explained on the grounds of an oncogenic stimulus. Interaction with the tumour microenvironment, as well as inherent characteristics of the tumour cell of origin are likely a source of this heterogeneity. In this thesis we describe the use of a novel mouse model which integrates Cre-Lox mediated and Tet-regulated gene expression. This system in combination with germline and somatic strategies has enabled us to interrogate how the state in glial development and the region in the brain where transformation occurs influence the process of gliomagenesis. The findings of this thesis suggest that the state of glial development at which a mutation is introduced is an important determinant of gliomagenesis. In support of this, we showed that early progenitors in the radial glial lineage are more susceptible to transformation than those, which have committed to a gliogenic lineage and are presumably further along in the process of differentiation. Highlighting the interplay between genetic alterations and the molecular changes that accompany the process of differentiation. Despite findings that suggest that neurogenic regions of the adult brain are more susceptible to transformation, we show that this is not always the case and instead, transformation is dependent on an interaction between specific combinations of genetic mutations and susceptible cell types regardless of the region of origin. Results from this thesis highlight the need to view the tumourigenic process of gliomas in the context of normal brain development as the cell context of oncogene expression may determine the phenotype and biologic aggressiveness of the tumour. Thus, the results of genetic or epigenetic alterations leading to brain tumours may be quite different in different cells of the hierarchy, suggesting unique treatment targets and strategies depending on the cell of origin.

Glioblastoma Multiforme (GBM) is the most common and lethal of human primary central nervous system tumours, with a median survival of 14-16 months despite surgery, radiation and chemotherapy. A reason for this dismal prognosis is insufficient understanding of the ontogeny of GBMs, which are highly heterogeneous at a pathological level. This pathological diversity, between and within GBMs as well as varying grades of gliomas, is not fully explained on the grounds of an oncogenic stimulus. Interaction with the tumour microenvironment, as well as inherent characteristics of the tumour cell of origin are likely a source of this heterogeneity. In this thesis we describe the use of a novel mouse model which integrates Cre-Lox mediated and Tet-regulated gene expression. This system in combination with germline and somatic strategies has enabled us to interrogate how the state in glial development and the region in the brain where transformation occurs influence the process of gliomagenesis. The findings of this thesis suggest that the state of glial development at which a mutation is introduced is an important determinant of gliomagenesis. In support of this, we showed that early progenitors in the radial glial lineage are more susceptible to transformation than those, which have committed to a gliogenic lineage and are presumably further along in the process of differentiation. Highlighting the interplay between genetic alterations and the molecular changes that accompany the process of differentiation. Despite findings that suggest that neurogenic regions of the adult brain are more susceptible to transformation, we show that this is not always the case and instead, transformation is dependent on an interaction between specific combinations of genetic mutations and susceptible cell types regardless of the region of origin. Results from this thesis highlight the need to view the tumourigenic process of gliomas in the context of normal brain development as the cell context of oncogene expression may determine the phenotype and biologic aggressiveness of the tumour. Thus, the results of genetic or epigenetic alterations leading to brain tumours may be quite different in different cells of the hierarchy, suggesting unique treatment targets and strategies depending on the cell of origin.

Anti-cancer drugs gain access to solid tumors via the blood, and must penetrate tissue to reach all viable cancer cells. This thesis aims to compare the distribution of anticancer drugs in normal tissues and tumours, to examine whether drug distribution is modifiable and quantifiable in solid tumours, and to determine whether extracellular drug distribution can be improved by modifying intracellular drug distribution. The time-dependent spatial distribution of three anticancer drugs, doxorubicin, mitoxantrone and topotecan, were studied in normal tissues and tumours. Ten minutes after drug administration, there was fairly uniform distribution in the heart, kidney and liver whereas drug distribution within tumours was limited to perivascular regions. Doxorubicin distribution in P-glycoprotein (PgP) over-expressing tumours was compared to that in wild-type tumours and changes in distribution were evaluated with the use of PgP inhibitors. There was better doxorubicin distribution in PgP-over-expressing tumours compared to wild-type tumours, and pretreatment of PgP-over-expressing tumours with PgP inhibitors decreased doxorubicin distribution. These data suggest that reduced uptake of drug into cells may enhance extracellular drug distribution, and the dual effects of PgP inhibitors (increased drug uptake in proximal cells, but poorer drug distribution) may explain, in part, why these agents have not provided clinical benefit. The effect of the proton pump inhibitor pantoprozole on intracellular and extracellular drug distribution was determined. Pantoprazole increased endosomal pH in cells, leading to less sequestration of doxorubicin within them, and increased the toxicity of doxorubicin for cultured cells. In wild-type MCF7 tumours, pretreatment with pantoprazole enhanced doxorubicin distribution and tumour growth delay without apparent increase in toxicity. These studies have led to initiation of a phase I clinical trial of pantoprazole and doxorubicin for patients with solid tumours. The work completed in this thesis demonstrates that drug distribution can be modified and that these changes can be quantified, and may correlate with improved anti-tumour effects. Improving drug distribution through the use of proton pump inhibitors may be an effective strategy to improve chemotherapeutic efficacy.

The ubiquitin-like proteins (Ubls) are small polypeptides that function as post-translational modifiers. Like ubiquitin, most Ubls are covalently attached to a lysine residue on target proteins. The small ubiquitin-related modifiers (SUMO) play important roles in a number of critical biological processes, such as proliferation and regulation of the cell cycle, yet their specific cellular functions have remained poorly understood. Like ubiquitin, SUMO proteins can also form oligomeric “chains”, but the functions of these structures were even less well understood. To this end, I created the first spectral library for the identification of Ub/Ubl proteins and Ub/Ubl chain linkages in mass spectrometry experiments. This tool has dramatically improved our ability to use MS to analyze the contents of biological samples for Ub and Ubls, and to identify specific types of Ub and Ubl chains in model organisms. I also used MS to conduct the first comprehensive SUMO system protein-protein interactome in any organism. In total, 452 high confidence protein-protein interactions were detected for S. cerevisiae SUMO system proteins, encompassing a total of 321 interacting partners. Yeast SUMO system components were found to interact with proteins involved in a number of different biological processes, and my mapping effort increased the number of known SUMO system interacting partners >50-fold. This study revealed that a number of transcriptional co-repressors and chromatin remodelling proteins interact physically with specific SUMO system components, with a clear division of labour between SUMO system enzymes. Finally, I conducted the first global analysis of SUMO chain function, using a combination of genetic, high-content microscopy, and high-density transcriptomics screens. Consistent with my interactomics work, this study demonstrated that inhibition of SUMO chain synthesis leads to severe chromatin condensation defects, which in-turn leads to chromosome missegregation, unscheduled transcription of stress-and nutrient-regulated genes, and aberrant intragenic transcription. Together, my work thus revealed a major role for the SUMO system in the maintenance of higher order chromatin structure and transcriptional repression.

Tumorigenesis is driven by the sequential accumulation of genetic lesions within a cell, each which confer the cell with traits that enable its abnormal growth. The result is a mass of dysregulated cells, or tumor, which, upon further mutation, may spread, or metastasize, to other organs of the body. The dissemination of tumor cells makes treatment difficult, and thus confers cancer with its associated lethality. Over the past 30 years, the RAS genes have been critical in teaching us the mechanisms underlying the molecular progression of cancer. RAS is mutated in 33% of all cancers and is often an early event in its stepwise progression. As a result, the RAS genes are widely accepted as ‘drivers’ or ‘initiators’ of human tumorigenesis. Unfortunately, efforts directed at targeting RAS in the clinic have as of yet been unsuccessful. This has triggered a need to identify genes that are required for RAS tumorigenesis that are therapeutically tractable. My research has focused on deciphering the potential role of the Rho GEF Arhgef2 in RAS-mediated tumorigenesis. I have found that Arhgef2 is a bona fide transcriptional target of RAS and is upregulated in human tumors harboring RAS mutations. Importantly, depletion of Arhgef2 in RAS-mutated cells inhibits their survival, proliferation, and tumor growth in murine models. In search of the mechanism underlying the requirement of Arhgef2 in RAS tumorigenesis, I have uncovered a novel function for Arhgef2 as a positive regulator of a central RAS pathway, the mitogen-activated protein kinase (MAPK) pathway. Thus, Arhgef2 is part of a positive feedback loop in which RAS-dependent increases in Arhgef2 expression results in the amplification of RAS signaling. Moreover, Arhgef2 confers tumor cells with properties favoring their malignant conversion, thereby implicating Arhgef2 in the formation of metastases. Together, these studies suggest that Arhgef2 plays an important role at multiple stages of tumorigenic progression and may therefore be a promising therapeutic target in RAS-mutated tumors.

This work is a part of a large project in our laboratory that is aimed toward characterization of prokaryotic transcription regulators from different families and their interactions with small-molecule effectors. My study was focused of IclR family of transcriprion regulators, specifically on its founding member Isocytrate Lyase Regulator (IclR) from E.coli and AllR regulator from E.coli, which share 42% sequence identity with IclR. I used a combination of biophysical, biochemical and structural biology techniques to explore the mechanisms by which IclR and AllR interact with their effectors. I performed site-directed mutagenesis experiments in order to research the role of individual amino acids in interaction of AllR regulator with its previously identified effector glyoxylate and to test whether oligomerization plays a role in effector-induced signal transduction by AllR. Using differential light scattering, which allows high-throughput screening of small molecules for thermostabilization of proteins, I identified potential effctors for the IclR regulator. The physiological relevance of these candidate molecules was tested in-vitro and in-vivo and their interaction with IclR was characterized by Isothermal Titration Calorimetry and X-ray Crystallography.

Recent evidence suggests that radiation may have a significant effect on tumour vascu lature in addition to damaging tumour cell DNA. It is well established that endothelial cells are among the first cells to respond after administration of ionizing radiation in both normal and tumour tissues. It has also been suggested that microvascular dysfunction may regulate tumour response to radiotherapy at high doses. However, due to limitations in imaging the microcirculation this response is not well characterized. Advances in high-frequency ultrasound and computation methods now make it possible to acquire and analyze 3-D ultrasound data of tumour blood flow in tumour micro-circulation. This thesis outlines the work done to test the hypothesis that single dose 8 Gy radio- therapy produces changes in tumour blood vessels which can be quantified using high- frequency power Doppler ultrasound. In addition, the issue of reproducibility of power Doppler measurements and the relationship between histopathology and power Doppler measurements have been examined.

Breast cancer is currently the most common diagnosed cancer among women and a significant cause of death. Breast density is considered a significant risk factor and an important biomarker influencing the later risk of breast cancer. Therefore, ongoing epidemiological studies using MRI are evaluating quantitatively breast density in young women. One of the challenges is segmenting the breast in order to calculate total breast volume and exclude non-breast surrounding tissues. This thesis describes an automatic 3D breast volume segmentation based on 3D local edge detection using phase congruency and Poisson surface reconstruction to extract the total breast volume in 3D. The boundary localization framework is integrated on a subsequent atlas-based segmentation using a Laplacian framework. The 3D segmentation achieves breast-air and breast-chest wall boundary localization errors with a median of 1.36 mm and 2.68 mm respectively when tested on 409 MRI datasets.

B-cell acute lymphoblastic leukemia (B-ALL) represents a collection of diseases that are categorized into subtypes based on the presence of recurrent cytogenetic abnormalities. These abnormalities often result in the expression of oncogenic drivers that denote a standard- or high-risk for relapse. Currently, survival rates boarder 40% for adult patients and relapses are often observed in patients lacking high-risk markers. Thus, there is an unmet need for biomarkers that can identify all high-risk leukemia, and development of novel therapies based on a better understanding of the molecular drivers of B-ALL. To address this need, I designed a multi-parameter phospho-flow cytometry platform and characterized basal and cytokine-potentiated signaling in adult B-ALL samples. I identified patterns of cytokine-responsiveness across B-ALL patients that correlated with the presence of high-risk oncogenic drivers. Furthermore, I demonstrated that small-molecule inhibitors could abrogate cytokine-induced signaling in high-risk patients suggesting these inhibitors may compliment current chemotherapeutic protocols.