Medulloblastoma is a devastating disease of the cerebellum, and the most common solid pediatric malignancy of the central nervous system. Recently, transcriptome-wide profiling has dissected medulloblastoma from one single disease into four disparate molecular subgroups – namely WNT, SHH, Group3 and Group4. Distinct genomic, cytogenetic, mutational and clinical spectra associated with these subgroups highlight the pressing need for targeted therapies, of which encouraging preliminary results have been generated. While the promise of personalized medicine is within our reach, improved understanding of the molecular mechanisms driving pathogenesis is critical to this process. The intent of my PhD thesis research was to characterize the molecular mechanisms contributing to medulloblastoma pathogenesis, and the clinical impact of these aberrations. Through a combinatorial use of genetic and epigenetic profiling, next-generation sequencing and bioinformatics analyses we have identified subsets of tumors with transcriptional signatures that influence their clinical properties. Furthermore, our results have shed light on the establishment of the normal cerebellar cytoarchitecture, identifying a physiological glutamate gradient with critical implications to both cerebellar development and disease. This thesis stresses the importance of interrogating medulloblastoma in a subgroup-specific manner. Our findings demonstrate the utility of pursuing an integrated (copy number, mutational, transcriptional and epigenetic) molecular approach, to further our understanding of the pathobiology of medulloblastoma. Finally, we propose rationale therapeutic targets that may improve the treatment of aggressive variants of this disease.

Medulloblastoma is a devastating disease of the cerebellum, and the most common solid pediatric malignancy of the central nervous system. Recently, transcriptome-wide profiling has dissected medulloblastoma from one single disease into four disparate molecular subgroups – namely WNT, SHH, Group3 and Group4. Distinct genomic, cytogenetic, mutational and clinical spectra associated with these subgroups highlight the pressing need for targeted therapies, of which encouraging preliminary results have been generated. While the promise of personalized medicine is within our reach, improved understanding of the molecular mechanisms driving pathogenesis is critical to this process. The intent of my PhD thesis research was to characterize the molecular mechanisms contributing to medulloblastoma pathogenesis, and the clinical impact of these aberrations. Through a combinatorial use of genetic and epigenetic profiling, next-generation sequencing and bioinformatics analyses we have identified subsets of tumors with transcriptional signatures that influence their clinical properties. Furthermore, our results have shed light on the establishment of the normal cerebellar cytoarchitecture, identifying a physiological glutamate gradient with critical implications to both cerebellar development and disease. This thesis stresses the importance of interrogating medulloblastoma in a subgroup-specific manner. Our findings demonstrate the utility of pursuing an integrated (copy number, mutational, transcriptional and epigenetic) molecular approach, to further our understanding of the pathobiology of medulloblastoma. Finally, we propose rationale therapeutic targets that may improve the treatment of aggressive variants of this disease.

No description available.

No description available.

Epithelial ovarian cancer (EOC) is difficult to detect in early stage disease, resulting in a high mortality rate. The molecular events underlying EOC development remain largely unknown. Chromosome 3 exhibits frequent deletions and rearrangements in EOC by cytogenetic analysis. In addition, loss of heterozygosity (LOH) mapping of matched ovarian tumour and constitutional DNA samples exhibits specific regions of chromosome 3 loss involving distinct regions: 3p25-p26, 3p24 and a region proximal to 3p14. Thus, chromosome 3p loss points to the location of tumour suppressor genes (TSG) implicated in tumourigenesis, based on Knudson's 'two-hit' model and the paradigm of the classical TSG. The dissertation hypothesis states at least one TSG implicated in EOC is located on chromosome 3p. A novel complementation approach based on the transfer of normal chromosome 3 fragments into OV-90, a tumourigenic EOC cell line harbouring LOH of the 3p arm, was used to generate functional evidence for chromosome 3p TSGs. Three hybrids exhibited complete suppression of tumourigenic potential based on the inability to form colonies in soft agarose, spheroids in cell culture, and tumours in nude mice xenograft models. While all hybrids had acquired various chromosome 3 regions, they all shared in common a 3p12-pcen interval, suggesting at least one common gene may have affected the suppression of tumourigenicity in the OV-90-derived hybrids. Twelve known/hypothetical genes mapping to 3p12-pcen region were characterized based on gene expression and mutation analysis following a classical model for TSG inactivation. To establish the relevance to EOC, gene expression of candidates was investigated in primary cultures of normal ovarian surface epithelial cells and both malignant serous and benign serous tumour samples. The gene expression and genetic analysis identified seven TSG candidates, none of which appeared to be mutated or transcriptionally silenced based on classical mechanisms of TSG inactivation in OV-90, thus suppression of tumourigenicity may have resulted from the functional complementation of one more haploinsufficient 3p12-pcen genes. Several genes (GBE1, VGLL3, ZNF654 ) appeared underexpressed in malignant tumours and these findings suggest the intriguing possibility that more than one 3p12-pcen gene was involved in the suppression of tumourigenicity in OV-90, and by extension, EOC.

The blood brain barrier (BBB) remains a major challenge to the advancement and application of systemic anti-cancer therapeutics into the central nervous system. The structural and physiological delivery constraints of the BBB significantly limit the effectiveness of conventional chemotherapy, thereby making systemic administration a non-viable option for the vast majority of chemotherapy agents. Furthermore, the lack of specificity of conventional systemic chemotherapy when applied towards malignant brain tumors remains a major shortcoming. Hence novel therapeutic strategies that focus both on targeted and enhanced delivery across the BBB are warranted. In recent years nanoparticles (NPs) have emerged as attractive vehicles for efficient delivery of targeted anti-cancer therapeutics. In particular, gold nanoparticles (AuNPs) have gained prominence in several targeting applications involving systemic cancers. Their enhanced permeation and retention within permissive tumor microvasculature provide a selective advantage for targeting. Malignant brain tumors also exhibit transport-permissive microvasculature secondary to blood brain barrier disruption. Hence AuNPs may have potential relevance for brain tumor targeting. However, the permeation of AuNPs across the BBB has not been well characterized, and hence is a potential limitation for successful application of AuNP-based therapeutics within the central nervous system (CNS). In this dissertation, we designed and characterized AuNPs and assessed the role of polyethylene glycol (PEG) on the physical and biological properties of AuNPs. We established a size-dependent permeation profile with respect to core size as well as PEG length when AuNPs were assessed through a transport-permissive in-vitro BBB. This study was the first of its kind to systematically examine the influence of design on permeation of AuNPs through transport-permissive BBB. Given the significant delivery limitations through the non-transport permissive and intact BBB, we also assessed the role of magnetic resonance imaging (MRI) guided focused ultrasound (MRgFUS) disruption of the BBB in enhancing permeation of AuNPs across the intact BBB and tumor BBB in vivo. MRgFUS is a novel technique that can transiently increase BBB permeability thereby allowing delivery of therapeutics into the CNS. We demonstrated enhanced delivery of AuNPs with therapeutic potential into the CNS via MRgFUS. Our study was the first to establish a definitive role for MRgFUS in delivering AuNPs into the CNS. In summary, this thesis describes results from a series of research projects that have contributed to our understanding of the influence of design features on AuNP permeation through the BBB and also the potential role of MRgFUS in AuNP permeation across the BBB.

Hematopoietic cells are recruited to and co-opted by the growing tumor making expansive tumor growth possible. Although several cell types become associated with the growing tumor, macrophages play a fundamental role. The movement of macrophages across the basement membrane and through the extracellular matrix to the tumor site requires the activation of proteases, such as plasmin, at their cell surface. The proteolytic aspect of macrophage recruitment may represent an exploitable aspect of tumor growth in terms of therapeutic strategies. Here I show that the S100A10 protein facilitates the infiltration of macrophages into the site of tumor growth by stimulating the generation of the protease plasmin at their surface. Using a mouse model in which wild-type (WT) and S100A10-null mice are inoculated with tumor cells, a decrease in tumor-associated macrophages (TAMs) and greatly diminished tumor growth in tumors grown in S100A10-null mice was observed. Although tumor growth in S100A10-null mice could be restored by intraperitoneal injection of WT macrophages, S100A10-null macrophages only restored tumor growth when directly injected into the tumor. Lastly, selective depletion of macrophages from a WT mouse by liposome encapsulated clodronate treatment resulted in similar tumor growth deficits as in the S100A10-null mouse. These results highlight a new role for the S100A10 protein in the recruitment of TAMs to the tumor site and demonstrate a potential therapeutic strategy in which the tumor associated cells may be targeted.

Due to its escalating prevalence, obesity is becoming a leading cause of morbidity and mortality worldwide. Obesity is a complex health problem accompanied by metabolic abnormalities and low-grade inflammation that increases the risk for developing comorbidities including type 2 diabetes. Recent evidence supports a role for fat (adipose) tissue derived factors, called adipokines, in the development of obesity and obesity-related metabolic pathologies. Chemerin is an adipokine that mediates immune and metabolic effects through the chemokine-like receptor 1 (CMKLR1). Chemerin is secreted as an inactive proform, prochemerin, which subsequently undergoes enzymatic cleavage into multiple chemerin products that differentially activate CMKLR1. Multiple studies have reported elevated total chemerin (a combination of prochemerin and various chemerin products) in obese humans suggesting chemerin involvement in obesity pathophysiology. However, the observational nature of these human studies have restricted them from identifying specific forms of chemerin that are elevated in obesity and the mechanisms that govern them. Herein, I have reported that the levels of both serum total chemerin and chemerin products capable of activating CMKLR1 are elevated in obese mice and in wild type mice following treatment with an obesity-associated inflammatory mediator tumor necrosis factor-? (TNF?). Likewise, cultured adipocytes produced active chemerin under basal conditions and highly active chemerin following TNF? treatment as measured by CMKLR1 activation. The current belief is that prochemerin circulates through blood primed for activation by immune and fibrinolytic enzymes present within injured tissues. My results challenge this theory, identifying adipocytes as cells alone produce and proteolytically activate chemerin. Under basal conditions, a balance between activating serine proteases and deactivating aminopeptidases governed the amount of CMKLR1-activating chemerin formed by adipocytes. Treatment of adipocytes with TNF? elevated the levels of serine proteases elastase and tryptase, which cumulatively shifted the proteolytic balance toward the production of chemerin products that highly activate CMKLR1. Taken together, my results are the first to identify that local TNF? triggers increased adipocyte production of chemerin providing an explanation for the elevated concentrations of chemerin in obese animals and humans. Furthermore, adipocyte processing represents a novel mechanism that likely governs the amount and type of circulating chemerin in obesity.

In Canada, colorectal cancer is the second leading cause of cancer death for both men and women. There are many different factors that contribute to the progression and spread of the disease. However, increasing evidence now suggests that the tumor microenvironment plays a paramount role in these processes. CD26 is a multifunctional, cell-surface glycoprotein that has intrinsic enzyme activity, binds adenosine deaminase and interacts with the extracellular matrix. Through its various functions it serves to constrain cancer progression. For example, it is known to cleave CXCL12, the ligand for CXCR4. The CXCL12:CXCR4 axis is normally involved in cancer metastasis by promoting cancer cell migration, invasion and proliferation. Down-regulation of CD26 is observed in certain cancers - this has been shown in vitro to occur in response to certain soluble mediators. The first part of this study looked at the effects of glucose and its metabolic product lactate on CD26 expression in colorectal carcinoma cells. Our study showed that CD26 expression is lower in cancer cells that are grown in low-glucose, high-lactate conditions, which replicates the situation within a tumor. The second part of this study examined the effect of adenosine, a purine nucleoside, on colorectal carcinoma cells and supportive stromal cells - cancer-associated HS675.T fibroblasts (CAFs) and Met-5a mesothelial cells. Adenosine increased the proliferation of CAFs and increased CXCL12 mRNA in both stromal cell lines. It also increased MMP-13 mRNA in stromal cells as well as colorectal cancer cells, suggesting that adenosine may promote progression and metastasis through various mechanisms. The last section focused on the ability of cellular products and 3-dimensional tissue topology to coordinate and affect the behaviour of the different cell populations. Here we show that secretory products from colorectal cancer cells promote CAF proliferation but inhibit mesothelial cell proliferation, and are also able to modulate MMP-13 expression. Finally, certain responses are enhanced in multicellular spheroids. In conclusion, the tumor microenvironment represents a major consideration in the treatment of solid tumors. Our data suggest that various soluble mediators, such as adenosine, may have therapeutic implications in cancer treatment and might represent novel targets for future research.

No description available.