In this presentation it is demonstrated that the unique magnetic properties of superparamagnetic cobalt-spinel ferrite nanoparticles can be employed in several novel applications. A method to selectively capture and remove pathogens from infected organisms to improve longevity is presented. Evidence is provided to show that automated methods using modified forms of hemofiltration or peritoneal dialysis could be used to eliminate the particle/pathogen or particle/infected cell conjugates from the organism postoperatively. It is shown that disparately functionalized nanoparticles can be used in concert as drug carrier and release mechanisms. Lastly, we provide preliminary evidence to support the use of magnetic nanoparticles for controlling reaction kinetics.

The identification of genes that are directly involved in tumor initiation and maintenance is instrumental for understanding the phenotypic variation of cancer and ultimately designing crucial therapeutic drugs to treat this disease. In recent years, the completed genome sequence of humans and cancers has markedly enhanced cancer gene identification. The overall goal of this dissertation is to develop a warehouse of statistical tools for identifying cancer genes with growingly increasing sequence data. These tools are founded on the latest discoveries for the genetic and developmental roots of cancer formation, including somatic mutations, aneuploid induction, epigenetic modifications, transgenerational imprinting, copy number variants, and host-tumor genetic interactions. New statistical methods and algorithms will be developed to integrate each of these discoveries. By comparing the difference in the DNA structure and sequence between the human and cancer genomes, a disequilibrium model has been formulated to identify and test the genetic mutations or “drivers” that cause cancer. A quantitative model is derived to unravel the aneuploidy control of cancer and estimate the genetic effects of aneuploid loci on cancer risk. Using a commonly used three-generation design, a two-stage hierarchical model is developed to estimate and test the transgenerational alteration of genetic effects and identify genetic imprinting effects due to different parental origins of the same allele. This hierarchical model allows the characterization of genetic interactions between additive and dominant effects and imprinting effects over generations. Cancer susceptibility may be controlled not only by host genes and mutated genes in cancer cells, but also by the epistatic interactions between genes from the host and cancer genomes. A model was derived to estimate genome-genome interactions of host DNA and cancer DNA. Models for cancer gene identifications require the solution of missing data problems given the fact that cancer genes and their incidence in a natural population cannot be observed directly. For this reason, I have built up the models within the mixture model framework. The maximum likelihood approaches, implemented with the EM algorithm, have been derived to provide the estimates of genetic parameters related to mutation rates, chromosome duplication rates, genetic imprinting, genetic interactions, and haplotype frequencies. I have performed various sets of computer simulation to investigate the statistical properties of the new models in terms of power, estimation precision, and false positive rates. A series of practical computational issues, including convergence rates and choices of initial values, are discussed. I have also formulated various testable hypotheses about the frequencies of genetic mutations and the effects of host genes, cancer genes, and their interactions on cancer susceptibility. This dissertation provides a most complete set of statistical models for cancer gene identification thus far in the literature. The biological relevance and statistical sophistication of these models will make them practically useful to unlock the genetic secrets of cancer.

Tumor motion limits the accuracy of radiation therapy. Positron emission tracking (PeTrack) is a technique that can track tumors through detecting annihilation gammas from implanted positron emission markers. This thesis focuses on detector development for PeTrack. Due to the intense scattered x rays from a Linac, scintillator (BGO) afterglow and detector gating, i.e. turning off the detector during the intense x-ray pulse, need to be addressed. The evaluation of BGO showed very low afterglow. A gating circuit was designed, optimized and tested. Energy resolution of the detector is better than 25% (FWHM) with optimal gating parameters. A data acquisition system for PeTrack was set up and calibrated successfully. The first PeTrack prototype was developed and evaluated. The prototype was able to localize two positron emitting markers with an average precision of 0.16-0.21 mm, and an average accuracy of 0.6 mm on distance between the two markers.

Kidney cancer is a complex disease comprising several types of renal carcinomas, which are classified in different subtypes based on their histological characteristics. A small number of cases of renal cancers are due to hereditary predispositions and nearly all the knowledge on the molecular pathogenesis of kidney cancers was learned by the investigation of these hereditary forms of renal carcinomas. In this thesis, we studied two hereditary diseases predisposing to the development of kidney cancer, von Hippel Lindau (VHL) and Birt-Hogg-Dube (BHD) syndromes, and their causative genes, VHL and FLCN respectively. First, we investigated the role of the extracellular matrix (ECM) regulation by VHL during tumorigenesis and angiogenesis, and we demonstrated that inactivation of the VHL-ECM assembly pathway results in highly vascularized tumors with a disrupted ECM. We concluded that loss of the ECM assembly promotes and maintains tumor angiogenesis by providing a way for new blood vessels to invade the tumor tissue. In the second part of this thesis, we developed a novel VHL mouse model to investigate the possible cooperation between VHL and p53 during tumorigenesis. We observed that inactivation of both tumor suppressor genes accelerate the formation of liver hemangiomas and splenic hemangiosarcomas. Furthermore, concomitant deletion of VHL and p53 abolished the development of lymphoma usually associated with loss of p53. Our results indicate that the phenotypes arising following the inactivation of VHL and p53 is organ-dependent. Finally, to study the pathogenesis of the BHD syndrome, we developed a new mouse model using an established embryonic stem cell line. We described the murine Flcn expression pattern and noticed that homozygous disruption of Flcn was embryonically lethal early during development. Furthermore, we observed a continuum of kidney lesions from renal tubules hyperproliferation to rare adenoma. FLCN tumor suppressor role was also substantiated using a human kidney cancer cells system. Altogether, these studies have confirmed a role for the VHL-ECM pathway during tumor angiogenesis and have led to the development of two mouse models to study the molecular mechanisms linked to VHL and FLCN during the formation of various kidney tumors.

Mature microRNAs miRNAs) are short 19-24 nt), non-protein-coding ribonucleic acids that play very important roles in the regulation of gene expression in animals and plants. miRNAs mainly bind to the 3 untranslated regions of target mRNAs to cause translational blockade or transcript degradation. Although miRNAs have been implicated in growing number of diseases, their protein targets and the specific biological functions of these targets remain largely unknown. Computational prediction of miRNA targets provides an alternative approach to assign biological functions. Although the experimental validation of miRNA target genes increases dramatically, majority of miRNA targets are still unknown and bioinformatic algorithms remain the key means of predicting putative miRNA targets. The principles of miRNA target predictions are based on sequence complementary, conservation across species, thermodynamic stability, site accessibility and inverse relationship between the expression profiles of miRNAs and predicted target mRNAs. Here we use partial least square PLS) and sparse partial least square SPLS) methods to predict miRNA targets from miRNA and mRNA microarray data. Based on the inverse relationship between miRNA and mRNA, we selected two sets of differentially expressed miRNAs and mRNAs from human colon cancer microarray data. The first set consisted of 71 upregulated mRNAs and 31 downregulated miRNAs and the second set consisted of 56 downregulated mRNAs and 2 upregulated miRNAs. Using PLS and SPLS methods, we detected significant inverse interactions/associations between miRNA and mRNA. Then we compared these miRNA target genes with the four other widely used miRNA target prediction programs: TargetScan 5.1, PicTar, miRanda and miRBase. We identified a set of miRNA targets predicted by PLS and/or Sparse PLS that were also predicted by TargetScan5.1, PicTar, miRanda and miRBase through union of them or intersection combinations. We also used our predicted miRNA target genes to explore miRNA-mediated biological networks or pathways in human cancer.

As life expectancy grows longer in developed nations, the possibility of falling prey to some form of cancer becomes more and more likely. More significantly, while many cancers have become curable or may be sent into remission by palliative treatments, other forms of cancer become metastatic and there is often little that can be done to overcome the disease. Therefore, in an effort to better understand the process of metastatic cancer, researchers in the fields of medicine, biology, epidemiology, clinical investigation, biomathematics, and biostatistics work together to find ways to extend the lives of these cancer patients. Understanding of this process can be facilitated through the use of mathematical, statistical, and stochastic models that approximate progress in the varying stages of this often devastating disease. This thesis will synthesize two existing models for the distribution of sizes of detectable metastases.

Many tumor suppressor proteins act to blunt the effects of mitogenic signaling pathways. Loss of function mutations in the Merlin tumor suppressor underlie Neurofibromatosis type 2 NF2) a familial autosomal dominant cancer syndrome. Studies of Drosophila suggest that Hippo Hpo) is required for inhibition of cell proliferation mediated by dMer, the orthologue of human Merlin. Mammalian sterile 20-like kinase-2 Mst2) is a mammalian Hpo orthologue； and numerous studies implicate Mst2 as a tumor suppressor. Mst2 is negatively regulated by the proto oncoprotein Raf-1 in a manner independent of Raf-1s kinase activity. We sought to determine if, in mammalian cells, Merlin could positively regulate Mst2. We also sought to determine if Mst2, in addition to being negatively regulated by Raf-1, itself might reciprocally regulate Raf-1. In contrast to findings from Drosophila, we find no evidence that mammalian Merlin positively regulates mammalian Mst2. Instead, surprisingly, RNAi silencing of Mst2 leads to elevated inhibitory phosphorylation of Raf-1 at Ser259 and impaired Raf-1 kinase activity. Consequent to this, ERK pathway activation and cell proliferation are attenuated. Phosphatase 2A PP2A) dephosphorylates Raf-1 Ser259 in response to mitogens. Interestingly RNAi silencing of Mst2 triggers a striking proteasome-dependent decrease in the levels of the catalytic subunit of PP2A PP2A-C). A similar effect is achieved upon silencing of large tumor suppressor Lats1/2) direct downstream substrates of Mst2. Our studies reveal a more complex role for Mst2 than previously thought. The Mst2-Lats1/2 pathway, by maintaining PP2A-C levels, may, in some situations, positively affect mitogenic signaling. Chapter II of this thesis presents unpublished and preliminary data indicating that Merlin may function as a tumor suppressor through negative regulation of mTORC1 complex and promotion of autophagy.

The Wnt/beta-catenin pathway is a key pathway involved in the regulation of proliferation and differentiation within many tissues, including the epithelium of the intestine and of the skin. Wnt signaling is implicated in stem cell renewal. Deregulation of Wnt/beta-catenin signaling is crucial early event in colorectal tumorigenesis. Earlier work in our lab demonstrated that Kruppel-like factor 4 (KLF4) interacts with beta-catenin in vivo, repressing Wnt signaling and inhibiting tumor growth. KLF4 is an anti-proliferative transcription factor expressed in differentiated epithelial cells in the intestine. Previous studies clearly establish KLF4 as a tumor suppressor in colorectal cancer. Expression of KLF4 is downregulated in colorectal tumors, and heterozygous deletion of the Klf4 allele in a mouse model of colorectal cancer results in the formation of approximately 50% more tumors. In addition, KLF4 is important in stem cell programming. KLF4, in combination with three additional transcription factors, is sufficient to reprogram differentiated fibroblasts into embryonic stem cells. This dissertation focuses on the molecular mechanisms of KLF4-mediated transcription, both in the context of KLF4-mediated activation and KLF4-mediated repression. I demonstrate that the N-terminal transactivation domain of KLF4 recruits the co-activator p300/CBP to the promoter of the positively-regulated gene IAP, resulting in increased histone acetylation. On the negatively regulated gene, Cyclin B1, I demonstrate that KLF4 recruits HDAC3, decreasing histone acetylation. In addition, I show that KLF4 is acetylated by p300/CBP, and that acetylation of KLF is important in the activation of target genes as well as its ability to inhibit proliferation. I demonstrate that KLF4 inhibits Wnt/beta-catenin signaling by blocking beta-catenin-mediated recruitment of p300/CBP to Wnt-regulated genes. Finally, I show that acetylation of beta-catenin is important in its ability to interact with p300/CBP and that KLF4 inhibits beta-catenin acetylation. These studies provide significant insight into the molecular mechanisms of KLF-mediated transcription, and will prove useful in the development of targeted therapies for colorectal cancer as well as providing a deeper understanding of the mechanisms behind stem cell reprogramming.

Yttrium-90 90Y) microsphere therapy is being utilized as a treatment option for patients with primary and metastatic liver cancer due to its ability to target tumors within the liver. The success of this treatment is dependent on many factors, including the extent and type of disease and the nature of prior treatments received. Metabolic activity, as determined by PET imaging, may correlate with the number of viable cancer cells and reflect changes in viable cancer cell volume. However, contouring of PET images by hand is labor intensive and introduces an element of irreproducibility into the determination of functional target/tumor volume FTV). A computer-assisted method to aid in the automatic contouring of FTV has the potential to substantially improve treatment individualization and outcome assessment. Commercial software to determine FTV in FDG-avid primary and metastatic liver tumors has been evaluated and optimized. Volumes determined using the automated technique were compared to those from manually drawn contours identified using the same cutoff in the standard uptake value SUV). The reproducibility of FTV is improved through the introduction of an optimal threshold value determined from phantom experiments. Application of the optimal threshold value from the phantom experiments to patient scans was in good agreement with hand-drawn determinations of the FTV. It is concluded that computer-assisted contouring of the FTV for primary and metastatic liver tumors improves reproducibility and increases accuracy, especially when combined with the selection of an optimal SUV threshold determined from phantom experiments. A method to link the pre-treatment assessment of functional PET based) and anatomical CT based) parameters to post-treatment survival and time to progression was evaluated in 22 patients with colorectal cancer liver metastases treated using 90Y microspheres and chemotherapy. The values for pre-treatment parameters that were the best predictors of response were determined for FTV, anatomical tumor volume, total lesion glycolysis, and the tumor marker, CEA. Of the parameters considered, the best predictors of response were found to be pre-treatment FTV &le；153 cm3, ATV &le；163 cm3, TLG &le；144 g in the chemo-SIRT treated field, and CEA &le；11.6 ng/mL.

Le cancer du sein affecte des milliers de canadiennes a chaque annee. Ainsi, la recherche pour une meilleure comprehension et pour le developpement de nouvelles idees au sujet du traitement et de la prevention de ce type de cancer continue a etre primordiale. Chez certains cas de cancer du sein, on retrouve la presence d’une proteine nommee mammaglobine 1 (MGB1). MGB1 appartient a la famille des secretoglobines et elle forme un complexe avec une autre secretoglobine, soit la lipophiline B (LPB). Les profils d’expression complets de MGB1 et LPB ne sont pas encore bien determines, mais il est certain qu’elles sont exprimees au niveau de cellules mammaires saines et surexprimees au niveau de cellules mammaires cancereuses. Les fonctions du complexe MGB1/LPB sont cependant encore inconnues. Neanmoins, dans les dernieres annees, certaines etudes ont souleve une premiere avenue a suivre dans le but d’elucider le role de MGB1/LPB； elles ont demontre que la presence de MGB1 dans des tumeurs mammaires primaires semble refleter un meilleur pronostic pour ce type de cancer. Le but de cette etude est alors d’elucider davantage les implications du complexe MGB1/LPB au niveau du cancer du sein. Plus precisement, elle examine si le complexe MGB1/LPB est capable d’affecter les phenotypes cancereux suivants : proliferation, apoptose et pouvoir d’invasion cellulaire au niveau des cellules mammaires. Elle verifie egalement si MGB1 peut influencer l’expression de LPB ou vice-versa. Globalement, l’importance de ce projet est d’approfondir nos connaissances sur ce complexe qui semble jouer un role important dans le processus biologique des cellules mammaires cancereuses. En resume, cette etude demontre d’une part que la co-expression de MGB1 et LPB dans les tissus mammaires doit dependre de facteurs de transcription communs, car ces deux proteines ne peuvent pas stimuler mutuellement leur expression. D’autre part, le complexe MGB1/LPB semble demontrer une tendance a augmenter le taux de proliferation et diminuer le pouvoir d’invasion de la matrice extracellulaire des cellules mammaires cancereuses. En somme, cette etude a permis la fondation de theories a considerer dans le but d’elucider les implications de MGB1/LPB au niveau du cancer du sein. En effet, les resultats eventuels seront peut-etre les premiers pas vers le developpement d’une nouvelle strategie pour combattre le cancer du sein.