Sushi, von Willebrand factor type A, epidermal growth factor and pentraxin domain containing 1 (SVEP1) is an extracellular matrix protein which may bind to cell surface molecules such as integrins. A non-synonymous single amino acid polymorphism in the Svep1 gene is associated with a 14% increased risk of coronary heart disease, a 13% higher risk of type 2 diabetes and a 1mmHg increase in systolic blood pressure. Expression of the SVEP1 gene is increased in the kidney in the Cyp1a1mRen2 rat model of diabetes and hypertension previously developed in our lab. SVEP1 is also known to be upregulated in human diabetic nephropathy and is upregulated in rodent models of renal fibrosis. I hypothesized that Svep1 played a role in renal fibrosis, diabetes and blood pressure. Hence, the primary goal of this thesis was to investigate the role of SVEP1 and in the pathogenesis of diabetes, hypertension and renal fibrosis. Svep1 gene expression is increased in the kidney in the DOCA-salt-angII-uninephrectomy model of hypertension and following UUO. SVEP1 hemizygous mice showed no differences in expression of pro-fibrotic genes after UUO compared to wildtype littermates. No overt metabolic phenotype was exhibited by the Svep1 hemizygous mice, however there was a significant decrease in fat depot weights after high fat diet (HFD) and a significant increase in blood glucose concentrations during the glucose tolerance test at the 12 week time point in hemizygous Svep1 mice compared with wild-type controls. After telemetry analysis of blood pressure no difference was seen in blood pressure but SVEP1+/-animals had an increased heart rate of 100 beats per minute compared to wildtype animals. Svep1 expression is increased in the kidney in models of hypertension and fibrosis, however loss of one Svep1 allele did not alter the severity of fibrosis in the UUO model or significantly alter glucose tolerance after high fat diet. However, the high fat diet experiment was a pilot study and should be repeated with a larger number of animals. In addition, generation of a mouse with the human point mutation could determine the mechanisms by which this extracellular matrix protein confers risk of diabetes and hypertension.

The primary motor cortex (M1) is a key brain area for the generation and control of motor behaviour. Output from M1 can be driven in part by long-range inputs from a collection of thalamic nuclei termed the motor thalamus (MTh), but how MTh input shapes activity in M1 and forelimb motor behaviour remains largely unresolved. To address this issue, we first defined the 3D anatomical coordinates of mouse forelimb motor thalamus (MThFL) by employing conventional retrograde and virus-based tracing methods targeted to the forelimb region of M1 (M1FL). These complimentary approaches defined MThFL as a ~0.8 mm wide cluster of neurons with anatomical coordinates 1.1 mm caudal, 0.9 mm lateral to bregma and 3.2 mm below the pial surface. Thus, MThFL incorporates defined areas of the ventrolateral, ventral anterior and anteromedial thalamic nuclei. To investigate the importance of M1FL and MThFL during skilled motor behaviour, we developed and optimised a quantitative behavioural paradigm in which head-restrained mice execute forelimb lever pushes in response to an auditory cue to receive a water reward. Forelimb movement trajectories were mapped using high-speed digital imaging and multi-point kinematic analysis. We inactivated both M1FL and MThFL of mice performing this motor behaviour using a pharmacological strategy, which in both cases resulted in a significant reduction in task performance. Inactivating M1FL significantly affected forelimb coordination and dexterity, resulting in erratic motion and posture. In contrast, mice with MThFL inactivated displayed a reduction in total motor output, although correct posture was maintained. We performed extracellular recordings in MThFL of expert-level mice, demonstrating that motor thalamic output during execution of task was dominated by a robust response to the onset of the auditory cue. Cue-evoked responses were also observed in motor thalamic neurons of naive mice. We have developed a novel solution to the stability problem encountered when performing whole-cell patch-clamp recordings from the motor cortex of head-restrained mice performing forelimb motor behaviour, and present preliminary recordings maintained through the execution of forelimb behaviour.

SK3 channels contribute to membrane repolarization and hyperpolarization that leads to both relaxation of smooth muscle and vascular branching. These two distinct properties are intensified in the SK3T/T mice possibly influencing pregnancy by dampening uterine contractions and causing dysfunctional placental development. SK3T/T mice have delayed or hindered parturition, suggesting a role for SK3 channels in labor contractions (Chapter 2 & 3). Based on these findings, we hypothesized that SK3 channel expression must be reduced late in normal pregnancy to enable the uterus to produce the forceful contractions required for parturition. The mechanism(s) downregulating this channel in the uterus during pregnancy is unknown. The SK3 gene promoter region contains two Specificity Protein (Sp) binding sites; Sp1, a transcription factor that enhances transcription of genes in response to estrogen, and Sp3, a factor that competes for the same binding motif as Sp1 to reduce gene expression (Chapter 4). SK3 channels may also be involved in the vascular remodeling that occurs during pregnancy. The SK3 channel is present in vascular endothelial cells and overexpression of this channel leads to abnormal vessel branching and an increase in vessel diameter. During pregnancy, the vascular system must adapt to accommodate dramatic increases in blood volume necessary to sustain the developing fetus. Overexpression of SK3 channels could produce abnormalities in the placental vascular network, similar to the abnormal vessel branching and vasodilatation found in the mesenteric circulation, thus leading to poor fetal outcome (Chapter 5). The aim of this research was to determine the function of the SK3 channel in pregnancy by focusing on its role in myometrial contractility in addition to identifying its role in remodeling the maternal vasculature and its impact on placental blood flow and fetal demise.

Renin is an enzyme that catalyzes the rate-limiting step in the production of angiotensin peptides, and is thus a key regulator of processes controlled by angiotensin such as blood pressure, hydromineral balance, and metabolism. Our laboratory and others have previously identified a novel isoform of renin (icRen) which, as a result of the utilization of an alternate first exon, lacks the signal peptide and first third of the pro-segment of classical secreted renin (sRen). This alternate icRen isoform thus remains within the cytoplasm of the cell, but is constitutively active. Here, we report that while sRen is the predominant form of renin expressed in most tissues during development, icRen is the predominant form of renin within the adult brain. Thus, we hypothesized that sRen and icRen play distinct physiological roles in adult mice. To examine this hypothesis, we have utilized the Cre-LoxP system to selectively delete either isoform globally or within selected cell types such as neurons and glia. We have successfully developed a "sRen-flox" model, in which endogenous mouse sRen isoform can be selectively deleted, while not affecting endogenous icRen production. Breeding these mice against the E2A-Cre, Nestin-Cre, and GFAP-Cre mouse lines resulted in global-, neuronal-, and glial-specific knockouts of sRen, respectively. Physiological characterization of resulting mice has uncovered postnatal lethality, hypotension, renal atrophy, vascular dysfunction and decreased body weight and white adipose in the global knockouts. Depletion of sRen from only neuronal or glial cells does not appear to alter any of these phenotypes at baseline. From these data, we conclude that while peripheral sRen is of primary importance to blood pressure regulation, hydromineral balance, and metabolism, central expression of this isoform is unimportant. Further, comparison of our results to published findings from global total renin knockout models indirectly supports a role for icRen in the brain. We are currently in the process of generating icRen-flox and subsequent knockout mice, which will be useful models to directly analyze the physiological role(s) of icRen.

Calcitonin gene-related peptide (CGRP) has been strongly implicated in the pathophysiology of migraine. CGRP levels are elevated during a migraine attack. Injection of CGRP can trigger a delayed migraine-like headache in migraineurs. Finally, CGRP receptor antagonists are effective antimigraine therapeutics. Consequently, a CGRP-sensitized mouse, nestin/hRAMP1 was genetically engineered to conditionally express a subunit of the CGRP receptor, hRAMP1, in neurons and glia. In response to CGRP, nestin/hRAMP1 mice demonstrated a significant decrease in time in the light zone of a dim light-dark box compared to vehicle-treated nestin/hRAMP1 mice and CGRP-treated control mice. This reflects photophobia-like behavior. Photophobia is a common symptom of migraine, where light exacerbates the headache pain. Furthermore, CGRP decreased motility in the dark zone, which may reflect exacerbation of pain by movement that is often experienced during a migraine. Wildtype mice have also demonstrated this CGRP-induced behavior, but required bright light and habituation to the chamber. While there is a difference in sensitivity in this assay between wildtype and nestin/hRAMP1 mice, it demonstrates that endogenous CGRP receptors are sufficient to convey this behavior. A common antimigraine drug, rizatriptan, attenuated the CGRP-induced behaviors in wildtype mice validating the assay as a migraine model. To explore the relative contributions of CGRP receptors on neurons versus glia, synapsin/hRAMP1 transgenic mice were genetically engineered to express hRAMP1 in neurons only. In contrast to the nestin/hRAMP1 mice, the synapsin/hRAMP1 mice did not show CGRP-induced light aversion upon naïve exposure to a dim chamber. This suggests that neuronal overexpression of hRAMP1 is insufficient to convey a heighted sensitivity to CGRP in the light aversion assay. As a first step to understanding the mechanism underlying CGRP-induced light aversion, a non-behavioral assay was developed to measure photic blink reflexes by measuring orbicularis oculi EMG responses in mice. Bright light increased orbicularis oculi activity, and an air puff induced a blink response. Interestingly. CGRP and bright light increased the duration of squinting following the air puff-induced blink. This pilot suggests that the trigeminal system plays a key role in mediating CGRP-induced light sensitivity. Overall, these studies propose a potential model for the mechanisms involved in migraine and photophobia in which CGRP likely acts through endogenous CGRP receptors on neurons and glia in the trigeminal system to trigger light sensitivity.

Pancreatic ductal adenocarcinoma (PDAC) is an incurable, highly metastatic cancer resistant to current treatments. A better understanding of the genetic basis of PDAC progression is urgently needed to improve treatment options. The ARF tumor suppressor is inactivated in ~45% of PDAC. My thesis lab identified a new, uncharacterized ARF binding protein, Partner of ARF isoform 1A (Parf-1A). This thesis explores the hypothesis that Parf-1A plays an important role in PDAC and ARF tumor suppressor signaling Initial studies sought to develop a novel mouse xenograft model of PDAC metastasis that would expedite testing of putative PDAC genes. Human PDAC cell lines stably expressing luciferase were generated and introduced by intracardiac injection into immunodeficient mice to model hematogenous dissemination of cancer cells. Tumor development was monitored non-invasively by bioluminescence imaging and found to recapitulate PDAC tumor formation and metastatic distribution. The model was validated by the ability of ARF to suppress PDAC cancer cell migration in vitro and reduce tumor cell colonization in vivo; establishing a new bioluminescent mouse model for rapidly assessing the significance of suspected PDAC genes. Using human PDAC cell lines and tumor specimens, we investigated the role and significance of Parf-1A to PDAC. RNAi analyses demonstrated Parf-1A is required for PDAC cell survival, proliferation and resistance to the PDAC therapeutic, oxaliplatin. PDAC cells are ARF-null; therefore these tumor promoting activities of Parf-1A were independent of ARF. Notably, immunohistochemical analyses of Parf-1A in human PDAC tumors showed Parf-1A expression is a prognostic marker of poor survival in PDAC patients. These data suggest Parf-1A is a novel biomarker of PDAC and potential target for anticancer therapy. Other studies tested how Parf-1A influenced ARF signaling. Parf-1A depletion and overexpression showed it inhibits ARF anti-proliferative activity by mobilizing ARF from the nucleus (where it is functional) into the cytoplasm. These data show Parf-1A is a new inhibitor of ARF. Considered with findings that Parf-1A can act independent of ARF to promote PDAC tumorigenesis, such results suggest Parf-1A is a novel oncoprotein that acts through multiple pathways to facilitate tumorigenesis. Thus, Parf-1A may have broad relevance to many types of human cancers.

The gene PTPN11 was identified in the early 1990s, and encodes the non-transmembrane protein tyrosine phosphatase SHP2. SHP2 is expressed ubiquitously in cells, and plays an important role in cancer. Unlike most phosphatases, SHP2 positively regulates several signaling pathways including the Ras/MAPK and Src signaling pathways and acts as a proto-oncogene. SHP2 is also a cancer essential gene in certain types of carcinomas, and promotes growth, survival, and epithelial to mesenchymal transformation. Gain of function (GOF) SHP2 mutations are known leukemic oncogenes, and have been identified to a smaller extent in solid tumors as well. Currently, the roles of SHP2 in lung carcinoma are not fully understood. While GOF SHP2 mutations have been detected in lung cancer, their contributions to cellular transformation had not been established. In addition, SHP2 is known to promote EGF growth factor receptor (EGFR) signaling. Since GOF EGFR mutations induce transformation of lung epithelial cells, it is possible that SHP2 plays a role in promoting GOF EGFR mutant driven tumorigenesis. The objective of this dissertation is to determine whether SHP2 can act as an oncogene in lung epithelial cells and whether SHP2 inhibition can affect GOF EGFR mutant induced lung cancer. To achieve these aims, we generated two novel doxycycline (Dox) inducible transgenic mouse models which express either the GOF SHP2E76K leukemic oncogene or the dominant negative SHP2CSDA mutant under the control of the Clara cell secretory protein (CCSP) promoter to regulate transgene expression to type II pneumocytes. To determine whether SHP2 plays a role in promoting GOF EGFR mutant signaling, we started by disrupting SHP2 function in vitro. Two non small lung cancer cell lines were used for this project: HCC827 carries the LREA deletion in exon 19, and H1975 co-expresses the EGFRL858R point mutation and the EGFRT790M gatekeeper mutation. After SHP2 PTP inhibition or knock-down by shRNA and siRNA, both cell lines exhibited decreased cell proliferation and reduced levels of pErk1/2 and c-Myc. Based on these results, we acquired a transgenic mouse line which expresses the EGFRL858R mutant under the control of the tet-O promoter and generated bitransgenic CCSP-rtTA/tetO-EGFRL858R and tritransgenic CCSP-rtTA/tetO-EGFRL858R/tetO-SHP2CSDA mice to study the effects of the dominant negative SHP2CDSA mutant on EGFRL858R mediated carcinogenesis in vivo. After 4, 6, and 8 weeks of Dox induction, pErk1/2 and pSrc levels were increased in the lungs of bitransgenic mice compared to wild type controls. Both kinases were suppressed by SHP2CSDA expression in tritransgenic mice. In addition, SHP2CSDA expression delayed tumor onset and prevented progression to a more aggressive phenotype. Tritransgenic mice also developed a smaller tumor burden compared to bitransgenic animals. These results suggest that SHP2 is critical for GOF EGFR mutant mediated lung tumorigenesis and describe a new role of SHP2 as a potential therapeutic target for the development of novel NSCLC drugs. Once we generated our CCSP-rtTA/tetO-SHP2E76K transgenic mouse model, we administered Dox for one month and found that SHP2E76K expression upregulates pErk1/2, pSrc, pGab1, c-Myc and Mdm2 levels in the lungs of bitransgenic mice compared to controls. After six to nine months of Dox induction, SHP2E76K expression caused formation of lung adenomas and adenocarcinoma. We then took advantage of the reversible feature of our mouse model to test whether lung tumors are dependent on sustained SHP2E76K expression for survival. MRI analysis of lung adenocarcinomas showed full regression of the lung tumors after Dox withdrawal. Histological evaluation of lung tissues revealed residual hyperplastic lesions as well as evidence of necrosis, while biochemical analysis showed that pGab1, pErk1/2, pSrc and c-Myc returned to basal levels. These results demonstrate that sustained SHP2E76K expression is required for lung tumor maintenance. Moreover, this data describe a novel function of SHP2E76K as an oncogene in lung carcinoma.

<p>Glioblastoma multiforme (GBM) is the most common form of primary brain tumor in humans. Its aggressive and infiltrative growth into the brain, and, at best, only partial sensitivity to radiotherapy and chemotherapy, renders it extremely difficult to treat and survival remains dismal. </p><p>Growth factors, such as platelet-derived growth factor (PDGF) and epidermal growth factor (EGF), and their corresponding receptors are seen in glioma tissue, suggesting the presence of autocrine stimulatory loops. <i>PDGFB</i> and a mutated EGF receptor were also identified as cellular homologues of two oncoviruses, thereby indicating a role in tumorigenesis. This thesis presents a brain tumor model in mice, developed using a <i>PDGFB</i> coding retrovirus to induce overexpression of PDGF-B in neonatal mouse brain. Immature tumors, with histological characteristics of primary brain tumors developed at relatively high frequencies. Mice injected with a non-coding retrovirus did not develop tumors, indicating the crucial role of PDGF stimulation in this system. Tumor cells were also shown to continue to depend on PDGF stimulation when cultured <i>in vitro</i>. </p><p>In human glioblastomas, growth factor receptor signaling is present in conjunction with defects in cell cycle arrest pathways. When the <i>PDGFB</i>-virus model was used with <i>p53</i> or <i>Ink4a-Arf</i> deficient mice, tumors arose with shorter latency and higher frequency. Loss of <i>p53</i> or <i>Ink4a-Arf</i> seemed to facilitate signaling through the PI3K/Akt pathway. Thus, a functional role for the co-existence of p53 loss of function and PDGF signaling in a subset of gliomas is presented. </p><p>Human GBM samples were collected and analyzed with respect to expression and activation of the EGFR and PDGFRα. Most tumors expressed the both receptors at moderate to high levels, but high activation of either receptor seemed mutually exclusive.</p>

Glioblastoma multiforme (GBM) is the most common form of primary brain tumor in humans. Its aggressive and infiltrative growth into the brain, and, at best, only partial sensitivity to radiotherapy and chemotherapy, renders it extremely difficult to treat and survival remains dismal. Growth factors, such as platelet-derived growth factor (PDGF) and epidermal growth factor (EGF), and their corresponding receptors are seen in glioma tissue, suggesting the presence of autocrine stimulatory loops. PDGFB and a mutated EGF receptor were also identified as cellular homologues of two oncoviruses, thereby indicating a role in tumorigenesis. This thesis presents a brain tumor model in mice, developed using a PDGFB coding retrovirus to induce overexpression of PDGF-B in neonatal mouse brain. Immature tumors, with histological characteristics of primary brain tumors developed at relatively high frequencies. Mice injected with a non-coding retrovirus did not develop tumors, indicating the crucial role of PDGF stimulation in this system. Tumor cells were also shown to continue to depend on PDGF stimulation when cultured in vitro. In human glioblastomas, growth factor receptor signaling is present in conjunction with defects in cell cycle arrest pathways. When the PDGFB-virus model was used with p53 or Ink4a-Arf deficient mice, tumors arose with shorter latency and higher frequency. Loss of p53 or Ink4a-Arf seemed to facilitate signaling through the PI3K/Akt pathway. Thus, a functional role for the co-existence of p53 loss of function and PDGF signaling in a subset of gliomas is presented. Human GBM samples were collected and analyzed with respect to expression and activation of the EGFR and PDGFRα. Most tumors expressed the both receptors at moderate to high levels, but high activation of either receptor seemed mutually exclusive.

Hepatitis C Virus (HCV) is one of the most important causes of chronic liver disease, affecting more than 170 million people worldwide. The mechanisms of hepatitis C pathogenesis are unknown. Viral cytotoxicity and immune mediated mechanisms might play an important role in its pathogenesis. HCV infection and alcohol abuse frequently coexist and together lead to more rapid progression of liver disease, increasing the incidence and prevalence of cirrhosis and hepatocellular carcinoma. The cytopathic effect of HCV proteins, especially the core, E1 and E2 structural proteins, which induce liver steatosis, oxidative stress and cell transformation may be amplified by alcohol abuse. The purpose of this study was to characterize the liver damage mechanisms induced by HCV structural proteins and alcohol and to determine the potential molecular mechanism(s) that may promote chronic, progressive liver damage. A transgenic mouse model expressing HCV core, E1 and E2 was used to investigate whether alcohol increased HCV RNA expression. Real-time RT-PCR analysis of genes involved in lipid metabolism and transport confirmed their abnormal expression in the alcohol-fed transgenic mice. In addition, light and electron microscopy analysis were performed on liver tissues of transgenic mice on an alcoholic diet versus those on a normal diet, in order to identify histological changes. The severe hepatopathy in HCV transgenic mice was exacerbated by alcohol. Mitochondria and endoplasmic reticulum had severe abnormalities in the electron microscopy analysis. The second part of this study focused on adaptive immune responses, which may also play an important role in HCV pathogenesis. I focused my analysis on dendritic cells (DC), which have been the main suspects to explain immune impairment in HCV infection. Their powerful antigen-presenting function allows them to stimulate the antiviral response of CD4+ and CD8+ T cells, the effector cells of the immune system. This unique function of the DC makes them possible targets for immune evasion by the Hepatitis C virus. In this study, DCs were generated from mouse bone marrow cells. I investigated their maturation capacity in the presence of structural proteins of HCV. The impact of HCV core/E1/E2 polyprotein on DCs cytokine expression and ability to activate T-cell lymphocytes was also analyzed. A dysfunctional CD4 T cell response was observed after exposure of DCs to core/E1/E2 polyprotein, indicating inefficient CD4 priming, which might lead to chronic HCV infection in humans. The presence of the core/E1/E2 polyprotein reduced the DC maturation capacity and the expression of certain cytokines (IL-12, IFNg, IL-6, MCP-1) important for stimulation and chemotaxis of T cells and other immune cells. My studies contribute to the understanding of HCV pathogenesis and may have implications to the development of better therapies for HCV infection.