<p>Bipolar disorder, or manic-depressive illness, is a common psychiatric illness affecting 1.5% of the population. It is a chronic illness that requires long-term (often lifetime) pharmacotherapy. The need for lifetime treatment with mood stabilizers (i.e. lithium, valproate and carbmazepine) and high relapse rates suggests that changes in gene expression may be involved in the mechanism of action of these drugs. Using differential display, we identified the 78-kilodalton glucose-regulated protein, GRP78, as being at valporate-regulated gene in both rat brain and cultured cells. GRP78 belongs to a family of resident endoplasmic reticulum proteins, referred to as teh ER stress proteins, which includes GRP94 and calreticulin. These proteins act as both calcium binding proteins and molecular chapterones within the endoplasmic reticulum and when overexpressed protect the cell against cytotoxic insults. In addition to GRP78, we were able to show that GRP94 and calreticulun are also upregulated in a concentration-and-time-dependent manner in response to valporate. This effect was shown to be either drug-or drug class-specific, as treatment with other psychotropic drugs did not elicit a similar response. In patient samples, all three ER stress proteins were increased in postportem temporal cortex from depressed subjects to who died by suicide. These results suggest a possible cytoprotective role for mood stabilizers, since studies have shown that valporate as well as lithium can regulate the expression of proteins with known cytoprotective properties. Indeed, when rat hippocampal neurons were treated for 7 days with the mood stabilizers, lithium, valporate and carbamzepine, a significant reduction in NMDA-mediated cytoplasmic vacuolization could be quantified using transmission electron microscopy. In conclusion, the results of this thesis contribute to a growing number of studies that propose neuroprotective for mood stabilizing drugs, and suggest that valporate might act in part by regulating ER stress proteins.</p> / Doctor of Philosophy (PhD)

<p>The alterations in renal vascular structure and function, and their role in the development and maintenance of hypertension were examined in Kyoto Wistar spontaneously hypertensive rats (SHR) and Wistar Kyoto normotensive controls (WKY). A study of the renal vascular bed in SHR with established hypertension and age matched normotensive WKY indicated that when the isolated kidney was perfused at a variety of flow rates, under maximally relaxed conditions, the renal vascular resistance (RVR) was similar between SHR and WKY. Consistent with the above finding, morphometric measurements of light and electron micrographs indicated that the lumen diameter of relaxed main renal, interlobar, arcuate and interlobular arteries, as well as the preglomerular arterioles was similar in SHR and WKY. The cross-sectional areas of total intima, endothelium, subendothelial space, internal elastic lamina and total adventitia, as well as the volume fraction of axons, nerve sheath cells, fibroblasts, collagen and fluid filled space within the adventitia were only modestly altered in SHR. However, with the exception of the preglomerular arterioles, the media of all the renal arterial classes of SHR exhibited an increase in smooth muscle cell (SMC) cross-sectional area and volume that was produced by SMC hypertrophy and/or hyperplasia, while the extracellular space surrounding the SHCs was increased in both arteriolar and pre-arteriolar vessels.</p> <p>Based on these structural alterations, it was hypothesized that, if the mass of the arterial media is increased, contraction from the adventitial side would tend to push the media of the thicker hypertensive vessel into the lumen to a greater degree than the thinner walled WKY vessel. Under relaxed conditions, the RVR would be expected to be similar between SHR and WKY; however, during contraction RVR should be elevated to a greater degree in SHR than WKY. To test the above hypothesis, pharmacological studies were undertaken. Consistent with the model, at maximal relaxation the RVR was similar in SHR and WKY, while contraction of the renal vascular bed with infused norepinephrine (HE), BaCl₂ angiotensin II, or by stimulating the periarterial nerves produced a larger elevation of RVR in SHR. Aside from a modest increase in HE sensitivity within the renal vasculature of WKY, the contractile sensitivity to the various agents was not altered when SHR and WKY were compared. These studies indicated that the nerve mediated contractile responses within the renal vasculature were mediated by alpha₁ and dopamine receptors, and the proportion of the maximal response attributed to each receptor was similar in SHR and WKY.</p> <p>Similar alterations, but of lesser magnitude, as those present in SHR with established hypertension were found to occur in prehypertensive SHR. The RVR at maximal relaxation was similar to that present in WKY, while the lumen diameter of the main renal, interlobar and cortical arteries was not modified between the two groups. All renal arteries of prehypertensive SHR that were studied exhibited an increase in the cross-sectional area ratio of arterial wall (intima + media) in relation to the lumen, and an increased number of SMC layers within the media. Consistent with the proposed model, when the renal vasculature of prehypertensive SHR was maximally contracted by infusing NE or by stimulating the periarterial nerves (under conditions where the presynaptic uptake of NE was blocked) the amplitude of RVR change was higher in prehypertensive SHR than WKY. As in SHR with established hypertension, the contractile sensitivity of the renal vasculature to NE was modestly increased in WKY.</p> <p>To further test if such alterations occur independently of high blood pressure, hydralazine (an antihypertensive drug that crosses the placental barrier) was fed to female SHR prior to, and during, pregnancy, and subsequently to newborn rats from birth to 21 weeks of age. These animals were compared to similarly treated WKY and nontreated SHR and WKY controls. Treated SHR maintained normal blood pressure throughout the treatment period. The in utero and post-natal normalization of blood pressure in SHR had virtually no effect in altering the renal vascular wall thickness. Within most of the arteries studied, SHR with normalized blood pressure had similar cross-sectional quantities of media and SMC layers as were present in untreated SHR, and greater quantities than that present in either control or treated WKY. The withdrawal of hydralazine from 26 week old in utero and post-natally treated SHR resulted in the re-establishment of hypertension within two days of withdrawal to the levels that were present in control SHR.</p> <p>These results suggest that the thickening of the renal vascular wall in SHR could be of etiological importance in the initiation and maintenance of high blood pressure in SHR, and that such changes are not a secondary modification produced by the elevation of blood pressure.</p> / Doctor of Philosophy (PhD)

<p>Direct assessments of the relative contributions of the major energy releasing pathways in human skeletal muscle during heavy exercise are difficult to obtain due to the invasive measurements required. With an isolated muscle preparation the muscles' environment is carefully controlled and all metabolic measurements are directly obtained. For this reason the isolated perfused rat hindquarter model, previously used to study resting muscle metabolism, was developed to examine the metabolism and performance heavily contracting skeletal muscle.</p> <p>Energy calculations based upon measurements of O₂ uptake (aerobic metabolism), lactate production (anaerobic glycolysis) and CP hydrolysis (alactic anaerobiosis) were made during 20 minutes of repetitive tetanic stimulation. During the initial 5 minutes of stimulation isometric tension production was high but fatigued rapidly and anaerobic involvement in energy production was large (30%), especially in the fast-twitch glycolytic muscle fibers. Muscle glycogenolysis provided the majority of substrate for both anaerobic glycolysis and aerobic metabolism. During the final 15 minutes of stimulation aerobic metabolism dominated (90%) while 60% of peak tension was held, mainly by the fast-twitch oxidative, glycolytic muscle fibers. Glycogen utilization was minimal and intramuscular triacylglycerol became the dominant fuel for oxidative metabolism, contributing 62% of the energy produced.</p> <p>Perfusions with acidotic mediums (metabolic and respiratory) reduced muscle glycogenolysis and lactate accumulation by 35% during the initial 5 minutes of stimulation. The decreased glycolytic flux reduced the availability of carbohydrate substrate for aerobic metabolism and O₂ uptake decreased. The associated reduction in energy release produced an increased rate of tension decay. Total energy release and tension production were also reduced during acidosis in the final 15 minutes of stimulation. The decreased glycolytic flux appeared to be due to an earlier fall in muscle pH during acidosis and subsequent inhibition of key regulatory enzymes such as phosphorylase and phosphofructokinase. However an alternate hypothesis is that acidosis exerted a direct negative effect on the excitation-contraction coupling mechanism, thereby reducing the need for energy production.</p> / Doctor of Philosophy (PhD)

<p>The studies contained in this thesis describe physical, biochemical and immunochemical characterization of human malignant melanoma cell surface antigens defined by a monkey antiserum and a mouse monoclonal antibody (MoAb).</p> <p>The monkey anti-CaCL 73-36 antiserum identified, melanoma-associated surface antigens (MAAs) that were shed from cultured melanoma cells. The shed MAAs appeared to be similar to shed HLA-A,B,C antigens in terms of flotation on KBr, sedimentation in sucrose gradient, and association with B2-microglobulin.</p> <p>The MoAb 140.240, which reacted with a melanoma-specific oncofetal antigen identified an epitope on an 87kd molecule that was integrally associated with melanoma cell plasma membrane. This molecule was a monomeric sialoglycoprotein that originated from a 77kd precursor polypeptide (p77). The precursor was converted to an intermediate 83kd giycopolypeptide (gp83) which in turn was further glycosylated to yield the mature 87kd glycopolypeptide (gp87).</p> <p>Both gp87 and p77 were detectable in the chase medium of melanoma cells metabolically labelled in the presence of tunicamycin (TM), but had molecular weights slightly larger than their cellular counterparts. Comparison of tryptic peptide maps of cellular and shed gp87 showed complete overlapping, except that the shed form contained two additional methionine-containing peptides. Surface radioiodination of TM-treated cells showed that p77 was also expressed on the cell surface.</p> <p>The purification of gp87 from membrane lysate and spent medium of cultured melanoma cells was achieved by a procedure involving ion-exchange, gel filtration, and antibody affinity column chromatography. Compared with the starting materials, the procedure yielded a purification factor of >3,300 and 28.6% recovery for cellular gp87, and a purification factor of >1,200 and 41.5% recovery for shed gp87. On SDS-PAGE, the purified preparations gave a single band of 87kd.</p> <p>Immunochemical analysis of MoAb 96.5 to a 97kd MAA (p97) developed in another laboratory showed that gp87 and p97 were identical molecules, although the epitope recognized by MoAb 140.240 was distinct from that recognized by MoAb 96.5.</p> <p>The biological and clinical significance of these findings in relation to the reports of other investigators on human melanoma-associated antigens are discussed.</p> / Doctor of Philosophy (PhD)

<p>Adenovirus type 5 is able to oncogenically transform cells growing in culture. These transformed cells show different degrees of transformation. The interest of my studies was to establish the presence and characterize the structure of virus DNA in adenovirus 5 transformed cell lines and to look for a possible correlation between the integrated viral sequences and the phenotype of the cells. For this purpose, cell lines transformed by wild type (virions or DNA) and by host range mutants (virions) were analyzed for their viral DNA content and for their transformed phenotype.</p> <p>More than the left 8% was always present in virion transformed cells. Cells transformed by host range mutants of complementation Group I generally contained a larger fraction of the genome than did their counterparts transformed by wild type virus. In some host range transformed cells, virtually the entire viral DNA molecule was found colinearly integrated.</p> <p>In the case of the cells transformed by wild type DNA or virions, it was not possible to correlate any particular phenotype with a specific integration pattern. The same pattern was found in the tumorigenic derivatives of a non oncogenic cell line (293) as in the parental line. Studies with cells transformed by Group I host range mutants showed that partial or complete integration of viral DNA into transformed cells was always associated with the production of some viral proteins and with the induction of a partially transformed, non oncogenic phenotype.</p> <p>An interesting finding was the limited number of inserts in cell lines isolated from uncloned populations within a small number of passages after transformation. This suggested the possibility of a limited number of sites available for transformation or alternatively, that a few cells rapidly overgrew the rest. In an attempt to answer this question and to obtain more information on the integration process, DNA was extracted from semipermissive rat cells at different times shortly after infection. Adenovirus 5 wild type and host range mutants from Groups I and II were used in individual experiments. The extracted DNA was alnalyzed for the presence and state of viral DNA. New forms of intracellular viral DNA, which might be intermediates for integration, were found. A fragment having the size of both ends of the viral DNA joined together, and which hybridizes with both ends when these are used separately as probes, was detected in Southern blots after digestion of rat cell DNA with different restriction enzymes. This structure was detected earlier after infection with host range 1 (Group I), than after infection/with host range 6 (Group II) or with wild type, and was found also in Hela cells and in two rat cell lines after infection with host range J or wild type. Studies followed to determine the nature of this new arranged viral DNA provided evidence for covalent head to tail joining and for the formation of circular Ad5 DNA molecules.</p> / Doctor of Philosophy (PhD)

<p>Biochemical and structural abnormalities have been reported in postmortem brain tissue from patients with mood disorders. Studies of the molecular pharmacology of drugs used in the treatment of mood disorders have led to a reinterpretation of earlier models of neuropathology in these diseases. Noradrenergic and serotonergic hypotheses have been expanded to include postsynaptic intracellular signal transduction pathways, regulation of gene expression, and synaptic plasticity. Because much of this evidence was obtained from postmortem brain, the experiments in this study also used postmortem tissue to examine the neuropathology of mood disorders. Human postmortem brain tissue was obtained from the Stanley Consortium Neuropathology Foundation and consisted of pieces of temporal and occipital cortex, and slices of prefrontal cortex and hippocampus from patients with bipolar affective disorder (BD), major depressive disorder (MDD), schizophrenia (SCZ) and controls (N = 15 per group). Components of the cAMP system were examined, as were potential target transcription and neurotrophic factors. Morphological consequences arising from altered cAMP signalling such as sprouting and DNA fragmentation were also investigated. There was a trend towards blunted temporal cortex adenylyl cyclase activity in subjects with mood disorders, and decreased occipital cortex Gas in lithium-treated BD subjects. Temporal cortex CREB level~ were decreased in antidepressant-untreated MDD subjects compared to controls, and normal in those subjects treated by antidepressants at the time of death. Temporal cortex CREB levels were decreased in BD patients treated with anticonvulsants relative to those not treated by anticonvulsants at the time of death. When assessing the effect of suicide on cAMP signalling, subjects who died as a result of suicide had lower temporal cortex CREB levels and CRE-binding than subjects who died of other causes. This effect was most evident in the MDD group. In hippocampus, BDNF levels were increased in subjects treated by antidepressants compared to subjects not treated by these medications at the time of death. This finding was more pronounced in the MDD group. Hippocampal studies also showed that BD subjects had increased mossy fibre staining relative to controls and other diagnoses, suggesting increased sprouting of the dentate gyrus axons. These postmortem findings are consistent with recent animal and cell models of antidepressant and mood stabilizer pharmacology. The changes in post-receptor signalling and hippocampal sprouting are also consistent with current conceptualizations of the neurobiology of mood disorders. These studies support the use of postmortem brain tissue as a clinically relevant research model, and add to the growing literature elucidating the pathophysiology of mood disorders and the molecular pharmacology of their treatments.</p> / Doctor of Philosophy (PhD)

<p>The epidemic rise in the prevalence of allergy and asthma, primarily in the "developed" world, has impelled prolific nodes of inquiry into the epidemiology, aetiology, immunology and management of these syndromes. While studies in human patients have revolutionized pharmacological treatment of asthma and allergy, and have also intimated some of the environmental agents and conditions conducive to disease expression in susceptible populations, only animal models of asthma afford the experimental flexibility upon which detailed in vivo analysis of immunology and pathogenesis depends. Because asthma arises through airway mucosal contact with allergens, chemical pollutants and/ or infectious agents, an authentic animal model of asthma should preserve the airway as the interface of incipient contact with antigen and, by extension, as the immune microenvironment that conditions allergic sensitization. This heuristic is particularly relevant when considering questions about the immunomodulatory effects of local, anti-inflammatory intervention. The research documented in this thesis investigates several immunomodulatory concepts- including pharmacological intervention (Chapter 2), costimulatory molecule blockade (Chapter 3) and chemokinetic manipulation of cell trafficking (Chapters 4 and 5)-in a murine model of airway mucosal sensitization to an innocuous antigen. The salient message informed by these studies is that the outcome of an immune-inflammatory response is very much a reflection of the airway microenvironment in which the immune system initially processes antigen. Of substantive clinical interest, these data indicate that the efficacy of acute, therapeutic intervention must be reconciled with the status of the antigen-specific response once treatment has ceased.</p> / Doctor of Philosophy (PhD)

<p>Experimental and clinical evidence implicate the β-integrin subunit and its associated intracellular effector ILK in the initiation and progression of human breast cancer. Roles for these proteins in promoting the oncogenic process, however, have not been demonstrated in vivo. As a result, experiments were designed to test the tumour-promoting properties of β-integrin and ILK in the mammary glands of transgenic mice. The first of these experiments involved the targeted ablation of a conditional allele of β-integrin in a transgenic mouse model of human breast cancer. Using this approach, it was found that the expression of β-integrin is required for oncogeneinduced transformation of the mammary gland epithelium in vivo. This requirement for β1-integrin expression was observed during both the initiation of tumourigenesis, as well as for maintaining the growth of established tumours. In addition, a block in tumour cell proliferation following ablation of β1-integrin expression was found to be associated with the suppression of F AK autophosphorylation, providing a molecular mechanism underlying the requirement for β1-integrin expression during tumourigenesis. The second experiment was designed to test the oncogenic properties of ILK, through the establishment of transgenic mice overexpressing ILK in the mammary gland epithelium. Following the induction of a hyperplastic mammary gland phenotype, these mice developed solid mammary tumours showing evidence of epithelial-to-mesenchymal transition, confirming that ILK overexpression can contribute to mammary tumourigenesis in vivo. Since expression of both β-integrin and ILK have been reported in samples of aggressive human tumours, these results may have clinical significance to the treatment of human breast malignancies.</p> / Doctor of Philosophy (PhD)

<p>Overexpression of the Neu/ErbB-2 receptor tyrosine kinase is implicated in the genesis of human breast cancer. Previous studies have observed elevated levels of endogenous ErbB3 protein in Neu-induced tumors. Although it has been suggested that the aberrant co-expression of Neu and ErbB3 may play a critical role in the induction of human breast tumors, the biological significance and the molecular mechanism of ErbB3 regulation during Neu-mediated tumorigenesis remains unclear. The results of this thesis demonstrate that the ability of ErbB3 to be constitutively phosphorylated and to regulate cell cycle progression through the activation of the PI-3K/mTOR pathway fully depends on the activity of Neu. Both PI-3K and mTOR in turn are involved in sustaining high levels of ErbB3 protein by increasing its stability. A mutant of ErbB3 unable to recruit and activate PI-3K (ErbB3-6F), blocks Neu-induced transformation, disrupts the mTOR/4EBP1 pathway and leads to apoptotic cell death. This strongly suggests that ErbB3 's role in these tumors is to provide cell survival signals by recruiting the p85 regulatory unit of PI-3K and coupling overexpressed Neu to the PI-3K/mTOR/4EBPl pathway. This thesis further demonstrates that the mTOR inhibitor rapamycin, delays the onset and inhibits the growth of Neu-induced mammary tumors in transgenic mice. The rapamycin-induced tumor inhibition correlates with downregulation of ErbB3, S6K, 4EBP1 and cyelin D1. I also show that the ErbB3-6F mutant exhibits antitumor activity in vivo by preventing tumor growth and facilitating tumor regression. Therefore, the antitumor activity of the rapamycin analogue, CCI-779 in human breast cancer may be mediated, in part, through the inactivation of ErbB3 thereby inhibiting the critical cell survival signals necessary for transformation. Taken together these results strongly suggest that Neu requires ErbB3 to drive mammary transformation and that the ErbB2/ErbB3 heterodimer may function as an oncogenic unit to recruit distinct yet complimentary signaling pathways that cooperate during mammary tumor progression. The results of this thesis provide a molecular basis for the selective targeting of the ErbB3/PI-3K/mTOR signaling pathway in the treatment of HER2-mediated human breast malignancies in which HER3 is overexpressed.</p> / Doctor of Philosophy (PhD)

<p>Embryonic studies clearly demonstrate that MyoD and Myf5 are required for specification of the myogenic lineage. Moreover, MyoD plays an essential role in initiating expression of myogenin and inducing differentiation. However, the factors and mechanisms which temporally regulate expression of myogenic genes are not fully understood. Furthermore, the controversy of whether cell cycle withdrawal precedes induction of MyoD transcriptional activity, or MyoD activity induces cell cycle withdrawal, still exist. In this thesis, I demonstrate that although MyoD transcriptional activity is regulated by cell cycle regulators, such as pRb, cdk4 and cyclin D 1, these molecules are functioning in a cell cycle-independent manner. In addition, MyoD regulates cell cycle progression, and more specifically S-phase entry, by regulating the stability of cyclin E. Moreover, MyoD regulates expression of cyclin Dl indirectly by modulating activation of NF-lCB. Importantly, I demonstrate how the E-protein, HEB, regulates myogenic gene expression in an isoform-specific, MRF-specific, and promoterspecific manner. In the absence of HEB, MyoD is unable to effectively induce expression of myogenin, and differentiation is inhibited. Therefore, ubiquitously expressed factors, such as pRb and HEB, are able to regulate the restricted pattern of gene expression in myogenic differentiation.</p> / Doctor of Philosophy (PhD)