Vascular Endothelial Growth Factor (VEGF) has been shown to be an absolute requirement for ovarian follicle development. Although VEGF is commonly regarded primarily as an angiogenic factor, granulosa cells are a major site of VEGF synthesis in the follicle and they express VEGF receptors (VEGFR1 and VEGFR2). Further, the development of the dominant follicle is characterised by a substantial increase in granulosa cell expression of VEGF and its receptors. In spite of this, potential non-angiogenic effects of VEGF in these follicles have not been elucidated. The objective of the three studies described in this thesis was to use an in vitro bovine granulosa cell model to investigate the roles of VEGF during development of the dominant follicle. In addition, in light of evidence in other cell types, potential interactions between VEGF signalling and heterotrimeric protein signalling in these follicles were also investigated. In the first study, granulosa cells were obtained from healthy follicles with diameters of 4 to 8 mm (corresponding to just before the selection of a dominant follicle during a follicular wave) or 9 to 14 mm (encompassing all developmental stages of a dominant follicle) and exposed to a range of VEGF concentrations (1 to 100 ng/ml) encompassing concentrations found naturally in bovine dominant follicles. VEGF at 1 ng/ml, but not at higher concentrations (P > 0.1), induced significant proliferation of bovine granulosa cells from 4 to 8 mm follicles (P = 0.024) and increased the proliferative response of these cells to FSH (P = 0.045). VEGF also induced a dose-dependent increase in ERK1/2 activation by granulosa cells from 4 to 8 mm follicles (P < 0.03) but did not have any effect on expression of the steroidogenic enzyme, CYP11A1, by these cells (P > 0.1). VEGF, at a dose of 1 ng/ml (P = 0.003), but not at higher doses (P > 0.1), induced an increase in COX-2 expression by granulosa cells from 9 to 14 mm follicles. In addition, LH stimulation of both ERK phosphorylation (P < 0.05) and COX-2 expression (P < 0.05) in granulosa cells from 9 to 14 mm follicles were prevented (P > 0.1) by specific inhibition of VEGFR2, indicating that VEGF may mediate COX-2 responses to LH in these cells. The second study sought to examine the expression of heterotrimeric G-protein á subunits and PLCâ isoforms by real-time PCR and westen blotting in bovine granulosa cells throughout follicle development to identify specific molecular components of heterotrimeric G-protein pathways that may functionally interact with intracellular VEGF signals. Results showed that GNAS, GNA11 and GNAI2 were all expressed at significantly (P < 0.05) higher levels in granulosa cells of pre-ovulatorysize follicles (10.0 to 13.9 mm) than in cells from smaller follicles (2.0 to 5.9 mm and 6.0 to 9.9 mm). In addition, all PLCB isoforms except PLCB2 were expressed in bovine granulosa cells with PLCB3 being more abundant than PLCB1 and -4. Levels of PLCB3 in granulosa cells from pre-ovulatory-size follicles were much higher (>16-fold; P < 0.005) than in smaller follicles. Immunocytochemical analysis revealed that PLCB3 was located primarily in the cytoplasm, whereas PLCB1 was distributed primarily in the nucleus. These results identified Gs, Gq/11, Gi2 and PLCâ3 as candidates for cross-talk between VEGF and heterotrimeric G-protein signalling during the development of the dominant follicle. The potential involvement of these molecules on VEGF-induced responses in granulosa cells from 9-14 mm follicles was investigated in the third study by determining the effects of specific inhibitors of Gi (pertussis toxin, PTX) or Gq/11 (YM-25489) or PLCB3 siRNAs on VEGF-induced p-ERK. Results showed a 2.3 fold mean increase in p-ERK in response to VEGF in the absence of G protein inhibitors (P < 0.0001) but a VEGF response that was completely or partially abolished, respectively, in the presence of PTX (P > 0.8) or YM-25489 (1.6-fold mean increase relative to untreated controls; P = 0.039). LH induced a 1.6 fold increase in p-ERK1/2 (P < 0.02) and this response was prevented by pre-incubation with PTX (P > 0.4) or YM-25489 (P > 0.5). In contrast, similar EGF-induced phosphorylation of ERK (about 5-fold relative to controls) occurred in the absence (P < 0.003) or presence of PTX (P < 0.003) or YM-25489 (P < 0.003). Transfection of granulosa cells with 3 siRNAs targeting PLCB3 that had been previously validated by western blotting and immunocytochemistry had no effect (P = > 0.7) on phosphorylation of ERK in response to VEGF, LH or EGF in granulosa cells. In conclusion, taken together, these results suggest novel roles of VEGF in stimulating granulosa cell proliferation and expression of COX-2 in bovine dominant follicles and implicate VEGF in synergising and/or mediating the effects of gonadotrophins in these cells. In addition, these results indicate a requirement for Gi2 and Gq/11 in VEGF activation of ERK1/2 and induction of the above responses in granulosa cells.

La perméabilité vasculaire est une caractéristique cruciale de l’angiogenèse. Les acteurs principaux sont les cellules endothéliales qui la régulent en réponse à divers facteurs perméabilisant, tels que le « Vascular Endothelial Growth Factor » (VEGF). Dans le contexte pathologique du cancer, les cellules tumorales produisent de grandes quantités de VEGF qui stimulent la perméabilité, ce qui leur permet d’infiltrer le réseau vasculaire. Il est connu que la tyrosine kinase Src contrôle cette modulation de la perméabilité. Puisque notre laboratoire a préalablement démontré que la phosphatase de type récepteur (PTP) DEP-1 est impliquée dans l’activation de Src en réponse au VEGF, nous avons émis l'hypothèse que DEP-1 pourrait aussi jouer un rôle dans la perméabilité des cellules endothéliales. Grâce à des expériences de transfections d’ARN interférant, nous démontrons que DEP-1 est important pour la régulation de la phosphorylation de la VE-Cadhérine, un médiateur critique de la perméabilité. L’impact de DEP-1 sur la dissociation de jonctions intercellulaires est également démontré par microscopie à immunofluorescence de cellules endothéliales. DEP-1 est également nécessaire à l’augmentation de la perméabilité induite par VEGF in vitro. Deux résidus tyrosine retrouvés dans la queue carboxy-terminale de DEP-1 sont essentiels à l’activation de Src en réponse au VEGF. Suite à la transfection d’un plasmide encodant DEP-1 muté pour ces deux résidus, nous démontrons aussi leur implication dans la régulation de la perméabilité in vitro par DEP-1. Ces travaux permettent ainsi d’approfondir nos connaissances sur un nouveau régulateur potentiel de la perméabilité vasculaire. / Endothelial cell permeability is a crucial step of angiogenesis. The main actors behind permeability are endothelial cells who accomplish this in response to permeabilizing factors, most notably Vascular Endothelial Growth Factor (VEGF). In a pathological context, migrating tumor cells produce great quantities of VEGF that stimulate an increase of vascular permeability, which allows them to intravasate into the vasculature. Src has been shown to mediate this process. Our laboratory has previously shown that the protein tyrosine phosphatase DEP-1 is involved in the regulation of VEGF-dependant activation of Src. These data thus suggested that DEP-1 might play a role in endothelial cell permeability. Here, we show through siRNA experiments that DEP-1 is important for the regulatory phosphorylation of VE-Cadherin which is critical for the induction of permeability. The impact of DEP-1 on intercellular junction dissociation is also demonstrated through immunofluorescence microscopy of endothelial cells. We further show that DEP-1 is absolutely required for the VEGF-dependent increase of permeability as illustrated by in vitro permeability assay on siRNA-transfected endothelial cells. Finally, we show that tyrosine residues in DEP-1’s carboxy-terminal tail, which are crucial for mediating Src activity in response to VEGF, are implicated in VEGF-dependant increase in permeability by transfecting plasmids coding for DEP-1 mutants of these tyrosine residues. These findings shed light on a novel potential key regulator of in vivo permeability.

Intimal hyperplasia (IH) is defined as the abnormal migration and proliferation of smooth muscle cells with associated deposition of extracellular matrix in the intimal layer. It is a natural response to endovascular injury induced by procedures such as angioplasty, stent implantation, or atherectomy. Research on the molecular pathways and mediators has led to the discovery of a variety of substances aimed to interrupt or attenuate IH. Heparin, low-molecular-weight heparin, aspirin, corticoids, angiotensin-converting enzyme inhibitors, cyclosporin, vascular endothelial growth factor (VEGF) and other agents have been investigated. However, none of these agents has been used with marked success at the clinical level. Therapeutic angiogenesis studies have demonstrated the potential of heparin binding angiogenic growth factors such as VEGF and basic fibroblast growth factor (bFGF/FGF-2) to treat ischemic heart diseases. Studies on endothelial nitric oxide synthase (eNOS) gene transfer in vivo showed attenuated IH caused by constitutive generation of nitric oxide (NO) via the NOS pathway. FGF-2 increased VEGF mRNA levels in single-cultures of rabbit smooth muscle cells (SMC) and also promoted NO production from endothelial cells (EC). Therefore, we hypothesize that FGF-2 mediates SMC inhibition through the NOS pathway. In order to elucidate the influence of these growth factors, we employed an appropriate SMC-EC co-culture system.Studies on SMC-EC interactions have been established in various in vitro co-culture systems. However, there exist only few co-culture systems in which the structure of a vessel wall is imitated. A direct co-culture model was used in this study to determine the effect of growth factors on the SMC and EC proliferation. In the following study we investigate the effects of VEGF, FGF-2 and FGF-2+VEGF on porcine aorta smooth muscle and endothelial cells. Addition of the higher concentrations (>10 ng/ml) of FGF-2 to the SMC-EC direct co-culture greatly reduced smooth muscle cell numbers and cell cycle S-phase, as judged by propidium iodide DNA analysis using flow cytometry. We also observed that coadministration of FGF-2 with VEGF did not show any difference on SMC proliferation compared to control. These data demonstrate the potent regulatory capabilities of FGF-2 on smooth muscle cell inhibition. Nitric oxide which is generated by the enzyme NOS is hindered by the addition of NOS inhibitor, NG-Methyl-L-arginine acetate (L-NMMA). Utilizing L-NMMA we found that FGF-2 mediated smooth muscle cell inhibition does not follow the NOS pathway. This study is intended to understand the interactions of combination of therapeutic growth factors on vascular cells. The current study is a first step towards an overall goal of setting up an in vivo porcine model for clinical treatment of IH using FGF-2.

Thèse numérisée par la Direction des bibliothèques de l'Université de Montréal.

Thèse numérisée par la Direction des bibliothèques de l'Université de Montréal.

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

Thèse numérisée par la Direction des bibliothèques de l'Université de Montréal.

Ce travail a analysé l’expression du facteur de croissance angiogénique, le VEGF-A, et de son récepteur VEGFR-1, dans le cancer prostatique localisé. Dans la première partie, nous avons mesuré le taux plasmatique de VEGF-A chez 100 patients opérés d’un cancer prostatique localisé. Nous avons également mesuré l’expression tissulaire du VEGF-A sur les pièces opératoires. Il n’y avait pas d’association entre le VEGF-A plasmatique et les facteurs pronostiques du cancer prostatique. Cependant, l’expression du VEGF-A était corrélée au score de Gleason (p=0,01). Dans la deuxième partie, la même cohorte de patients a été utilisée. L’expression tissulaire du VEGFR-1 a également été mesurée. Les patients ont été suivis avec des dosages réguliers du PSA. Durant le suivi, 14 patients ont eu une récidive biologique. Ni le taux plasmatique de VEGF-A (p=0,25), ni l’expression tissulaire du VEGF-A (p=0,38), ni l’expression tissulaire du VEGFR-1 (p=0,34) n’étaient associés au risque de récidive biologique. Dans la troisième partie, nous avons mesuré l’expression tissulaire du VEGF-A et du VEGFR-1 sur les pièces opératoires de 40 patients opérés d’un cancer prostatique localisé. L’expression tissulaire du VEGF-A était significativement plus importante chez les patients ayant eu une progression tumorale après l’intervention que chez les patients n’ayant pas récidivé (p=0,046). En revanche, celle du VEGFR-1 était identique dans les deux groupes. L’expression tissulaire du VEGF-A était le facteur prédictif de progression tumorale le plus significatif. / This study analysed the expression of angiogenic growth factor, VEGF-A and its receptor, VEGFR-1 in localized prostate cancer. In the first part, we measured the plasma levels of VEGF-A in 100 patients operated with radical prostatectomy for clinically localized prostate cancer. We also measured the tissue expression of VEGF-A using ELISA on the surgical specimen. There were no associations between plasma levels of VEGF-A and the usual prognostic factors of prostate cancer. However, the tissue expression of VEGF-A correlated with Gleason score (P = 0.01). In the second part, we used the same patients group. Patients were prospectively followed with regular PSA determinations.14 patients had a biochemical recurrence. Neither plasma level of VEGF-A (P =0.25) nor tissue expression of VEGF–A (P=0.38) and its receptor VEGFR–1 (p=0,34) were associated with the risk of biochemical recurrence after radical prostatectomy. Finally, we measured the tissue expression of VEGF-A and VEGFR-1 on the surgical specimens of 40 patients who underwent radical prostatectomy for clinically localized prostate cancer. The tissue expression of VEGF-A in patients who experienced progression was significantly higher than in those who remained free of recurrence (P=0.046). However, the expression of VEGFR-1 was similar in both groups. In logistic analysis, the expression of VEGF-A was the most significant predictor of tumor progression. These results suggest that the tissue expression of VEGF-A has a prognostic impact in clinically localized prostate cancer.

ABSTRACT Vascular permeability and angiogenesis are crucial events in the rodent and human uterus in early pregnancy and are regulated by vascular endothelial growth factor (VEGF) and prostaglandins liberated from arachidonic acid by cyclooxygenase 2 (COX2). These events coincide with the typical morphological features of the receptive uterus and are regulated by synchronized release of ovarian hormones (oestrogen and progesterone). However, administration of follicle stimulating hormone (FSH) and human chorionic gonadotropin (hCG), commonly used in assisted reproduction, affect the synchrony of the hormonal milieu, particularly by increasing oestrogen levels. This causes detrimental changes to the uterine morphology and affects vascular permeability at the site of implantation. In the present study, the expression of COX2 and VEGF was compared between control and hyperstimulated rat uteri during the peri-implantation period using immunohistochemistry and Western blot analysis. While in control pregnant rats COX2 and VEGF immunolocalization occurred in the luminal epithelial cells and stroma on consecutive days, strong immunolocalization of COX2 and VEGF occurred in the luminal epithelial cells but was inhibited in the stroma of the hyperstimulated rats. This appears to have resulted in the suppression of stromal decidualization and vascular permeability. Western blot analysis did not show any results. This may be due to low concentrations of the protein in the sample. Since vascular permeability and angiogenesis are critical to the process of implantation and are influenced by VEGF and COX2, disturbance of the pattern of these two proteins by hyperstimulation may contribute to the low implantation rate in IVF programes. immunohistochemistry and Western blot analysis. While in control pregnant rats COX2 and VEGF immunolocalization occurred in the luminal epithelial cells and stroma on consecutive days, strong immunolocalization of COX2 and VEGF occurred in the luminal epithelial cells but was inhibited in the stroma of the hyperstimulated rats. This appears to have resulted in the suppression of stromal decidualization and vascular permeability. Western blot analysis did not show any results. This may be due to low concentrations of the protein in the sample. Since vascular permeability and angiogenesis are critical to the process of implantation and are influenced by VEGF and COX2, disturbance of the pattern of these two proteins by hyperstimulation may contribute to the low implantation rate in IVF programes.

Le facteur vasculaire de la croissance endothélial A (VEGF-A) joue un rôle essentiel dans un grand nombre de processus physiologiques et pathologiques incluant les maladies cardiovasculaires, inflammatoires et cancéreuses. Ainsi, au cours de cette thèse, nous avons développé une approche intégrée et de génomique fonctionnelle sur le rôle du VEGF-A de manière à mieux rendre compte de la diversité des processus dans lesquels il est impliqué en ajoutant dans le panel des investigations des phénotypes autour des hormones thyroïdiennes, acteurs essentiels de la physiopathologie humaine. L’objectif final de cette thèse était d’identifier des polymorphismes impliqués dans la variation plasmatique du VEGF-A et des produits de son expression protéique et génique utiles en médecine personnalisée. Pour ce faire des études sur des phénotypes intermédiaires comme les molécules d’adhésion et d’inflammation, le syndrome métabolique et les lipides communs aux pathologies précédemment citées ont été réalisées dans la cohorte STANISLAS, constituée de sujets sains, ainsi que dans d’autres populations de réplication de l’équipe. Des études dans une population pathologique, visant à explorer les liens avec les hormones thyroïdiennes et les thyroïdites ont été réalisées avec une population Tunisienne, ‘cas- témoins’, les patients étant traités pendant 6 mois pour une maladie de Graves ou de Hashimoto. .Nos résultats principaux sont au final les suivants : 1. Association de l'isoforme VEGF-A145 avec les ARNm d'ICAM-1, de sélectine L et de TNF-α. 2. Association entre les taux plasmatiques de VEGF-A et ceux d’ICAM-1 et de sélectine E. 3. Interactions épistatiques entre les variants du VEGF-A pour les taux de sélectine E, de TNF-α, d'ICAM-1 et d'IL-6. 4. Association significative entre le rs4416670 et les niveaux de l'ARNm de la sélectine-L. 5. Association significative entre le rs10738760 et le risque de syndrome métabolique. 6. Association entre rs6921438 et HDL et LDL. 7. Interaction entre rs4416670 et hypertension pour la variation d'apolipoprotéine E. 8. Corrélation positive entre taux plasmatiques de VEGF-A et de FT4 chez les sujets contrôles. 9. Association négative entre FT4 plasmatique et VEGF-A145 dans la population totale des patients avec thyroïdite et chez les contrôles. 10. Niveaux de VEGF-A165 plus élevés chez les patients traités pour une maladie de Graves par rapport aux sujets contrôles. 11. Niveaux de VEGF-A165b plus élevés chez les patients traités pour une maladie de Hashimoto par rapport à ceux traités pour une maladie de Graves. 12. Niveaux de VEGF-A165b plus élevés chez tous les patients en hypothyroïdie après 6 mois de traitement par rapport à ceux en euthyroïdie. 13. Niveaux de VEGF-A189 plus faibles chez tous les patients avec thyroïdite en comparaison avec les sujets contrôles. 14. Association entre les niveaux de VEGF-A145 et le statut thyroïdien uniquement chez les patients traités pour une maladie de Hashimoto. 15. Association entre la positivité des anti-TPO et les taux de VEGF-A165 dans la population totale des patients avec thyroïdite et chez les contrôles. 16. Association entre l’allèle A du rs10738760 et un sur-risque de maladie de Graves 17. Association du rs692148 (allèle A) avec un sur-risque de maladie de Graves. 18. Association de l’allèle T du rs4416770 avec un risque accru d’hyperthyroïdie indépendamment du diagnostic initial et du traitement des patients. 19. Association de l’allèle A du rs10738760 avec des taux circulants de FT3 dans la population totale des patients avec thyroïdite et contrôles. Cette thèse montre à travers ces nombreux résultats de médecine stratifiée l’implication centrale de la génomique du VEGF-A dans la régulation de processus physiologiques et pathologiques. La mise en évidence de son rôle dans la régulation des hormones thyroïdiennes pourrait qualifier le VEGF comme un nouveau dénominateur commun des maladies cardiovasculaires, inflammatoires et [...] / The vascular endothelial growth factor (VEGF-A) plays a key role in a large variety of physiological and pathological processes, including cardiovascular and inflammatory diseases and cancer. Consequently, during this thesis, we have developed a functional genomics and integrative approach of the VEGF-A roles in order to better describe the diversity of the processes in which it is implicated by adding in the investigation panel phenotypes the thyroid hormones, which are essential in the human physiopathology. The final objective of this work was to identify the VEGF-A plasma variability linked polymorphisms and the VEGF-A expression products useful for personalized medicine. Therefore, studies on common among diseases intermediate phenotypes such as adhesion, and inflammation molecules, metabolic syndrome and lipids have been conducted on the STANISLAS cohort (healthy volunteers), as well as on other replication-directed populations of the team. Studies in a pathological population aiming to investigate links with thyroid hormones and thyroiditis have been conduced in a Tunisian “case-control” population with cases treated for thyroiditis during 6 months. Here are our main results: 1. Association between VEGF-A145 and ICAM-1, L-selectin and TNF-α mRNA levels. 2. Association between plasma levels of VEGF-A and ICAM-1 and E-selectin. 3. Epistasis interactions between VEGF-A variants for E-selectin, TNF-α, ICAM-1 and IL-6 plasma levels. 4. Association between rs4416670 and L-selectin mRNA levels. 5. Association between the rs10738760 and the risk for metabolic syndrome. 6. Association between rs6921438 and HDL and LDL plasma levels. 7. Interaction between rs4416670 and hypertension for plasma variation of apolipoprotein E. 8. Positive correlation between plasma levels of VEGF-A and FT4 in Tunisian controls. 9. Negative association between plasma FT4 levels and VEGF-A145 levels in the entire Tunisian population (patients and controls). 10. Increase of VEGF-A165 levels in Grave’s patients in comparison with controls. 11. Increase of VEGF-A165b levels in Hashimoto patients in comparison with Graves patients. 12. Increase of VEGF-A165b levels in hypothyroid patients in comparison with hyperthyroid patients after 6 months of treatment. 13. Decrease of VEGF-A189 levels in all after 6 months of treatment thyroiditis patients in comparison with controls. 14. Association between VEGF-A145 and thyroid status, only in Hashimoto patients. 15. Association between anti-TPO antibodies positivity and VEGF-A165 in the entire Tunisian population (patients and controls). 16. Association between the A allele of rs10738760 and an increased risk of Graves’ disease. 17. Association of the A allele of rs692148 with an increased risk for Grave’s disease. 18. Association of the T allele of rs4416770 with a higher risk of hyperthyroidism regardless of the initial diagnosis and treatment of patients. 19. Association of A allele of rs10738760 with higher FT3 plasma levels in the entire Tunisian population (patients and controls). This thesis shows, through an important number of results of stratified medicine, the central implication of VEGF-A genomics in the regulation of physiological and pathological processes. The identification of its role in the regulation of thyroid hormones could highlight the role of VEGF-A as a new common denominator of cardiovascular, inflammatory and oncological diseases that has to be clinically validated