L'angiogénèse est un phénomène vital dans de nombreux processus physiologiques tels que le développement embryonnaire ou la cicatrisation. Mais c'est aussi un processus pathologique impliqué dans la croissance des tumeurs malignes, dans le développement des métastases et dans des maladies telles que l'endométriose, la dégénérescence Maculaire liée à l'âge (DMLA) et l'arthrose. La néovascularisation pathologique est dense et désordonnée contrairement à la néovascularisation physiologique, ce qui en fait une cible thérapeutique attrayante. Les agents perturbateurs vasculaires (VDA) sont une nouvelle classe de molécules bioactives qui détruisent sélectivement la néovascularisation tumorale et arrêtent l'apport sanguin aux tumeurs solides, provoquant une nécrose tumorale étendue. Parmi les VDA, les agents se liant à la tubuline sont principalement représentés. Le processus d'angiogénèse au sein de l'arthrose ainsi que le mode d'action des VDA nous amène a envisager l'utilisation de ses derniers en tant que nouveau traitement anti-arthrosique. Dans la recherche de nouveaux scaffolds VDA, nous avons d'abord comparé les activités biologiques de plusieurs VDA. Une classe en particulier a attiré notre attention sur laquelle seules quelques études avaient été menées. Cela nous a permis de sélectionner des pistes avec un espace chimique exploitable pour concevoir des scaffolds originaux. Cette thèse présentera les analogues qui ont été synthétisés, leur viabilité cellulaire sur les HUVEC (cellules endothéliales de la veine ombilical humaine) et leur activité sur l'inhibition de la polymérisation de la tubuline (ITP). / Angiogenesis is a vital phenomenon in many physiological processes such as embryonic development or wound healing. But it is also a pathological process involved in the growth of malignant tumors, in the development of metastases and in diseases such as endometriosis or age-related macular degeneration (AMD) and osteoarthritis. Pathological neovasculature is dense and disordered in contrast to physiological neovasculature, which makes it an attractive therapeutic target. Vascular Disrupting Agents (VDA) are a new class of bioactive molecules that selectively destroy tumour neovasculature and shut down blood supply to solid tumours, causing extensive tumour cell necrosis.1 Among small molecule VDAs, tubulin-binding agents are mostly represented. The process of angiogenesis within osteoarthritis as well as the mode of action of VDA leads us to consider the use of its last as a new anti-osteoarthritis treatment. In the search for novel VDA scaffolds, we first compared the biological activities of several VDA. One class in particular attracted our attention on which only a few studies had been conducted. This allowed us to select leads with exploitable chemical space to design original scaffolds. These thesis will present analogs which were synthesized, their cell viability on HUVEC (human umbilical vein endothelial cells) and their activity on tubulin polymerization inhibition (ITP).

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

La coronaropathie est une cause majeure de mortalité, motivant la recherche de stratégies limitant le remodelage cardiaque ou stimulant la néovascularisation après un infarctus du myocarde (IDM). Ce travail vise à étudier chez la souris adulte le rôle, sur la fonction cardiaque, de gènes régulant l'angiogenèse et le métabolisme cellulaire en modulant leur expression endothéliale en conditions basales ou en post-IDM (après ligature coronaire) : PPARβ/δ, Wt1, Cyp51 et Dnmt2. Les paramètres échocardiographiques ont été mesurés pré et post-IDM, des analyses histochimiques réalisées, et l’expression de gènes cibles comparée selon le génotype. La surexpression de PPARβ/δ stimule l'angiogenèse basale, causant une hypertrophie ventriculaire gauche (VG). En post-IDM, elle induit un remodelage VG pathologique et majore la taille de l'IDM, posant la question des interactions entre endothélium et cardiomyocytes. En post-IDM, l'invalidation de Wt1 limite l'angiogenèse coronaire, majore le remodelage VG et la taille de l'IDM. A l'état basal, l'invalidation de Cyp51 est à l'origine d'une insuffisance cardiaque dilatée, via une perméabilité vasculaire accrue et une activation endothéliale. La modification de la composition membranaire en stérols peut expliquer la dysfonction de l'endothélium, modifiant ses interactions avec les cardiomyocytes. Ainsi, Cyp51 possède un rôle essentiel dans la structure et la fonction cardiaque, ouvrant le champ de son étude en post-IDM. Enfin, l'expression de Dnmt2 est indispensable pour limiter l'hypertrophie cardiaque, via le contrôle de l'activité de l'ARN polymérase II par la méthylation de l’ARN non codant Rn7sk. / Coronary heart disease is a major cause of mortality, explaining the increasing interest in therapeutics targeting cardiac remodeling and neovascularization after myocardial infarction (MI). Using endothelial expression modulation in adult mice in basal or post-MI conditions (after coronary artery ligation), this work studied several genes involved in angiogenesis and cardiac metabolism, PPARβ/δ, Wt1, Cyp51 and Dnmt2, and their role in cardiac function. Echocardiographic structural and functional parameters were measured before and after MI, histochemistry analyses performed, and target genes expression compared between different genotypes. PPARβ/δ basal overexpression resulted in an increased angiogenesis and cardiac hypertrophy. After MI, it caused MI expansion through increased cardiac remodelling. This discrepancy raises the issue of communication between endothelial cells and cardiomyocytes. Endothelial Wt1 expression is essential for cardiac repair after MI : deletion was responsible for neovascularization impairment, poorer cardiac remodeling and MI enlargement. Endothelial Cyp51 expression is necessary for basal cardiac structure and function. After Cyp51 deletion, membrane and cell junction disorganization caused increased vascular permeability and endothelium activation, resulting in dilated cardiomyopathy. The accumulation of toxic oxysterols or lack of cholesterol might account for endothelial dysfunction, through abnormal endothelial cells to cardiomyocytes signalling. Dnmt2 deletion caused cardiac hypertrophy. through methylation of non-coding RNA Rn7sk and control of RNA polymerase II activity.

Tableau d’honneur de la Faculté des études supérieures et postdoctorales, 2007-2008. / Les nerfs et les vaisseaux sanguins sont souvent associés. De plus, il a été démontré que les nerfs périphériques peuvent induire les vaisseaux sanguins cutanés à se différencier dans un phénotype artériel. Nous avons développé par génie tissulaire le premier modèle d'étude tridimensionnel permettant l'analyse in vitro de la formation de pseudocapillaires (capillary-like tubule; CLT) sur un réseau nerveux préétabli. Les cellules neurales ont induit une augmentation de 27 % du nombre de pseudocapillaires formé dans les tissus reconstruits. Cet effet était aboli par l'utilisation de K252a, un inhibiteur de TrkA, B, et C, les récepteurs respectifs du NGF, du BDNF et du NT-3. De plus, nous avons démontré qu'en ajoutant 10 ng/ml de NGF, 0.1 ng/ml de BDNF, 15 ng/ml de NT-3 et 50 ng/ml of GDNF en absence de cellules neurales, une augmentation majeure de 40 à 80% du nombre de CLT était induite. Collectivement, nos résultats suggèrent que les cellules neurales peuvent induire l'angiogenèse en sécrétant des facteurs neurotrophiques. De plus, nous démontrons pour la première fois que le NT-3 et le GDNF peuvent induire l'angiogenèse in vitro.

No description available.

A. Background<p><p>Skin wound healing (WH) is a dynamic and extremely determinate process of cellular, humoral and molecular mechanisms which begins directly after wounding and can last for years. WH is described as is an intricate process in which the skin (or another organ-tissue) repairs itself after injury. The process of skin WH occurs through the actions of an interplay of cells, growth factors and cytokines leading to wound closure.<p><p>WH occurs in three precisely and highly programmed phases: the inflammatory phase (day 0 to day 7) followed by the proliferative phase or vasculogenic phase (day 7 to day 21) and finally the remodeling phase (2 days - up to 2 years). For a successful healing, all three phases must occur in the proper sequence and time frame.<p><p>Many factors can interfere with one or more phases of the WH process, thus causing improper or impaired healing. The proliferation phase, in particular, requires the participation of various cells types such as fibroblasts, endothelial cells (ECs) and endothelial progenitor cells (EPCs), to produce a healthy well-vascularized granulation tissue for epithelization and wound closure.<p><p>A.1 Wound Healing And Obesity<p><p>In 2008, over 1.4 billion adults, 20 and older, were overweight. Of these, obesity has been shown to affect over 500 million people (OMS website). Moreover, the prevalence of obesity continues to rise, and by 2018, it is estimated that obesity will cost $ 347 billion annually.<p><p>Each year, in the US, approximately 33 million overweight and obese patients undergo surgery. Obesity causes a number of known health problems and increased post-surgical complications such as wound infection, dehiscence, hematoma and seroma. Surgeons anecdotally report WH complications among obese patients; however, little research has been conducted to investigate the mechanisms mediating impaired obesity-related WH. <p><p>Some previous work on diabetic patients and diabetic mice showed an imbalance between pro-oxydant and anti-oxydant genes as well as impaired EPCs proliferation and tube formation during the WH process. More then a hundred cytologic factors have been found to impair WH in the type 2 diabetic patient. It is a very complex and multifactorial problem involving decreased growth factors secretion, impaired keratinocyte and fibroblast functions, impaired EPs function, alteration of the macrophage function and granulation tissue synthesis, etc. <p><p>Based on these findings and because obesity is associated with the development of type 2 diabetes, we hypothetize that, impaired balance between pro-apoptotic/anti-apoptotic and pro- oxydant /anti-oxydant genes is involved in impaired WH. Furthermore, we hypothetize that impaired EPCs function leads to the perturbation of the proliferation phase of obesity impaired WH.<p><p>A.2. Wound Healing and Age<p><p>The world population is aging; by 2030, nearly 20% of Americans, (± 72 million people), will be 65 years old and older. In 2010, 17% of the European population was over the age of 65. By 2060, it is projected that the share of those aged 65 and over will rise to 30%, accounting for more then 150 million people. (ec.europa.eu) These aging subjects undergo an increasing number of surgical procedures: in the past two decades, the percentage of surgeries in patients over 65 has doubled to nearly 40%.<p>As a corollary, it is well established knowledge that elderly WH is impaired. However, little is known about the underlying mechanisms of age-related impaired WH.<p><p>As previously mentioned, adult BM-derived EPCs contribute to peripheral tissue repair and regeneration. In light of the abundant literature suggesting that neovascularization is impaired in the elderly, we characterize a novel model of senile cutaneous WH and investigate the role that vasculogenesis plays in the pathogenesis of age related impaired WH.<p>Aged mice colonies have traditionally been the model for aged small mammalian research, however, the ability to use a readily-available transgenic mouse model with features of accelerated aging would aid in the exploration of targeted therapies and a great number of age-related investigations.<p><p>We hypothesize that the Hutchinson-Gilford Progeria Syndrome (HGPS) Zmpste24 deficient (Zmpste24-/-) mouse mimics physiological ageing and can be used as a novel model for the study of senescent WH. We further hypothetized that impaired balance between pro-apoptotic/anti-apoptotic and pro-oxydant /anti-oxydant genes as well as impaired EPCs function are responsible for the impairment of the proliferative phase, leading to overall impaired WH.<p><p>A.3 Aims<p><p>Recently, a great deal of research has been directed at understanding the critical factors inducing poorly healing wounds. However, a lot remains unclear.<p><p>It is now well accepted that new blood vessel formation occurs not only by angiogenesis (blood vessels formation from a preexisting network of capillaries), but also by vasculogenesis (blood vessels formation from BM SCs recruitment) and that EPCs contribute to as much as 25% of new blood vessels formed in healing tissues4. They are mobilized from the BM in response to injury and production of local cytokines, are incorporate into wounds and play an integral role in systemic tissue repair. <p><p>Based on this finding, we hypothesized that co-morbidities related impaired WH may be due, in part, to decreased EPCs number, migration/homing, and/or function resulting in impaired vasculogenesis. Because age and/or obesity have been shown to be one of the most common predictors of altered WH, we decided to focus on these two parameters.<p><p>Following a bedside to bench approach the purpose of this work was to 1) develop coherent and translatable models of co-morbidity digging in the physiologic/pathologic mechanisms underlying altered healing in obese and senile mice; 2) develop targeted therapeutics to improve impaired WH.<p><p>B. Material and Methods<p><p>B.1 Human Model<p><p>Since obesity impairs WH and BM EPCs are important for tissue repair, we hypothesize that obesity- impaired WH is due, in part, to impaired EPCs mobilization, trafficking, and function. Peripheral blood was obtained from non diabetic, obese (BMI > 30, n = 25), and non obese (BMI < 30, n = 17) subjects. Peripheral blood human EPCs were isolated, quantified, and functionally assessed.<p>As for aged impaired WH, EPCs of aged subjects have already been found to have decreased adhesion, migration and proliferative properties as well as being decreased in number in elderly patients undergoing surgery compared to younger patients.<p><p>B.2. Mice Models<p><p>Two models of WH were developed and characterized.<p>In order to isolate the effect of obesity on EPCs and WH, OB non-diabetic female TallyHo/JngJ mouse were selected (Female mice don’t express hyperglycemia and hyperinsulinemia). Female SWR/J non-OB mice were used as control mice. In order to limit variables, TallyHO/JngJ obese mice were selected over other OB mice that exhibit a polygenic type of obesity (Jackson Laboratory Website). By selecting this mouse model, we have excluded in our selection of the ideal model common confounding factors such as hyperglycemia, hyperinsulinemia, immune disorders.<p><p>Zmpste24 is a metalloproteinase involved in the maturation of lamin A (LmnA), an essential component of the nuclear envelope. When Zmpste24 or LmnA are knocked-out, mice exhibit profound nuclear architectural abnormalities and histopathological defects that phenocopy an accelerated aging process. Of crucial importance, the lamin-A dependent nuclear alterations seen in Zmpste24-deficient mice have also been found in human physiological aging. We defined the utilization of the Hutchinson-Gilford Progeria Syndrome (HGPS) Zmpste24 deficient (Zmpste24- /-) mouse as a novel model for the study of senescent WH (controls used were C57BL/6J mice).<p><p>B.3. Wounding Model and Data Collection<p><p>All mice group underwent wounding using a stented wound model developed in our laboratory and previously published. Briefly, paired 6-mm circular, full-thickness wounds extending through the panniculus carnosus were made on the dorsal skin of the mouse. An O-ring, 12-mm splint made of silicone sheeting was then sutured to the skin around the wound. To minimize wound contraction and reliably recapitulated the granulation and re-epithelialization seen in human WH by secondary intention. Time to wound closure was measured using standardized digital photographs taken on days 0, 7, 14, and 21. Wound closure was calculated as a percentage of the original wound.<p><p>For each model, EPCs were harvested, quantified by flow-cytometry and their function tested. Wounds were harvested at various time points and RNA, DNA and protein analysis were conducted. Finally immunohistochemistry to assess epidermal thickness, vascularity and WH were also realized.<p><p>In a second step, after characterization of the models, local (using targeted siRNA gel) and systemic therapies (using AMD3100, a PC mobilizer) were applied on the wounds and compared to controls. WH was monitored. We conducted the previously mentioned analysis (RT-PCR, ELISA and DNA analysis) on the harvested samples.<p><p>All values are expressed as a mean ± standard error of mean (SEM). The number of mice per treatment group was determined using G*Power (G*Power©, Melbourne, Australia) to provide a power greater than 0.80. Student T test was realized to compare two groups among each other.<p><p>C. Results<p><p>C.1. Human EPCs Have Impaired Function<p><p>There was no difference in the number of baseline circulating human EPCs in non-diabetic OB and non-OB<p>subjects, but EPCs from OB subjects had impaired adhesion (p<0.05), migration (p<0.01), and proliferation (p<0.001).<p><p>C.2. Obesity and Wound Healing<p><p>TallyHo/JgnJ OB mice demonstrated significantly impaired healing when compared to SWR/J control mice. They healed at an average of 28 ± 2 days (p<0.05). Post-wounding circulating EPCs were quantified and wounds were analyzed. Circulating EPCs recruitment is impaired in wounded TallyHo/JngJ mice and their wounds shown significantly decreased new blood vessel formation through decreased HIF-1α/SDF-1α signaling (p<0.05). Their wounds are characterized by increased apoptosis, increased DNA damage and impaired pro-/anti-oxydant balance. Immunonistochemistry and histology showed decreased vascular vessels in TallyHo/JngJ wounds and thinner epidermal thickness.<p><p>In the local treatment phase, local p53 silencing consistently improved WH to a nearly normal healing time (wounds healed in 18 ± 2 days, p<0.05). sip53 treatment showed a significant decrease in pro-apoptotic markers (p53, Bax, PUMA p<0.05) and a significant increase in angiogenic markers (VEGF, SDF-1α, HIF-1α) with increased blood vessel formation and decreased DNA damage.<p><p>C.3. Age and Wound Healing<p><p>In these experiments, we show that not only is Zmpste24-/- WH impaired when compared to C57BL/6J mice (Zmpste24-/- mice healed at average 40 days ± 2 days p<0.05) at all time points but that they also showed decreased vascularity and proliferation in the wound bed (p<0.05).<p><p>Histological analysis was performed utilizing hematoxylin and eosin staining to assess epidermal thickness, CD31 immunofluorescence to assess vascular density, p53 and caspase 3 to assess apoptosis, 8’OHdG staining to assess DNA damage and PCNA to assess proliferation. Epidermal thickness was significantly decreased in Zmpste24-/- animals compared to WT as well as vascular density, and proliferation in Zmpste24-/- wound tissue (p<0.05). <p><p>Circulating vasculogenic EPCs recruitment was impaired in Zmpste24-/- mice and their wounds showed significantly decreased new blood vessel formation through decreased HIF-1α/SDF-1α signaling (p<0.05). Zmpste24-/- wounds are characterized by increased apoptosis and an abnormal rise in ROS.<p>In the treatment phase, local p53 silencing consistently improved healing by more then a two fold (18 ± 2 days). VEGF production was significantly increased and pro-apoptotic factors were significantly downregulated in siRNA-treated Zmpste24-/- mice (p<0.05). DNA damage due to ROS production was also shown to be significantly decreased following treatment. Our results suggest a vasculogenic dysfunction in wound closure and showed that the specific knock down of p53 significantly improves WH.<p><p>Because EPCs showed impaired function, lower peripheric blood counts and impaired SDF-1α/HIF-1α signaling, we hypothesized that improving their mobilization by using a progenitor cell mobilizer, AMD3100, known to mobilize SCs from the BM, in a systemic treatment phase will improve WH. Peripheral blood counts were significantly increased and time to wound closure significantly decreased (20 days ± 2, p<0.05). Vasculogenic markers and anti- apoptotic molecules were upregulated compare to non-treated animals.<p><p>D. Conclusions<p><p>Obesity impaired wound closure is a complex problem with many contributory factors. Our results suggest that obesity impairs the BM-derived EPCs response to peripheral injury and this, in turn, impairs wound closure. This impairment is associated with decreased new blood vessel formation and increased DNA damage leading to an increase in the p53 pathway. We also demonstrate that targeted siRNA therapy can partially rescue impaired WH due to obesity. Based on these results we support the encouraging argument that, WH and closure has the potential be improved through specific local and systemic therapies in vivo in our rodent model and that further studies are needed to support this in a clinical environment.<p><p>Impaired WH due to ageing is a complex phenomenon that is partially understood. We demonstrate that the Zmpste24-/- transgenic knockout mouse provides a model for age-related WH investigation. Zmpste24-/- animals heals their wounds with significant delays, showed impaired EPCs mobilization following wounding through an impaired HIF-1α/SDF-1α pathway and increased apoptosis. Furthermore, WH can be improved through specific local siRNA therapy and systemic stem cell mobilization therapies.<p><p>Our results suggest strong similar patterns between obesity and ageing in the way they mediate WH impairments trough (premature) ageing. Our encouraging endeavor to bring WH back to baseline in these diseased models underlines the possibility to reverse the microenvironment alterations and improves EPCs contribution to the WH process. Because EPCs are involved in virtually every tissue repair process happening in the human body, we hope that this work will lead the way for new research in various fields in medicine to improve wound care and quality of life of patients. / Doctorat en Sciences biomédicales et pharmaceutiques / info:eu-repo/semantics/nonPublished

Notre travail a permis tout d’abord d’investiguer les effets des nucléotides extracellulaires sur les cellules<p>dendritiques (DCs) qui sont des cellules présentatrices d’antigènes capables d’initier et de réguler la<p>réponse immunitaire. Afin d’avoir une vue globale de l’action des nucléotides extracellulaires sur les DCs,<p>un profil d’expression génique de l’ATPgS – dérivé stable de l’ATP - a été réalisé par microarray dans les<p>cellules dendritiques dérivées de monocytes (MoDCs).<p>Notre groupe a préalablement montré que malgré que l’ATP est considéré comme un signal de danger, il<p>confère des propriétés immunosuppressives aux DCs (Marteau et al, 2005). Nous nous sommes focalisés<p>sur des régulations géniques pouvant être mises en relation avec un action anti-inflammatoire de l’ATP.<p>Nous avons ainsi démontré que l’ATP était capable d’inhiber la sécrétion des chimiokines MCP-1 et MIP-<p>1a initiée par l’action du LPS, ce qui a pour conséquence de diminuer la capacité des DCs à recruter des<p>monocytes ou d’autres DCs. Ce travail a fait l’objet d’une publication en tant que premier auteur<p>(Horckmans et al, 2005).<p>Un grand nombre d’autres gènes régulés liés à la réponse immune et à l’inflammation a été identifiée<p>dans le profil microarray de l’ATPgS. Nous avions notamment pu identifier une augmentation de la<p>sécrétion de VEGF-A en réponse à l’ATP, amplifiée en présence de LPS. Cette régulation est extrêmement<p>intéressante au vu de l’action immunosuppressive du VEGF sur les DCs. Par ailleurs, cette régulation<p>pourrait constituer un lien entre les DCs et l’angiogénese. Ce travail a fait l’objet d’une publication en tant<p>que premier co-auteur dans la revue Journal of Immunology (Bles et al, 2007).<p>En conclusion, nos données nous ont ainsi permis de montrer que les nucléotides adényliques peuvent<p>avoir par leur action sur les cellules dendritiques une action anti-inflammatoire voire pro-angiogénique,<p>en inhibant le recrutement leucocytaire et une action immunosuppressive en stimulant la sécrétion de<p>VEGF.<p> / Doctorat en Sciences biomédicales et pharmaceutiques / info:eu-repo/semantics/nonPublished

No description available.

Une des pathologies les plus répandues dans les pays développés est la dégénérescence maculaire liée à l’âge (DMLA). Elle se présente sous différentes formes dont la plus sévère est caractérisée par la présence de formation de néo-vascularisation provenant de la choroïde (CNV) migrant vers la rétine. Malheureusement, le diagnostic peut être seulement posé une fois que les symptômes apparaissent et que la vision est déjà affectée. Récemment, un vif engouement s’est porté sur les microARN (miARN), devenus des molécules d’intérêt pour le développement de biomarqueurs et représentant également un important potentiel thérapeutique. Nous avons donc émis l’hypothèse que l’identification de miRNA dans un contexte de DMLA avec CNV permettrait de caractériser leurs rôles, tant au niveau de l’identification de biomarqueurs que de nouvelles cibles thérapeutiques. Le premier objectif de notre projet était d’identifier chez l’humain une signature de miARN spécifique à la forme humide de la DMLA. Pour ce faire, nous avons réalisé un criblage des miARN dans l’humeur vitrée de patients. Nous avons observé une augmentation des niveaux de miR-146a et une diminution de miR-106b et miR-152 spécifique à la DMLA. Cette signature fut confirmée dans le plasma de cette même cohorte de patients. D’autre part, l’exploration de bases de données AMD Gene Consortium (AGC) et Ingenuity Pathway Analysis (IPA) a démontré une relation entre les miARN détectés et des mutations génétiques associées à la forme DMLA avec CNV. En effet, nous avons identifié un SNP (single-nucleotide polymorphism ou SNP) (rs1063320) dans un site de liaison de miR-152 du gène HLA-G. Le second objectif était d’explorer la possibilité d’utiliser, du point de vue thérapeutique, un des miARN identifiés au 1er objectif. Nous avons, tout d’abord, quantifié l’expression de miR-146a, miR-106b et miR152 dans un des modèles classiques de la DMLA avec CNV chez la souris (brûlures par laser). Le candidat retenu fut miR-106b, puisqu’en plus d’être impliqué dans l’angiogenèse, son expression dans le modèle animal reflétait de plus près celle obtenue chez les patients. Suite à ces résultats, nous avons: exploré A) le mécanisme influençant la diminution de l’expression de miR-106b et B) l’effet de sa surexpression sur l’angiogenèse. D’abord en A), nous avons observé une activation de la voie de PERK dans notre modèle animal de CNV induite par laser menant à une diminution de l’expression de MCM7 et du polycistron miR-106b~25. . Ensuite en B), nous avons pu observer l’effet anti-angiogénique de miR-106b sur la migration des cellules endothéliales in vitro et sur la formation de NV in vivo. L’action anti-angiogénique de miR-106b pourrait ainsi avoir un potentiel thérapeutique important sur la formation de la CNV dans la DMLA. / Age-related macular degeneration (AMD) is a leading cause of blindness worldwide affecting individuals over the age of 60. The neovascular form (NV AMD) is characterized by choroidal neovascularization (CNV) and is responsible for the majority of central vision impairment. Unfortunately, diagnostic of AMD can only be done after symptoms have appeared and loss of visual field occurred. Recently, there is a growing interest in microRNAs (miRNAs) for their eventual use in the development of new biomarkers or new therapeutic strategies. Our hypothesis is that wet AMD is associated with specific signature of several miRNAs that can be used as biomarkers for the disease. These miRNAs can be harnessed for therapeutic interventions. Our first objective was to identify a specific signature of miRNAs for wet AMD by using non-biased microRNA arrays and individual TaqMan qPCRs. We profiled miRNAs in the vitreous humour and plasma of patients with NV AMD. We identified a disease-associated increase in miR-146a and a decrease in miR-106b and miR-152 in the vitreous humour, which was reproducible in plasma. Moreover, miR-146a/miR-106b ratios discriminated patients with NV AMD with an area under the Receiver Operating Characteristic curve (ROC AUC) of 0,977 in vitreous humour and 0,915 in plasma, suggesting potential for a blood-based diagnostic. Furthermore, using the AGC and IPA database, we mapped a NV AMD-associated single nucleotide polymorphism (SNP) (rs1063320) in a binding site for miR-152-3p in the HLA-G gene. Our second objective was to explore the therapeutic potential of a specific miRNA (identified in objective 1). First, we explored the expression levels of miR-146a, miR-106b and miR-152 in the laser burn mouse model. We demonstrated that levels of miR-106b were significantly decreased in this mouse model of CNV. We divided this objective in two parts: Part A, explore mechanisms causing miR-106b downregulation and part B, study the therapeutic potential of miR-106b on the inhibition of angiogenesis and CNV formation. First, we showed that expression of the miR-106b-25 cluster is negatively regulated by the ER stress pathway of protein kinase RNA-like ER kinase (PERK) and a reduction in levels of MCM7, the host gene of miR-106b. �����Second, we demonstrated that therapeutic delivery of miR-106b to the retina with lentiviral vectors protects against aberrant retinal neovascularization in two distinct mouse models of pathological neo-vascularization. Results from this study suggest that miRNAs, such as miR-106b, have the potential to be used as multitarget therapeutics for conditions characterized by aberrant retinal neovascularization.

No description available.