Tumors are composed not only of malignant cells, but also of various types of normal cells, including vascular cells and infiltrating immune cells, which drive tumor development and progression. The tumor vasculature is abnormal and dysfunctional due to sustained tumor angiogenesis driven by high levels of pro-angiogenic factors. Proteins differentially expressed in tumor vessels affect vascular function and the tumor microenvironment and may serve as targets for therapy. The tumor is also a site of sustained chronic inflammation. The recruitment and activation of inflammatory cells significantly influence tumor progression and regression. Targeting molecules regulating tumor angiogenesis and inflammation in the tumor microenvironment is therefore a promising strategy for the treatment of cancer. This thesis is aiming to understand and investigate the molecular regulation of these two processes in tumors. αB-crystallin is a heat shock protein previously proposed as a target for cancer therapy due to its role in increasing survival of tumor cells and enhancing tumor angiogenesis. In this thesis, we demonstrate a novel role of αB-crystallin in limiting expansion of CD11b+Gr1+ immature myeloid cells in pathological conditions, including tumor development. In addition, we show that αB-crystallin regulates leukocyte recruitment by promoting expression of adhesion molecules ICAM-1, VCAM-1 and E-selectin during TNF-α-induced endothelial activation. Therefore, targeting of αB-crystallin may influence tumor inflammation by regulating immature myeloid cell expansion and leukocyte recruitment. Abnormal, dysfunctional vessels are characteristic of glioblastomas, which are aggressive malignant brain tumors. We have identified the orphan G-protein coupled receptor ELTD1 as highly expressed in glioblastoma vessel and investigated its role in tumor angiogenesis. Interestingly, deficiency of ELTD1 was associated with increased growth of orthotopic GL261 glioma and T241 fibrosarcoma, but did not affect vessel density in any model. Further investigation is warranted to evaluate whether ELTD1 serves a suitable vascular target for glioblastoma treatment. Anti-angiogenic drugs targeting VEGF signaling is widely used in the clinic for various types of cancer. However, the influences of anti-angiogenic treatment on tumor inflammation have not been thoroughly investigated. We demonstrate that VEGF inhibits TNF-α-induced endothelial activation by repressing NF-κB activation and expression of chemokines involved in T-cell recruitment. Sunitinib, a small molecule kinase inhibitor targeting VEGF/VEGFR2 signaling increased expression of chemokines CXCL10, CXCL11, and enhanced T-lymphocyte infiltration into tumors. Our study suggests that anti-angiogenic therapy may improve immunotherapy by enhancing endothelial activation and facilitating immune cell infiltration into tumors.

Chemotactic cytokines, also known as chemokines, direct the migration of leukocytes following their interaction with seven transmembrane domain receptors that are part of the chemokine receptor family (Bachelerie et al. 2014). Chemokines are key players in cancer progression and the regulation of cancer-related inflammation. Atypical chemokine receptors (ACKRs) represent a subset of proteins belonging to the family of chemokine receptors but unable to signal through conventional cascades. ACKRs have recently emerged as important molecular players in health and diseases (Massara et al. 2016). They affect chemokine availability and function and impact many pathophysiological events, including the tumorigenesis process (Sjöberg et al. 2020). Chemerin is a nonchemokine chemoattractant for dendritic cell subsets, macrophages, and natural killer cells (Valérie Wittamer et al. 2003). Chemerin is the natural ligand for the receptors CMKLR1 (ChemR23), GPR1, and CCRL2. Chemerin expression is frequently downregulated in human tumors. The chemerin/CMKLR1 axis has been linked to immunity and inflammation as well as to cancer and angiogenesis. However, the exact function of CCRL2 in physiological and pathological processes remains poorly characterized. CCRL2 shares up to 40% homology with other C-C chemokine receptors in addition to many structural and functional similarities with the family of ACKRs, such as the lack of conventional G protein-mediated signaling and the inability to induce functional responses. CCRL2 is expressed by different cell types, such as endothelial cells and various leukocyte populations, and its expression is strongly upregulated by inflammatory signals. CCRL2 acts as a chemerin presenting molecule to cells expressing functional chemerin receptors (CMKLR1 and possibly GPR1) (Zabel et al. 2008). We have demonstrated that the expression of bioactive chemerin by tumor cells delays the growth of tumors in vivo, and a similar tumor growth delay is observed when chemerin is expressed in the skin of the host mice. In these tumors, the neoangiogenesis process is impaired, resulting in hypoxia, necrosis, and growth delay. A similar phenotype is observed for tumor cells growing in CCRL2 KO mice. In contrast, in a chemical carcinogenesis model (DMBA/TPA), the development of papillomas is accelerated in CCRL2 KO mice. In the present study, we studied the role of chemerin in angiogenesis and further investigated the effect of CCRL2 on the chemerin/CMKLR1 axis in tumorigenesis by testing tumoral cell lines overexpressing or invalidated for CCRL2. Moreover, we investigated whether CCRL2 is involved in the proliferation, migration, clonogenicity, and spheroid formation capacity of B16 melanoma and LLC carcinoma cells. Firstly, our results showed that chemerin exerted a strong anti-angiogenic effect in a bead sprouting assay, whereas we could not detect pro- or antiangiogenic properties of chemerin in various other assays. We demonstrated that the overexpression of CCRL2 significantly inhibited tumor growth in vivo, partially dependant on chemerin/CMKLR1 axis and independent of GPR1 expression. Also we showed that CCRL2 invalidation restored the tumor growth delay observed in CCRL2 KO mice. Importantly, we validated that CCRL2 expression in tumors affected the proportion of blood vessels, and resulted in a larger hypoxic and necrotic areas. CCRL2 expression did not impact the proliferation, migration and clonogenicity of B16 and LLC cells, but it strongly affected the spheroid formation capacity of B16 melanoma cells with a potential effect on the adhesion processes. Taken together, these results indicate that chemerin/CMKLR1/CCRL2 axis is significantly affecting the tumor growth by regulating angiogenesis, and CCRL2 is considered as a negative regulator of tumorigenesis. / Les cytokines chimiotactiques, également appelées chimiokines, dirigent la migration des leucocytes suite à leur interaction avec des récepteurs à sept domaines transmembranaires (Bachelerie et al. 2014). Les chimiokines sont des acteurs clés dans la progression du cancer et la régulation de l'inflammation liée au cancer. Les récepteurs atypiques de chimiokines (ACKR) représentent un sous-ensemble de protéines appartenant à la famille des récepteurs de chimiokines mais incapables de signaler via les cascades conventionnelles. Les ACKRs ont récemment été reconnus comme des acteurs moléculaires importants en physiologie et physiopathologie (Massara et al. 2016). Ils affectent la disponibilité et la fonction des chimiokines, et ont un impact sur de nombreux événements physiopathologiques, y compris le processus de tumorigenèse (Sjöberg et al. 2020). La chémérine est une protéine chimioattractante non apparentée aux chimiokines, active sur différentes populations leucocytaires, dont les cellules dendritiques, les macrophages et les cellules natural killer (Valérie Wittamer et al. 2003). La chémérine est le ligand naturel des récepteurs CMKLR1 (ChemR23), GPR1 et CCRL2. L'expression de la chémérine est fréquemment diminuée dans les tumeurs humaines. Le rôle de l’axe chémérine/CMKLR1 a été montré dans l'immunité et l'inflammation ainsi que le cancer et l'angiogenèse. Cependant, la fonction exacte de CCRL2 dans les processus physiologiques et pathologiques reste mal caractérisée. CCRL2 partage jusqu'à 40 % d'homologie avec d'autres récepteurs de C-C chimiokines, en plus de nombreuses similitudes structurelles et fonctionnelles avec la famille des ACKRs, telles que l'absence de signalisation médiée par les protéines G et l'incapacité d’induire des réponses fonctionnelles. CCRL2 est exprimé par différents types cellulaires, tels que les cellules endothéliales et diverses populations de leucocytes, et son expression est fortement augmentée par les signaux inflammatoires. CCRL2 agit uniquement en régulant les concentrations locales de chémérine, et en présentant le ligand à des cellules exprimant des récepteurs fonctionnels de la chémérine (CMKLR1 et potentiellement GPR1) (Zabel et al. 2008). Nous avons démontré que l'expression de la chémérine bioactive par les cellules tumorales retarde la croissance des tumeurs in vivo, et un retard similaire de la croissance tumorale est observé lorsque la chémérine est exprimée dans la peau des souris hôtes. Dans ces tumeurs, le processus de néoangiogenèse est altéré, entraînant une hypoxie, une nécrose et un retard de croissance.Un phénotype similaire de croissance retardée de lignées tumorales est observé chez les souris CCRL2 KO. En revanche, dans un modèle de cancérogenèse chimique (DMBA/TPA), le développement des papillomes est accéléré chez les souris CCRL2 KO. Dans la présente étude, nous avons étudié le rôle de la chémérine dans l'angiogenèse et l'effet de CCRL2 sur l'axe chémérine/CMKLR1 dans la tumorigenèse en testant des lignées cellulaires tumorales surexprimant ou invalidées pour CCRL2. De plus, nous avons étudié si CCRL2 est impliqué dans la prolifération, la migration, la clonogénicité et la capacité de formation de sphéroïdes de cellules tumorales de mélanome B16 et de carcinome pulmonaire (LLC). Premièrement, nos résultats ont montré que la chémérine exerçait un fort effet anti-angiogénique dans un test d’angiogenèse sur billes (bead sprouting assay), alors que nous n'avons pas pu détecter les propriétés pro- ou anti-angiogéniques de la chémérine dans divers autres tests. Nous avons démontré que la surexpression de CCRL2 inhibait significativement la croissance tumorale in vivo, un effet partiellement dépendant de l'axe chémérine/CMKLR1 et indépendant de l'expression de GPR1. Nous avons également montré que l'invalidation de CCRL2 dans les cellules tumorales supprimait le retard de croissance tumorale observé chez les souris CCRL2 KO. Surtout, nous avons validé que l'expression de CCRL2 dans les tumeurs affectait la proportion de vaisseaux sanguins et résultait en des zones hypoxiques et nécrotiques plus grandes. L'expression de CCRL2 n'a pas eu d'impact sur la prolifération, la migration et la clonogénicité des cellules tumorales B16 et LLC, mais elle a fortement affecté la capacité de formation de sphéroïdes par les cellules de mélanome B16 avec un effet potentiel sur les processus d'adhésion cellulaire. En conclusion, notre étude a permis de mettre en évidence les effets inhibiteurs de l'axe chemerin/CMKLR1/CCRL2 sur la croissance tumorale tout en régulant l'angiogenèse, et de montrer que CCRL2 peut être considéré comme un régulateur négatif de la tumorigenèse. / Doctorat en Sciences biomédicales et pharmaceutiques (Médecine) / info:eu-repo/semantics/nonPublished

No description available.

<p>To this day, cancer remains the leading cause of mortality worldwide1. A major contributor to cancer progression and metastasis is tumor angiogenesis. The formation of blood vessels around a tumor is facilitated by the complex interplay between cells in the tumor stroma and the surrounding microenvironment. Understanding this interplay and its dynamic interactions is crucial to identify promising targets for cancer therapy. Current methods in cancer research involve the use of two-dimensional (2D) monolayer cell culture. However, cell-cell and cell-ECM interactions that are important in vascularization and the three-dimensional (3D) tumor microenvironment cannot accurately be recapitulated in 2D. To obtain more biologically relevant information, it is essential to mimic the tumor microenvironment in a 3D culture system. To this end, we demonstrate the utility of poly(ethylene glycol) diacrylate (PEGDA) hydrogels modified for cell-mediated degradability and cell-adhesion to explore, in 3D, the effect of various tumor microenvironmental features such as cell-cell and cell-ECM interactions, and dimensionality on tumor vascularization and cancer cell phenotype. </p><p>In aim 1, PEG hydrogels were utilized to evaluate the effect of cells in the tumor stroma, specifically cancer associated fibroblasts (CAFs), on endothelial cells (ECs) and tumor vascularization. CAFs comprise a majority of the cells in the tumor stroma and secrete factors that may influence other cells in the vicinity such as ECs to promote the organization and formation of blood vessels. To investigate this theory, CAFs were isolated from tumors and co-cultured with HUVECs in PEG hydrogels. CAFs co-cultured with ECs organized into vessel-like structures as early as 7 days and were different in vessel morphology and density from co-cultures with normal lung fibroblasts. In contrast to normal lung fibroblasts (LF), CAFs and ECs organized into vessel-like networks that were structurally similar to vessels found in tumors. This work provides insight on the complex crosstalk between cells in the tumor stroma and their effect on tumor angiogenesis. Controlling this complex crosstalk can provide means for developing new therapies to treat cancer.</p><p>In aim 2, degradable PEG hydrogels were utilized to explore how extracellular matrix derived peptides modulate vessel formation and angiogenesis. Specifically, integrin-binding motifs derived from laminin such as IKVAV, a peptide derived from the α-chain of laminin and YIGSR, a peptide found in a cysteine-rich site of the laminin β chain, were examined along with RGDS. These peptides were conjugated to heterobifunctional PEG chains and covalently incorporated in hydrogels. The EC tubule formation in vitro and angiogenesis in vivo in response to the laminin-derived motifs were evaluated. </p><p>Based on these previous aims, in aim 3, PEG hydrogels were optimized to function as a 3D lung adenocarcinoma in vitro model with metastasis-prone lung tumor derived CAFs, HUVECs, and human lung adenocarcinoma derived A549 tumor cells. Similar to the complex tumor microenvironment consisting of interacting malignant and non-malignant cells, the PEG-based 3D lung adenocarcinoma model consists of both tumor and stromal cells that interact together to support vessel formation and tumor cell proliferation thereby more closely mimicking the functional properties of the tumor microenvironment. The utility of the PEG-based 3D lung adenocarcinoma model as a cancer drug screening platform is demonstrated with investigating the effects of doxorubicin, semaxanib, and cilengitide on cell apoptosis and proliferation. The results from drug screening studies using the PEG-based 3D in vitro lung adenocarcinoma model correlate with results reported from drug screening studies conducted in vivo. Thus, the PEG-based 3D in vitro lung adenocarcinoma model may serve as a better tool for identifying promising drug candidates than the conventional 2D monolayer culture method.</p> / Dissertation

Les facteurs de croissance endothéliaux (VEGFs) sont produits par les tumeurs qui sont hypoxiques. Principaux responsables de la néo-vascularisation pathologique, ils régulent le stroma tumoral. Les nouvelles stratégies qui ciblent et inhibent le VEGF ouvrent vers la thérapie anti-cancéreuse moderne. Elles ont pour but de contrôler l’angiogenèse tumorale plutôt que la détruire. Le défi est donc de piéger sélectivement le VEGF produit en excès, dans le microenvironnement tumoral, sous l’effet de l’hypoxie. La thèse présentée dans ce manuscrit est consacrée à la réalisation d’une nouvelle stratégie ciblante par l’intermédiaire de cellules, aussi appelée « Cheval de Troie ». Elle combine dans la même entité, une unité de ciblage et un système de délivrance spécifique d’un gène/molécule thérapeutique. Dans le but d’adresser la thérapie aux cellules cancéreuses sans affecter les cellules saines, un modèle de cellules endothéliales de type précurseur (CEPs) a été utilisé comme cellules ciblantes capables d’atteindre spécifiquement le site tumoral. Les CEPs ont été « armées » pour exprimer un gène thérapeutique chargé d’inhiber le VEGF. La neutralisation a été obtenue par la production d’une forme soluble du récepteur-2 du VEGF (VEGFR2 soluble), agissant comme inhibiteur. Caractéristique des tumeurs solides se développant, l’hypoxie a été choisie pour déclencher/éteindre l’expression et la sécrétion du VEGFR2 soluble, en introduisant, en amont du gène thérapeutique, une séquence régulatrice : HRE. Adressé au site tumoral par les CEPs, le régulateur de l’angiogenèse qu’est la forme soluble du VEGFR2, est exprimé de manière conditionnée et réversible, à l’hypoxie. Ceci ouvre à de nouvelles stratégies de normalisation contrôlée et stable des vaisseaux tumoraux en vue de l’utilisation de thérapies combinées. / Vascular endothelial growth factors (VEGFs) are over-expressed upon hypoxia in solid tumors. Major actors directing pathologic neo-vascularisation, they regulate the stromal reaction. Novel strategies that target and inhibit VEGF bring promise to modern anti-cancer therapies. They aim to control rather than destroy tumor angiogenesis. Consequently, the challenge is to selectively trap VEGFs, over-produced upon hypoxia, in the tumor microenvironment. The thesis presented in this manuscript focuses on the design of a novel cell-based targeting strategy, so-called “Trojan Horse”, combining in the same engineered entity, a targeting unit and a specific drug/gene delivery system. Aiming to address the therapy to cancer cells without affecting healthy cells, a model of endothelial precursor cell (EPCs) was used as targeting cell able to reach specifically the tumor site. EPCs were “armed” to express a therapeutic gene to inhibit VEGF. Trapping was attempted based on the production of a soluble form of the VEGF receptor-2 (sVEGFR2) as a candidate inhibitor. Hypoxia, a hallmark of developing solid tumors, was chosen to turn on/off the sVEGFR2 expression and secretion by introducing, upstream of the therapeutic gene, a hypoxia response element (HRE) regulating sequence. Properly addressed by the EPCs to the tumor site, such angiogenesis regulator as the soluble form of VEGFR2 is, was chosen to be expressed in a hypoxia-conditioned and reversible manner. This opens new strategies for a stably controlled normalization of tumor vessels in view of adjuvant design for combined therapies.

<p>Tumor angiogenesis is critical to tumor growth and metastasis, yet much is unknown about the role vascular cells play in the tumor microenvironment. A major outstanding challenge associated with studying tumor angiogenesis is that existing preclinical models are limited in their recapitulation of in vivo cellular organization in 3D. This disparity highlights the need for better approaches to study the dynamic interplay of relevant cells and signaling molecules as they are organized in the tumor microenvironment. In this thesis, we combined 3D culture of lung adenocarcinoma cells with adjacent 3D microvascular cell culture in 2-layer cell-adhesive, proteolytically-degradable poly(ethylene glycol) (PEG)-based hydrogels to study tumor angiogenesis and the impacts of neovascularization on tumor cell behavior. </p><p>In initial studies, 344SQ cells, a highly metastatic, murine lung adenocarcinoma cell line, were characterized alone in 3D in PEG hydrogels. 344SQ cells formed spheroids in 3D culture and secreted proangiogenic growth factors into the conditioned media that significantly increased with exposure to transforming growth factor beta 1 (TGF-β1), a potent tumor progression-promoting factor. Vascular cells alone in hydrogels formed tubule networks with localized activated TGF-β1. To study cancer cell-vascular cell interactions, the engineered 2-layer tumor angiogenesis model with 344SQ and vascular cell layers was employed. Large, invasive 344SQ clusters developed at the interface between the layers, and were not evident further from the interface or in control hydrogels without vascular cells. A modified model with spatially restricted 344SQ and vascular cell layers confirmed that observed 344SQ cluster morphological changes required close proximity to vascular cells. Additionally, TGF-β1 inhibition blocked endothelial cell-driven 344SQ migration. </p><p>Two other lung adenocarcinoma cell lines were also explored in the tumor angiogenesis model: primary tumor-derived metastasis-incompetent, murine 393P cells and primary tumor-derived metastasis-capable human A549 cells. These lung cancer cells also formed spheroids in 3D culture and secreted proangiogenic growth factors into the conditioned media. Epithelial morphogenesis varied for the primary tumor-derived cell lines compared to 344SQ cells, with far less epithelial organization present in A549 spheroids. Additionally, 344SQ cells secreted the highest concentration of two of the three angiogenic growth factors assessed. This finding correlated to 344SQ exhibiting the most pronounced morphological response in the tumor angiogenesis model compared to the 393P and A549 cell lines. </p><p>Overall, this dissertation demonstrates the development of a novel 3D tumor angiogenesis model that was used to study vascular cell-cancer cell interactions in lung adenocarcinoma cell lines with varying metastatic capacities. Findings in this thesis have helped to elucidate the role of vascular cells in tumor progression and have identified differences in cancer cell behavior in vitro that correlate to metastatic capacity, thus highlighting the usefulness of this model platform for future discovery of novel tumor angiogenesis and tumor progression-promoting targets.</p> / Dissertation

Background: Neuroblastoma is the most common extracranial solid tumor in children. The survival rate in high-risk neuroblastoma is less than 50 % despite intensive multimodal therapy, and there is thus an immense need for new treatment options. In a previous preclinical study conducted at Uppsala University, treatment with sunitinib was found to inhibit tumor growth and angiogenesis in an orthotopic model of high-risk neuroblastoma.  Aim: The present study aimed to further explore the effect of sunitinib on tumor stroma, focusing on whether it was possible to detect and quantify tumor-associated neutrophils (TANs) in tumor sections from the above-mentioned study using immunohistochemistry (IHC). Methods: Tissue sections from formalin-fixated paraffin-embedded tumors were stained with anti-Ly-6G/Ly-6C, anti-ITGAM, or hematoxylin-eosin, and the number of granulocytes was quantified manually using a light microscope. An independent samples two-tailed t-test was used for statistical analysis. Results: The average number of granulocytes increased by 40 % in animals treated with sunitinib compared to control animals (p = 0.003) in hematoxylin-eosin stained tumor sections of orthotopic neuroblastomas. The results from the staining with anti-Ly-6G/Ly-6C or anti-ITGAM, on the other hand, were impossible to quantify due to the high background staining despite the concentration of antibody used. Conclusion: In conclusion, this report indicates that the density of TANs in an orthotopic murine neuroblastoma model is increased by treatment with sunitinib. However, to confirm this result, the study should be repeated once a reliable IHC method for the detection of TANs has been developed.

No description available.

L’objectif de cette thèse est de développer l’imagerie de transfert de saturation des agents de contraste lipoCEST pour la détection de l’angiogenèse dans un modèle souris de tumeur cérébrale U87. Un lipoCEST offrant un seuil de sensibilité in vitro de 100 pM est optimisé afin de répondre aux contraintes de l’imagerie CEST in vivo. Grâce à la mise en place d’un dispositif expérimental dédié à l’imagerie CEST, nous évaluons les performances des lipoCEST pour détecter de façon spécifique l’angiogenèse tumorale. Nous montrons pour la première fois qu’il est possible de détecter un lipoCEST in vivo dans un cerveau de souris suite à une injection intraveineuse. De plus, l’utilisation d’un lipoCEST fonctionnalisé avec un peptide RGD permet de cibler spécifiquement l’intégrine ανβ3 surexprimée lors de l’angiogenèse tumorale. L’association spécifique du RGD-lipoCEST est confirmée grâce à des données d’immunohistochimie et de microscopie de fluorescence. Enfin, dans le but de tendre vers un protocole d’imagerie moléculaire par IRM-CEST, nous mettons en place un outil de quantification des lipoCEST. Cet outil repose sur la modélisation des processus d’échange de protons in vivo. Grâce à la prise en compte des inhomogénéités de champs B0 et B1 qui peuvent se révélées être délétères pour le contraste CEST, nous démontrons que la précision de notre outil de quantification est de 300 pM in vitro. La quantification des données CEST acquises chez la souris U87 permet d’estimer à 1,8 nM la concentration maximale en RGD-lipoCEST liés à leur cible moléculaire. / The study aimed at developing saturation transfer imaging of lipoCEST contrast agents for the detection of angiogenesis in a U87 mouse brain tumor model. A lipoCEST with a sensitivity threshold of 100 pM in vitro was optimized in order to make it compatible with CEST imaging in vivo. Thanks to the development of an experimental setup dedicated to CEST imaging, we evaluated lipoCEST to detect specifically tumor angiogenesis. We demonstrated for the first time that lipoCEST visualization was feasible in vivo in a mouse brain after intravenous injection. Moreover, the integrin ανβ3 overexpressed during tumor angiogenesis can be specifically targeted using a functionalized lipoCEST with RGD peptide. The specific association between the RGD-lipoCEST and its target ανβ3 was confirmed by immunohistochemical data and fluorescence microscopy. Finally, in order to tend to a molecular imaging protocol by CEST-MRI, we developed a quantification tool of lipoCEST contrast agents. This tool is based on modeling of proton exchange processes in vivo. By taking into account both B0 and B1 fields inhomogeneities which can dramatically alter CEST contrast, we showed that the accuracy of our quantification tool was 300 pM in vitro. The tool was applied on in vivo data acquired on the U87 mouse model and the maximum concentration of RGD-lipoCEST linked to their molecular targets was evaluated to 1.8 nM.

Au cours des dernières années, des thérapies anti-cancéreuses ciblant l'angiogenèse tumorale ont été développées et ont démontré un bénéfice en terme de survie globale pour les patients atteints de certains cancers métastatiques. Cependant, dans de nombreux cas, les tumeurs acquièrent des résistances échappent au traitement. Le développement de nouveaux composés anti-angiogène est donc une réelle nécessité pour être proposés en seconde ligne thérapeutique. Dans ce travail, notre objectif était d'identifier de nouvelles molécules anti-angiogènes par le criblage à haut débit, de la chimiothèque académique de l'Université de Grenoble. Nous avons adapté le test de blessure sur cellules endothéliales au format des plaques de 96 puits et avons identifié une famille de molécules qui inhibent spécifiquement leur fermeture. L'activité anti-angiogène de la molécule leader (COB223) a été confirmée dans des modèles d'angiogenèse tridimensionnels in vitro, et, chez la souris, dans un modèle d'angiogenèse sous-cutanée. Nous avons testé l'activité anti-tumorale de COB223 dans un modèle de xénogreffe chez la souris et observé une diminution significative de la taille des tumeurs dans les souris traitées. A la recherche de son mécanisme d'action, nous avons observé que COB223 inhibe la prolifération cellulaire et diminue les phosphorylations de MEK et Raf, de ERK1/2 induites par le VEGF-A dans les cellules endothéliales. Nous avons également montré que COB223 n'inhibe pas les phosphorylations du VEGFR2 et de PLC. D'après ces résultats, nous proposons que la cible de COB est localisée dans la voie de signalisation VEGF/ PLC /PKC/ERK entre PKC et Ras. / Several anti-tumoral therapies targeting angiogenesis have been developed over the recent years and have demonstrated benefits for several metastatic cancers. However, in many cases, resistances to these treatments appear over time, allowing tumor escape. The development of new anti-angiogenic compounds is thus dramatically urged in order to propose second-line anti-angiogenic treatments. In this work, our aim was to identify new anti-angiogenic compounds through high throughput screening of the academic library from the University of Grenoble. We adapted the endothelial cell scratch assay to 96-well plates. We identified a family of molecules that specifically inhibited endothelial cell migration. The anti-angiogenic activity of the leader molecule (COB223) was confirmed in vitro in 3D cellular models of angiogenesis and in vivo using a mouse model of subcutaneous sponge implantation. We tested the anti-tumoral activity of COB223 on a mouse xenograft model. We observed that tumor growth was significantly reduced in treated mice correlated with decreased microvessel density. In search for its mechanism of action, we observed that COB223 inhibits cell proliferation and reduces VEGF-A-induced phosphorylation of MEK and ERK1/2 in endothelial cells. We also showed that COB223 did not affect VEGFR2 and PLC phosphorylation but reduces Raf phosphorylation responsible for its activity. These results allow us to propose that the molecular site of action of COB223 is located in the VEGF/ PLC /PKC/ERK pathway, between PKC and MEK.