Le mélanome est le plus agressif et le plus sévère des cancers cutanés du fait de son fort potentiel métastatique. Pourtant à ce jour, aucun biomarqueur pour la détection précoce du mélanome n'est unanimement reconnu. Notre groupe a récemment démontré que le métabolisme du céramide est fortement altéré dès les premiers stades de la maladie, conduisant à l'augmentation de la production d'un dérivé du céramide, la sphingosine 1-phosphate (S1P). La S1P est sécrétée par les cellules du mélanome et a été identifiée comme une molécule majeure du remodelage du microenvironnement tumoral, qui favorise la progression du cancer. De façon physiologique, la S1P circulante se trouve majoritairement sous forme liée aux protéines de haute densité (HDLs), aux protéines de basse et très basse densité (LDLs et VLDLs) ainsi qu'à l'albumine. Les cellules de mélanome pourraient produire de la S1P non liée qui pourrait rendre compte des effets produits par ce lysosphopholipide sur les cellules du microenvironnement tumoral suite à sa fixation sur les récepteurs S1PR présents à la surface des cellules stromales. Ainsi, cette forme libre de S1P pourrait représenter un nouveau biomarqueur pour la détection précoce du mélanome. Cependant, il n'existe à l'heure actuelle aucun moyen permettant de la quantifier. Le but de ce travail interdisciplinaire a été de développer un nouveau capteur basé sur un polymère à empreintes moléculaires (MIP) dans le but de quantifier la S1P libre dans le sang de patients atteints de mélanome. Cette étude a été réalisée entre l'équipe " Ingénierie pour les sciences du vivant (ELiA) " du Laboratoire d'Analyse et d'Architecture des Systèmes (LAAS), et l'équipe " Sphingolipides, métabolisme, mort cellulaire et progression tumorale " du Centre de Recherche en Cancérologie de Toulouse (CRCT), en étroite collaboration avec l'équipe " Biomimétisme et structures bioinspirées " de l'Université Technologique de Compiègne (UTC). Dans un premier temps, nous avons synthétisé un nouveau MIP dirigé contre la S1P par une méthode de thermopolymérisation en masse. Nous avons caractérisé puis optimisé ce MIP en effectuant des mesures de spectrométrie de masse couplée à la chromatographie en phase liquide et des mesures de spectroscopie de fluorescence. Le MIP a été comparé à un NIP (Non Imprinted Polymer) et exposé à des analogues de la S1P afin d'évaluer sa sélectivité. Dans un second temps, en vue de l'utilisation d'un MIP en tant que couche sensible d'un futur capteur et pour anticiper son immobilisation et sa structuration sur le transducteur, nous avons mis au point un nouveau MIP photopolymérisable en 2D. Ce MIP a d'abord été structuré en motifs par photolithographie sur des surfaces de silicium puis validé par des mesures de microscopie de fluorescence. Le MIP a également été structuré sous la forme de couches minces sur les surfaces actives de capteurs de Microbalance à Cristal de Quartz (QCM) dans le but de le valider par cette méthode sans marquage. Enfin, nous avons exploré l'utilisation d'une fibre optique recouverte d'une couche de MIP photopolymérisé dans le but de détecter, par spectroscopie infrarouge, la liaison de la S1P avec le MIP à la surface de la fibre. / Melanoma is the most aggressive and severe form of cutaneous cancer due to its high metastatic potential. However, to date, no marker for the early detection of melanoma has been unanimously accepted. Our group has demonstrated that ceramide metabolism is strongly altered in melanoma, leading to the overproduction of sphingosine 1-phosphate (S1P), one of its derivatives. S1P is secreted by melanoma cells and has been identified as a critical molecule of tumor microenvironment remodeling that supports cancer progression. Physiologically, circulating S1P is predominantly linked to high density lipoproteins (HDLs), low and very low density lipoproteins (LDLs and VLDLs), as well as albumin. Melanoma cells produce unbound S1P that could be responsible for the effects induced by this lysophospholipid on the tumor microenvironment, as a result of its binding to S1PR receptors present on the surface of stromal cells. Thus, secreted tumor S1P could represent a new biomarker for the early detection of melanoma. However, there are currently no means to quantify it. The goal of this interdisciplinary work was to develop a new sensor based on a Molecularly Imprinted Polymer (MIP) in order to quantify unbound S1P present in the blood of melanoma patients. This study has been conducted between the "Engineering for Life science Applications (EliA)" group at the Laboratory for Analysis and Architecture of Systems (LAAS) and the "Sphingolipids, metabolism, cell death and tumor progression" group at the Cancer Research Center of Toulouse (CRCT), in strong collaboration with the team "Biomimetism and Bioinspired Structures" of the University of Technology of Compiègne (UTC). First, we synthesized a new MIP against S1P employing a bulk thermopolymerization approach. The resulting MIP was characterized and optimized by performing both mass spectrometry and fluorescence spectroscopy measurements. It was compared to a Non Imprinted Polymer (NIP) and exposed to S1P analogues to assess its selectivity. Second, in order to use the MIP as the sensitive layer of a future sensor and prepare its immobilization and structuration onto a transducer, we synthesized a new surface photopolymerizable MIP. This MIP was first structured by photolithography onto silicon substrates and validated by fluorescence microscopy measurements. The MIP was also structured as a thin layer onto Quartz Crystal Microbalance (QCM) chips in order to validate its binding capacities using this label-free method. Finally, the use of a MIP-coated optical fiber as an infrared sensor was explored, with the aim of detecting S1P in blood using Attenuated Total Reflectance (ATR) spectroscopy.

Organoboron compounds are ubiquitous in organic chemistry. Fundamental transformations utilizing organoboron compounds are a necessary addition to any organic chemist's synthetic toolbox. In addition to their extensive use as synthetic intermediates, organoboron compounds are being increasingly studied for their material and medicinal properties. Excitingly, significant advances have been made over the years towards the synthesis of a wide variety of organoboron substrates. In the case of vinylboronic acids, synthesis primarily occurs through cis addition of boron reagents across triple bonds. However, methods affording trans addition products are scarce. Furthermore, many current methods rely on the use of expensive and toxic transition-metal catalysts. Herein, we describe the development of trans addition of boron reagents across polarized triple bonds to afford novel vinylboronic acids. Emphasis is placed on the transition metal-free nature of the reactions as well as the regio- and stereoselectivity observed in the products. In addition, the synthetic utility of the resulting trans addition products is demonstrated in the synthesis of biologically relevant molecules. We first describe the Brønsted base-mediated trans silaboration of propiolamides in which two functional groups with orthogonal reactivity are simultaneously installed. We then go on to describe an organocatalyzed trans hydroboration of propiolate esters as well as a complementary Brønsted base-mediated trans hydroboration reaction of propiolamides. To conclude this portion, we demonstrate how the products from the previous methods can be used to synthesize difluoroborylacrylamides which possess unique and versatile reactivity. Herein we disclose the first small-molecule inhibitors of the sphingosine-1-phosphate (S1P) transporter spinster homolog 2 (SPNS2). While little is known in regard to the structure and function of SPNS2, previous studies have demonstrated the vital role SPNS2 plays in S1P mediated processes and have identified SPNS2 as a potential clinical target. For example, SPNS2 is critical to S1P-mediated lymphocyte egress from primary lymphoid tissues. Thus, small molecule inhibition of SPNS2 represents a novel therapeutic strategy for the treatment of autoimmune disorders such as multiple sclerosis. In this study, we report the discovery of small molecule inhibitors that display low micromolar activity using a novel yeast-based SPNS2 assay. Inhibitor structure-activity-relationship studies led to the discovery of the imidazole-based amine inhibitor 7.54. Furthermore, administration of 7.54 to mice recapitulates the lymphopenic phenotype observed in previous SPNS2 knockout studies. / Doctor of Philosophy / Boron-containing compounds are important in organic chemistry and are involved in the synthesis of a variety of materials and medicines used in everyday life. As such, the ability to efficiently and sustainably prepare boron-containing compounds has far reaching consequences. Access to an important class of boron-containing compounds known as vinylboronic acids has previously been established; however, product selectivity is often limited to what is referred to as cis addition products. Furthermore, access to the corresponding trans addition products is often limited to processes involving expensive transition metal catalysts that produce environmentally toxic waste. Herein, novel transition metal-free trans addition processes are described for preparing vinylboronic acids. In addition, the application of the resulting products is demonstrated through the synthesis of biologically relevant compounds. Sphingosine-1-phosphate (S1P) is an important signaling lipid that is involved in a variety of physiological processes. Improper balance in the amount of S1P in the body is associated with a variety of disease states such as autoimmunity and cancer. Two drugs that inhibit S1P-mediated processes have been approved by the FDA, fingolimod (Gilenya®) and siponimod (Mayzent®). However, there are drawbacks to targeting the S1P receptor directly, including dose-limiting side effects that are associated with these drugs. Consequently, recent efforts have focused on developing new ways to control the effects of S1P. Herein, we describe the discovery and development of the first reported inhibitors of the S1P transporter, spinster homolog 2 (SPNS2). A library of compounds was synthesized and tested for SPNS2 inhibition. The resulting structure-activity-relationship studies led to the discovery of the imidazole-based propanamine derivative 7.54. Furthermore, we demonstrate the potential of SPNS2 inhibition to control the effects of S1P in mice. These studies provide a foundation for future SPNS2-based drug discovery that will hopefully lead to the development of improved therapies for the treatment of autoimmune disease and cancer.

Résumé confidentiel / Confidential abstract

Résumé confidentiel / Confidential abstract

La Sphingosine 1-phosphate (S1P) est un lipide bioactif qui joue un rôle majeur dans de nombreux processus tels que le développement vasculaire, la circulation des cellules immunitaires et la régulation de l’intégrité vasculaire. L’absence de S1P circulante engendre une augmentation constitutive de la perméabilité vasculaire ainsi qu’une sensibilité accrue à l’anaphylaxie systémique passive, cependant l’implication respective des différentes sources de la S1P au cours de ces processus, demeure méconnue. Ceci est plus particulièrement vrai pour les plaquettes qui constituent une réserve importante de S1P. La S1P d’origine plaquettaire ne contribue pas au pool de S1P plasmatique, il n’est libéré que lors de l’activation plaquettaire. Nous avons étudié le rôle de la S1P d’origine plaquettaire au sein des vaisseaux sanguins, au cours du développement embryonnaire et chez des souris adultes, en présence ou en absence d’autres sources de S1P circulante, notamment des globules rouges. Par délétion des gènes codant pour les sphingosine kinases 1&2 au sein des megakaryocytes, nous avons généré des souris dont les plaquettes sont incapables de synthétiser et donc de sécréter la S1P. Ces plaquettes sont par ailleurs incapables, in vitro, de promouvoir le maintien de la barrière endothéliale. Néanmoins, une déficience en S1P d’origine plaquettaire n’augmente pas le risque d’hémorragie au cours du développement embryonnaire, dans un contexte d’inflammation, d’irradiation, ou de traumatisme ; cela suggère que la S1P d’origine plaquettaire n’est pas indispensable au cours de ces processus. Cependant, nous avons découvert que la résistance au choc anaphylactique est compromise en l’absence de S1P d’origine plaquettaire. L’aspirine bloquant la sécrétion de S1P par les plaquettes, un traitement à l’aspirine induit également une sensibilisation au choc anaphylactique de façon similaire à celle observée dans le cas d’une déficience en S1P d’origine plaquettaire. Il semblerait que dans ce contexte, la S1P d’origine plaquettaire joue plutôt un rôle pour promouvoir le tonus vasculaire via S1P2, que de maintenir l’intégrité vasculaire. / Sphingosine 1-phosphate (S1P) is a bioactive lipid that plays key roles in vascular development, immune cell trafficking and regulation of vascular integrity. Lack of circulating S1P leads to constitutive vascular leak and sensitivity to passive systemic anaphylaxis, but the relative roles of different sources of S1P in these processes is unclear. This is especially true for platelets, which carry large amounts of S1P but do not contribute S1P to plasma unless activated. We have addressed roles of platelet S1P in developing and mature murine vessels in the presence or absence of other circulating S1P sources, notably red blood cells. By deletion of the genes encoding sphingosine kinases 1&2 in megakaryocytes, we generated mice with platelets that were greatly impaired in their ability to secrete S1P. These platelets were also unable to promote endothelial barrier function in vitro. Nevertheless, platelet S1P deficiency did not sensitize to bleeding during embryonic development, or in association with inflammation, irradiation or traumatic injury, suggesting that platelet S1P is dispensable for these processes. Instead, we find that recovery from anaphylactic shock is impaired in the absence of platelet-derived S1P. Consistent with the ability of aspirin to block S1P release from platelets, it impaired recovery from anaphylactic shock to a similar degree as platelet S1P deficiency. Instead of protecting vascular integrity, platelet-derived S1P appeared to act via the promotion of vascular tone through S1P2 in this context. Collectively, we find that S1P is necessary for the capacity of platelets to promote endothelial barrier function in vitro, but that this capacity becomes redundant in the presence of red blood cell-derived S1P in vivo. Nevertheless, during anaphylactic shock, both sources of S1P are necessary for complete recovery. Blocking S1P release with aspirin, an over-the-counter anti-inflammatory drug, may impair the protective functions of platelet S1P.

Les surcharges en fer, qu’elles soient d’origine génétique ou secondaire, favorisent la baisse de densité minérale osseuse et par conséquent l’apparition d’une ostéoporose. Des liens entre surcharge en fer et perte osseuse ont pu être établis in vivo et in vitro, néanmoins les mécanismes mis en jeu, notamment sur la cellule ostéoblastique, restent incomplètement caractérisés. Notre objectif a donc été de préciser les mécanismes cellulaires conduisant à l’altération du phénotype et de l’activité ostéoblastique observée en présence d’un excès de fer. La réalisation préalable d’une étude transcriptomique sur la lignée ostéoblastique humaine MG-63 nous a permis d’identifier plusieurs gènes susceptibles de voir leur niveau d’ARNm régulé par le fer. Il a été fait l’hypothèse que ces différents gènes pouvaient être impliqués dans la survenue des pertes osseuses observées au cours des surcharges en fer. Ainsi l’expression du gène SPNS2, dont la protéine permet l’export de la Sphingosine-1-Phosphate (S1P), a été identifiée comme potentiellement induite par un excès de fer. Les relations entre l’expression du gène SPNS2 et un excès de fer ont ainsi été investiguées et les résultats obtenus ont mis en évidence une augmentation fer-dépendante de l’ARNm du gène SPNS2 dans la lignée MG-63, non retrouvée dans d’autres types cellulaires. Cette caractérisation nous a ainsi conduits à déterminer, dans la lignée MG-63, l’impact fonctionnel d’une exposition au fer sur l’export cellulaire de la S1P. Nous avons donc pour cela mis au point une méthode d’étude basée sur une stratégie « fluxomique » nous permettant d’évaluer l’efflux de la S1P au moyen d’un outil de spectrométrie de masse. Nos résultats objectivent une diminution des capacités de synthèse et d’export de la S1P en présence de fer et ceci malgré la surexpression du gène SPNS2. La diminution concomitante de l’expression du récepteur S1PR1 et du gène COL1A1 codant pour la chaîne α du collagène de type I suggère un impact fonctionnel de la baisse de concentration en S1P extracellulaire sur la cellule MG-63. La mise en évidence, dans un modèle ostéoblastique, d’une altération fer-dépendante de l’axe de signalisation S1P/S1PR ouvre de nouvelles perspectives quant à la compréhension des mécanismes mis en jeu lors des pertes osseuses associées aux surcharges en fer. / Osteoporosis may complicate genetic or secondary iron overload as reported in clinical and animal studies. However, the mechanisms leading to disrupted bone homeostasis are still to be fully elucidated. In vitro, iron exposure of both osteoblast and osteoclast cell models induces phenotypic and functional impairment, but the molecular mechanisms of iron excess on bone cell physiology are not well characterized, particularly in osteoblast. Our objective was to study the impact of iron overload on osteoblast biology and characterize the molecular mechanisms involved. Transcriptomic analysis previously performed by our group on MG-63 osteoblast-like cell-line to identify iron-modulated genes revealed that expression of SPNS2 gene, which encodes a transporter for the signaling lipid sphingosine 1-phosphate (S1P), is potentially induced by iron. The purpose of this work was to characterize the SPNS2 iron-related regulation and analyze its potential impact on S1P efflux and the S1P/S1PR signaling pathway in MG-63 cells. Our findings showed that iron exposure induces a dose-dependent increase of SPNS2 mRNA levels in MG-63 osteoblast-like cells that was not observed in hepatocyte and enterocyte cell models. We then performed a fluxomic assay on MG-63 cells to investigate iron potential impact on S1P efflux. Unexpectedly, our data showed that extracellular S1P levels were decreased in presence of iron excess and its associated SPNS2 upregulation. Furthermore, based on the observed iron associated S1PR1 and COL1A1 decrease, the defect in S1P export system seems to have functional consequence on MG-63 cells. These results suggest that iron may affect osteoblast S1P/SPR signaling and potentially alter a wide range of bone processes, thus participating in bone impairment in situations of chronic iron overload. These data open a new door for the understanding of mechanisms involved in iron-induced osteoporosis.

Excessive microvascular permeability is a serious complication involved in traumatic injury and inflammatory diseases. Alcohol intoxication can exacerbate the physiological derangements produced by microvascular endothelial barrier dysfunction in such disease conditions. Sphingosine-1-phosphate (S1P) has known endothelial barrier-protective properties, and has been shown to ameliorate microvascular leakage in a model of combined alcohol intoxication and hemorrhagic shock and resuscitation (HSR). However, whether the barrier-protective properties of S1P extend to endothelial cells of the blood-brain barrier (BBB) is unclear. The mechanisms of S1P-induced barrier protection during alcohol intoxication or HSR are also unknown. In the current study, we tested the hypothesis that S1P could enhance endothelial barrier during alcohol intoxication or hemorrhagic shock by preserving the integrity of junction proteins and the endothelial glycocalyx, and protecting mitochondrial function. Cultured primary human brain microvascular endothelial cell (HBMEC) monolayers were used to characterize endothelial-specific mechanisms of S1P protection of the BBB during alcohol treatment. Transendothelial electrical resistance (TER) and apparent permeability coefficients for albumin, dextran-4 kDa, and sodium fluorescein were used as indices of barrier function. Junctional localization was determined by immunofluorescence confocal microscopy. We also used an established in vivo rat model of conscious HSR and assessed microvascular leakage, endothelial glycocalyx integrity, and mitochondrial function by intravital microscopy. Cultured rat intestinal microvascular endothelial cell (RIMEC) monolayers were used to test the ability of S1P to protect against glycocalyx shedding and endothelial barrier dysfunction caused by direct disruption of mitochondrial integrity due to inhibition of mitochondrial complex III. The results show that alcohol significantly impaired HBMEC TER and increased solute permeability, which was reversed with application of S1P after alcohol treatment. Alcohol caused the formation of gaps between cells. Treatment with S1P (after alcohol) increased junctional localization. Our in vivo results show that S1P protects against HSR-induced hyperpermeability, preserves the expression of adherens junctional proteins, and protects against glycocalyx degradation. S1P treatment during HSR also protects against mitochondrial membrane depolarization. Besides that, S1P protects RIMECs against mitochondrial dysfunction-induced endothelial barrier dysfunction and glycocalyx degradation by acting through mitochondrial complex III. Our results indicate that S1P may be useful for restoring BBB function during alcohol intoxication. Moreover, S1P protects against HSR-induced mitochondrial dysfunction in endothelial cells, which in turn improves the structure of the endothelial glycocalyx after HSR and allows for better junctional integrity to prevention of excess microvascular permeability.

Osteocytes have been proposed as the mechanosensory cells during the process of bone adaption. In this thesis, a microfluidics chamber system (MCS) device was designed, fabricated and tested as a means to maximally simulate the in vivo osteocytic ultrastructure and reproduce the in vivo shear stress experienced by osteocyte, providing an ideal platform for in vitro study on osteocyte mechanotransduction. By employing a micropipette aspiration technique, single osteocyte adhesion and osteocytic process formation were achieved on PDMS with MCS structure. In this study, the involvement of sphingosine-1-phosphate (S1P) signaling pathway in osteocytes responding to oscillatory fluid flow (OFF) was also examined. Firstly, MLO-Y4 osteocytes like cells were demonstrated to express integrated and functional S1P cascade. By modulating S1P cascade components and testing a series of cellular outcomes, it was indicated that exogenous S1P, endogenous S1P and S1P receptor S1P2 were involved in the regulation of loading induced osteocytic responses.

Osteocytes have been proposed as the mechanosensory cells during the process of bone adaption. In this thesis, a microfluidics chamber system (MCS) device was designed, fabricated and tested as a means to maximally simulate the in vivo osteocytic ultrastructure and reproduce the in vivo shear stress experienced by osteocyte, providing an ideal platform for in vitro study on osteocyte mechanotransduction. By employing a micropipette aspiration technique, single osteocyte adhesion and osteocytic process formation were achieved on PDMS with MCS structure. In this study, the involvement of sphingosine-1-phosphate (S1P) signaling pathway in osteocytes responding to oscillatory fluid flow (OFF) was also examined. Firstly, MLO-Y4 osteocytes like cells were demonstrated to express integrated and functional S1P cascade. By modulating S1P cascade components and testing a series of cellular outcomes, it was indicated that exogenous S1P, endogenous S1P and S1P receptor S1P2 were involved in the regulation of loading induced osteocytic responses.

L'angiogenèse, soit la formation de nouveaux vaisseaux sanguins à partir de capillaires préexistants, est un processus essentiel au développement et à la croissance des tumeurs. Bon nombre de protéines ont été étudiées afin d'élucider les voies de signalisation impliquées dans l'angiogenèse, dont la métalloprotéase membranaire de type 1 (MT1-MMP). Cette protéine est reconnue pour jouer un rôle crucial dans la migration cellulaire, détruisant la matrice extracellulaire pour permettre aux cellules de migrer et ce, dans différents types cellulaires. Cependant, les mécanismes impliqués dans le contrôle de son activité demeurent incompris. Dans la présente étude, nous avons observé, en utilisant des procédures d'immunoprécipitation et de microscopie confocale, que la stimulation des cellules endothéliales de veines ombilicales humaines (HUVECs) avec la sphingosine-1-phosphate (S1P), un lipide qui induit la migration des cellules endothéliales (CEs), provoque le transfert de la MT1-MMP à la périphérie cellulaire et son association avec p130Cas (Crk-associated substrate). p130Cas est une protéine d'arrimage impliquée dans les voies de signalisation de la motilité cellulaire et est également reconnue pour se relocaliser dans des replis membranaires suite à une stimulation à la S1P. Ces résultats suggèrent fortement que l'identification du complexe MT1-MMP/p130Cas au front principal des CEs migrantes pourrait être un mécanisme efficace par lequel la protéolyse péricellulaire est reliée à l'activation des voies de signalisation, permettant la migration coordonnée des CEs au cours de l'angiogenèse. ______________________________________________________________________________ MOTS-CLÉS DE L’AUTEUR : Angiogenèse, MT1-MMP, p130Cas, Migration, Sphingosine 1-phosphate.