Les endosulfatases (H-Sulf 1 et 2) catalysent l’hydrolyse des sulfates en position 6 des résidus glucosamine des chaines d’héparane sulfate. Elles modulent ainsi l’activité des héparanes sulfate vis à vis de nombreuses protéines. Plusieurs études ont montré que l’endosulfatase H-Sulf 2 est produite en quantité plus importante dans de nombreux cancers (poumons, sein, pancréas etc …). La réduction de l’activité de cette enzyme chez des souris malades permet la diminution de la tumeur ainsi que le prolongement de la durée de vie. Ce travail de thèse vise à synthétiser une série d’inhibiteurs spécifiques des endosulfatases humaines. Une étude préliminaire a été menée sur un monosaccharide portant une fonction sulfamate, cette fonction étant connue pour inhiber les sulfatases. Ce projet de thèse repose sur la synthèse d’oligosaccharide de type héparane sulfate de tailles différentes portant tous cette fonction inhibitrice. Pour ce faire il a fallu mettre au point une stratégie itérative incluant des réactions de glycosylations stéréosélectives. Puis l’ouverture régiosélective d’acétal de benzylidène ainsi que la sulfamoylation ont été examinées l’optimisation des fonctionnalisations et des déprotections sont encore en cours afin de compléter cette librairie de molécules inhibitrice des endosulfatases humaines. / Human endosulfatase (H-Sulf 1 and 2) catalyse heparan 6-O-sulfate hydrolysis. Thanks to this hydrolysis they change the heparan sulfate proteoglycan activity. Many studies have shown that H-Sulf 2 are surexpressed in many cancer (lung, breast, pancreas etc...) The decrease of the activity of this enzyme in the case of the mouse allows a decrease of the size of the tumor and an extension of the mouse life. The aim of this thesis is to synthetise a library of endosulfatase specific inhibitors. A preliminary study was made on a monosaccharide carrying a sulfamate. This fonctional group have shown that it is able to inhibate the sulfatases. This project thesis is to synthetise heparan sulfate oligsaccharides carrying this sulfamate moeity. We apply an iterative process to make this oligosaccharide, thanks to a high stereoselective glycosylation. Finaly a reductive opening of benzylidène acetals, a sulfamoylation have been examinated, functionalization and deprotection are still in progress for get the full library.

L’AVC représente la première cause d’handicap acquis chez l’adulte. L’AVC ischémique, représentant 87% des AVCs, est une pathologie complexe dont le premier facteur de risque aggravant est l’hypertension artérielle. À l’heure actuelle les seuls traitements disponibles sont la thrombolyse et la thrombectomie. Cependant, ces traitements présentent de nombreuses contre-indications et effets secondaires limitant leurs applications chez les patients. L’objectif des travaux menés dans cette thèse est l’évaluation d’un traitement pharmacologique, le RGTA (ReGeneraTing Agent), combiné ou non à un traitement cellulaire utilisant les cellules souches mésenchymateuses (CSMs), chez des rats normo- et hyper-tendus. Les résultats obtenus dans cette thèse montrent qu’à la suite d’une ischémie cérébrale, les traitements évalués offrent une neuroprotection et une récupération fonctionnelle persistantes, chez les animaux noromo- et hyper-tendus. Cette récupération est expliquée par la réduction du volume lésionnel, par une meilleure plasticité cérébrale (angiogenèse, neurogenèse), ainsi par la potentialisation de l’effet des CSMs par le RGTA. En conclusion, nos études démontrent l’efficacité d’une thérapie robuste de neurorprotection chez le rongeur à la suite d’une ischémie cérébrale. / Stroke is the leading cause worldwide of adult severe disability. The limited available treatments for ischemic stroke, which accounts for 87% of strokes, makes it necessary to develop new therapeutical approaches. Stroke is a complex pathology and chronic hypertension (CAH) represents the first aggravating risk factor for ischemic stroke. At the present time, the only two available treatments for ischemic stroke, thrombolysis and thrombectomy, present several side effects limiting their clinical use. Here we evaluate the effect of a molecular RGTA (ReGeneraTing Agent) based therapy combined or not to a cellular therapy based on the use of mesenchymal stem cells (MSCs) for the treatment of ischemic stroke in normo- and hyper-tensive rats. The results demonstrate that the evaluated therapies confer a long lasting neuroprotection accompanied by animals’ functional recovery. Further analysis suggest that RGTA enhances brain plasticity (angiogenesis, and neurogenesis), protects the extracellular matrix structure, and potentiates MSCs’ beneficial effects. In conclusion, our studies demonstrate the efficacy of a molecular and cellular combined therapy conferring a persistent neuroprotection and functional recovery for the treatment of ischemic stroke.

Ce travail de thèse a eu pour thème principal le protéoglycane syndecan-1 au cours du myélome multiple, une hémopathie maligne caractérisée par la présence d'un clone de plasmocytes tumoraux au sein de la moelle osseuse. Syndecan-1 est un élément majeur de la physiopathologie du myélome multiple, ce protéoglycane étant au coeur d'un réseau complexe d'interactions moléculaires conditionnant le devenir des cellules plasmocytaires tumorales.Les chaînes de glycosaminoglycanes présentes sur le core protéique de syndecan-1sont responsables d'une grande partie de son activité. Nous avons ainsi caractérisé, par une approche transcriptomique, 100 gènes codant les protéines impliquées dans la synthèse et la modification de ces chaînes. Nous avons de cette manière identifié des cibles moléculairesen vue de moduler, voire d'inhiber leur activité.Dans le but d'identifier les métalloprotéinases des familles ADAM et ADAMTS susceptibles d'interagir avec syndecan-1, nous avons réalisé l'étude du profil d'expression des gènes codant ces reprolysines et leurs inhibiteurs dans les cellules de la différentiation lymphocytaire B, les cellules plasmocytaires normales et tumorales ainsi que dans l'environnement médullaire.Dans une dernière partie, nous avons évalué l'efficacité d'une approche d'inhibitiondes chaînes héparanes sulfates via l'utilisation de l'héparine. Nous observons que certaines lignées myélomateuses sont inhibées par l'héparine et ses dérivés et que ces mêmes lignées sont stimulées par l'antidote de l'héparine, le sulfate de protamine. Les mécanismes mis enjeu sont en relation avec la modulation de la biodisponibilité des facteurs permettant la croissance des cellules. / Multiple myeloma is a hematological malignancy characterized by the expansion of aclone of malignant plasma cells in the bone marrow compartment. Syndecan-1 is a majorproteoglycan involved in a complex network of molecular interactions in multiple myelomaphysiopathology. As heparan sulfate and chondroitin sulfate chains are the bioactive components ofsyndecan-1, we first analysed the signature of genes encoding 100 proteins involved in thesynthesis of these chains, from precursor uptake to post-translational modifications, usingAffymetrix microarrays.In order to identify the metalloproteinases belonging to ADAM and ADAMTS familiespotentially implicated in the interactions with syndecan-1, we performed a gene expressionprofile focused on the genes encoding these reprolysines and their inhibitors.In a last part, we evaluated the efficacy of an inhibitory approach based on theutilization of heparin in human myeloma cell lines in vitro, inhibitory effects being in relationwith a modulation of the biodisponibility of heparin-binding factors.This work led us to identify targets of interest in relation with syndecan-1 biology inmultiple myeloma. They could be used to design new therapeutic strategies.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

Ce travail de thèse a pour objectif la simplification de la préparation d’un nouveau type d’inhibiteur d’entrée du VIH conçu, synthétisé et validé dans le cadre d’une collaboration entre l’équipe de Glycochimie Moléculaire et Macromoléculaire dont je dépends, l’Institut de Biologie Structurale de Grenoble et l’Institut Pasteur de Paris. Ce prototype est constitué d’un mime fonctionnel de CD4 lié de façon covalente à un fragment dodécasaccharidique d’Héparane Sulfate dont la synthèse est complexe. Nous avons donc proposé de préparer des oligomaltosides sulfatés afin de déterminer s’ils pouvaient se comporter comme des mimes d’Héparane Sulfate.Dans un premier temps, nous avons mis au point la synthèse, en huit étapes et 38 % de rendement global, d’un précurseur trisaccharidique oligomérisable à partir de maltotriose, un trisaccharide biosourcé commercial. Au cours de ce travail, nous avons résolu trois points particulièrement délicats : l’allylation de l’extrémité réductrice du maltotriose, l’installation d’un groupement paraméthoxybenzylidène en position O-4III et O-6III et la protection sélective des positions O-6I et O-6II par un groupement silylé. Les optimisations menées nous ont permis de limiter la formation de produits secondaires, d’augmenter le rendement de chaque étape et de pouvoir mener sans problème cette synthèse sur une échelle d’une dizaine de grammes. Dans un deuxième temps, le précurseur trisaccharidique a été transformé en différents accepteurs et donneurs de glycosyle dont les comportements dans différentes conditions de glycosylation ont été étudiés. Nous avons ainsi pu démontrer qu’une activation des donneurs sous forme de trichloroacetimidate conduisait à des rendements faibles de part la formation d’une quantité importante des produits de réarrangement en trichloroacétamides anomériques. Une activation sous forme de N-Phényltrifluroacétimidate a permis de résoudre ce problème, sans toutefois que les rendements en soient toujours augmentés. En effet, nous avons pu montrer que la nature du groupement protecteur en O-6I du donneur a une influence déterminante sur l’issue de la réaction de glycosylation, tant au niveau de sa stéréosélectivité que de son rendement. Un groupement encombré ou un ester en O-6I du donneur est ainsi indispensable pour avoir une bonne stéréosélectivité alpha. Le meilleur rendement obtenu est, pour le moment, de 56 %. Des optimisations en cours permettront d’augmenter le rendement et de préparer les oligomaltosides sulfatés visés dans un avenir proche afin de tester leur activité biologique. / This work aims at simplifying the preparation of a new type of HIV entry inhibitor, conceived, synthesized and validated within a collaboration between our group, the "Institut de Biologie Structurale" (Grenoble) and the Institut Pasteur (Paris). This prototype is composed of a CD4 functional mimetic linked to a dodecasaccharide fragment of Heparan Sulfate, whose synthesis is complex. In order to determine if Heparan Sulfate may be replaced by simpler sulfated oligosaccharides, we decided to prepare a set of sulfated oligomaltosides.To this goal, we first optimized the synthesis of an oligomerizable maltotrioside building block in eight steps and 38% global yield from maltotriose, a commercial and biosourced trisaccharide. In this work, we had to address three major points: the allylation of the reducing end of maltotriose, the introduction of a paramethoxybenzylidene group between positions O-4III and O-6III and the selective protection of the remaining primary positions O-6I and O-6II by a silylated protecting group. Each step has been optimized to minimize the amount of secondary products and thus to enhance its yield. The resulting synthesis was thus shown to be highly reproducible up to ten grams scale.Then, glycoside acceptors and donors were prepared from the oligomerizable maltotrioside building block and we studied their behaviors in glycosylation reactions. We found that trichloroacetimidate activation led to poor glycosylation yields, due to the competitive formation of trichloroacetamidyl glycoside rearrangement product. Gratifyingly, N-phenyltrifluroacetimidate activation solved the rearrangement problem, but yields sometimes remained low. Indeed, we were able to demonstrate that the nature of the protecting group in position O-6I of the donor strongly influenced both the stereoselectivities and yields of the glycosylations: a bulky or ester group is needed in this position to obtain a full alpha stereoselecticity. To date, the highest yield obtained is 56 %.Ongoing optimizations will allow us to enhance the yields and to prepare the targeted sulfated oligomaltosides in a near future in order to test their biological activity.

<p>Heparan sulfate (HS) is a linear polysaccharide, located on the surface and in the extracellular matrix of most cells, that regulates functions of numerous proteins. HS-protein interaction is mainly mediated by sulfate groups found in N-sulfated (NS) regions of the HS, but may also involve rare HS substituents such as N-unsubstituted glucosamine (GlcNH₂) residues. The location of GlcNH₂ in an HS-epitope recognized by the monoclonal antibody 10E4, that specifically stains the prion lesions in scrapie-infected murine brain, suggests an involvement of GlcNH₂ in prion disease and other amyloid-related disorders. HS in general is strongly associated with amyloidosis, including Alzheimer’s disease (AD). Therefore, the aims of this thesis were to structurally characterize GlcNH₂-containing HS sequences found in native tissues, to further study HS epitopes recognized by 10E4, and to investigate the possible role(s) of GlcNH₂ and other HS structures in binding to amyloid β peptide (Aβ) (core material in AD plaque lesions, also stained by 10E4).</p><p>The GlcNH₂ content (0.7-4% of total disaccharide units) varied between HS from different tissues. Most GlcNH₂ units were found in poorly modified N-acetylated (NA-) or NA/NS-domains, located toward the polysaccharide-protein linkage region.</p><p>Binding of human cerebral cortex HS to Aβ(1–40) monomers requires N-, 2- and 6-O-sulfation of HS, while binding to Aβ fibrils requires N- and 2-O-sulfation only. GlcNH₂ units do not appreciably contribute to interaction with Aβ. Aβ fibril-binding HS domains also bind to fibroblast growth factor 2 (FGF-2), indicating that Aβ (neurotoxic) and FGF-2 (neuroprotective) may compete for common binding sites in HS. However, Aβ had no effect on FGF-2-induced MAPK signaling in NIH 3T3 fibroblasts.</p><p>Continued studies on 10E4-antigenic HS epitope(s) showed that binding of 10E4 to the previously identified antigenic tetrasaccharide, ∆UA-GlcNH₂-GlcA-GlcNAc, requires the nonreducing hexuronic acid (∆UA) to be 4,5 unsaturated (induced by lyase cleavage), and thus is artificial. Further studies are needed to clarify the potential involvement of GlcNH₂ in 10E4-recognition of the native HS epitope(s).</p>

<p>In the extracellular space, many specialized proteins are located to support cells and to mediate cell-cell signalling. One class of such molecules is heparan sulfate (HS) proteoglycans, which are proteins with different properties and locations but all of them decorated with long unbranched HS polysaccharide chains. During biosynthesis the HS chains are modified by sulfation and a C5-epimerase converts some glucuronic acid residues to iduronic acid. The patterns of the modifications vary distinctly between tissues and developing stages and give HS chains different affinity for biologically important proteins. Thus, the regulation of HS biosynthesis is likely to influence a wide variety of biological events.</p><p>This thesis focuses on the biosynthesis of HS in animals with targeted disruptions in genes important for HS production. The N-deacetylase N-sulfotransferase (NDST) is a key enzyme in HS biosynthesis, directing other modifications. We show that NDST isoforms have very different roles in HS biosynthesis. Inactivation of NDST1 affects HS biosynthesis in all tissues. In embryonic liver HS from NDST1-/- mice the N-sulfation was decresed with twothirds, while the absence of NDST2 did not affect HS structure. In the absence of NDST1 in the liver, however, NDST2 is active in HS N-sulfation. </p><p>In a study of embryonic stem cells lacking both NDST1 and NDST2, no N-sulfate groups could be detected. 6-O-sulfate groups were, however, still present at half of its normal level. This was an unexpected finding since 6-O-sulfotransferases have been thought to be strictly dependent on N-sulfate groups for substrate recognition.</p><p>By adapting an automated method for HS analysis to mammalian tissues, we could extend our analyses to more tissues and other transgene models. We also developed a protocol to create a sensitive “fingerprint” of HS structure. With these methods we could determine the individual HS structure of different mouse tissues. </p>

<p>Heparan sulfate (HS) is a ubiquitous polysaccharide on the cell surface and in the extracellular matrix. HS is an important actor in the regulation of cell signaling, especially in the developing embryo. In combination with cell culture and biochemical experiments, <i>in vivo</i> studies of genetically modified animals have pointed out the sulfation pattern of HS as highly important for binding of ligands, their receptors and other signaling modulators.</p><p>The sulfation pattern of an HS chain is gained by several modifying steps, performed by multiple enzymes during biosynthesis in the Golgi apparatus. By alterations of sulfation pattern, and the amount of sulfate groups, a cell can regulate the binding properties of its HS to different molecules. The most highly sulfated form of HS is called heparin, and can only be found intracellularly in mast cells.</p><p>This thesis describes the phenotypes and the alterations in HS/heparin biosynthesis of two genetically modified mouse strains deficient in N-deacetylase/N-sulfotransferase-1 (NDST1) and -2 (NDST2) respectively. We have found NDST1 to be important for correct sulfation of HS and that NDST2 is crucial in heparin biosynthesis. NDST2 deficient mice completely lack heparin and therefore have a severe mast cell phenotype. NDST1 deficient mice produce undersulfated HS and show several developmental disturbances. Some NDST1 embryos die in utero while the rest die neonatally due to breathing difficulties. Defect brain, eye and skeletal development has also been observed while some organs, such as the liver, appear to be largely unaffected. Several phenotypes are similar to defects seen in other mouse strains with impaired fibroblast growth factor and bone morphogenetic protein signaling, among others. This suggests the phenotypes of NDST1 deficient embryos to be of a multi factorial origin, in complete accordance to the many signaling pathways HS is suggested to modulate.</p>

<p>Heparan sulfate (HS) proteoglycans consist of sulfated HS chains covalently bound to core proteins. They are ubiquitously expressed, on the cell surface and in the extracellular matrix, throughout the body. During biosynthesis the HS chain is modified to generate a highly variable pattern of sulfated residues, able to interact with a wide variety of ligands, such as growth factors, morphogens and extracellular matrix molecules. The presence of HS proteoglycans is crucial during various developmental processes as they are involved in generation of morphogen gradients and influence the function of several growth factor pathways essential for tissue assembly and differentiation.</p><p>In this thesis the phenotypes of two mouse strains, deficient in different isoforms of the HS biosynthetic enzyme N-deacetylase/N-sulfotransferase (NDST) have been analyzed. In addition, NDST deficient embryonic stem (ES) cells have been analyzed with regard to HS structure and differentiation capacity. Mice deficient in NDST1 die peri-natally. The embryos display an overall low-sulfated HS and several developmental defects, with a lung phenotype as the predominant cause of death. Mice deficient in NDST2 lack sulfated heparin in connective tissue type mast cells while HS structure is unaltered. These results indicate that NDST1 is the isoform mainly responsible for HS biosynthesis during development. However, NDST1/2 deficient embryos do not survive beyond E5.5 and have a greatly disturbed morphology, suggesting that NDST2 has an essential role during early embryonic development. HS synthesized by NDST1/2 deficient ES cells had a total lack of N-sulfate groups while, interestingly, about half of the 6-O-sulfate groups remained. This result was unexpected since 6-O-sulfotransferases have been thought to be strictly dependent on N-sulfate groups for substrate recognition. Further characterization of the NDST1/2 deficient ES cells during in vitro differentiation demonstrated that the expression pattern of markers for all three germ layers was disturbed. In addition, it was demonstrated that NDST1 is not needed for mast cell development, that lack of NDST2 results in abnormal mast cells and that no mast cells is formed from NDST1/2 deficient ES cells.</p>

Heparan sulfate (HS) is a linear polysaccharide, located on the surface and in the extracellular matrix of most cells, that regulates functions of numerous proteins. HS-protein interaction is mainly mediated by sulfate groups found in N-sulfated (NS) regions of the HS, but may also involve rare HS substituents such as N-unsubstituted glucosamine (GlcNH2) residues. The location of GlcNH2 in an HS-epitope recognized by the monoclonal antibody 10E4, that specifically stains the prion lesions in scrapie-infected murine brain, suggests an involvement of GlcNH2 in prion disease and other amyloid-related disorders. HS in general is strongly associated with amyloidosis, including Alzheimer’s disease (AD). Therefore, the aims of this thesis were to structurally characterize GlcNH2-containing HS sequences found in native tissues, to further study HS epitopes recognized by 10E4, and to investigate the possible role(s) of GlcNH2 and other HS structures in binding to amyloid β peptide (Aβ) (core material in AD plaque lesions, also stained by 10E4). The GlcNH2 content (0.7-4% of total disaccharide units) varied between HS from different tissues. Most GlcNH2 units were found in poorly modified N-acetylated (NA-) or NA/NS-domains, located toward the polysaccharide-protein linkage region. Binding of human cerebral cortex HS to Aβ(1–40) monomers requires N-, 2- and 6-O-sulfation of HS, while binding to Aβ fibrils requires N- and 2-O-sulfation only. GlcNH2 units do not appreciably contribute to interaction with Aβ. Aβ fibril-binding HS domains also bind to fibroblast growth factor 2 (FGF-2), indicating that Aβ (neurotoxic) and FGF-2 (neuroprotective) may compete for common binding sites in HS. However, Aβ had no effect on FGF-2-induced MAPK signaling in NIH 3T3 fibroblasts. Continued studies on 10E4-antigenic HS epitope(s) showed that binding of 10E4 to the previously identified antigenic tetrasaccharide, ∆UA-GlcNH2-GlcA-GlcNAc, requires the nonreducing hexuronic acid (∆UA) to be 4,5 unsaturated (induced by lyase cleavage), and thus is artificial. Further studies are needed to clarify the potential involvement of GlcNH2 in 10E4-recognition of the native HS epitope(s).