No description available.

La O-fucosylation, catalysée par Pofut1, est une glycosylation rare qui consiste en l’ajout d’un fucose O-lié sur la sérine ou la thréonine d’une séquence consensus (C2X4(S/T)C3), portée par un domaine EGF-like (ELD) d’une glycoprotéine membranaire ou sécrétée. Notre analyse de la lignée murine Pofut1cax/cax, hypomorphe pour le gène Pofut1, a révélé une hypertrophie musculaire post-natale associée à une diminution du pool de cellules satellites. Ce phénotype est en partie associé à un défaut d’interaction entre les récepteurs NOTCH hypo-O-fucosylés des myoblastes dérivés de cellulessatellites (MDCS) et leurs ligands DSL, ce qui aboutit à une plus faible activation de la signalisation Notch. D’autres protéines potentiellement impliquées dans la myogenèse peuvent également être la cible de POFUT1. C’est notamment le cas de la protéine Wnt inhibitory factor 1 (WIF1), qui dispose de cinq ELDs, dont deux sont potentiellement aptes à recevoir un O-fucose (ELDs III et V). Par une approche phylogénétique, nous avons montré la conservation de ces deux sites de O-fucosylation et de deux sites de N-glycosylation chez la plupart des bilatériens. Nos expériences démontrent l’occupationde tous ces sites, excepté le site de O-fucosylation de l’ELD V, chez la protéine WIF1 murine. La capacité de l’ELD III, produit de manière isolée, à recevoir un fucose O-lié a été démontrée après O-fucosylation in vitro, par l’association de cycloaddition azide-alcyne assistée au cuivre (CuAAC) et de spectrométrie de masse en mode MRM. Cette nouvelle approche expérimentale a par la suite été standardisée et sa sensibilité évaluée en comparant deux autres ELDs (ELDs 12 et 26 de NOTCH1) connus pour être O-fucosylés mais présentant des affinités différentes pour POFUT1. De façonsurprenante, l’ELD V de WIF1 ne peut être O-fucosylé, probablement en raison d’un clash stérique entre cet ELD et POFUT1, prévenant ainsi leur interaction. L’analyse de la protéine WIF1 entière a confirmé les résultats obtenus sur les ELDs isolés et démontre l’occupation des deux sites de N-glycosylation. Enfin, nos résultats montrent également l’importance de ces deux N-glycanes, mais également celle du O-fucose de l’ELD III, pour une sécrétion optimale de la protéine WIF1 murine. / The, Pofut1-catalyzed O-fucosylation, is a rare glycosylation which consists of the addition of an O-linked fucose to the serine or threonine of a consensus sequence (C2X4(S/T)C3), carried by an EGF-like domain (ELD) of a membrane or secreted glycoprotein. Our analysis of the murine line Pofut1cax/cax, hypomorphic for the Pofut1 gene, revealed post-natal muscle hypertrophy associated with a decrease in the satellite cell pool. This phenotype was partly associated with a lack of interaction between hypo-O-fucosylated NOTCH receptors of satellite cell-derived myoblasts (SCDM) and their DSL ligands, which resulted in a lower activation of Notch signaling. Other proteins potentially involved in myogenesis may also be the target of POFUT1. This is indeed the case for the protein Wnt inhibitory factor 1 (WIF1), which has five ELDs, whose only two are potentially able to receive an O-fucose (ELDs III and V). Using a phylogenetic approach, we showed in most bilaterians that these two O-fucosylation sites and two N-glycosylation sites were conserved. Our experiments showed theoccupation of all these sites, except for the O-fucosylation site of murine WIF1 protein ELD V. The ability of the ELD III, produced as an isolated protein, to receive O-linked fucose was demonstrated after an in vitro O-fucosylation by combination of copper-catalysed azide-alkyne cycloaddition (CuAAC) and MRM-mass spectrometry. This new experimental approach was then standardized and its sensitivity was evaluated by comparing two other ELDs (NOTCH1 ELDs 12 and 26) known to beO-fucosylated but with different affinities for POFUT1. Surprisingly, WIF1's ELD V could not be O-fucosylated, probably due to a steric clash between this ELD and POFUT1, thus preventing their interaction. The analysis of the full-length WIF1 protein confirmed our results obtained with isolated ELDs and demonstrated the occupation of the two N-glycosylation sites. Finally, our results also showed the importance of these two N-glycans, but also the importance of ELD III’s O-fucose, foroptimal secretion of the murine WIF1 protein.

Photodynamic therapy (PDT) is an emerging therapy for the treatment of cancer and various other human disorders. 5-Aminolaevulinic acid (ALA) is a simple natural product that is of great interest for PDT because it can be converted within cells via the haem biosynthetic pathway to the photosensitiser, protoporphyrin IX (PpIX). ALA-PDT has become a first line clinical approach for the treatment of cancerous and precancerous skin lesions (e.g Bowen’s disease, basal skin carcinomas, and actinic keratosis) that would otherwise require significant conventional surgery. However, ALA being a zwitterion suffers from poor lipid solubility and at the same time has stability issues at physiological or alkaline pH. The work herein describes some novel strategies to enhance the delivery of ALA to specific cell types using targeted ALA dendrimeric prodrugs. Specifically, it describes the synthesis of molecules consisting of branched units with 3 or more copies of ALA attached to a central core structure (e.g. gallic acid) using copper-catalysed azide-alkyne click chemistry (CuAAC). Selective delivery of the dendrimeric ALA cargo was achieved by attachment of a homing peptide to an independently addressable functional group on the prodrug core. As proof of concept of this approach, systems were prepared containing a peptide that allows selective targeting of the epidermal growth factor receptor (EGFR) which is overexpressed in a variety of tumours. Targeted ALA delivery and PpIX production was studied with these prodrugs in EGFR-expressing breast adenocarcinoma cells (MDA-MB-231 cells) and a peptide-targeted derivative with 9 ALA units was found to have enhanced PDT efficacy compared to an equimolar dose of ALA. Other targeting units that have been attached to these dendrimeric ALA prodrugs include biomolecules such as vitamin E, thymidine (a nucleoside) and a glucose derivative. Additionally, strain-promoted azide-alkyne cycloadditions (SPAAC) of the same EGFR-targeting peptide with some classical photosensitisers were also investigated and biological studies in EGFR-overexpressing cell lines were carried out. Lastly, a group of cell penetrating peptide-ALA conjugates have been synthesised via CuAAC as a novel approach for targeted ALA delivery.

Resorcin[4]arenes and their rigid structures “cavitands” offer a unique molecular structure scaffolds suitable for attaching and orienting multi-ligands on their upper rims. Their remarkable structure properties, the relative ease of chemical modification on their either rims or through the multiple phenolic hydroxyl (-OH) groups, manipulating their cavity size and the role of spatial directionality provided by their cores, led synthetic chemist to explore their utilities in the synthesis of valuable resorcin[4]arene/cavitand assemblies with wide range of applications in many research areas such as supramolecular chemistry, host-guest chemistry, glycoconjugates, polymer chemistry, host-guest chemistry, micro-reactors in chemical catalysis and others. In particular, in the field of glycoconjugates, the synthesis of multi-directional glycoresorcin[4]arenes have found valuable uses in glycobiology, i.e. lectins and cells recognitions. Recently, the spatial directionality of β-D-glucopyranoside based on the resorcin[4]arene rigid structure “cavitand” was first described by our group in the synthesis of resorcin[4]arene cavitand glycoconjugates (RCGs). The efficacy of RCGs was established as phase catalysts in order to evaluate their abilities catalyzing different organic transformations in aqueous environment as a green approach in organic syntheses. RCGs showed a variety of advantages such as low catalytic loading and faster conversions for water-insoluble hydrophobic substrates in aqueous media. This dissertation consists of four chapters. Chapter 1 describes the design and the synthesis of a family of spatially directional resorcin[4]arene cavitand glycoconjugates (RCGs) by coupling β-D-glucopyranose moieties on the resorcin[4]arene cavitand upper rims via multiple 1,4-disubstituted 1,2,3-triazolethe linkages achieved from the well-known click reaction, the Cu(I)-catalyzed azide-alkyne cycloaddition (CuAAC). The multiple directional β-D-glucopyranoside units on the cavitand cores are capable of multiple H-bond interactions resulting in a significant molecular host system, i.e. “pseudo-saccharide cavity” capable of accommodating different guest molecules, especially hydrophobic organic species, in aqueous environment raising the advantages of these amphiphilies for appreciable applications as molecular vessels and micro-reactors in an ideal atmosphere “water”. In Chapter 2, the utility of RCGs as efficient phase transfer catalysts for thiocyanation and thiazole formation in water is described. The catalytic activity (1 mol% loading) of RCGs enhancing the formation of both thiocyanate and thioazole species in aqueous media is a direct evidence of their capability acting as micro-reactors in water. Also, the recoverability and the reusability of RCGs for thiocyanation and thizaole formation have been examined. Additionally, the influence of the flexibility/rigidity and the orientation of the multiple β-D-glucopyranose moieties on the RCGs’ catalytic efficciency in thiocyanation and thiazole formation has been investigated. In Chapter 3, the design and the synthesis of an octopus-like structure resorcin[4]arene glycoconjugate (RG) is described. RG consists of eight β-D-glucopyranoside units attached to a flexible open system resorcin[4]arene core. The flexibility of the resorcin[4]arene assembly and the presence of the multiple β-D-glucopyranosides provides a wider-type of pseudo-saccharide bucket capable of encapsulating larger molecules in aqueous environment. RG was evaluated for catalyzing the CuAAC in water without the use of any added organic solvent. RG was remarkably efficient in catalyzing the CuAAC reactions in aqueous atmosphere for a variety of small/bulky and hydrophilic/hydrophobic azido and alkyne substrates. In Chapter 4, the construction of resorcin[4]arene cavitand based-carcerands (RCCs) via tandem CuAAC reactions is described. The CuAAC reaction was utilized as an efficient coupling method for the complimentary azido- and alkyne-cavitand halves via multiple 1,4-disubstituted 1,2,3-triazole linkers. Unlike the most reports of covalently linked carcerand synthesis, which are accomplished at very dilute concentration, the fast kinetics of the CuAAC reaction enabled carcerand synthesis at much higher concentration and in high yield.

L'une des préoccupations de la chimie moderne est de développer des procédés éco-compatibles : une tendance consiste à remplacer les solvants organiques par les Liquides Ioniques (LI). Ainsi ces travaux décrivent l'utilisation des LI dans deux réactions à économie d'atomes : l'addition de Michael et la réaction de cycloaddition 1,3-dipolaire catalysée (ou non) par un sel de cuivre (CuAAC). La première partie des travaux relate l'emploi de Liquides Ioniques Super-Acides (LISA), connus pour générer une activation électrophile. Une optimisation du LISA a été effectuée sur une réaction-modèle, puis l'application à d'autres électrophiles et nucléophiles a été étudiée. La pureté du LISA influe sur la chimiosélectivité : s’il est partiellement hydrolysé, la réaction d'annélation de Robinson devient prépondérante. La version asymétrique du processus a été abordée, montrant qu'il est possible d'obtenir un excès énantiomérique à partir de dérivés d'acides aminés. La seconde partie de l'étude a permis de mettre au point une synthèse de (triazolylméthyl)vinylphosphonates à partir d'un acétoxyméthylvinylphosphonate selon une procédure monotope reposant sur la cycloaddition 1,3-dipolaire dans différents LI. Nous avons ensuite montré que le LI joue aussi le rôle d'activateur pour cette réaction. / In the context of sustainable chemistry, an alternative to conventional organic solvents is the use of ionic liquids. These works reported here aims to describe the use of ionic liquids (IL) in two atoms economy reactions, namely the Michael addition reaction and 1,3-dipolar cycloaddition catalyzed (or not) by a copper salt (CuAAC). In the first part of the work reports the use of Super-Acid Ionic Liquids (SAIL), which initiate an electrophilic activation. After optimization of SAIL on a reaction model, application to other electrophiles and nucleophiles is discussed. The purity of SAIL affects the chemioselectivity: if the SAIL is partially hydrolyzed, a Robinson annulation predominates. The asymmetric version of the process is investigated, showing that it is possible to carry out an enantioselective reaction with amino acid derived SAIL. The second part of the study deals with setting up an original synthesis of (triazolylmethyl)vinylphosphonate from acetoxymethylvinylphosphonate using a one-pot procedure involving a 1,3-dipolar cycloaddition in different LI. We further showed that the ionic liquid can also act as an activator for this reaction.

O-GlcNAcilação, ou modificação por O-GlcNAc, consiste na glicosilação de proteínas envolvidas em processos celulares fundamentais, e a sua desregulação tem sido associada a importantes doenças tais como diabetes tipo 2, doenças neurodegenerativas, cardiovasculares e câncer. A O-GlcNAcilação é regulada por duas enzimas: O-GlcNAc transferase (OGT) e O-GlcNAcase (OGA). Muitos inibidores da OGA descritos apresentam falta de seletividade entre a OGA e as isoenzimas Hex A e B lisossomais, o que pode ocasionar acúmulo de gangliosídeos no interior dos lisossomos e doenças neurodegenerativas. Desta forma, visando a obtenção de inibidores mais seletivos, foram propostos neste estudo a síntese e avaliação biológica de compostos derivados de N-acetilglicosamina, contendo na posição C-2 o anel 1,2,3-triazólico 1,4-di-substituído com diferentes cadeias laterais, na maioria contendo anéis aromáticos na sua extremidade, e a investigação da influência da extensão da cadeia entre os anéis na cavidade da OGA. Foram obtidos os derivados triazólicos 1, 2 (no estudo anterior) e 8a-j (no presente trabalho), em rendimentos bons a moderados, por meio da estratégia de click chemistry, envolvendo a reação de CuAAC entre o intermediário azido glicopiranosídeo 6, e dez diferentes alcinos, dos quais cinco foram previamente sintetizados (9 - 11, 14 e 15), e cinco disponíveis comercialmente. Para a síntese dos precursores acetilênicos 9 - 11, 14 e 15 foram utilizadas diferentes estratégias sintéticas de acordo com cada alcino, como, por exemplo, reações de substituição nucleofílica e aminação redutiva. Os derivados triazólicos foram obtidos como mistura de anômeros ?:? na proporção de 10:1, respectivamente. Devido à seletividade da enzima OGA apenas para substratos ?, os compostos 8a-h foram purificados por CLAE-DAD para separação das misturas. Posteriormente, foi possível realizar a cristalização do azido 6 em etanol, com consequente separação do anômero ? deste intermediário, o que auxiliou na etapa sintética dos derivados 8i e 8j como anômeros puros. Os ensaios de MTT não evidenciaram citotoxicidade para os compostos, avaliados em 1,0 ?M. Os ensaios de western blot e enzimáticos dos derivados 1, 2 e 8a-j demonstraram que apenas os compostos 1, 8i e 8j foram ativos para OGA, com valores de IC50 de 0,49, 0,52 e 0,72 ?M, respectivamente. A partir deste resultado, foi possível sugerir que a extensão ideal da cadeia ligada entre os anéis triazólico e aromático é de dois carbonos (1), podendo acomodar um heteroátomo de nitrogênio (8i) ou oxigênio (8j). Ademais, os ensaios enzimáticos dos compostos 1, 8i e 8j, avaliados para as Hex A e B, apresentaram IC50 de 550, 569 e 571 ?M, respectivamente, sugerindo alta seletividade destes compostos para a OGA em detrimento das isoenzimas lisossomais. Com o trabalho desenvolvido em parceria com a Dra. Nuria E. Campillo (CSIC) foram planejados novos derivados não-carboidratos capazes de realizar interações com importantes resíduos do sítio catalítico da OGA e com potencial atividade biológica. / O-GlcNAcylation, or O-GlcNAc modification, consists of the glycosylation of proteins involved in fundamental cellular processes, and its deregulation has been linked to important diseases such as type 2 diabetes, neurodegenerative, cardiovascular diseases, and cancer. O-GlcNAcylation is regulated by two enzymes: O-GlcNAc transferase (OGT) and O-GlcNAcase (OGA). Many OGA inhibitors described exhibit lack of selectivity between OGA and the lysosomal isoenzymes Hex A and B, which can result in the accumulation of gangliosides within the lysosomes and neurodegenerative diseases. Therefore, aiming to obtain more selective inhibitors, this study proposed the synthesis and biological evaluation of N-acetylglucosamine derivatives, containing at the C-2 position the 1,4-di-substituted 1,2,3-triazole ring bearing different side chains, most of them containing aromatic ring at their end, and the investigation of the side chain extension influence between the rings into OGA active site pocket. It was obtained the triazole derivatives 1, 2 (in the previous study) and 8a-j (in the current project) in good to moderate yield through the click chemistry strategy, involving the CuAAC reaction between azide glucopyranoside intermediate 6 and ten different alkynes, which five were previously synthetized (9 - 11, 14 and 15), and five commercial available. For the synthesis of the acetylenic precursors 9 - 11, 14 and 15 it was employed different synthetic strategies according to each alkyne, for example nucleophilic substitution reactions and reductive amination. The triazole derivatives were obtained as a mixture of ?:? anomers in the proportion of 10:1, respectively. Due to the OGA selectivity to ? substrates, compounds 8a-h were purified by HPLC-DAD in order to separate the mixtures. Later in this study, it was possible to obtain the crystallization of azide 6 in ethanol, resulting in the separation of the ? anomer from this precursor, which assisted in the synthetic step of the derivatives 8i and 8j as pure anomers. The MTT assays did not show cytotoxicity for the synthesized compounds, evaluated at 1.0 ?M. The western blot and the enzymatic assays, evaluated for compounds 1, 2 and 8a-j, demonstrated that only compounds 1, 8i e 8j were active towards OGA, with IC50 values of 0.49, 0.52 e 0.72 ?M, respectively. From this result, it was possible to suggest that the ideal side chain extension linked between the triazole and aromatic rings is of two-carbon atoms (1), being able to accommodate one heteroatom of nitrogen (8i) or oxygen (8j). Furthermore, the enzymatic assays, evaluated for compounds 1, 8i e 8j against Hex A and B, exihibted IC50 of 550, 569 e 571 ?M, respectively, suggesting a high selectivity of these compounds to OGA rather than the two lysosomal isoenzymes. In the project developed in partnership with Dra. Nuria E. Campillo (CSIC) it was designed new non-carbohydrates compounds capable of interacting with important residues of the OGA catalytic site with potential biological activity.

Deux types de matériaux membranaires ont été développés pour la séparation du CO2. Les premiers associent un matériau polymère de référence (Pebax®) à de nouveaux additifs pseudopeptidiques bioconjugués. Ces bioconjugués pseudopeptide-polymère ont été obtenus par une stratégie de synthèse de type « grafting-to » à partir de pseudopeptides 1:1[α/α-Nα-Bn-hydrazino] dimère et tétramère fonctionnalisés par une fonction alcyne. La partie oligomérique poly(acrylate de diéthylène glycol) (PEDEGA) a été synthétisée de façon contrôlée par « Single Electron Transfer Living Radical Polymerization » (SET-LRP) à partir d’un amorceur azido fonctionnel permettant un couplage direct par chimie « click » CuAAC. L'influence de ces additifs sur les propriétés de sorption et de séparation du CO2 a été analysée en termes de relations propriétés-morphologie-structure. Ces additifs originaux ont permis d’augmenter les performances de séparation du CO2 de la membrane référence ; en améliorant de 46% la perméabilité au CO2, tout en maintenant de bonnes sélectivités αCO2/N2 = 44 et αCO2/CH4 = 13. Par ailleurs, deux familles de poly(uréthane-imide)s (PUIs) à architecture contrôlée ont été développées dans l'objectif d'obtenir des matériaux avec des proportions très élevées en unités oxyde d'éthylène tout en évitant leur cristallisation. Des PUIs multi-blocs linéaires ont été synthétisés par polycondensation avec différentes tailles de bloc souple polyéther Jeffamine, correspondant à des taux massiques de phase souple variant entre 40% et 70%. Afin d'augmenter encore la proportion de phase souple jusqu'à la valeur très élevée de 85%, des PUIs multi-blocs greffés ont ensuite obtenus par une stratégie de "grafting-to" à partir d'un précurseur PUI comportant des fonctions alcyne latérales et d'oligomères azido-PEDEGA de différentes masses molaires. L’évolution de leurs performances de séparation du CO2 a été corrélée à leur proportion de phase souple et morphologie et à leur capacité d’interaction avec le CO2. Au taux maximal de phase souple (85%), on obtient des performances élevées (PCO2 = 196 Barrer ; αCO2/N2 = 39 et αCO2/CH4 = 12). Ainsi, par rapport au PUI greffable précurseur, le greffage d'oligomères PEDEGA a permis de multiplier la perméabilité jusqu’à 17 fois en maintenant la valeur élevée de αCO2/N2 constante et avec une diminution de seulement 30% de αCO2/CH4. En permettant d'obtenir une proportion très élevée de phase souple non cristalline, la stratégie de greffage s'est donc révélée exceptionnellement efficace pour améliorer les performances des copolymères multi-blocs PUI pour la capture du CO2. / In this PhD thesis, two types of membrane materials were developed for CO2 separation. The first ones associate a reference polymer material (Pebax®) with new pseudopeptidic bioconjugates additives. These pseudopeptide-polymer bioconjugates were obtained by a "grafting-to" synthetical pathway from alkyne-functionalized 1:1[α/α-Nα-Bn-hydrazino] dimer and tetramer pseudopeptides. Poly(diethylene glycol acrylate) (PEDEGA) oligomeric part was synthesized under controlled conditions using Single Electron Transfer Living Radical Polymerization (SET-LRP) from an azido-functionalized initiator allowing direct coupling via CuAAC "click" chemistry. The influence of these additives on CO2 sorption and separation properties was analyzed in terms of properties-morphology-structure relationships. These original additives allowed to enhance CO2 separation performances of the reference membrane, increasing CO2 permeability by 46%, and maintaining good selectivities αCO2/N2 = 44 et αCO2/CH4 = 13. In another work, two families of poly(urethane-imide)s (PUIs) with controlled architecture were developed for obtaining membrane materials with high content in ethylene-oxide units while avoiding their crystallization. Linear multi-blocks PUIs were first synthesized by polycondensation with different sizes of Jeffamine polyether soft block, corresponding to soft block contents varying from 40 to 70%wt. To further increase the soft phase content until a very high level (85%wt), grafted multi-blocks PUIs were obtained by a "grafting-to" strategy from an alkyne-functionalized precursor PUI and azido-PEDEGA oligomers with different molar weights. The evolution of their CO2 separation performances were correlated to their soft phase content, morphology and CO2 sorption ability. For the maximum soft phase content (85%wt), high performances were obtained for CO2 separation (PCO2 = 196 Barrer ; αCO2/N2 = 39 et αCO2/CH4 = 12). Compared to the precursor PUI, the grafting strategy allowed to increase CO2 permeability 17-fold. At the same time, the good selectivity αCO2/N2 was maintained with a slight decrease (-30%) of the selectivity αCO2/CH4. By allowing very high non-crystalline soft phase contents, the grafting strategy was exceptionally efficient in improving multi-block copolymer performances for CO2 capture

Cette thèse présente une nouvelle méthode d’assemblage de nanoparticules (NPs) magnétiques par une approche de chimie « click » sur des surfaces fonctionnalisées par des molécules organiques (self-assembled monolayers, SAMs). Le contrôle précis de la nanostructure des assemblages permet d’étudier les propriétés magnétiques collectives des NPs modulées par les interactions dipolaires. Un état de l’art des travaux décrits dans la littérature est présenté sur la synthèse de NPs d’oxydes métalliques, les enjeux et méthodes d’assemblage de NPs magnétiques ainsi que les propriétés magnétiques des NPs. Plusieurs types de NPs d’oxydes de fer présentant différentes tailles, morphologies (sphère, cube, tétrapode) et structures internes (coeur-coquille) sont synthétisés par la méthode de décomposition thermique La méthode développée d’assemblage des NPs consiste en une réaction dite « Copper (I) catalyzed alkyne-azide cycloaddition » (CuAAC), basée sur les interactions spécifiques entre des groupes alcyne et azoture présents à la surface des NPs et des SAMs. La fonctionnalisation des NPs et la préparation des SAMs sont également décrites. En modulant les conditions opératoires lors de la réaction d’assemblage, différentes nanostructures (films denses, NPs spatialement isolées ou chaînes de NPs) sont obtenues dans le but d’étudier les propriétés magnétiques collectives des NPs dans les assemblages. Un effet de la distance interparticule et de la mise en forme des NPs en 2D sur les interactions dipolaires sont mis en évidence. La réaction de click pour l’assemblage de NPs est ensuite perfectionnée, d’une part, la diminution du temps de la réaction d’assemblage en présence de radiations microondes, et d’autre part la préparation de films multicouches de NPs « one-pot » en contrôlant la réaction de click par électrochimie. Enfin des dispositifs magnéto-résistifs sont élaborés en déposant des assemblages denses de nanoparticules entre deux électrodes distantes d’une centaine de nanomètres. / This thesis aims to develop a new method to control the assembling of iron oxide magnetic nanoparticles (NPs) by “click” chemistry onto surfaces functionalized by organic molecules (self-assembled monolayers, SAMs). The high control on the nanostructure of NP assemblies results in the study of the collective magnetic properties. The state of the art, on the NPs synthesis, on the issues and methods of the assembling of magnetic NPs and on the magnetic properties of NPs films is established. The NPs of different sizes, morphologies and inner structures are synthesised by thermal decomposition. Self assembled monolayers (SAMs) are deposited on the substrates to address their surface for the NPs assembling. This assembling is achieved by specific interactions between alkyne and azide groups worn by the NPs and the SAM by theCopper Catalyzed Alkyne-Azide Cycloaddition (CuAAC) reaction which leads to covalent and irreversible anchoring of the NPs onto the substrate’s surface. Various nanostructures of NP assemblies (spatially isolated NPs, dense films, NPs chains) are obtained in order to study their magnetic properties. The interparticule distance and the dimensionality of the assembly (2D, 1D) are shown to strongly modulate dipolar interactions. The assembling by click chemistry is then improved by two means: (i) microwave activation to shorten the reaction time and (ii) control of the reaction by electrochemistry and 3D films preparation. This method is suitable for the elaboration of magnetoresitive devices by deposition of dense films of NPs between electrodes remote from a hundred of nanometers.

Les premiers iminosucres multivalents rapportés dans la littérature datent de 1999. Depuis, c’est plus d’une centaine de clusters de ce type qui ont été synthétisés et décrits une quarantaine de publications. L’obtention d’un premier effet multivalent fort sur une glycosidase, en 2010, a initié de nouveaux travaux visant à étudier et à comprendre son mécanisme et ses limites. Dans cette optique le présent travail de thèse a exploité plusieurs approches. La première partie décrit la synthèse de dendrons "cliquables" permettant de multiplier par trois ou par neuf la valence initiale des plateformes utilisées. La deuxième est une étude de la synthèse d’espaceurs rigides. La troisième est la préparation de plateformes modulables, des neo-cyclodextrines, pour obtenir un contrôle plus fin de la topologie des clusters. L’association de dendrons "cliquables" et de plateformes cyclopeptoïdes de dimensions contrôlées a abouti à un résultat sans précèdent en termes d’effet multivalent. Ainsi, le cluster 36-valent à ligands DNJ est un inhibiteur 170 000 fois plus fort que l’analogue monovalent correspondant sur l’alpha-mannosidase des pois sabre blanc (Jack Bean). / The first multivalent iminosugars were published in 1999. From this date, it’s more than a hundred of clusters that were synthesized and presented in about forty publications. In 2010, the first strong multivalent effect in glycosidase inhibition was obtained and prompted further studies of its mechanism and its limits. To reach these goals, this PhD work has developed different strategies. The first was to synthesize "clickable" dendrons which can lead to a multiplication of the initial valency of our scaffolds by three or by nine. The second approach was a study to obtain rigid linkers. The third one was the preparation of modular scaffolds, neo-cyclodextrins, in order to finely tune the topology of the resulting clusters. The combination of our "clickable" dendrons with cyclopeptoid scaffolds gave an unprecedented multivalent effect on glycosidase inhibition. The 36-valent DNJ-based cluster is indeed a 170 000-fold more potent inhibitor than the corresponding monovalent control for Jack Bean alpha-mannosidase.

La réaction de cycloaddition entre un azoture et un alcyne catalysée par le cuivre (I) (CuAAC) pour former un 1,2,3-triazole 1,4-disubstitué est très utilisée dans de nombreux domaines de la chimie. Cette réaction a très vite été utilisée en synthèse peptidique notamment du fait du caractère isostère du noyau triazole et de la liaison amide. Toutefois l'impact de l'insertion d'un triazole sur la structure secondaire de peptides n'a été que faiblement exploré. Ainsi, pour étudier l'effet d'un tel remplacement nous avons choisi deux peptides modèles structurés le premier modèle est un peptide linéaire de la famille des peptaibols et le second est un peptide cyclique la tamandarine B. Dans un premier temps, nous avons optimisé une voie de synthèse de peptaibols que nous avons appliqué à l'alaméthicine F50/5 et à un analogue de la bergofungine D. Pour cela, nous avons utilisé une synthèse peptidique en phase solide sous irradiation micro-ondes en profitant d'un cocktail de réactifs efficaces contenant du diisopropylcarbodiimide comme agent de couplage et de l'oxyma. Cette méthode de synthèse a été étendue à l'obtention d'analogues silylés de l'alamethicine en substituant dans différentes positions l'acide aminobutyrique par un résidu hydrophobe et encombré : la bis-triethylsilyl-dipropylglycine (TES-Dpg). Dans la deuxième partie nous avons développé la synthèse de dipeptides à motif triazole. Puis nous avons défini les conditions réactionnelles nécessaires pour leur utilisation en SPPS et synthétisé des peptides contenant plusieurs motifs triazoles. Ces dipeptides ont ensuite été utilisés pour réaliser un scan triazole sur les deux peptaibols modèles. Les études structurales par dichroïsme circulaire, RMN et les tests biologiques de différents analogues nous permettent de conclure que le triazole affecte la structure secondaire des peptaibols et par conséquence induit une perte d'activité. Ainsi il apparaît que le concept de triazole comme isostère de la liaison peptidique doit être employé avec prudence. Dans l'optique de comprendre l'impact de triazole sur l'activité et la structure secondaire de peptides cycliques, nous avons généré des analogues simplifiés de la tamandarine B un depsipeptide cyclique d'origine marine. Nous rapportons les résultats préliminaires de cette étude. / The cycloaddition reaction between an azide and an alkyne catalyzed by copper (I) (CuAAC) to form a 1,2,3-triazole 1,4-disubstituted is widely used in many areas of chemistry. This reaction was rapidily used in peptide synthesis because of the isosteric nature between the triazole ring and the amide bond. However, impact of the insertion of a triazole on the secondary structure of peptides was only scarcely explored. Thus, to study the effect of such a replacement we chose two models of structured peptides. The first models are linear peptides of the peptaibols family and the second is the cyclic peptide tamandarin B.First, we have optimized a peptaibols synthesis that was applied to alamethicin F50 / 5 and to an analog of bergofungin D. For this we used a solid phase peptide synthesis under microwave irradiation taking advantage of the efficient cocktail consisting of diisopropylcarbodiimide as a coupling agent and Oxyma. This synthesis has been extended to silylated analogues of alamethicin by substituting in different positions the aminobutyric acid by the hydrophobic and crowded residue: bis-triethylsilyl-dipropylglycine (Dpg-TES).In the second part we have developed the synthesis of dipeptides containing a triazole motif. Then we defined the reaction conditions necessary for their use in SPPS and synthesized peptides containing several triazoles rings. These dipeptides were then used to perform a triazole scan on the two peptaibols models. The structural studies by circular dichroism, NMR and biological tests of various analogs allow us to conclude that the triazole affect the secondary structure of peptaibols and consequently induces a loss of activity. Thus it appears that the concept of triazole as isosteric of the peptide bond should be used with caution.In order to understand the triazole impact on the activity and the secondary structure of cyclic peptides, we generated simplified cyclic analogues of tamandarin B a depsipeptide of marine origin. We report the preliminary results of this study.