L’analyse d’échantillons du Patrimoine Culturel est primordiale pour la compréhension des techniques, la conservation des œuvres et leur restauration. Ces échantillons sont rares et précieux et l’analyse doit être effectuée sur une faible quantité de matière, ce qui nécessite le développement et l’optimisation de méthodes analytiques appropriées. L’objectif de ce travail de thèse a donc été de développer de nouvelles méthodes analytiques dans le but d’étudier les glycoprotéines et les polysaccharides de gommes végétales des échantillons du Patrimoine Culturel. Les macromolécules ont été séparées par chromatographie d'exclusion stérique et électrophorèse sur gel de polyacrylamide modifié, révélant la présence de poids moléculaires très élevés pouvant atteindre jusqu’à 1 à 2 millions de Dalton. Une stratégie analytique innovante basée sur la spectrométrie de masse a permis d’obtenir des empreintes caractéristiques de chaques gommes. La stratégie analytique développée a été appliquée avec succès sur un échantillon d'aquarelle daté de 1870 du Metropolitan Museum of Art (New York, USA). / The analysis of Cultural Heritage samples is critical for the understanding of the arstists' technique, the conservation and restoration of artworks. These objects under investigation are rare and precious and the amount of sample available for the analysis is usually very low, thus implying the development and the optimization of adapted analytical methodologies. The objective of this PhD was to develop new analytical methods to study glycoproteins and polysaccharides from plant gums in Cultural Heritage samples.These macromolecules have been separated by size exclusion chromatography and modified polyacrylamide gels, which allowed to reveal the presence of proteins with molecular weights up to 1-2 million Dalton. A novel analytical strategy based on mass spectrometry allowed to obtain the caracteristic fingerprint of each plant gum. This developed method was successfully applied on a watercolor sample dated from 1870 supplied by the Metropolitan Museum of Art (New York, USA).

This thesis describes a new method for the synthesis of pseudodisaccharides containing a carbasugar analogue attached to a "true" sugar. The methodology is based on a Diels-Alder cycloaddition of vinyl sugars and appropriately substituted pyran-2-ones, followed by chemical manipulation of the resulting cycloadducts. The thesis also describes the synthesis of inhibitors of Golgi α-mannosidase II and glucokinase. The first chapter is a comprehensive survey of the reported synthetic routes to pseudodisaccharides from the literature. The results and discussions are presented in chapter 2. This chapter starts by discussion of the preparation of vinyl sugars and pyran-2-ones and the regio- and stereoselectivity of their cycloadditions. This is followed by reporting the chemical manipulations of these cycloadducts and the synthesis of a pseudodisaccharide. Cycloadducts are shown to lose carbon dioxide at elevated temperatures to afford dihydrobenzenes. The loss of the bridging carbon dioxide from the cycloadducts is experimentally and computationally investigated. The resulting dihydrobenzenes are shown to also be useful as precursors in the synthesis of pseudodisaccharides. The chemical manipulation of these dihydrobenzenes is used towards the synthesis of a pseudodisaccharide. The third and fourth chapters focus on the synthesis of new inhibitors of Golgi α-mannosidase II and glucokinase respectively. A range of 6-aminoglucose and mannose derivatives were prepared and tested for the inhibition of Jack bean α-mannosidase, but were found to lack any inhibition. Similarly, a range of 6-triazologlucose derivatives were prepared but were found to lack any cytotoxicity. The fifth chapter contains the details of the preparation, experimental procedures and spectroscopic characterisation of the synthesised chemical compounds. Rate calculations are reported in Appendix I and the X-ray crystallographic data are presented in the Appendix II.

The introduction of electron capture dissociation mass spectrometry in 1998 has provided a unique technique for the analysis of peptides and proteins, especially for the identification and localisation of posttranslational modifications. Despite many successes debate continues on the radical-based mechanism of ECD. This thesis explores ECD behaviour in a wide range of PTMs with the intention of furthering our knowledge of the radical-based mechanism. Studies were undertaken on the effect of 3-nitrotyrosine, which is an electron withdrawing group, on ECD. The presence of nitration dramatically decreases peptide backbone sequence coverage but results in the presence of abundant small neutral losses. The key finding is the insight provided into the hierarchy of the various proposed ECD mechanisms. ECD of cysteine bound modifications is shown to result in the fragmentation of the sulfur-modification bond and backbone sequence coverage is highly diminished when analysing S-nitrosopeptides. ECD behaviour of hydrogen-deficient radical peptides is highly dependent on gas-phase peptide structure, with electron capture typically resulting in an increase in charge-reduced precursor intensity. Comparisons of the intermolecular phospho-guanidinium bond strengths between phospho-serine, -threonine and -tyrosine were undertaken. ECD of these complexes results in the retention of the noncovalent bond allowing backbone sequence coverage.

Higher carbon chain sugars have gained increased interest recently; they are important building blocks of natural and unnatural products with biological properties. The synthesis of these higher sugar skeletons is commonly known to be achieved with the Wittig methodology which exploits phosphorus ylide chemistry. This method has been successfully used for the synthesis of the higher carbon sugars. The aim of this project was to synthesise ß,ß'-dioxo sugar-derived phosphorus ylides, a new class of ylides, as versatile intermediates to valuable higher carbon sugar derivatives and carbohydrate mimics. Model reactions were initially conducted; tetrahydro-2-furoic acid and tetrahydro-2H-pyran-4-carboxylic acid, compounds which are structurally similar to the precursor sugars, were identified as suitable model compounds. These compounds were converted to acyl chlorides and then converted to ß,ß'-dioxo phosphorus ylides precursors by acylation. The methodology proved successful and 8 examples were isolated. However, low yields were obtained due to the inevitable formation of triphenylphosphine oxide. The method was then extended to sugar derivatives, prepared using standard protecting group chemistry. It was found that acylation could be achieved using the simple acyl chloride route or peptide coupling methodology for sugar derivatives which were acid sensitive. ß,ß'-dioxo sugar-derived phosphorus ylides (16 examples) were successfully isolated in low yields. The oxidation and thermal reactivity of the ß,ß'-dioxo ylides were studied. Oxidation resulted in the successful synthesis of vicinal tricarbonyls, isolated as a mixture with the gem-diols (hydrates). The thermal decomposition of the ylides gave alkynes in moderate yields.

Iminosugars are polyhydroxylated alkaloids, and can be generally defined as sugar mimetics in which the endocyclic oxygen atom has been replaced with a basic nitrogen. A common affect of this atomic substitution is to bestow these compounds with the ability to inhibit various sugarprocessing enzymes; most significantly the glycosidases (glycoside hydrolases) which areintimately involved in a huge array of biological functions. Compounds which inhibit these enzymes concordantly possess much potential as medicinal agents for the treatment of a variety of diseases. Several iminosugars have already achieved market approval as drugs, and many more are promising candidates in the late stages of clinical development. As such there remains considerable interest in this class of compound, both in terms of the exploration of novel iminosugar structures, as well as the continual development of more efficient general methodology for their synthesis. The densely-packed functionality and stereochemical information present in iminosugars makes them challenging targets for asymmetric chemical synthesis, whereas carbohydrates are clearly very attractive as chiral-pool starting materials for this purpose. Indeed, the majority of the most successful syntheses of iminosugars use the latter approach, and such is the focus of this thesis. Chapter 1 presents a relatively brief introduction to iminosugars, including their types of structure, natural occurrence and biological mode of action. The rationale behind their use as therapeutic agents for the treatment of some significant disease targets is also discussed. Chapter 2 is concerned with the preparation of a number of novel polyhydroxylated azetidines, and their evaluation as glycosidase inhibitors. Such compounds represent an almost entirely neglected class of iminosugars within the literature. An overview of natural and synthetic products incorporating an azetidine motif is given, as well as a brief review of preparative methods and known azetidine iminosugars. A highly efficient and flexible method for the key azetidine ring formation is demonstrated by the cyclisations of 3,5-di-O-triflates of pentoses and hexoses, and of a 2,4-di-O-triflate of glucose, with various primary amines. In this manner, many azetidine triols and tetrols were prepared in good yield. Furthermore, this process is readily adaptable to the installation of added functionality to the azetidine scaffold, as demonstrated by the preparation of 1-acetamido analogues. The initial biological screening of these compounds showed a promising array of glycosidase inhibition, including that of selective inhibition of fungal enzymes. Chapter 3 describes a strategy with which to prepare all sixteen stereoisomers of a known piperidine iminosugar, alpha-homonojirimycin (alpha-HNJ), in a highly divergent manner from just four of the possible thirty-two 6-azidoheptitols using traditional chemical synthesis in tandem with biotechnological transformations. One half of the execution of this strategy is described in this thesis. Two 6-azidoheptitols were prepared from D-mannose, thereby providing access to four 6-azidoketoheptoses through a combination of microbial oxidation and enzymatic epimerisation. Catalytic hydrogenation of these 6-azidoketoheptoses furnished four diastereomeric mixtures of 2,6-iminoheptitols, with varying degrees of stereoselectivity. Purification of these mixtures allowed six 2,6-iminoheptitols to be isolated, two of which have never previously been tested for glycosidase inhibition. Significantly, one of them was found to be a potent and highly selectiveinhibitor of alpha-galactosidases, and may therefore be of interest in the treatment of Fabry disease.

Branched chain sugars display a varied and valuable range of biological activities. This thesis concerns the synthesis of 3,5-di-C-methyl-D-glucose, a potential inhibitor of glycogen phosphorylase (GP), and therefore a proposed therapeutic agent for type 2 diabetes. Chapter 1 looks at the occurrence of branched sugars in the natural world and current therapies for type 2 diabetes. Inhibition of GP is explored, and the molecular modelling studies which led to the design of the project target. Chapter 1 also looks into the development of new foodstuffs, the chemistry and biochemistry of imino sugars and branched hydroxy proline analogues. In Chapter 2, a range of different approaches to 3,5-di-C-methyl-D-glucose are investigated. Most of the initial investigations were carried out on the L-enantiomer, a readily available test system deriving from 2-C-methyl-D-ribono lactone. 2-C-Methyl-D-ribono lactone is synthesised rapidly from D-glucose in a one-pot reaction; as the key starting material for this work, the scalability of this process was investigated. One of the attempted syntheses of di-C-methyl glucose lead to the development of a route towards 3,5-di-C-methyl fructose, a novel dibranched ketose sugar. It was envisaged that through an enzymatic transformation, it might be possible to produce 3,5-di-C-methyl glucose stereoselectively. Synthesis of both enantiomers of 3,5-di-C-methyl glucose and mannose are reported, alongside results of GPb inhibition studies. Analysis of the preferred ring size of a range of di-C-methyl branched sugars and sugar lactones generated in this work is also presented. Chapter 3 explores the chemistry of 2,4-di-C-methyl-L-arabinono lactone, a key intermediate in the synthesis of 3,5-di-C-methyl-L-glucose. From this lactone a novel deoxy sugar, 2-deoxy-2,4-di-C-methyl-L-arabinono lactone, was generated. Routes towards a selection of imino sugars were explored, resulting in the synthesis of a methyl branched isofagomine analogue. A substituted aziridine was synthesised, from which a route to a di-C-methyl branched piperidine was proposed, and a pyrrolidine. Also presented is a synthesis of a dihydroxy di-C-methyl branched proline analogue. Detailed NMR analysis of several of the sugars generated in this work was carried out by Dr M. Wormald, of the University of Oxford Biochemistry department. The results of these investigations are presented in the Appendix. Throughout this work, the presence of quaternary centres has posed a problem with the assignment of relative configuration. As a result, this work has been greatly supported by X-ray crystallography, and the structures shown herein were wholly generated by me. Several other crystals were run during the course of this work, not all pertaining to these projects, and are provided in the CD appendix.

Etant donné le rôle des sucres dans de nombreux processus physiopatholologiques importants, le développement de nouveaux outils chimiques pour la glycobiologie est primordial. Une nouvelle approche, consistant à fonctionnaliser sélectivement la liaison C-H anomérique des sucres par insertion d’un métallocarbène, a été développée, offrant ainsi un accès aux α- et β-cétopyranosides. Celle-ci est basée sur l’utilisation d’un bromoacétate ancré en position 2 du sucre, qui joue un rôle clé au sein d’une séquence réactionnelle glycosylation stéréosélective/diazotransfert/fonctionnalisation. En effet, ce bromoacétate permet de contrôler la stéréosélectivité de la réaction de glycosylation par assistance anchimérique ; puis, il permet l’installation du précurseur de carbène afin de favoriser la quaternarisation de la position anomérique. La validation de l’approche sur des méthyl-glycosides modèles a ensuite permis son application à des disaccharides. Ces études ont par ailleurs mis en évidence la tolérance de la transformation vis-à-vis de plusieurs groupements protecteurs communs en chimie des sucres. Enfin, une étude mécanistique, effectuée sur des substrats deutérés, puis complétée par des calculs DFT, a mis en évidence que l’étape de fonctionnalisation impliquait un mécanisme concerté ou dissociatif selon l’orientation équatoriale ou axiale de la liaison C-H anomérique respectivement. / Due to the major roles of carbohydrates in numerous physiopathological processes, it is essential to develop new chemical tools for glycobiology. A new approach, consisting in the selective functionalization of the anomeric C-H bond of carbohydrates by insertion of a metal-carbene, has been developed, thus offering a new access toward α- and β-ketopyranosides. This is based on the use of a key bromoacetate grafted at position 2 of the sugar, which is the corner stone of a stereoselective glycosylation/diazotransfer/functionalization sequence. Indeed, this group firstly controls the stereoselctivity of the glycosylation step by anchimeric assistance; then, it allows the convenient installation of the carbene precursor to promote quaternarization of the anomeric position. The validation of the concept on model methyl-glycosides has allowed its application on disaccharides. In addition, the sequence has shown good tolerance toward many protecting groups commonly used in carbohydrate chemistry. On the basis of mechanistic studies involving deuterium labeled substrates and DFT calculations, we have shown that this functionalization step follows a concerted or stepwise process depending on the equatorial or axial orientation of the anomeric C-H bond, respectively.

Le diabète de type 2 est aujourd’hui une maladie de plus en plus répandu. En ce sens il est nécessaire de mettre au point de nouveau composé permettant d’inhiber la gluconéogenèse. C’est pourquoi nous nous sommes intéressé à la synthèse du FR225654 1, décaline présentant une activité hypoglycémiante in vivo après administration par voie orale et inhibiteur de la néoglucogenèse in vitro (IC50 = 1,1.10-7 M). Ce composé, isolé en 2005 du champignon Phoma sp N°00144 et jamais encore synthétisé à ce jour, possède un mécanisme d’action inconnu. La nécessité d’une synthèse par voie chimique convergente et flexible, permettant l’accès à des analogues, est donc évidente. La stratégie consistait à préparer une trans-décaline par le biais d’une réaction de Diels-Alder intramoléculaire à partir d’un triène précurseur. La combinaison de ces travaux a constitué une avancée importante dans le cadre de la synthèse du FR225654, un hypoglycémiant potentiel. La mise au point d’une synthèse convergente du précurseur de la réaction de Diels-Alder permettra notamment d’effectuer par la suite des modifications aisées en vue de la préparation d’une vaste gamme d’analogues simplifiés. A ce jour, le produit de cyclo-addition a été isolé et caractérisé, validant ainsi l’étape clé de la stratégie de synthèse. Ainsi, l’accès rapide au FR225654 est rendu possible et la synthèse d’analogues est maintenant envisageable. Les produits synthétisés feront l’objet d’une évaluation biologique, l’objectif ultime étant d’accéder à de nouveaux médicaments pour le traitement du diabète de type 2. / Type 2 diabetes mellitus (T2DM) is a growing worldwide health concern that is expected to afflict over 366 million people by 2030. FR225654 is a novel gluconeogenesis (GNG) inhibitor recently isolated from the culture broth of Phoma sp.. This compound selectively inhibits GNG of primary rat hepatocytes and shows highly hypoglycemic effects in several in vivo mouse models (80% decrease of glycemia). However, to date, the mechanism of action and molecular target remain unknown. From a structural point of view, FR225654 exhibits a highly oxygenated trans-decalin ring substituted by a β-keto-enol moiety and a side-chain bearing a conjugated carboxylic acid and a trisubstituted olefin. Project specific objectives were to design an efficient total synthesis which could also permit a straightforward access to diverse analogues. This feature would constitute a crucial step for the further understanding of Structure Activity Relationship of FR225654. The work consists in synthesizing separately a side chain and a trans-decalin core by means of an intramolecular Diels-Alder reaction from a precursor. To date, synthesis of the precursor has been achieved in 13 steps as well as the side chain. The Intramolecular Diels-Alder reaction has also been validated in order to accomplish the first total synthesis of FR225654.