Structures were determined for 48 new natural products and several related compounds by NMR methods. One new natural product and two unnatural product structures were determined by X-ray diffraction. Molecular mechanics calculations on two indoles related to the neurotransmitter serotonin and on some synthetic cyclophanes were used to gain information about their preferred conformations. Considerable time is wasted redetermining the structures of known natural products when they are encountered in new sources. To help alleviate this problem, a database which searches on proton NMR chemical shifts was developed.

<p>Early man dabbled with the use of plant extracts to cure ailments. This practice has been passed down from generation to generation and today more than 50% of the world'sdrugs are natural products or derivatives thereof. Scientists have thus established a branch of research called natural product research. This branch of research involves the identification and purification of secondary metabolites with a specific biological activity. The methodology involves the screening of plant products for a specific biological activity, purification of the biologically active natural product by separation technology and structure determination. The biologically active natural products is then further scrutinized to serve as a novel drug or lead compound for the development of a novel drug. This research exploited this research methodology.</p>

The work presented in this thesis mainly describes the discovery and development of methodology for the synthesis of anthecularin and anthecotulides, a family of unusual sesquiterpene lactones. Firstly, two 1,3-dipolar cycloaddition approaches toward anthecularin have been evaluated, using either oxidopyrylium ylide chemistry (Path A) or carbonyl ylides, generated by rhodium-catalysed decomposition of diazo ketones (Path B). Synthesis of the key precursor for the diazo strategy was achieved but unfortunately no desired cycloadduct was isolated. Secondly, an experimentally straightforward method to stereoselectively synthesise β-hydroxymethyl-α-methylene-γ-butyrolactones was developed using chromium or zinc. The synthetic utility of this methodology was demonstrated in syntheses of (±)-methylenolactocin, (±)-hydroxymatairesinol and, ultimately, (±)-hydroxyanthecotulide using a gold-catalysed Meyer-Schuster rearrangement. Finally, the first asymmetric synthesis of (+)-anthecotulide has been achieved, in 6 steps from commercially available materials. During this synthesis the absolute configuration was established. Furthermore, a novel rhodium-catalysed enantioselective ene-yne cycloisomerisation was used to form the α-methylene-γ-butyrolactone core.

This dissertation describes work towards two marine alkaloid natural products of the manzamine family. The total synthesis of (–)-nakadomarin A, via two conceptually different strategies is described along with the development of a novel nitro-Mannich-Mannich cascade reaction, which has been applied in a synthesis of the diazatricyclic core of the madangamines. A short and highly stereoselective synthesis of (–)-nakadomarin A has been developed. A sequential alkyne ring-closing metathesis/syn selective reduction strategy enabled the stereoselective construction of the Z-configured alkene in the fifteen-membered macrocycle of the molecule. ‘Matched’ catalyst and substrate control facilitated a highly diastereoselective nitro olefin Michael addition to fix two of the four stereocentres in one key step. Furthermore, a nitro-Mannich/lactamization cascade, furan N-acyl iminium ion cyclisation and alkene ring-closing metathesis enabled the total synthesis of the natural product in 19 steps. In collaboration with Prof. Amir Hoveyda and Prof. Richard Schrock, an alternative approach to a highly Z-selective macrocyclic ring closure in the synthesis of (–)-nakadomarin A has been developed. Three diene substrates were prepared and the Z-selectivity of alkene ring-closing metathesis investigated using a range of molybdenum MAP (monoaryloxide pyrrolide) catalysts. Initial studies using these catalysts produced promising results (Z:E = ~1.2:1), relative to commercially available ruthenium metathesis catalysts (Z:E = 2:3). Using a recently developed chiral tungsten MAP complex, alkene ring-closing metathesis was found to proceed in excellent yield (90%) with exceptional Z:E-selectivity (Z:E ~1.2:1) and requiring only low catalyst loadings (5 mol%). A novel nitro-Mannich-Mannich cyclisation cascade has been developed allowing access to highly functionalized piperidines. Application of this unprecedented mode of reactivity using a substrate derived from a Michael-addition of methylcyanoacetate to a functionalized cyclic nitro olefin allowed the preparation of a cis-fused 6,6’-bicycle in excellent yield. Further elaboration of this 6,6’-bicycle enabled the synthesis of the diazatricyclic core of the madangamine alkaloids to be achieved.

<strong>Chapters 1</strong> and <strong>2</strong> of this thesis focus on the application manganese(III) and copper(II)-mediated oxidative radical cyclisation of alkenyl amidomalonates to the formation of pyrrolidinone-lactones and their subsequent use in the total syntheses of highly bioactive natural products. A novel concise total synthesis of (-)-salinosporamide A based on the oxidative cyclisation previously developed in the group is presented in <strong>chapter 1</strong>. The second chapter discusses the work towards the pyrrolidinone core of the oxazolomycin. Each chapter contains its own introduction to set in context the presented results, which discusses the isolation and the biological activity of the two families of natural products. Previous synthetic work toward salinosporamide A and the oxazolomycin family achieved in the group and reported in the literature is also described in the introduction of each chapter. The third chapter of the thesis succinctly presents the extension of the scope of the manganese(III) and copper(II)-mediated oxidative radical cyclisation reaction. The optimisation, development and scope of the rapid access to fused THF-lactones via the cyclisation of alkenyl oxymalonates is described. Preliminary synthetic manipulations of the resultant bicyclic products to study the application possibilities of the new reaction in future complex molecules syntheses are also presented. The final conclusion gives a summary of the results obtained and introduces the proposed future work that may arise from these three areas.

Dendritic cells (DC) are one of the major antigen presenting cells (APC) of the body. They, by capture of antigen and cross-presentation of these antigens, activate dormant T-cells and co-activate B-cells. As such they regulate the immune system toward either a more humoral type immune response or a more cellular type immune response. These properties have made them very studied over the past decade and many works have focus on the development of vaccine or therapeutic using DCs as a target. However, most of these actual studies have been done by injection of in vitro pre-activated DCs. The major drawback of this technique is the use of non-natural and non individual specific DCs (monocytes derived DCs and/or stem cells DCs). That is why therapeutic carrier targeting specifically DCs has to be developed. To achieve this goal, specific molecules present at the surface of DCs and involved in the activation of the immune system has to be targeted. Among them, DC-Specific ICAM-3 Grabbing Non-integrin CD209 (DC-SIGN) is very specifically expressed only on one subset of DCs called interstitial DCs. This lectin has been proven to be one of the first contacts of the DCs with T-cells and to induce one major interaction for cells proliferation of dormant T-cells. The goal of the project is to design a probe that can be used in vivo and post-mortem to target DCs via DC-SIGN. Therefore, we can use these particles as a proof of concept in vivo and in vitro, record the immune response obtained with them in vivo and in vitro and design probes that can be used to induce specific immune response for future therapy development. Lewis sugars have been shown to be quite specific to DC-SIGN. Their syntheses have been carried out in our lab with a cyanomethylthio linker at their anomeric position. This linker, once activated as a 2-imino-2-methoxyethyl moiety, has permitted the attachment of the oligosaccharides at the surface of dextran-coated iron oxide MRI nanoparticle. These particles have been chosen for their powerful properties and the advantage of the technique they are used for. Indeed, as particles their sizes mimic pathogens and DCs would interact with them, as they will with pathogen. Moreover, many copies of each oligosaccharide could be attached at their surface enhancing the interaction of the particles with the targeted lectin via a multivalent effect. As a technique, MRI has the advantage to be recorded over a long period of time (compare to ¹⁸F PET for example), with a relatively low signal/noise ratio (compare to fluorescence techniques) and without being harmful. FITC fluorescent Lewis X nanoparticles have been actually design and characterised (size by DLS, number of sugar by particles by ICP or fluorescamine fluorescence assay and binding affinity by ELISA with DC-SIGN-Fc). They have been first tested in vitro with models cells (Raji and monocytes derived DCs) for specific uptake assays, where they exhibit specific uptake and internalisation. Lewis-x nanoparticles have also been tested in vivo in a rat model and have been shown to be retained in Lymph nodes compared to control particles. Post mortem analysis appears to demonstrate that these particles were internalised by rat DCs and transported in the centre of the lymph node known as the T-cell region. Finally, cytokines and CD86 concentration measurement have shown that upon internalisation of the nanoparticles, DC maturated. In addition, an antigenic OVA peptide epitope was attached to the surface of the nanoparticles for future T-cell proliferation experiments. It will allow the determination of the immune response expected. In summary, we have developed an immunogenic MRI-active probe that can target specifically DC-SIGN via the interaction with Lewis antigens present at the surface of the probe and trigger DC maturation.

<strong>Introduction</strong> The introduction describes the importance of arylamine compounds to society and provides a brief overview of the methods available for their synthesis. The application of metathesis catalysis to the de novo synthesis of heteroaromatic compounds is also described. <strong>Results and discussion</strong> The first section describes efforts towards the de novo synthesis of arylamines using a cross metathesis/oxidation protocol to form a 1,5-unsaturated dicarbonyl followed by an amine mediated cyclisation. The scope with respect to the 1,5-unsaturated dicarbonyl and amine is covered as well as the utility of some of the products. The section concludes with a modification of the Bohlmann Rahtz pyridine synthesis to furnish arylamines. The next section describes the applications of our methodology to the synthesis of naphthylamines, specifically using the palladium catalysed &alpha;-arylation reaction. A discussion of the α-arylation reaction is included as well as our efforts to explore the scope of the reaction. The third section follows our efforts to apply this methodologyy to the synthesis of five benzo[c]phenanthridine alkaloids including the first reported synthesis of maclekarpine B and C. The final section concludes with a discussion of our efforts towards the de novo synthesis of furans bearing a benzylic stereocentre.

This thesis details the development of manganese(III) acetate-mediated oxidative radical cyclisation methodology. In particular, the use of radicals to form complex, highly sterically congested and strained carbo- and heterocycles in a stereocontrolled manner is described. Chapter 1 gives a summary of the literature regarding three key areas relevant to this work. Radical reaction mechanisms are introduced, including the use of transition metals and lanthanides in C-centred radical cyclisations. The formation of highly sterically congested vicinal all-carbon quaternary stereocentres is also discussed. Finally, the use of radical cyclisation methodology for the synthesis of complex cyclic structures and applications in natural product total synthesis is examined. Chapter 2 gives an account of the manganese(III) acetate-mediated cyclisation of 5-pentenyl malonates bearing a terminal aryl group. The effects of the aryl group are tested with a range of electronically varied substituents. The formation of bi- and tricyclic cyclopentane-lactones bearing adjacent quaternary-quaternary-tertiary stereocentres is demonstrated. Chapter 3 demonstrates the synthesis of highly strained tricyclic bis-lactones. The metal complexes manganese(III) acetate and cerium(IV) ammonium nitrate are shown to give complementary stereoselectivity across a range of cyclisation substrates. Possible synthetic applications of tricyclic bis-lactones are also investigated. Chapter 4 details an asymmetric formal synthesis of the proteasome inhibitor salinosporamide A. An oxidative radical cyclisation forms the key heterocycle in Danishefsky’s synthesis of this biologically important molecule, and showcases the use of the radical chemistry in natural product synthesis. Full experimental details, selected NMR spectra, and X-ray crystallographic data are also provided.

This thesis describes the development of synthetic strategies towards the densely functionalised pyrrolidinone core of the polyene &beta;-lactone-&gamma;-lactam antibiotic oxazolomycin A. <strong>Chapter 1 The oxazolomycins</strong> The oxazolomycins, a unique class of biologically active molecules containing a spiro-fused β-lactone-γ-lactam ring system are introduced. The isolation, structural elucidation and biological properties of the oxazolomycins as well as those of the structurally related inthomycins are reviewed. <strong>Chapter 2 Previous syntheses</strong> The two total syntheses of neooxazolomycin and the synthetic approaches to the pyrrolidinone core of oxazolomycin A and KSM-2690 B are evaluated. <strong>Chapter 3 Project aims</strong> An outline of the synthetic strategy employed in this project and details of the novel retrosynthesis of the pyrrolidinone core of oxazolomycin A are discussed. <strong>Chapter 4 Synthetic studies towards the pyrrolidinone core of oxazolomycin A</strong> The synthetic studies carried out towards the pyrrolidinone core of oxazolomycin A are described in detail. The preparation of an advanced intermediate containing the five chiral centres, four of which are contiguous, was achieved in twenty steps as a single diastereomer and as a single enantiomer. <strong>Chapter 5 Synthetic studies towards the middle fragment of oxazolomycin A</strong> A novel synthetic approach to the diene fragment contained in oxazolomycin A is reported. The formal synthesis of a dienyl iodide, in four fewer steps than previously reported was accomplished. <strong>Chapter 6 Conclusions and future work</strong> A summary of the synthetic work reported in this thesis and proposals for future study are presented. <strong>Chapter 7 Experimental</strong> Full experimental procedures and characterisation of compounds are reported. <strong>Chapter 8 References</strong> A complete list of citations employed in the previous seven chapters is provided.

Bohlmann et al. reported the oxidative spirocyclisation of 2-(ω-hydroxyalkyl)furans under Clauson-Kaas conditions to furnish 1,6-dioxaspiro[4.5]dec-3-enes, thereafter termed the “Bohlmann cyclisation.” This thesis describes the development of an analogous aza-Bohlmann cyclisation. Treatment of 2-(ω-aminoalkyl)furans with m-CPBA or singlet oxygen generates hydroxy- or methoxybutenolides, respectively, which undergo spirocyclisation upon treatment with H₂SO₄ to generate [4.4]- and [4.5]-spiroaminoacetals. The axial/equatorial preferences of N-sulfonylspiroaminoacetals featuring a 3-O-isovaleryl or 3-O-benzyl substituent are described. Acid-catalysed equilibration revealed that in acetonitrile the axial isomer is thermodynamically favoured for both substrates. The first total synthesis of the spiroaminoacetal alkaloid pandamarilactone-1 is discussed, via an aza-Bohlmann cyclisation, in 13 steps and 3% overall yield from 4-pentyn-1-ol.