<p>This thesis describes the design and synthesis of compounds that are</p><p>intended to inhibit serine and aspartic proteases. The first part of the text deals with preparation of inhibitors of the hepatitis C virus (HCV) NS3 serine protease. Hepatitis C is predominantly a chronic disease that afflicts about 170 million people worldwide. The NS3 protease, encoded by HCV, is essential for replication of the virus and has become one of the main targets when developing drugs to fight HCV. The inhibitors discussed here constitute surrogates for the widely used <em>N</em>-acyl-hydroxyproline isostere designated 4-hydroxy-cyclopentene. The stereochemistry of the 4-hydroxy-cyclopentene scaffold was determined by nuclear overhauser effect spectroscopy (NOESY) and the regiochemistry by heteronuclear multiple bond correlation (HMBC). The scaffold was decorated with different substituents to obtain both linear and macrocyclic HCV NS3 protease inhibitors that display low nanomolar activity. The second part of the thesis describes the design and synthesis of potential aspartic protease inhibitors. The hydroxyethylene motif was used as a noncleavable transition state isostere. The synthetic route yielded a pivotal intermediate with excellent stereochemical control, which was corroborated by NOESY experiments. This intermediate can be diversified with different substituents to furnish novel aspartic protease inhibitors.</p> / Report code: LIU-TEK-LIC-2006:45

Structure-based drug design and combinatorial chemistry play important roles in the search for new drugs, and both these elements of medicinal chemistry were included in the present studies. This thesis outlines the synthesis of protease inhibitors against thrombin and the HCV NS3 protease, as well as the synthesis of a combinatorial library using solid phase chemistry.In the current work potent thrombin inhibitors were generated based on the D-Phe-Pro-Arg motif incorporating cyclopentene and cyclohexene scaffolds that were synthesized by ring-closing metathesis chemistry. A structure-activity relationship study was carried out using the crystallographic results for one of the inhibitors co-crystallized with thrombin. HCV NS3 protease inhibitors comprising the proline bioisostere 4-hydroxy-cyclopent-2-ene-1,2-dicarbboxylic acid were synthesized displaying low nanomolar activity. The stereochemistry and regiochemistry of the scaffolds were determined by NOESY and HMBC spectra, respectively. The final diastereomeric target compounds were isolated and annotated by applying TOCSY and ROESY NMR experiments. Furthermore, a 4-phenyl-2-carboxypiperazine targeted combinatorial chemistry library was synthesized to be used early in the lead discovery phase. This was done using a scaffold that was synthesized by palladiumcatalyzed aromatic amination chemistry and subsequently derivatized with eight electrophiles and ten nucleophiles.

Total synthesis of natural products is of contemporary interest in organic synthesis. One of the useful ways to synthesize the natural products is to originate from inexpensive chiral pool compounds abundantly available in nature. In this context, our research group is actively involved in the use of tartaric acid as the four carbon four hydroxy building block in the synthesis of a number of natural products of therapeutic importance. Our strategy relies on the utility of γ-hydroxy amides derived from tartaric acid involving a controlled addition of Grignard reagents and stereoselective reduction. We were successful in application o f this useful building block for the synthesis of a variety of natural products possessing varied functional groups (Chart-1). derived from tartaric acid in the synthesis of oxylipins such as pinellic acid and diyne containing natural products. Chapter 1 of the thesis describes the total synthesis of (+) pinellic acid 6 and (Z,8S,9S,10R)-8,9,10-trihydroxyoctadec-6-enoic acid 10. Key strategy in the synthesis of pinellic acid is elaboration of the aldehyde 3, derived from the γ-hydroxy amide 2 via Horner-Emmons-Wadsworth reaction to yield the α,β-unsaturated ketone 4. Stereoselective reduction of the ketone with (R)-BINAL-H produced the alcohol with requisite stereochemistry which was further extended to pinellic acid 6 (Scheme 1). Wittig homologation of the aldehyde 8 derived from γ-hydroxy amide 7 is the key step for the synthesis of the (Z,8S,9S,10R)-8,9,10-trihydroxyoctadec-6-enoic acid 10. Second chapter of the thesis deals with total synthesis of diyne containing natural products. In the first part of this chapter enantioselective synthesis of possible diastereomers of heptadeca1-ene-4,6-diyne-3,8,9,10-tetrol, a structure proposed for the natural product isolated from Hydrocotyle leucocephala, is accomplished. The alkyne precursors 13 and 14 were synthesized from the α-hydroxy ester 12 derived from γ-hydroxy amide 11 while the alkyne 17 is synthesized from the masked tetrol 16 derived from lactol 15 which was obtained from D-ribose. yne to assemble the diyne unit which was further elaborated to heptadeca-1-ene-4,6-diyne3,8,9,10-tetrol (Scheme 3). It was found that the NMR spectral data of the putative structure assigned for the natural product did not match with any of the diasteromers that were synthesized. This establishes that the structure proposed for the natural product is wrong and requires revision. OH OH OH 18 OH OH 19 OH OH 20 OH OH Scheme 3: Synthesis of diastereomers of heptadeca-1-ene-4,6-diyne-3,8,9,10-tetrol. [Part of this work is published: Prasad, K. R.; Swain, B. J. Org. Chem. (in press)] Second part of this chapter deals with the synthesis of panaxytriol 26 and panaxydiol 28. Key reaction in the synthesis of panaxytriol and panaxdiol is the coupling of bromoalkynes 25 and 27 with 3-silyloxy pent-1-en-4-yne and further elaboration to the triol and diol. The required alkynes were synthesized from the primary alcohol 24 which was obtained from the γ-hydroxy amide 11 involving a series of simple synthetic operations. (Scheme 4). (For structural formula pl see the abstract file)

The aim of this work was to investigate palladium-catalysed cyclisations of bromoenynamides in the synthesis of amidodienes, which on further reaction or oxidation can lead to a diverse range of heteroaromatic systems. Building upon work within the Anderson group on the palladium-catalysed cyclisations of bromoenynes, we have been able to successfully apply a palladium-catalysed carbopalladation/Stille coupling/electrocyclisation cascade to bromoenynamides and then further develop this to incorporate a Suzuki coupling, leading to bicyclic amidodienes which can undergo selective oxidation to a range of heteroaromatics including indolines, indoles, tetrahydroquinolines and benzazepines. During the investigations into the cascade cyclisation, a reductive cyclisation was discovered which enabled access to a range of monocyclic amidodienes. These could subsequently be subjected to a series of Diels-Alder cycloadditions (thermal, Lewis acid-catalysed, arynes) and oxidations to afford a complimentary range of heteroaromatic systems. Whilst this methodology was successful with bromoenynamides, extension of its application to bromoenynhydrazides, with the hope of accessing a relatively unusual range of heteroaromatic structures including indazoles, cinnolines and diazepines, proved to be more problematic. Finally, expansion of the cascade methodology into a fully intramolecular cyclisation, enabled studies towards the synthesis of the trikentrin family of natural products to be conducted.

The Heck reaction has become a fundamental reaction for synthetic organic chemists over the last half century and is utilised heavily in the fine chemical industry and for natural product synthesis. This thesis describes some of the applications of the Heck reaction to modern day organic synthesis. Introduction: This section presents an overview of the Heck reaction starting from its conception during the late 1960s to present day understanding. A variety of ligand classes are described along with commonly accepted catalytic cycles for their activity during the reaction. Results and Discussion: In the first part of the thesis, the use of a cross-metathesis/Heck reaction protocol to synthesise a range of 2,4,6-trisubstituted pyridines is described. Attempts were made to expand the scope of the methodology by employing vinyl Weinreb amides, but this proved unsuccessful for the synthesis of pyridines. Nevertheless, the Heck reaction on vinyl Weinreb amides worked efficiently and the scope of this arylation was explored. Following on, the functionalisation of the Weinreb amide products was studied to generate a range of enone products, some of which would be difficult to synthesise via direct Heck reaction on the respective precursor enone. In the second part of the thesis, previous syntheses of inthomycin B and inthomycin C are described. The synthesis of inthomycin B and inthomycin C were then attempted using an unprecedented Mukaiyama aldol/cross-metathesis based approach to generate the triene core of both natural products.

This thesis investigates a total synthesis of the incednine aglycon by utilising alkenylsilane reagents to assemble the pentaenyl and tetraenyl systems through cross-coupling reactions. The early chapters develop methodology to access both cyclic alkenylsiloxanes and functionalised (E)-alkenylsilanes by the controlled hydrogenation of alkynylsiloxanes and silylolefination of aldehydes, respectively, and culminate in the synthesis of a C6-C13 bis(alkenylsilane)incednine fragment (Scheme 1). The C1-C5 and C14-C23 coupling partners are synthesised in three and ten steps from propargyl alcohol and L-alanine methyl ester through phosphorous-based olefination strategies. In the final chapter we describe our first generation approach to incednine which entails orthogonal cross-couplings to construct the C5-C6 and C13-C14 bonds (Scheme 2).

This thesis describes studies towards the total synthesis of manzamine A (9), a marine alkaloid. Two routes are presented. The first route applied a novel palladium-catalysed arylative allene spirocyclisation cascade to the synthesis of manzamine A (9). In the first generation, a short route was developed to access the tricyclic ACE core 263a in only nine steps. The second generation applied the palladium-catalysed cascade to a similar system which utilised non-terminal allene pro-nucleophile 450 in an attempt to access a homologated derivative of the ACE core. The second route relied on a diastereoselective Michael addition between nitro-olefin 473 and 8,5-fused ring system 146 comprising rings C and E of manzamine A (9). Further elaboration of the Michael addition product enabled the synthesis of tetracyclic ABCE core precursor 464 to be carried out and preliminary investigations into ring B formation were investigated.

Asymmetric phase-transfer catalysis is a powerful technique that enables a wide range of transformations under mild conditions, often using inexpensive and environmentally benign reagents. By extending the applications of phase-transfer catalysis we have developed a highly diastereo- and enantioselective synthesis of functionalized indolines bearing two contiguous stereocentres, one of which is quaternary and all carbon, in a single synthetic step. The reaction proceeds with complete diastereoselectivity and with high levels of enantioselectivity (up to 99% ee). Despite the development of phase-transfer catalysis as a primary synthetic tool in organic synthesis, the mechanistic understanding of these reactions still remains a challenge, due mainly to the difficulty of studying the complex multi-phase systems. Therefore, a further aim of this project was to understand the reaction mechanism of our phase-transfer catalysed transformation. Investigations into the mechanism of our phase-transfer catalysed reaction have been carried out by studying the reaction kinetics. These have shown that the reaction follows a sigmoidal curve with an induction period present. A detailed kinetic investigation was carried out which demonstrated that an autocatalytic mechanism is operational.

This work describes the design and synthesis of a set of four organic molecules that are intended to hydrogen-bond to each other in a pairwise manner. The four hydrogen-bonding units, termed ‘A’, ‘B’, ‘C’ and ‘D’, when placed in solution together, are designed so that A binds only to B, and C binds only to D. Each unit does not bind to itself, nor to either of the other two units to which binding is not intended. For example, A binds to B, but not to A, C, or D. Each unit contains an array of four hydrogen-bonds for strong binding to its partner, is designed to be as rigid as possible, as non-tautomeric as possible, and utilises a staggered non-symmetrical architecture. Of the four intended compounds, three were successfully synthesised (A, B and D). Units B and D were soluble in CDCl₃, but Unit A was not. Therefore, the design and synthesis of Unit A was amended, and two variants of Unit A that are both soluble in CDCl₃ were successfully synthesised. ¹H NMR binding experiments were performed between Unit B and each of the two variants of Unit A. Their binding behaviour was described. A binding constant could not be calculated because the units did not bind in a 1:1 fashion.

Thesis (PhD)--Stellenbosch University, 2015. / ENGLISH ABSTRACT: Benzoxazoles are an important class of π-electron-excessive, benzene-fused heterocyclic compounds found in natural products and display a wide range of pharmacological applications. It is therefore a widely used starting scaffold for drug and agrochemical discovery programs. Other applications include: chiral auxiliaries in asymmetric reactions, chiral receptors for the resolution of racemic mixtures, fluorescent whitening dyes, various photochromic materials and as ligands for a wide range of catalytic reactions. Due to our interests in resorcinarenes, we came across 4-hydroxybenzoxazoles, a structural motif that has not been explored as potential asymmetric ligands. In this thesis it was attempted to investigate the synthesis, functionalisation and coordination chemistry of these compound class and finally look at a method of synthesising chiral 4-hydroxybenzoxazoles from amino acids. A small library of achiral 4-hydroxybenzoxazoles were synthesised in good yields. These compounds were then reacted with various transition metals, of which only the Pd-salts proved to return any usable compounds. The first structural evidence of the bonding of 4- hydroxybenzoxazoles was recorded from single crystal X-ray diffraction analysis of the coordination compounds that formed. Different coordination modes were recorded, depending on the ligand and the Pd-salt used. The PdCl2 compounds were also tested for catalytic activity with a Heck reaction, showing good conversions for the reaction between iodobenzene and styrene to form stilbene. Further examination pointed to the ligands playing an insignificant role in the reaction and the products possibly due to only the PdCl2’s reactivity. During this period it was also attempted to functionalise the phenol group with P(III) groups and repeat the coordination and catalytic studies. Efforts to synthesise these compounds were not successful, with oxidation of the P(III) to P(V) groups or degradation of these compounds. Efforts to synthesise these via phosphorous protection, utilising BH3 or the in situ trapping of the compounds with transition metals, were also not successful. During the trapping experiments the phosphinite and Pd-salt formed a re-arranged product that is a known and useful catalyst on its own. Finally a small library of chiral benzoxazoles and 4-hydroxybenzoxazoles were synthesised, starting from amino acids and utilising a Mitsunobu reaction to perform the ring closing. Antimicrobial tests with these compounds did not return any appreciable results. / AFRIKAANSE OPSOMMING: Bensoksasool is 'n belangrike klas van π-elektron-ryk, benseen-saamgesmelte heterosikliese verbindings wat in natuurlike produkte voorkom en 'n wye verskeidenheid van farmakologiese funksies vertoon. Dit is dus 'n baie algemene basis struktuur vir dwelm- en landbouchemiese ontdekkings programme. Ander gebruike sluit in: chirale ligande in asimmetriese reaksies, chirale reseptore vir die resolusie van rasemiese mengsels, fluoresserende verwittings kleurstowwe, verskeie fotochromiese materiaal en as ligande vir 'n wye verskeidenheid van katalitiese reaksies. As gevolg van ons belangstelling in resorsinarene, het ons op 'n strukturele motief afgekom wat nog nie ondersoek is as potensiële asimmetriese ligande nie, die 4- hidroksiebensoksasole. In hierdie tesis is gepoog om die sintese, funksionalisering en koördinasie chemie van hierdie klas verbindings te ondersoek en uiteindelik 'n metode te ontwikkel om die sintese van chirale 4-hidroksiebensoksasole vanaf aminosure te bewerkstellig. 'n Klein biblioteek van achirale 4-hidroksiebensoksasole was gesintetiseer in goeie opbrengste. Hierdie verbindings was toe behandel met verskeie oorgangsmetale, waarvan slegs die Pdsoute enige bruikbare verbindings gevorm het. Die eerste strukturele bewyse van die binding van die 4-hidroksiebensoksasole is aangeteken met behulp van enkelkristal X-straaldiffraksie ontleding van die koördinasieverbindings wat gevorm is. Verskillende koördinasie mode is aangeteken, afhangende van die ligand en die Pd-sout wat gebruik was. Die PdCl2 verbindings is ook vir katalitiese aktiwiteit met 'n Heck reaksie getoets. Die reaksie het baie goeie omskakeling gewys vir die reaksie tussen iodobenseen en stireen na stilbeen. Verdere ondersoeke het getoon dat die ligande nie ‘n beduidende rol in die reaksie speel nie en die produkte moontlik slegs as gevolg van die PdCl2 se reaktiwiteit is. Gedurende hierdie tydperk was daar ook probeer om die fenol groep met P(III) groepe te funksionaliseer. Met die uitgangstowwe sou die koördinering en katalitiese studies herhaal word. Pogings om hierdie verbindings te sintetiseer was nie suksesvol nie, met oksidasie van die P(III) na P(V) groepe of afbreking van hierdie verbindings. Pogings om dit te sintetiseer via fosfor beskermingstegnieke, deur gebruik te maak van BH3 of die in situ vasvang van die verbindings met oorgangsmetale, was ook nie suksesvol nie. Gedurende die vasvang eksperimente het die fosfien en PdCl2 ‘n herrangskikkings-produk gevorm wat op sy eie ‘n bekende en nuttige katalisator is. Ten slotte was 'n klein biblioteek van chirale bensoksasole en 4-hidroksiebensoksasole gesintetiseer, vanaf aminosure. Om die ringsluiting te bewerkstellig was ‘n Mitsunobu reaksie gebruik. Antimikrobiese toetse met hierdie verbindings het nie enige noemenswaardige resultate opgelewer nie.