Natural products are a rich source of lead compounds for treatment of cancer as well as other diseases. Researchers at the National Cancer Institute, as part of their continuing effort to discover anticancer agents from natural sources, created the 60 human tumor cell-line anticancer screen to test natural products for their potential against various types of cancer. Through this screening process a family of natural products called schweinfurthins was discovered to possess potent and differential activity. Of potentially great significance, the pattern of activity that the schweinfurthins displayed in the screen does not correlate with any currently used anticancer drug, indicating that the schweinfurthins likely act via a previously unknown mechanism or on a novel target. Our group has synthesized many of the natural schweinfurthins as well as numerous analogues in an effort to probe the pharmacophore and gain understanding of the key features that are important for potency as well as differential activity. During the course of these studies, it was discovered that the right-half of the molecule is most amenable for modifications. One potential modification to the schweinfurthins is to replace the resorcinol substructure seen in the right-half of the natural product with a heteroaromatic moiety such as a benzofuran or indole system. This change may produce analogues that are potentially more active, that contain motifs that are seen in many therapeutic drugs, and that have improved chemical stability relative to the natural products. With this goal in mind benzofuran and indole containing schweinfurthin analogues were synthesized. Once these compounds were prepared, it was found that such modifications were welltolerated, and in the case of the indole analogues activity in the 60 cell-line screen was equivalent to the corresponding natural product. In an effort to improve that activity, prenyl and geranyl side chains were added to the indole system, at both the C-2 and C-3 positions, to better match the structure of the natural schweinfurthins. In addition, analogues methylated selectively on the indole nitrogen or phenol were synthesized to improve stability. The impact of those modifications on the activity was tested, and potent compounds were found. The left-half of the schweinfurthins is prepared via a Lewis acid mediated cascade of a geranyl epoxide. The protecting group that is typically employed on the terminating phenol, a methoxymethyl ether or MOM group, is cleaved during the reaction. In the past preparation of an analogue that lacked a substituent at the C-5 position, it was found that the MOM cation released during the cyclization would participate in an electrophilic aromatic substitution reaction at the neighbouring position which resulted in the formation of a benzyl methyl ether. In order to probe the scope of this reaction and its potential utility in the synthesis of natural products, several geranyl epoxides with various "protecting groups" on the phenol were prepared and subjected to the cyclization conditions. These investigations have established that stabilization of the liberated cation determines the likelihood and regioselectivity of a tandem electrophilic aromatic substitution reaction.

There are many diverse classes of biologically active natural products containing chiral 1,5- or 1,5,7-polyol moieties, including the novel fibrinogen receptor antagonist tetrafibricin, a potential antiplatelet therapeutic drug to treat various arterial thrombotic diseases. There have been some elegant synthetic strategies developed to synthesize these challenging 1,5-diol motifs; however, many of them suffer from a variety of inherent limitations. To overcome many of these challenges, our group has developed an iterative configuration-encoded strategy to access 1,5-polyols with unambiguous stereocontrol by exploiting Julia–Kocienski couplings of enantiopure α-siloxy-γ-sulfononitrile building blocks with alcohol stereocenters previously established via asymmetric catalysis. Our method is a strategy level innovation that allows for the efficient and rapid access to all stereoisomers of a 1,5-polyol family from cheap and easily accessible reagents, without the need to determine the configuration of each alcohol stereocenter in the growing polyol chain. We were able to modify our configuration-encoded 1,5-polyol methodology to access the anti,syn-1,5,7-triol within the C15–C25 fragment of tetrafibricin with excellent selectivity by incorporating differentiable protection and merging this approach with the tactic of diastereoselective intramolecular conjugate addition via benzylidene acetal construction to access the syn-1,3-diol functionality. We also applied our iterative configuration-encoded strategy to the synthesis of the 1,5-polyol-containing C26–C40 fragment of tetrafibricin with excellent stereoselectivity by modifying our previous route to the C27–C40 segment. By overcoming the challenges associated with the reduction of α-siloxynitriles and extending the carbon chain to alter the subsequent Mukaiyama aldol coupling location, we were able to furnish the C26–C40 fragment with the correct protection and functionality for further coupling to the C15–C25 segment. With the C15–C25 and C26–C40 fragments in hand, we joined these segments via asymmetric BF3⋅OEt2-mediated Mukaiyama aldol construction with high 1,3-anti stereoinduction. We determined the preceding stereoselectivity by first using the simplified model C26–C40 fragment and found that replacing the TBDPS with TBS protection of the β-siloxy aldehyde increased the level of 1,3-anti induction. To complete the C15–C40 fragment of tetrafibricin, we performed an intramolecular hydroxyl-directed anti-reduction to furnish the desired anti,anti,anti-1,3,5,7-tetraol moiety. We were able to establish the configurations of these chiral alcohols using a battery of 2D NMR experiments. Finally, to complete the total synthesis of tetrafibricin, we have proposed a route to couple our C15–C40 fragment with the C8–C14 segment via a precedented asymmetric aldol reaction, followed by coupling to the known C1–C7 polyene fragment. With minor functional group transformations and a global deprotection, access to the natural product tetrafibricin should be achievable.

Salvinorin A [(2S,4aR,6aR,7R,9S,10aS,10bR)-9-(acetyloxy)-2-(3-furanyl)-dodecahydro-6a,10b-dimethyl-4,10-dioxo-2H-naptho[2,1-c]pyran-7-carboxylic acid methyl ester] is a trans-neoclerodane diterpene from the leaves of the hallucinogenic Mexican sage Salvia divinorum and has been identified as the principal psychoactive component in this plant of traditional spiritual importance. Salvinorin A is the most potent naturally occurring hallucinogen found so far and is reported to act selectively as a ƒÛ-opioid receptor agonist. Synthetic modification of the natural product has contributed to a number of proposed pharmacophores to identify the key structural features necessary for biological activity and a direct strategy for the asymmetric synthesis of the natural product is desirable since it allows access to a more diverse range of analogues. An ambitious retrosynthetic study of salvinorin A indicated the C(3)-heterosubstituted furan as an appropriate starting material for a Diels-Alder approach towards the ketone ring of the natural product. An expedient and high yielding methodology for the preparation of 3-furylamines is described, allowing the flexible introduction of alkyl substituents in the C(5) position. Optically pure ephedrine isomers have been explored as chiral amine auxiliaries and have been successfully attached as 3-furylamine substituents using the general methodology described. The 3-furylamines are electron rich dienes that are highly reactive towards Diels-Alder cycloaddition reactions with methyl acrylate. Diastereoisomers of the 7-oxanorbornane species methyl 1-methyl-5-oxo-7-oxa-bicyclo[2.2.1]heptane-2-carboxylate were prepared as new compounds from the hydrolysis of Diels-Alder cycloadducts and are functionalised bicyclic intermediates to access the ketone of the natural product. Diels-Alder reactions between the non-racemic (2S)-ephedrine-derived furans and methyl acrylate gave spiro-oxazolidine adducts that underwent hydrolysis to give the desired ketone. X-ray crystallography data for the derivatised cycloadduct established diastereoselectivity in favor of the (1S,4S)-enantiomer, as desired for the asymmetric natural product synthesis. A procedure for the ether cleavage of methyl 1-methyl-5-oxo-7-oxa-bicyclo[2.2.1]heptane-2-carboxylate was required to access the convergent precursor methyl 5-acetoxy-2-methyl-4-oxocyclohex-2-enecarboxylate. Successful C-O cleavage was achieved using Lewis-acid catalysis with BBr3 followed by mixing with the hindered base 2,4,6-collidine to yield methyl 5-hydroxy-2-methyl-4-oxocyclohex-2-enecarboxylate albeit only at high dilution. Acetylation proceeded in excellent yield in the same reaction vessel to give methyl 1-methyl-5-oxo-7-oxa-bicyclo[2.2.1]heptane-2-carboxylate in excellent yield. The devised synthetic pathway is shown to successfully construct the ketone ring of salvinorin A and stereoselectivity for the (1S,4S)-enantiomer can be achieved using the ephedrine derived furans as desired for the asymmetric natural product synthesis. The ƒÔ-lactone ring 6-(furan-3-yl)-5,6-dihydro-4-methyl-3-vinylpyran-2-one was derived from rudimentary precursors as a convergent reagent to introduce the lactone ring of salvinorin A. A short synthesis for the racemic compound is described starting from the aldol reaction between 3-furaldehyde and acetone to give the 3-furfurol, 4-(furan-3-yl)-4-hydroxybutan-2-one in quantitative yield. The 3-furfurol was reacted to form the ƒÑ-bromovinyl ester, 1-(furan-3-yl)-3-oxobutyl 2-bromobut-3-enoate using a deconjugation/esterification protocol with 2-bromobut-3-enoyl chloride. Intramolecular ring closure to the ƒÔ-lactone was achieved using a Reformatsky reaction and dehydration under acidic conditions yielded the racemic convergent precursor 6-(furan-3-yl)-5,6-dihydro-4-methyl-3-vinylpyran-2-one in high yield. A possible strategy for joining the ketone and lactone fragments for the total synthesis of salvinorin A is proposed.

In order to search for bioactive compounds, we have studied the chemical constituents from the organic extracts of two Formosan soft corals Sarcophyton crassocaule and Paralemnalia thyrsoides. This study had led to the isolation of forty-six natural compounds 1¡V46, including sixteen new cembrane¡Vtype diterpenoids, crassocolides A¡VP (1¡V16) and four known cembrane¡Vtype compounds 17¡V20 from S. crassocaule; nineteen new sesquiterpenoids and norsesquiterpenoids, paralemnone (21), isoparalemnone (22), paralemnol (23) and paralemnolins A¡VP (24¡V39), along with seven known compounds 40¡V46 from P. thyrsoides. The structures of these compounds were established by the detailed spectroscopic analysis (IR, MS, 1D¡B2D NMR) and by comparison with related physical and spectral data from other known compounds. The absolute configurations of 1-46 were determined using a modified Mosher's method for 1, 7, 22 and 27. The structures of 5, 21, 24 and 37 were further proven by X-ray diffraction analysis. The cytotoxicity of compounds 1-46 against a limited panel of cancer cell lines was also determined. Also, the activity of compounds 21-28, 35-37 and 41-42 to inhibit the pro-inflammatory iNOS and COX-2 protein expression of LPS-stimulated RAW264.7 macrophage cells has been estimated.

Fungal polyketides are a diverse class of natural products that possess many pharmacological properties, including anticancer properties. These properties are evident in the resorcylic acid lactones, a family of polyketides, including zearalenone and radicicol, which shows potent inhibition of tumour cell growth. The key step in the biosynthesis of these lactones is macrocyclization of a linear carboxylic acid into the macrolactone. This reaction is catalyzed by a polyketide synthase (PKS) thioesterase enzyme. Bacterial PKS thioesterases (TEs) have been extensively studied and their substrate specificity has been characterized in vitro. They are highly substrate selective for the macrocyclization reaction. Since Fungal PKS TEs show little sequence homology to bacterial TEs, we have begun investigating their substrate specificity. In particular we are examining the ability of fungal TEs to macrocyclize compounds with varying ring sizes, stereogenic configuration, and nucleophiles. Herein we present the synthesis of a number of diverse TE substrates and the in vitro macrocyclization results for the TEs from zearalenone and radicicol biosynthetic pathway with these substrates.

The field of metal-catalyzed C–H bond functionalizations is an incredibly vibrant and spans beyond the formations of biaryl motifs. The introduction chapter will cover the mechanistic aspects of the C–H bond functionalization with metal-carboxylate complexes. The mechanistic facets of this reaction will be the main conducting line between the different sections and chapters of the first part of this thesis. In the second chapter, will be described additives that can readily promoted C–H bond arylation of poorly reactive substrates. More specifically, we will revisit the intramolecular direct arylation reaction we will demonstrate the effect of pivalic acid as a co-catalyst by developing milder reaction conditions. In the third chapter we be described experimental and computational studies which suggested that the a single pathway might be involved in the palladium-catalyzed C–H bond functionalization of a wide range of (hetero)arene. Following this we will describe a general set of conditions for the direct arylation of wide range of heteroarenes. Also, we will present two different strategies to selectively and predictably arylate substrates containing multiple functionalizable C–H bonds. In the fourth chapter will be presented our efforts toward the development of new C–H bond functionalization methods in which we could apply our knowledge on the C–H bond cleavage and apply it to the formation of new scaffolds. The development of two new palladium-catalyzed methods were also described. In the fifth chapter, our effort toward the development of ligands to specifically promoted C–H bond cleavage will be presented. In the sixth chapter will be presented the latest results on the study of the mechanism of the C–H bond cleavage combining experimental and computational studies. In part B of this thesis will be presented our strategy toward the total synthesis of ginkgolide C that included two gold(I)-catalyzed reactions as key steps in the preparation of the spiro[4.4]nonane core of this natural product. The first studies on the feasibility of the key steps of the synthesis will be described.

Fungal polyketides are a diverse class of natural products that possess many pharmacological properties, including anticancer properties. These properties are evident in the resorcylic acid lactones, a family of polyketides, including zearalenone and radicicol, which shows potent inhibition of tumour cell growth. The key step in the biosynthesis of these lactones is macrocyclization of a linear carboxylic acid into the macrolactone. This reaction is catalyzed by a polyketide synthase (PKS) thioesterase enzyme. Bacterial PKS thioesterases (TEs) have been extensively studied and their substrate specificity has been characterized in vitro. They are highly substrate selective for the macrocyclization reaction. Since Fungal PKS TEs show little sequence homology to bacterial TEs, we have begun investigating their substrate specificity. In particular we are examining the ability of fungal TEs to macrocyclize compounds with varying ring sizes, stereogenic configuration, and nucleophiles. Herein we present the synthesis of a number of diverse TE substrates and the in vitro macrocyclization results for the TEs from zearalenone and radicicol biosynthetic pathway with these substrates.

The field of metal-catalyzed C–H bond functionalizations is an incredibly vibrant and spans beyond the formations of biaryl motifs. The introduction chapter will cover the mechanistic aspects of the C–H bond functionalization with metal-carboxylate complexes. The mechanistic facets of this reaction will be the main conducting line between the different sections and chapters of the first part of this thesis. In the second chapter, will be described additives that can readily promoted C–H bond arylation of poorly reactive substrates. More specifically, we will revisit the intramolecular direct arylation reaction we will demonstrate the effect of pivalic acid as a co-catalyst by developing milder reaction conditions. In the third chapter we be described experimental and computational studies which suggested that the a single pathway might be involved in the palladium-catalyzed C–H bond functionalization of a wide range of (hetero)arene. Following this we will describe a general set of conditions for the direct arylation of wide range of heteroarenes. Also, we will present two different strategies to selectively and predictably arylate substrates containing multiple functionalizable C–H bonds. In the fourth chapter will be presented our efforts toward the development of new C–H bond functionalization methods in which we could apply our knowledge on the C–H bond cleavage and apply it to the formation of new scaffolds. The development of two new palladium-catalyzed methods were also described. In the fifth chapter, our effort toward the development of ligands to specifically promoted C–H bond cleavage will be presented. In the sixth chapter will be presented the latest results on the study of the mechanism of the C–H bond cleavage combining experimental and computational studies. In part B of this thesis will be presented our strategy toward the total synthesis of ginkgolide C that included two gold(I)-catalyzed reactions as key steps in the preparation of the spiro[4.4]nonane core of this natural product. The first studies on the feasibility of the key steps of the synthesis will be described.

Ansamycin natural products have historically been a rich source of new drugs for the treatment of bacterial infections and cancer. The C17-benzene ansamycins in particular have shown excellent preclinical results as potential anti-fungal and anti-cancer medicines. However, their thorough clinical evaluation has been hampered by the absence of a concise synthetic strategy. In order to address this issue, recently developed hydrogenative carbon-carbon bond forming methods were applied toward a short total synthesis of C17-benzene ansamycins. This class of natural products provides a challenging testing ground for these methods while facilitating the further development of compounds which may be used as treatments for life threatening diseases. In the first synthetic approach to the C17-benzene ansamycins key bond formations include direct iridium catalyzed carbonyl crotylation from the alcohol oxidation level followed by chelation-controlled dienylation to form the stereotriad, which is attached to the arene via Suzuki cross-coupling. The diene-containing carboxylic acid is prepared using rhodium catalyzed acetylene-aldehyde reductive C-C coupling mediated by gaseous hydrogen. Finally, ring-closing metathesis delivers the cytotrienin core. The second approach toward triene-containing C17-benzene ansamycins resulted in the syntheses of trienomycins A and F, which were prepared in 16 steps (longest linear sequence) and 28 total steps. The C11-C13 stereotriad was generated via enantioselective ruthenium-catalyzed alcohol CH syn crotylation followed by chelation-controlled carbonyl dienylation. Finally, diene-diene ring closing metathesis to form the macrocycle. The present approach is 14 steps shorter (LLS) than the prior syntheses of trienomycins A and F, and eight steps shorter than any prior synthesis of a triene-containing C17-benzene ansamycin. / text

Vinigrol is a structurally unique diterpenoid natural product featuring a tricyclo[4.4.4.0.4a,8a]tetradecene carbon skeleton containing eight contiguous stereocenters and a challenging oxygenation pattern. Vinigrol has been demonstrated to possess a wide array of biological activities including tumor necrosis factor (TNF) antagonism, antihypertensive activity, and platelet aggregation inhibitory activity. Our first-generation plan for the synthesis of vinigrol utilized a cascade reaction sequence involving: (1) diastereoselective alkylation of an α-alkenyl-β-ketoester, (2) retro-aldol-aldol equilibration (3) anion-accelerated oxy-Cope rearrangement, and (4) transannular Dieckmann condensation to afford the bicyclo[5.3.1]undecene ring system of vinigrol in a single operation. Discoveries concerning the limitations of this process are disclosed. Our second-generation approach to vinigrol employed a cis-decalin substrate in an alternative cascade reaction sequence, which was expected to deliver the complete tricyclo[4.4.4.0.4a,8a]tetradecene carbon skeleton of vinigrol in one step. An unexpected deviation from the envisioned reaction pathway instead afforded an alternative tricyclic enol silane. / Chemistry and Chemical Biology