1    

 

 

 

Chemistry  of  Natural  Products  (Chem  135)  

Fatty  Acids  and  Polyketides  Report  on  Secondary  Metabolites  

2008-­‐30228    BARBA,  Bin  Jeremiah  D.      2008-­‐00116    BIADOMANG,  John  Kevin  Paulo  C.      2008-­‐68119    NARCISO,  Karina  Sophia  M.      8/5/2011    

2    

INTRODUCTION

FATTY ACIDS

Fatty acids are alkanoic acids usually encountered in nature as their ester derivatives collectively known as lipids. The majority of naturally occurring fatty acids have straight chains possessing an even number of carbon atoms. The most common of them consist of 10 to 20 carbon atoms. Palmitic acid (16C) and stearic acid (18C) are particularly widespread. These are saturated fatty acids, but unsaturated fatty acids are also commonly encountered like 9-octadecanoic acid (oleic acid). Saturated fatty acids with odd number of carbon atoms are also found in nature, particularly in bacteria and lower forms of plants and animals. There are also certain unusual types such as acetylenic acids, and branched saturated or unsaturated fatty acids.

Fatty acids can be classified according to a specific property. They may be classified based on saturation (saturated or unsaturated), number of carbon atoms (odd or even), length of carbon chain (2-6: short-chain, 6-10: medium-chain, 12-24 long-chain or above 24: very long-chain) and chain type (straight, branched, ring).

Saturated fatty acids are primary metabolites which occur most commonly in organisms. They have the distinct lack of double bonds between the carbon atoms such that the chain of carbon atoms is fully saturated with hydrogen.

Unsaturated fatty acids can be produced in two general routes which are the aerobic route which occurs in animals and plants, and the less common anaerobic route which occurs in some bacteria. The aerobic pathway directly introduces the double bond into a saturated fatty acid precursor by oxidative desaturation regulated by desaturase enzymes. Double bond is introduced generally between 9 and 10.

Figure 1. Oxidative desaturation of stearic acid

In the anaerobic route, chain elongation, reaction with a further malonyl, reduction to alcohol and dehydration gives the conjugated alkene and isomerization gives the non-conjugated alkene with cis double bonding.

3    

Figure 2. Anaerobic formation of oleic acid

The position in which new double bonds are introduced differs between animals and plants. Animals introduce new double bonds between existing double bonds and carboxyl group while plants do it between existing double bonds and the methyl terminus of the fatty acid.

Figure 3. Polyunsaturated FA (linoleic from oleic acid)

Most fatty acids are undoubtedly primary metabolites. With the considerable amount of structural diversity and limited natural distribution of certain fatty acids, they are more appropriately referred to as secondary metabolites. Ricinoleic acid is the 12-hydroxy derivative of oleic acid and is the major fatty acid found in castor oil, expressed from seeds of the castor oil plant (Ricinus communis; Euphorbiaceae). It is derived from oleic acid. Castor oil has a long history of use as a domestic purgative, but it is now mainly employed as a cream base. Undecenoic acid (Δ9-undecenoic acid) can be obtained from ricinoleic acid and is used in a number of fungistatic preparations. Epoxy fatty acids like vernolic acid have been found in substantial quantities in the seed oil of some plant species, including Vernonia galamensis and Stokesia laevis (both Compositae/Asteraceae). Vernolic acid is formed by from linoleic acid. Tuberculostearic acid is a C-methyl derivative of stearic acid found in the cell wall of Mycobacterium tuberculosis, the bacterium causing tuberculosis, and it provides a diagnostic marker for the disease. It is

4    

derived from oleic acid. From there, Sterculic acid, may also be derived, an unusual fatty acid containing a highly strained cyclopropene ring. Sterculic acid is present in the seed oil from Sterculia foetida (Sterculiaceae), and with similar cyclopropene acids, like malvalic acid, which is present in edible cottonseed oil from Gossypium species (Malvaceae). Malvalic acid is produced from sterculic acid, an inhibitor of the Δ9-desaturase which converts stearic acid into oleic acid and is potentially harmful to humans in that it can alter membrane permeability and inhibit reproduction. Chaulmoogric and hydnocarpic acids are uncommon cyclopentenyl fatty acids found in chaulmoogra oil, expressed from seeds of Hydnocarpus wightiana (Flacourtiaceae). Chaulmoogra oil provided for many years the only treatment for the relief of leprosy, these two acids being strongly bactericidal towards the leprosy infective agent Mycobacterium leprae. Lipostatin was isolated from the mycelium of Streptomyces toxytricini cultures and shown to possess marked inhibitory activity towards pancreatic lipase, the key enzyme for intestinal fat digestion. The β-lactone function in the lipophilic molecule irreversibly inactivates lipase by covalent reaction with a serine residue at the catalytic site; the reaction closely parallels that of serine residues with β-lactam antibiotics. Tetrahydrolipstatin (orlistat), obtained by catalytic hydrogenation of lipostatin, was selected for further development; it is more stable and crystallizes readily, although it is somewhat less active. Orlistat is now manufactured by total synthesis. These have been developed into an anti-obesity drug.

Many unsaturated compounds found in nature contain one or more acetylenic bonds, and these

are predominantly produced by further desaturation of olefinic systems in fatty acid-derived molecules. Acetylene group of polyacetylene natural products is believed to be introduced by further oxidative desaturation of an existing double bond. The first natural actelyene to be discovered was tariric acid in 1892. They are surprisingly widespread in nature and are found in many organisms, but they are especially common in plants of the Compositae/Asteraceae, the Umbelliferae/Apiaceae, and fungi of the group Basidiomycetes. These compounds tend to be highly unstable and some are even explosive if sufficient amounts are accumulated. Since only very small amounts are present in plants, this does not present any widespread hazard. Whilst fatty acids containing several double bonds usually have these in a non-conjugated array, molecules containing triple bonds tend to possess conjugated unsaturation. This gives the compounds intense and highly characteristic UV spectra, which aids their detection and isolation.

Some noteworthy acetylenic structures are Cicutoxin from the water hemlock (Cicuta virosa; Umbelliferae/Apiaceae) and oenanthotoxin from the hemlock water dropwort (Oenanthe crocata; Umbelliferae/Apiaceae) which are extremely toxic to mammals, causing persistent vomiting and convulsions, leading to respiratory paralysis or death; Falcarinol, a constituent of Falcaria vulgaris (Umbelliferae/Apiaceae), Oenanthe crocata, Hedera helix (Araliaceae), and several other plants, which is known to cause contact dermatitis in certain individuals when the plants are handled; panaxytriol, a characteristic polyacetylene component of ginseng (Panax ginseng; Araliaceae); and Wyerone from the broad bean (Vicia faba; Leguminosae/Fabaceae) which has antifungal properties.

Figure 4. Polyacetylenes

5    

Primary amides of unsaturated fatty acids have been characterized in humans and other

mammals, and although their biological role is not fully understood, they may represent a group of important signalling molecules. Oleamide, the simple amide of oleic acid, has been shown to be a sleep-inducing lipid, and the amide of erucic acid, erucamide, stimulates the growth of blood vessels. The ethanolamide of arachidonic acid, anandamide, appears to be the natural ligand for receptors to which cannabinoids bind. The herbal preparation echinacea is derived from the roots of Echinacea purpurea (Compositae/Asteraceae) and is used for its immunostimulant properties, particularly as a prophylactic and treatment for the common cold. At least some of its activity arises from a series of alkamides (also termed alkylamides), amides of polyunsaturated acids with isobutylamine. These acids are predominantly C11 and C12 diene-dynes.

Figure 5. Fatty Acid Amides

Oxygenated fatty acids, referred to as oxylipins, are derived principally from arachidonic acid in mammals which exhibit a diverse range of pharmacological effects including regulation of blood pressure, control of platelet aggregation, allergic responses and inflammation processes; and from linolenic acid in plants which display different functions such as wound healing, defense mechanism against pathogens and are components of cuticle. They are involved in responses to physical damage by animals or insects, stress and attack by pathogens.

Prostaglandins are a group of modified C20 fatty acids first isolated from human semen and

initially assumed to be secreted by the prostate gland. They are now known to occur widely in animal tissues, but only in tiny amounts. Prostaglandins of the A-, E-, and F-types are widely distributed in soft corals, especially Plexaura, though many corals produce prostaglandin structures not found in animals. They have been found to exert a wide variety of pharmacological effects on humans and animals. They are active at very low, hormone-like concentrations and can regulate blood pressure, contractions of smooth muscle, gastric secretion, and platelet aggregation. Their potential for drug use is extremely high, but it has proved difficult to separate the various biological activities into individual agents. Latanoprost, travoprost, and bimatoprost are recently introduced prostaglandin analogues which increase the outflow of aqueous humour from the eye. They are thus used to reduce intraocular pressure in the treatment of the eye disease glaucoma. Bimatoprost is an amide derivative related to prostaglandin amides (prostamides) obtained by the action of COX-2 on anandamide, the natural ligand for cannabinoid receptors. The pharmacological properties of prostamides cannot readily be explained simply by interaction with prostaglandin receptors; bimatoprost also inhibits prostaglandin F synthase. An unusual side-effect of bimatoprost is also being exploited cosmetically: it makes eyelashes grow longer, thicker, and darker.

6    

Figure 6. Prostaglandins

Figure 7. Latanoprost (left) and Prostamide E2 (right)

Isoprostanes represent a new class of prostaglandin-like compounds produced in vivo in humans

and animals by non-enzymic radical-mediated oxidation of membrane-bound polyunsaturated fatty acids independent of the COX enzyme. An isomer of PGF2α in which the two alkyl substituents on the five-membered ring were arranged cis rather than trans was detected in human urine and was the first of these compounds to be characterized. This compound was initially termed 8-iso-PGF2α, or 8-epi-PGF2α; however, as many more variants in the isoprostane series were discovered, it is now termed 15-F2t -IsoP. The first number refers to the position of the hydroxyl, F2 relates the compound to the prostaglandin class, and then t (trans) or c (cis) defines the relationship of the side-chain to the ring hydroxyls. The isoprostanes can be viewed as arising by a radical mechanism which resembles the enzyme-mediated formation of prostaglandins. Isoprostanes derived similarly from docosahexaeneoic acid (DHA), a major lipid component of brain tissue, have also been isolated, and these are termed neuroprostanes. Structurally related compounds are also found in plants; these are derived from α-linolenic acid and are termed phytoprostanes Interest in these isoprostanoid derivatives stems partly from the finding that certain compounds possess biological activity, probably via interaction with receptors for prostaglandins. For example, 15-F2t -IsoP is a potent vasoconstrictor and also aggregates platelets. A promising application relates to their origin via radical peroxidation of unsaturated fatty acids. Radicals are implicated in inflammatory and degenerative diseases such as atherosclerosis, cancer, and Alzheimer’s disease. Isoprostane analysis of urine or serum provides non-invasive monitoring of oxidative damage as an insight into these disease states.

Figure 8. A4 Neuroprostane (left) and Phytoprostane PPE1-1 (right)

7    

The thromboxanes were isolated from blood platelets, and whilst TXA2 showed high biological activity, TXB2 was effectively inactive. TXA2 is a potent stimulator of vasoconstriction and platelet aggregation. Aggregation of blood platelets to form a clot or thrombus is caused by increasing cytoplasmic calcium concentrations and thus deforming the platelets, which then fuse together. TXA2 has the opposite effect to PGI2, and presumably the development of thrombosis reflects an imbalance in the two activities. Both compounds are produced from the same precursor, PGH2, which is converted in the blood platelets into TXA2, and in the blood vessel wall into PGI2. Thromboxanes A3 and B3 have also been isolated from blood platelets. These are derived from Δ5,8,11,14,17-eicosapentaenoic acid and relate structurally to prostaglandins in the 3-series.

The leukotrienes are involved in allergic responses and inflammatory processes. An antigen–

antibody reaction can result in the release of compounds such as histamine or materials termed slow reacting substance of anaphylaxis (SRSA). These substances are then mediators of hypersensitive reactions such as hay fever and asthma. Structural studies have identified SRSA as a mixture of LTC4, LTD4, and LTE4. These cysteine-containing leukotrienes are powerful bronchoconstrictors and vasoconstrictors, and induce mucus secretion, the typical symptoms of asthma. LTE4 is some 10–100-fold less active than LTD4, so that degradation of the peptide side-chain represents a means of eliminating leukotriene function. LTB4 appears to facilitate migration of leukocytes in inflammation, and is implicated in the pathology of psoriasis, inflammatory bowel disease, and arthritis. The biological effects of leukotrienes are being actively researched to define the cellular processes involved. This may lead to the development of agents to control allergic and inflammatory reactions. Drugs inhibiting the formation of LTC4 and LTB4 are in clinical trials, whilst montelukast and zafirlukast have been introduced as orally active leukotriene (LTD4) receptor antagonists for the prophylaxis of asthma.

Figure 9. Leukotrienes

POLYKETIDES

Polyketides constitute one of the major classes of natural products grouped together purely on biosynthetic grounds. They are mostly produced by various microorganisms, fungi, some plants and marine invertibrates. This heterogenous group of compounds comprising of polyethers, polyenes, polyphenols, and macrolides is mainly derived from one of the simplest building blocks available in nature – acetic acid.

The original meaning of polyketide was a compound with polyketomethylene groups, which were said to contain multiple ketene groups. Also included were compounds derived from polyketomethylene structures, for example, by addition or loss of water or by decarboxylation. Loss of water typically produced cyclization of the carbon skeleton and therefore made possible further extensive modification. Eventually, it was learned that the central carbon skeleton of polyketides was formed by iterative decarboxylative condensations of malonic acid (or substituted malonic acid) thioesters, yielding

8    

polyacetate, polypropionate, or polybutyrate chains. Consequently, the contemporary polyketide concept is based in biosynthesis to which common underlying biosynthetic processes have so much potential variation that strikingly different chemical end products are produced.

From ancient times, some polyketides were known indirectly by their biological activities. The purgative materials in cascara, rhubarb, and senna are usually polyketide-derived anthracenes. Traditional antispasmodics, long used in the Middle East to treat angina, contain significant amounts of the polyketide khellin (5,8-dimethoxy-2-methy1-6,7-furanochromone). Perhaps the most famous example of a polyketide in history was the use of coniine-containing hemlock to execute Socrates in BCE 399. The pharmaceutical potential of polyketides has been increasingly appreciated and exploited in modem times as antibacterials (e.g. erythromycin, tetracycline, and tylosin), immunosuppressants [e.g. mycophenolic acid, rapamycin and tacrolimus (FK 506)], anticancer agents (e.g. daunomycin), antifungal agents (e.g. amphotericin and griseofulvin), cholesterol-lowering agents (e.g. lovastatin), and veterinary products (e.g. avermectin and monensin) .

The term ‘polyketide’ was termed by the organic chemist, John Norman Collie in 1907 to describe a class of aromatic molecules produce synthetically while studying pyrones. Figure 10 shows that upon treatment with aqueous sodium hydroxide, the pyrone (1) was converted to a phenolic compound, orcinol (3). Collie proposed that the triketone (2) was an intermediate. He also noted that the substitution pattern of the hydroxyl groups on alternate carbons of the aromatic ring of orcinol was characteristic of many phenolic natural products. He proposed that such poly-β-ketones might be produced in living cells as biosynthetic precursors of phenolic compounds with 1,3 pattern of hydroxyl groups and that the proposed polyketone intermediates could be hypothetically considered as polymers of ketene (4). He therefore coined the term polyketide to refer to both the polyketones and the derived phenolic natural products. Although Collie’s his biosynthetic assumption was advance of his contemporaries, his ideas failed to give an impact and the term was not used until more than 50 years later.

Figure 10. Collie's chemical studies on polyketones and polyketides

In mid-1950s, Arthur J. Birch independently had the same general idea by which time molecules

and chemical processes occurring in living cells were better known. Birch was able to extend the proposal with the suggestion that the proposed polyketones might arise in nature by condensation of acetate units. Unaware of the term polyketide, other natural product chemists coined the term ‘acetogenin’ for these classes of natural products. With the availability of radiolabelling, Birch was able to test his acetate hypothesis by feeding 14C labeled acetate to cells of a fungus, Penicillium patulum, that produce 6-methylsalicylic acid (6-MSA). In his hypothesis, four acetate units are linked head to tail to generate a triketo-acid and then one of the resulting keto groups is reduced to hydroxyl group. Generation of a carbanion at the β-keto residue would then allow an aldol condensation to form a six-membered carbocycle. A sequence of dehydration and enolisation reactions would then lead to the aromatic natural product. The resulting isotopically labeled natural product molecules confirmed his hypothesis. This was done and confirmed with many other phenolic systems as well.

9    

Figure 11. Biosynthetic study pf 6-MSA biosynthesis

Both terms were interchangeably used then but it was polyketides which gained universal

acceptance.

There has been dynamic search for new natural products since the middle of last century and with it the discovery of the number and diversity of polyketide structures increased. Their exact roles in their biological origins are not all completely elucidated. Certain polyketides function as pigments, virulence factors, infochemicals (ie. pheromones) or defense mechanisms (ie. toxins). Polyketide derivatives have provided important therapeutics for various clinical uses such as antibiotics, immunosuppresants, antiparasitics, cholesterol-lowering and antitumoral agents. But they have also been known for food-spoiling and virulence capacity.

Figure 12. Polyketide structures

Total chemical synthesis of polyketides is highly complex and thereby challenging; however, the

structural and functional diversity that results from controlled assembly acetate and propionate remains remarkable. On the other hand, with the accumulating information and literature which pertains to polyketide biosynthesis, design and manipulation of its biosynthetic pathways have been made for the production of polyketides especially in the search novel drug candidates.

Birch was able to produce a unified hypothesis for the genesis of both aliphatic and aromatic polyketides based partly on mechanistic reasoning and on analogy with fatty acid biosynthesis. Polyketide biosynthesis has a lot in common with fatty acid biosynthesis in terms of chemical mechanisms involved

10    

in chain extension and employed precursors such as acetyl CoA and malonyl-CoA units. In general, both are constructed by repetitive decarboxylative Claisen thioester condensations of activated acyl starter unit with malonyl-CoA-derived extender units. This is repeated until a defined chain length is established after which the thioester bound substrates are released from the enzyme complex. However, polyketide synthase (PKS) differs from fatty acid synthase (FAS) such that PKS uses a broader range of biosynthetic building blocks and assembles them in more variety. FAS catalyze a full reductive cycle after each elongation whereas reduction remains optional for PKS. The differences between PKS and FAS constitute a metabolic branch point between primary and secondary metabolism at which case both pathways may have diverged at an early stage during evolution.

Polyketide diversity results from events that occur before, during and after chain assembly. The primary determinants are the type and number of biosynthetic building blocks employed. Basic substitution patterns are based on β-keto processing reactions and the resulting stereochemistry as well as other reactions such as α- and β-alkylations. After the synthesis of the backbone, the chain is released from PKS by lactonization, hydrolysis, other nucleophilic attacks or reductive release, which may be followed by core cyclization. The resulting carbon skeleton may undergo secondary cyclizations, C-C cleavage and rearrangement reaction to give rise to different carba- and heterocycles. Finally, tailoring reactions such as biotransformations (C, O and N glycosylations, alkylations, acyl transfers, hydroxykation and epoxidation) and other known modifications (halogenations, transamination, nitrile formation and desaturation) decorate the polyketide structure.

Complex polyketides are mainly derived from acetate/malonate and propionate/methylmalonate and polyketide biosynthesis usually initiated by loading acetyl-CoA onto the synthase. There are, however, other alternative starter units and various activation and loading strategies.

Despite diversity, the overall chemical aspects of chain construction are effectively the same such that they are produced by Claisen reactions involving thioesters and a basic chain composed of a linear sequence of C2 units is constructed. Even when the starter unit is modified, or methylmalonyl-CoA is used as an extender unit instead of malonyl-CoA, the fundamental chain is effectively the same and is produced in the same mechanistic manner. Therefore, notwithstanding the various building blocks employed, these compounds are all still considered as derived via the acetate pathway.

Flavonoids are the largest group of polyketides with more than 6,000 known compounds. They occur mostly in glycosylated form and are often accumulated in the vacuole. Their main intermediates are chalcones which serve as precursors of all the other subgroups. Flavones, flavonols, anthocyanins, proanthocyanins and flavan-3,4-diols occur nearly ubiquitously in higher plants. Flavonoid subgroups are limited to certain species like isoflavonoids which are produced mainly in leguminous plants (Fabaceae). Flavonoids serve as pigments, radical scavengers (antioxidants) signaling molecules and defense compounds. Two phenylpropanoids from Humulus lupulus are 8-prenylnaringenin, the strongest phytoestrogens known so far and xanthohumol, a promising cancer chemopreventive agent.

Figure 13. Flavonoids

Macrolides are a large family of compounds, mostly with antibiotic activity, characterized by a

macrocyclic lactone (sometimes lactam) ring. Rings are commonly 12-, 14-, or 16-membered, though polyene macrolides are larger in the range 26–38 atoms. The largest natural macrolide structure discovered has a 66-membered ring. The macrolide antibiotics are composed principally of propionate units, or mixtures of propionate and acetate units. An example is Erythromycin A from Saccharopolyspora erythraea, a valuable antibacterial drug which contains a 14-membered macrocycle composed entirely of

11    

propionate units, both as starter and extension units, the latter via methylmalonyl-CoA. Erythromycin activity is predominantly against Gram-positive bacteria, and the antibiotic is prescribed for penicillin-allergic patients. It is also used against penicillin-resistant Staphylococcus strains, in the treatment of respiratory tract infections, and systemically for skin conditions such as acne. It is the antibiotic of choice for infections of Legionella pneumophila, the cause of legionnaire’s disease. Erythromycin exerts its antibacterial action by inhibiting protein biosynthesis in sensitive organisms. Avermectins have no antibacterial activity, but possess anthelmintic, insecticidal, and acaricidal properties, and these are exploited in human and veterinary medicine. The avermectins are also 16-membered macrolides, but their structures are made up from a much longer polyketide chain (starter plus 12 extenders) which is also used to form oxygen heterocycles fused to the macrolide.

Figure 14. Zooxanthellamide Cs (ZAD-Cs)

Polyene antifungals are a group of macrocyclic lactones with very large 26–38-membered rings. They are characterized by the presence of a series of conjugated E double bonds and are classified according to the longest conjugated chain present. Medicinally important ones include the heptaene amphotericin B and the tetraene nystatin. There are relatively few methyl branches in the macrocyclic chain. The polyenes have no antibacterial activity but are useful antifungal agents. Their activity is a result of binding to sterols in the eukaryotic cell membrane; this action explains the lack of antibacterial activity, because bacterial cells do not contain sterol components.

Figure 15. Amphotericin B, a polyene antifungal

Tetracyclines are formed from the condensation of eight C2 units and a malonamide starter. It is

produced by many Streptomyces species from an unusual group of bacteria, the actinomyces. Examples are terramycin, aureomycin and tetracycline which are broad spectrum antibiotics.

12    

Figure 16. Tetracyclines

Macrolide formation does not always occur, and similar acetate/propionate precursors might also be expected to yield molecules which are essentially linear in nature. An example is Mupirocin, an antibiotic used clinically for the treatment of bacterial skin infections and for controlling Staphylococcus aureus, particularly methicillin-resistant Staphylococcus aureus (MRSA), when other antibiotics are ineffective. It has an uncommon mode of action, binding selectively to bacterial isoleucyl-tRNA synthase, which prevents incorporation of isoleucine into bacterial proteins. Mupirocin is produced by Pseudomonas fluorescens.

Figure 17. Mupirocin (linear polyketide)

Polyether structures are found in some toxins produced by marine dinoflagellajtes, which in turn, are taken up by shellfish, thus passing on their toxicity to the shellfish. Examples of these are okadaic acid, brevetoxins and ciguatoxins.

13    

Figure 18. Brevetoxins (Polyethers)

When cyclizations of the growing poly-β-keto chain occurs, aromatic or phenolic polyketides may

be formed. An example of which is Mevastatin (Compactin), which is produced by cultures of Penicillium citrinum and Penicillium brevicompactum, and was shown to be a reversible competitive inhibitor of HMG-CoA reductase, dramatically lowering sterol biosynthesis in mammalian cell cultures and animals, and reducing total and low-density lipoprotein (LDL) cholesterol levels, and Lovastatin, a slightly more active version of Mevastatin.

Figure 19. Aromatic poleketides

14    

METHODOLOGY The isolation of fatty acids and polyketides follows the staple extraction-purification-analysis procedure to extract other secondary metabolites. Usually, secondary metabolites only constitute 0.01% dry weight of the plant. The specific origin of the natural products varies. They may be unstable and even present itself as complex mixture. The metabolites produced vary in its source and the environment they were cultivated in. For microorganisms, the metabolites produced depend on the medium in which the microbe is grown and other fermentation details. For fungi, the metabolites can be found in the mycelium or they are excreted in the broth or medium. Insects and marine organisms may have stored and modified compounds obtained from the food they intake. The basic isolation scheme follows crushing the biological material with a solvent, sequential extraction of different fractions of the sample, separation of specific compounds of the same nature. Solvents used for extraction vary in polarity. For lipid material, specifically, non-polar solvents are used such as petroleum ether. Commercial extraction can also be utilized such as by the use of kits and complicated apparatuses. Different extraction procedures include steam distillation and super-critical carbon dioxide as solvent. From obtaining the crude extract, acidic, basic, and neutral fractions are separated in order to somehow purify the target metabolite. A solution of the extract in organic solvent is shaken with an organic base such as aqueous sodium hydrogen carbonate to remove carboxylic acids as their water-soluble sodium salts. Sodium hydroxide solutions can be used to extract the more weakly acidic phenols. Treatment of the extract solution with an acid, such as dilute hydrochloric acid, will remove the bases such as the alkaloids as their salts. With the extraction of the acidic and basic fractions, the neutral compounds remain behind in the organic phase. Ester hydrolysis, auto-oxidation, and rearrangement may occur in the extraction. This may lead to artifacts, compounds arising from a natural product by human intervention. Separation of the compounds in solution must then be chemically mild. Separation is needed in order to characterize and examine minor components because many natural products normally occur in closely related species as they have mono, di, and trihydroxy derivatives.

Figure 20. General schematic of lipid extraction and analysis

15    

When it comes to secondary metabolite extraction, the solvent has to diffuse into cells, solubilize the metabolites, and diffuse out of the cells enriched in the extracted metabolites. The extraction method should be exhaustive in the sense that it should be fast, simple, reproducible, and extract as much metabolites as possible. CRUDE EXTRACTION Extraction methods are employed to obtain the secondary metabolite being considered. These methods should be classified in terms of the source of the metabolites, whether they be solid or liquid. For more physical sources, the sample may undergo maceration, ultrasound-assisted solvent extraction, percolation, Soxhlet extraction, pressurized solvent extraction, and extraction under reflux. Maceration involves pounding and pulverizing the source of the metabolite in a suitable solvent. This is then followed by filtration and centrifugation. However, this method is time consuming and can consume large volumes of solvent, thus, leading to potential loss of metabolites. Some compounds cannot be extracted as efficiently if they are poorly soluble at room temperature. However, there is less risk for the degradation of thermolabile metabolites in maceration. Ultrasound-assisted Solvent Extraction applies ultrasound to plant powder in a vial that is placed in an ultrasonic bath. Mechanical stress in induced by the production of cavities. This increases the solubilization of the metabolites in the extraction solvent, thus, improving extraction yield. This method is applied for the extraction of intracellular metabolites from plant cell cultures. In percolations, the powdered source or plant material is soaked in a solvent in a percolator. If the plant powder is too fine or if the material swells excessively, it can clog the percolator. There is also a chance that the material is not distributed homogeneously that the solvent may not be able to reach all areas and the extraction shall be incomplete. This process needs a lot of time and solvent. Soxhlet extraction is one of the more convenient ways in which to extract metabolites. The plant powder is placed in a thimble in an extraction chamber where a suitable solvent is added and heated under reflux. This process is a continuous process. It is involves less time and solvent. Degradation of thermolabile compounds and formation of artifacts may happen with the use of the Soxhlet apparatus as the system is constantly heated at the boiling point of the solvent. Pressurized solvent extraction is used for rapid and reproducible initial extraction because the powdered material is placed in an oven. The solvent is pumped from a reservoir to fill the cell where the sample is located. The solvent is heated and pressurized at programmed levels for a certain period of time. Nitrogen gas is used to purge the sample. High temperatures and pressures increase the penetration of solvent into the material, which improves the solubility of the metabolite in the sample. This process only needs a small amount of solvent, and repeated extractions can be performed. Extraction under reflux such as steam distillation collects the vapors condensed when the sample that is immersed in a solvent is heated until boiling point. Using this method risks degradation of thermolabile components. For liquid sources, the samples undergo partitioning between immiscible or miscible solvents because the sample is actually a complicated mixture of several compounds with varying chemical and physical properties. The metabolites can be extracted with solvents of increasing polarity to obtain different fractions that can be further resuspended and in other solvents. This will enable the extraction of the natural product in its pure form.

16    

Figure 21. Basic schematic of liquid-liquid extraction

LIPID EXTRACTION Simple lipids, such as free or unesterified fatty acids are often part of large aggregates in storage tissues. Complex lipids such as esterified fatty acids and polyketides are much more difficult to extract because they are constituents of membranes and are closely associated with protein and polysaccharides. Weak hydrophobic or Van der Waals forces, hydrogen bonding, and ionic bonding are the most common intermolecular forces of attraction involved when the lipids are linked to cellular components. The role of the solvent used for extraction is then to have two main characteristics: to readily dissolve the lipid, and to overcome the interaction between the lipids and matrix tissue. Lipids with functional groups of low polarity, such as the fatty acids, can be extracted by hydrocarbons, such as hexane, cyclohexane, toluene, and chloroform. They tend to be less soluble in solvents such as alcohols. The polar lipids of the other hand, such as polyketides, are only sparingly soluble in hydrocarbon solvents. The more common solvent used for the extraction of polyketides is ethyl acetate. A solvent mixture composed of isopropanol-hexane (2:1 by volume) is often used for its limited toxicity. It has yet to be tested on its versatility with tissues. A more efficient solvent combination would be of chloroform-methanol (2:1 by volume) to extract lipids from animal, plant, and bacterial tissues. It is found that more lipids are extracted when the tissue is first extracted with methanol alone, then chloroform. More than one extraction may be carried out and rehydration may be necessary for lyophilized tissues. Lipid extracts are never obtained pure. They tend to contain different contaminants such as sugars, amino acids, urea, and salts. In order for the extract to be analyzed, the contaminants must be removed. This can be done by washing the crude extract with chloroform-methanol mixture (2:1 v/v) and equilibrating it with ¼ of its volume in saline solution. The mixture will partition into two layers, the lower phase being that of the chloroform-methanol-water layer, which contains the lipids, while the upper phase contains the contaminants. Aside from this, some lipids can be physically trapped in the tissue matrix. It is helpful to utilize water because it assists extraction by means of swelling the biopolymers. Sometimes denaturation techniques, whose target molecules are those of the cell walls, must be employed before a thorough lipid extraction is possible.

17    

The lipid extraction methods can be classified into three categories: batch solvent extraction, semi-continuous solvent extraction, and continuous solvent extraction. Batch solvent extraction involves mixing the sample and the solvent in a suitable container, such as a separatory funnel. Continuous shaking of the container with the solvent and sample is applied before the solvent and residues are separated by gravity, filtration, or centrifugation. This technique is applied with solvent mixtures that form a polar and non-polar phase. The non-polar phase usually contains the extracted lipids. Semi-continuous solvent extraction involves extraction of the lipids via repeated washing or percolation with an organic solvent under reflux. The most common example of such extraction is by the use of Soxhlet extraction method. Continuous solvent extraction is similar to semi-continuous solvent extraction, however, instead of the solvent building around the sample, it trickles through it. This reduces the amount of time the extraction must be carried out. However, channeling of the solvent can occur, meaning it can preferentially take certain routes through the sample. This makes the extraction inefficient. The most common method used is for this type of extraction is the Goldish method. AUTOXIDATION Polyunsaturated fatty acids and polyketides, once extracted, must be protected because they autoxidize rapidly in air. When these lipids are autoxidized, they may not be able to be used for analysis by chromatography. Autoxidation happens when free radicals attack the sample or molecule and is exacerbated by light and metal ions. This radical reaction is autocatalytic. Linoleic acid is oxidized twenty times as much as oleic acid. Cytosporacin, a highly unsaturated polyketides, is quite at risk for autoxidation. A double bond in fatty acids and polyketides can increase the rate of destruction by two to three-fold. Antioxidants are usually added in order to protect the lipid extracts. Butylated hydroxy toluene (BHT) is a common synthetic antioxidant used in order to limit the autoxidation of the lipids to be extracted. The antioxidant should be added at a level of 10 to 100 mg/L relative to the lipid concentration. When a large amount of antioxidant is added, however, they may act as pro-oxidants. They can be removed with the solvents by evaporation in a steam of nitrogen. The lipids should be handled in an atmosphere of nitrogen. They solvents are best removed by means of a rotary evaporator at a certain temperature. PREPARATION OF DERIVATIVES OF FATTY ACIDS Prior to analysis and even to aid in separation, fatty acid components of lipids must be converted to low molecular weight non-polar derivatives. This improves peak shape and resolution when analyzed by gas chromatography. Thus, it is necessary to consider which derivatives should be prepared and what method should be employed to produce such derivative. Hydrolysis/Saponification Gas chromatography analysis of free fatty acids is difficult because they are not very volatile. Saponification or hydrolysis reduces the molecular weight of the fatty acid and increases its volatility. Saponification or hydrolysis is also employed when the fatty acids are required in free form. This can be done by heating the lipid mixture under reflux with an excess of dilute aqueous ethanol alkaline and the products are recovered. The mechanism of the saponification of the fatty acids involves the cleavage of the ester bond between the fatty acid and the glycerol moiety under heat. The products are the fatty acids, diethyl et her-soluble non-saponifiable material, and water-soluble hydrolysis products. This

18    

process frees fatty acids for a lipid mixture with non-saponifiable materials such as sterols, alcohol, hydrocarbons, pigments, vitamins, and other non-saponifiable materials. This method is usually followed by methylation Methylation and Esterification Acid-catalyzed esterification and transesterification Methylation of fatty acids reduces the polarity and increases the volatility of the said lipid. Esterification of free fatty acids and transesterification of fatty acids linked by ester bonds to glycerol and cholesterol are done by heating them with a large amount of methanol in the presence of and acidic catalyst. The most common and mildest reagent used is anhydrous hydrogen chloride in methanol. This is prepared by bubbling hydrogen chloride gas into dry methanol. The general mechanism of fatty acids by alcohols in the presence of a suitable acid catalyst involves the protonation of the acids to give an oxonium ion, which then undergoes an exchange reaction with an alcohol to give the intermediate. The intermediate loses a proton to become an ester.

Scheme 1 Acid catalyzed esterification

Scheme 2 Acid catalyzed transesterification

A solvent composed of 1-2% (v/v) concentrated sulfuric acid in methanol transesterifies lipids in the same manner and rate as with using methanol hydrogen chloride. This method cannot be prescribed when dealing with polyunsaturated fatty acids because sulfuric acid is a strong antioxidant. Long reflux times and high concentration of sulfuric acid can lead to the formation of colored by products and the destruction of polyenoic fatty acids. Boron trifluoride-methanol is a powerful acidic catalyst for the esterification of fatty acids. This can be used for a wide range of lipid classes, but general longer reaction times are necessary than with free fatty acids. However, boron trifluoride-methanol can produce methoxy artefacts across the double bonds when applied to unsaturated fatty acids. Also, artifact formation is commonly seen with aged reagents.

19    

Base catalyzed Transesterification In base catalyzed transesterification, esters form an anionic intermediate in the presence of an alcohol anion. This intermediate can dissociate back to the original ester or form the new ester. In the presence of a large excess of the alcohol from which the anion was derived, the equilibrium point of the reaction will be displaced until the sole product is the new ester. An unesterified fatty acid is converted to a carboxylate ion in a basic solution, and this is not subject to nucleophilic attack by alcohols or bases because of its negative charge. Esterification cannot occur.

Scheme 3 Base catalysed transesterification

Sodium or potassium methoxide in anhydrous methanol is the most useful basic transesterifying agents. These are prepared by dissolving the clean metals in anhydrous methanol. Under normal conditions, no isomerization of double bonds in polyunsaturated fatty acids, but prolonged or careless use of basic reagents can cause alterations of fatty acids. Sometimes, an additional solvent such as toluene or tetrahydrofuran is needed to solubilize non-polar lipids. Methyl iodide reacts with the sodium or potassium salts of fatty acids in the presence of a polar aprotic solvent such as dimethylacetamide to form methyl esters. Alkaline transesterification reagents under mild conditions do not affect amide-bound fatty acids. Diazomethane Diazomethane reacts rapidly with unesterified fatty acids to give methyl esters, but does not allow transesterification of other lipids. This reaction happens in the presence of a methanol catalyst. Diazomethane is prepared in ethereal solution by the action of alkali on a nitrosamide.

Scheme 4 Reaction of diazomethane with a fatty acid

SEPARATION METHODS When a semi-purified extract of the lipids are finally obtained, they must be separated into lipid classes. This is how one retrieves fatty acids and polyketides specifically. Further more, the lipids cannot be fully analyzed if they are not separated further to narrower classifications of lipids. Distillation Distillation was initially used to separate mixtures of fatty acids and esters derived from natural fats. This technique is most commonly applied due to their thermolability. With distillation, the fatty acids can be separated into groups depending of their chain lengths. This is because the difference in boiling points or volatility of fatty acids depends on the difference in chain length of fatty acids. The fatty acid vapors are

20    

condensed in the surface condensers. To remove low-volatile light end components, the main distillate is taken out as a middle fraction and the light ends as the top fraction. Distillation also makes it easier to obtain amounts of individual fractions that are large enough to be used for further chemical and biological purposes. The disadvantage to distillation is that one cannot distinguish between saturated and unsaturated compounds of the same chain length. Kugelrohr distillation is one of the most often used methods of distillation. It is a short-path vacuum distillation apparatus used to distill small amounts of compounds with high boiling points under reduced pressure. It distills with minimal hold-up and sample loss, and it removes color and particulates. It can be used for sublimation and solvent evaporation. For polyketides, however, not much has been said regarding the use of distillation as a means of separation of the classes or kinds of polyketides. A study made by Appendino, et. al, however used fractional distillation under vacuum and normal air condenser to further purify Ethyl 6-Methylsalicylate after its biosynthesis. Crystallization Saturated and unsaturated fatty acids can also be separated by means of salt-solubility methods because these lipids form salts with metallic ions, whose solubilities in water and organic solvents vary with nature of the metallic ion and the chain length, degree of unsaturation, and other characteristics of acid radicals. The most widely used salt separation method is the use of lead salts or soap of fatty acids in ether. Ethanol can be substituted to ether. Ethanol usage is preferred for preparative purposes. This simple method is applied because of the direct precipitation of solid fatty acids because the lead salts of the fats crystallize. They can then be filtered with great rapidity and ease. The solid or saturated acids are regenerated in insoluble lead salts by boiling the fatty acids with dilute HCl. The cake of solid fatty acids is separated and the aqueous layer is extracted with ether. Low-temperature crystallization can be used to separate fatty acids and monoesters, and also for the separation of glycerides and other lipid substances. This method utilized the melting point characteristic of fatty acids, which depends on the chain length and degree of unsaturation. As the chain length increases, the melting point of fatty acids increases. For long chain fatty acids, however, unsaturation results in a decrease of melting point. At low temperatures, short chain fatty acids crystallize and polyunsaturated fatty acids can be isolated from the rest of the fatty acids. Organic solvents such as hexane and acetone facilitate separation of fatty acids during crystallization. Polyketides can be separated by low-temperature crystallization as well by using ether or hexane. Usually, the extract crystallizes upon standing at 4C. Urea-Fatty Acid complexes revolutionized techniques in lipid chemistry. It was found that urea forms crystalline inclusion compounds with straight-chain compounds in water, methanol, and ethanol. Fatty acids are included in such classification. The inclusion compounds are combinations of two or more molecules, wherein one is crystalline framework of the other. The longer the chain length, the more stable the urea complex. The degree of unsaturation of the compound increases with the increasing deviation in behavior from the normal saturated compound. Thus, with these found relationships, it can be said that saturated fatty acids form urea complexes preferentially to monounsaturated, mono- preferentially to di unsaturated, and so on. Using urea-fatty acid complexes in the separation of lipids into their classes reduced the free fatty acid content of glycerides to 1% or less. Remarkable properties of urea complexes involve their resistance towards autoxidation, and that one is able to tell the length of the carbon chain in terms of the formation of the urea complexes and their dissociation temperature. After the urea-fatty acid

21    

complexes fractionated, they can be easily decomposed by warming, this liberating the unsaturated fatty acids. There has yet to be any recorded study on the use of urea complexation with the separation of polyketides from a mixture.

Figure 22. Urea complexation of free fatty acids

Chromatography Adsorption chromatography is used to separate several unsaturated fatty acids using charcoal, silicic acid, alumina or even filter paper. It can separate lipids of different polarity. The insoluble polar material such as the silicic acid is pack into a long, thin glass column, and the lipid mixture, usually in chloroform solution, is applied at the top of the column. The polar lipids bind to the polar phase, but the neutral lipid pass directly through the column and emerge in the first chloroform wash. The polar lipids are then eluted, in order of increasing polarity, by means of washing the column with solvents of increasing polarity. Two of the most commonly used methods of column chromatography are Thin Layer Chromatography (TLC) and Column Chromatography. Thin-layer chromatography employs the same principle with adsorption chromatography. A thin layer of silica gel with the sample lipids dissolved in chloroform or a solvent is placed in a chamber saturated with an organic solvent. When the solvent rises on the plant, it takes the lipid with it. The less polar lipids move fastest. The plate can be dyed with iodine fumes or rhodamine. It allows the lipid to fluoresce, which make them easily detectable. The regions containing the separated lipid can be scraped from the plate and recovered by extraction with an organic solvent.

Figure 23. Separation of methyl ester derivative of unsaturated FA by TLC on silica gel with 10% (w/w) silver nitrate. Mobile phase is hexane-diethyl ether, where plate A is 9:1 (v/v) and plate B is

2:3 (v/v)  

22    

Column chromatography is more often used for the purification of liquid. An impure or crude sample solution is loaded on a column of adsorbent, such as silica gel or alumina. An organic solvent of the eluent flows down the column. The components of the sample solution separate from each other by partitioning between the stationary packing material and the mobile eluent. Depending on the polarity of the solvents used, the pattern of the migration of the lipids in the system will be the same as TLC. Flash column chromatography is a rapid form of column chromatography that is based on optimized pre-packed columns where in solvent is pumped at a high flow rate. It is a quick and easy way to separate complex mixtures of compounds. This method utilizes a plastic column filled with some form of solid support such as silica gel. The sample is placed on the top of support. The rest of the column is filled with an isocratic solvent, which enables the sample to run through the column in the presence of pressure or pumps to speed up the separation. Reversed phase partition chromatography involves the partition of the lipid mixture between a polar mobile phase and a non-polar stationary phase. Reversed-phase paper chromatography is useful when dealing with very small samples. It is necessary to impregnate the paper with various materials such as latex, hydrocarbons, or silicones, which support the non-polar phase. This chromatographic method is able to separate the even-numbered C6-C12 fatty acids when methanol and acetone or methanol and water are used as mobile solvents. If acetone and water was used as mobile solvents, reversed polarity partition chromatography can separate C4-C29 fatty acids. High Performance Liquid Chromatography is a chromatographic technique that utilizes different stationary phases, a pump that mobilizes the mobile phase/s and analyte through the column, and a detector to provide a characteristic retention time for the analyte with the stationary and mobile phases. HPLC operates at ambient temperature so the degradation of heat-sensitive functional groups is prevented. Adsorption chromatography or HPLC on columns of silica gel is used to determine which fatty acids and polyketides have polar functional groups, especially oxygenated moieties, i.e. hydroperoxides. If handled with care, isomers differing in the position of hydroperoxy or hydroxy groups on an aliphatic chain can be separated. Chiral chromatography makes it possible to resolve enantiomeric functional groups, and hydroperoxy, or hydroxy groups on a variety of commercial chiral stationary phases. Gas Chromatography uses a high-boiling liquid as stationary phase and inert gas as the mobile phase to analyze compounds that can be vaporized without decomposition. It separates volatile components of a mixture according to their relative tendencies to dissolve the inert material in the chromatography column, and to volatilize and move through the column, carried by a current of an inert gas such as helium or nitrogen. It can be used to separate free fatty acids from 1 to 12 carbon atoms. Improvement of this technique was done by methylating the fatty acids prior to separation. The volatility and separation at temperatures below 250C of the fatty acids was enhanced. The fatty acyl methyl esters are loaded onto the gas-liquid chromatography column, and the column is heated to volatilize the material. The more soluble the molecules are in the column material, the more they partition to the column itself. The less soluble are carried by the stream of helium and emerge first from the column. The order of elution will depend on the nature of the solid adsorbant in the column and the boiling point of the components of the liquid mixture. As the chemicals exit the column, they are detected and identified electronically. To date, gas chromatography is not utilized as much in the separation and analysis of polyketides.

Silver Ion Chromatography, whether it is in the form of TLC or HPLC, is able to give a better resolution of trans- and cis- forms of a given positional isomer of a mono-unsaturated fatty acid. This technique uses silver ions that are bound to an ion-exchange support. The principle behind this type of

23    

chromatography is like that of the salt formation of unsaturated organic compounds with metal ions. The complexes are of the charge-transfer type, meaning the unsaturated compound acts as an electron donor and the silver ion as an electron acceptor. The bond formation in consideration is that of a s type bond between the 2pπ of an olefinic double bond and the free 5s and 5p orbitals of the silver ion, and a π acceptor backbone between the occupied 4d orbitals of the solver ion and the free antibonding 2pπ* orbitals of the olfenic bond. Using silver ion chromatography, one can conclude that unsaturated acyclic and carbocyclic compounds form more stable complexes than do aromatic, the stability decreases with increasing chain length, the stability decreases with an increasing number of substituents at the double bond in the order: RCH=CH2 > R2C=CH2 > cis RCH=CHR > trans RCH=CHR > R2C=CHR > R2C=CR2.,

and the stability increases when a hydrogen from a molecule of the olefinic bond is replaced wit a deuterium or tritium because or greater electron release. Moreover, the cis/trans rule and the cahain operates universally, with cis olefinic double bonds forming stronger complexes in comparison with the corresponding trans molecules, and the longer the chain length, the lower the stability of the complexes formed. Aside from the complexation of unsaturated fatty acid to the silver ion, the supporting material is also a factor to the migration order of lipids. The most widely used support material, silica gel, possess appreciable polarity and absorption activity. The elution of the lipids in the systems is a product of a mixed retention mechanism. This means that the position of the double bond when the unsaturated lipid complexes with the silver ions will influence the conformation of the molecule. Thus, the strength at which the molecule adheres to the support will also depend on the interaction between the functional groups (i.e. FAME) and the silanol moiety.

Figure 24. Principle behind Silver Ion Chromatography

Figure 25. The Dewar model of interaction between a silver ion and an olefinic double bond

MOST COMMONLY USED EXTRACTION PROCEDURES/PROTOCOLS FOR FATTY ACIDS The "Folch" Procedure The composition of the solvent is an 8:4:3 ratio of chloroform, methanol, and saline solution. The tissue sample can be homogenized with methanol and chloroform immediately, or with methanol first, then chloroform. It has been said that the latter produces better results. Two to three extractions can be

24    

performed. Generally, there is no need to heat the solvents, but sometimes, it is necessary for wet bacterial cells. However, the extractability of the tissue is variable and depends on the nature of the tissue and the lipids. The Bligh and Dyer Method This is the second more commonly used extraction procedure. The tissue sample (using the endogenous water from the wet tissue as the ternary component), chloroform, and methanol were combined to give a single-phase system for homogenization. After filtering, the residue was re-homogenized with fresh chloroform to ensure that the simple lipids were extracted completely. The combined organic layers were added to fresh saline solution to produce a biphasic mixture. The chloroform layer produced contained the lipid. Extraction of Plant Tissues The lipids in plant material are very liable because they are able to undergo extensive enzyme catalyzed degradation when extracted with chloroform-methanol mixtures. Preliminary extraction with propan-2-ol will overcome such degradation. n-Propanol is able to separate fatty acid amylose inclusion complexes or those linked via ionic or hydrogen bonding to hydroxyl groups of the starch components. After the initial extraction, re-extraction of the solid residue with chloroform-methanol solution may be done. After evaporation of the solvent, the crude extract can be taken up in fresh chloroform-methanol and given a "Folch" wash to eliminate non-lipid contaminants. IDENTIFICATION OF FATTY ACIDS AND POLYKETIDES Spectroscopy Fatty acids and polyketides may be subjected to IR spectroscopy in free state, bounded state, or as methyl ester derivatives to determine important and distinguishing functional groups. The esterified version is preferred for fatty acids because a band due to the free carboxyl group between 10 to 11 µm may obscure a number of other important features in the spectra. Sharp bands at 5.75 µm and 8.6 µm are due to the esterified carbonyl function. For free fatty acids, the first of the bands are displaced to 5.9 µm and broad bands are found at 3.5 µm and 10.7 µm. cis-Double bonds give rise to small bands at 3.3 µm and 6.1 µm. trans-Double bands give rise to a sharp peak at 10.3 µm. The remaining bands are absorption frequencies of the hydrocarbon chain. Using Raman spectroscopy, structural features in unsaturated fatty acids and polyketides are identified. It exploits the characteristic vibrational frequencies of chemical bonds, which is influence by which atoms are bound, the bond’s saturation, and its molecular environment. The characteristic Raman bands are found for cis-double bonds (1656 cm-1), trans-double bonds (1670 cm-1), triple bonds (2232 and 2291 cm-1), and terminal triple bonds (2120 cm-1). Ultraviolet spectroscopy is used to detect or confirm the presence of fatty acids and polyketides containing conjugated double bond systems or chromophores and to observe chemical or enzymatic isomerization of fatty acid and polyketide double bonds in which conjugated systems are found. The greater the double bonds in the conjugated system, the greater bands seen at specific wavelengths. Different geometrical isomers have different spectra. The greater the number of trans-double bonds, the higher the extinction coefficient, and the shorter the wavelengths of the maximum band.

25    

Nuclear Magnetic Spectroscopy has been applied to the identification of lipid structures in terms of detection and location of double bond systems in fatty acid chain and in their methyl ester derivatives and in polyketides. In proton NMR, the number of peaks and their respective positions in the spectra aids in determining the conjugation of the fatty acids and polyketides. The integration of the signals assists in confirming the assignments to particular protons. Free hydroxyl groups gives rise to two separate signals that are caused the –OH molecule itself, its intensity and position that may vary due to hydrogen bonding effects, and the CHO- proton. Methyl braches on aliphatic chains do not give signals that are helpful in locating their positions, unless the branch is immediately adjacent to either end of the molecule. Cis- and trans- isomers of unsaturated fatty acids are readily distinguished by the use of lanthanide shift reagents, and 13C NMR spectroscopy has been suggested as a means of quantification of trans-unsaturation in lipid mixtures. Carbon NMR however, is somehow insensitive to many purposes, so it is best to be couple with mass spectrometric studies for better analysis. Mass Spectrometry Mass spectrometry is invaluable in the analysis of fatty acid derivatives and polyketides because it gives the location of double bonds, molecular weight, retention times, branched functional groups, and oxygenated substituents in the fatty acid sample. This method is able to give information on the fatty acid and polyketides with regard to its structure – whether they be saturated strain-chains, unsaturated, branched chain, carbocyclic, oxygenated, and other miscellaneous classification of such lipid class.

Scheme 5 Example of fatty acid fragmentation

Atmospheric pressure ionization (API) mass spectrometry, a variation on the commonly employed MS in conjunction with liquid chromatography (LC-MS), provides information on the behavior of polyketides under API. The introduction of API sources extends the range of analytes that can be studied by MS to those that are high in molecular weight, thermally labile, and polar. It gives a spectrum that is rich in fragments, thus rich in structural information when properly interpreted. API uses the high flow rates of the typical HPLC procedure without losing a large part of the fraction/sample to waste. The mobile phase that contains the eluting analyte is heated to high temperatures, sprayed with high flow rates of nitrogen by means of an electrospray ionization technique (ESI), and the entire aerosol cloud is subjected to a corona discharge that creates ions. Skimmers are used to transmit ions to the high vacuum region and subjected to the mass analyzer.

Figure 26. Schematic of API-MS

26    

Isotopic Labeling Isotopic labeling is a method to tracking the passage of a sample through a system or a process by labeling or including less abundant isotopes in its chemical composition. When the less common isotopes are detected later on the in the study, system, or process, it will indicate that it came from the labeled substance. It will be easy to detect the presence of labeling isotopes because of their difference in weight and the consequences of having molecules that contain the isotopes and thus, different vibrational modes. Mass spectrometry and IR spectroscopy can be used to detect the presence of such isotopes. Isotopic labeling can also be used to study chemical reactions. Specific atoms from the reactant are replaced by their isotope. By means of NMR, it will be easy to detect where a particular molecular fragment in the reactant ends up as a particular fragment of one of the products. For fatty acids, for example, the elucidation of the mechanism of the acid-catalyzed esterification begs the question of which starting product the oxygen atom of the product water molecule comes from or which oxygen is in the ester. Isotopic labeling can be done to answer this equation, in which the alcohol is labeled with the oxygen isotope 18O.

Scheme 6 Esterification with an 18O-labelled alcohol

For polyketides, isotopic labeling is particular used in structure elucidation of a sample that has undergone biosynthesis. Polyketides are made up of acetate units. Methyl branchings may occur from the incorporation of propionate instead of acetate or the result of a subsequent transfer of a methyl group of S-adenosyl methionine (SAM). This can be determined by fermenting Phoma betae Fr. from a media that contained 13C-labeled substrate [1-13C] acetate, [2-13C] acetate, or [Me-13C] methionine. Each betaenone B, synthesized in one of the nutrient media were individually investigated by 13C NMR spectroscopy. Results showed that betaenone B is made up of 13C-labeled acetate units. Moreover, it has been proven that the methyl branching were due to the methyl transfer from SAM instead of the incorporation of propionate, as the corresponding 13C NMR signals of the methyl groups were amplified after the cultivation in the nutrient medium containing [Me-13C] methionine

Scheme 7 Biogenesis of betaenone B

27    

Chemical Degradative Procedures Chain-length Determination Catalytic hydrogenation to form saturated compounds can help in the determination of the chain length of unsaturated fatty acids. The resulting solution can then be identified positively by gas chromatography or high performance liquid chromatography. In order to obtain the best results, it is preferred to carry out the reaction on the single fatty acid of interest. Location of Double Bonds Oxidative fission is generally employed to determine positions of double bonds in alkyl chains, followed by gas chromatographic identification of the products. The most commonly used methods are von Rudloff oxidation and ozonolysis followed by oxidative or reductive cleavage of the ozonide. Permanganate-periodate oxidation or von Rudloff oxidation involves oxidizing the methyl ester of the unsaturated fatty acid or polyketides in tert-butanol solution by a solution containing a small amount of potassium permanganate with a larger amount of sodium metaperiodate. The sodium metaperiodate regenerates the permanganate as it is reduced. The whole of the solution is buffered by potassium carbonate. When the reaction is complete, the solution is acidified, excess antioxidant is destroyed by sodium bisulfite, and the products are extracted with diethyl ether. The products are monocarboxylic and dicarboxylic acids. These acids are then methylated for further analysis. The resulting solution can then be identified positively by gas chromatography or high performance liquid chromatography. Ozonolysis and Reductive or Oxidative Cleavage is the process of producing ozonides as ozone attacks olefins. The produced ozonide can be cleaved reductively to aldehydes and aldehyde-esters by reagents such as dimethyl sulfide in methanol. The process of breaking carbon-carbon double bonds with ozonolysis involves the O3 molecules bonding to both molecules at the separation point. Both molecules, when separated, will have one of four functional groups where the split occurred: an alcohol, a deprotonated alcohol, an aldehyde, or carboxylic acid. The presence of these functional groups when analyzed by IR NMR, Mass Spectrometry, or other analytical methods that can detect such species can be used as a marker or indicator of the location, and number of double bonds found in the fatty acid sample. Special procedures for configurational isomers or for polyunsaturated fatty acids and polyketides must be employed because of the complexity brought by the structures of such class of lipids, i.e. cis- and trans- configuration of the double bond system. Reduction of the compounds prior to oxidation overcomes such complexity. Hydrazine is known for the reduction of fatty acids, without causing any double bond migration or stereomutation. With this reagent, monoenes are formed. The isomers are then found with double bonds in each of the positions in which they were present in the original polyunsaturated fatty acid. The cis- and trans-Monoenes are separated by silver ion chromatography, and the structures are identified fully by analytical methods aforementioned. Location of Other Functional Groups Isolated triple bonds do not exhibit particularly distinctive features when examined by spectroscopic techniques other than Raman spectroscopy. A TLC spray composed 4-(4’-nitrobenzyl)-pyridine (5%) in acetone gives off a violet color with acetylenes. The position of the triple bond can be determined by mass spectrometry. Permanganate-periodate reagent readily cleaves triple bonds. Though

28    

at a slower rate, ozone can react with triple bonds and the products of the reductive cleavage are mono- and dibasic acids rather than aldehydes and aldehyde-esters, so that double and triple bonds in a single fatty acids can be differentiatied. Allenic groups have distinctive IR and NMR spectra. They also exhibit a marked optical activity. The position of the group in the fatty acid chain can be determined by partial reduction by hydrazine and oxidation of the monoene fragments by ozone. Oxygenated functional groups, such as keto, hydroxyl, or epoxyl in fatty acid chains and polyketides can be determined in terms of their presence and location by spectroscopic value. Gas chromatography can be used to detect the presence of hydroxyl groups before and after the preparation of volatile derivatives such as acetates of trimethylsilyl ethers. The fatty acid chain or polyketide is hydrogenated to eliminate any multiple bonds before free hydroxyl or epoxyl groups are converted to iodide by means of iodine and red phosphorous. Hydrogenolysis with zinc and hydrochloric acid in methanol removes the iodine atom in the aliphatic chain. The keto groups will be reduced to hydroxyl groups by the action of sodium borohydride. The resulting saturated straight-chain fatty acid can be analyzed via gas chromatography. Mass spectrometry can be used to determine the position of the hydroxyl group. Epoxyl groups are cleaved with periodic acid in halogenated solvents or in diethyl ether. The position of the ring is established by GC analysis. Furanoid fatty acids that have been separated by TLC, can be detected by spraying the sample with a 2% solution of tetracyanoethylene in acetone. This will produce blue spots on a yellow background. Cyclopropane fatty acids can be mistaken for normal saturated fatty acids in silver ion chromatography. Cyclopropane fatty acids however can form methoxy-derivatives with boron trifluoride-methanol reagent. This product can then be converted to methyl-branched fatty acids by vigorous catalytic hydrogenation. The derivatives produced may be characterized by mass spectrometry. Methyl branches are comparatively inert to most chemical reagents. Thus, they are difficult to be located by chemical means. They positions can be determined chemically by oxidizing the lipids with acidic potassium permanganate. QUALITATIVE TESTING FOR FATTY ACIDS Solubility Test The degree of solubility of a lipid or fatty acid will depend on the relative amounts of the elemental composition, and structural positions of the sample. The lipids with high hydrocarbon content, such as fatty acids, are relatively non-polar in nature and insoluble in water. The ionic character of a lipid can be altered by changes in the pH. If the sample, such as a fatty acid, contains a strong acid, hydrolysis of some of the ester bonds will occur. The products of such hydrolysis may have solubility properties different from the original molecule Emulsification Test Emulsification is the reduction of the surface tension between two normally immiscible liquids or the prevention of two immiscible liquids to form two distinct layers. A lipid with a non-polar hydrocarbon portion and a polar portion are usually good emulsifying agents. The non-polar portion solubilizes with the less polar layer, while the polar portion will be attracted to the polar molecules in the solution. Since fatty acids have polar and non-polar portions, they should be good emulsifying agents.

29    

Halogenation Test for Unsaturated Lipids The presence of unsaturated fatty acids may be determined by halogenation because double bonds have the capability to undergo addition reactions. The halogen is able to add itself across the double-bonded carbon atoms of the lipid. This will result in the saturation of the carbon-carbon bonds. The amount of halogen, such bromine, added to a lipid molecule will give a quantitative measure of the degree of unsaturation in the lipid.

30    

ELUCIDATION OF STRUCTURE Polyketide synthases (PKSs) produce an array of natural products with diffferent biological

activities. Despite the diversity in terms of structure of polyketides, the PKSs share a common chemical strategy for the assembly of polyketides. Initially, it loads a starter molecule onto an active site cysteine, a decarboxylative condensation reaction extends the polyketide chain. Repeated elongation steps continues until a desired lenght is achieved, and the product is released. Elongation occurs either a modular or iterative approach. The ability of PKSs to use various starter groups, e.g. acetyl, propionyl, or other aliphatic and aromatic substrates, generates a broad range of molecular scaffolds. Polyketide formation is an obscure process until today. Regulation of the final length of the polyketide, production of novel polyketides, and varying specificities of PKSs have led to different methods to characterize and differentiate products and its respective PKSs. Plant specific PKSs, which are essential for the biosynthesis of anti-microbial phytoalexins, and anthocyanin floral pigments, have been investigated regarding structural control of its generated polyketides. 2-Pyrone synthase (2-PS) and chalcone synthase (CHS) are plant specific PKSs which has 74% amino acid sequence identity were both elucidated regarding its products to form a molecular basis of its molecule selectivity and control of polyketide length. In the experimentation, polyketide structure elucidation was done using LC/MS/MS method. Standard assay conditions were performed usgin p-coumaroyl-CoA or acetyl-CoA with malonyl-CoA as the extender molecule. Reactions were quenched with 5% acetic acid and extracted with ethyl acetate. Analysis of products were conducted using a single quadrupole mass spectrometer coupled to a Zorbax SB-C18 column. High performance liquid chromatography were done by 0 to 100% gradient system of methanol in water within 20 minutes. The flow rate was 0.25 ml min-1.

Figure 206. Naringenin produced by CHS

31    

Figure 21. CTAL generated by CTAS

Figure 229. CHS T197L reaction product

32    

Figure 230. Methylpyrone formed by 2-PS

Figure 3124. CHS reaction product

33    

CHEMICAL SYNTHESIS

FATTY ACIDS

Most of the common fatty acids can be isolated from natural sources and can be purchased commercially at purity of 70-95%. However there are certain times when chemical synthesis becomes necessary when the acids are not readily isolated from natural sources due to lack of rich sources (i.e. myristoleic acid); when they are not known to occur naturally such as the many stereoisomers of octadecenoic acid; and when they are required in an isotopycally labeled form (2H, 3H, 13C, 14C) for the study of reaction mechanism and biochemical processes.

Some acids can be made by chain extension of readily available starting acids which work well

with saturated FA and some unsaturated ones. For unsaturated acids, the two most common reaction sequences involve the use of acetylenic intermediates or the Wittig reaction. The former is based on the ability of the acetylene (ethyne) to be alkylated once or twice and the ability of the triple bond to be partially reduced stereospecific olefinic compounds. The reactivity of alkynes can be successfully extended to give polyenes. One such route to methylene-interrupted polyenes involves the reaction between propargyl bromides and ethynyl compounds as organometallic derivatives – mainly sodium, lithium and magnesium – in the presence of cuprous salt:

Figure 3225. Acetylation Method

Further unsaturation may already be present in one or both R groups. Poly-ynoic acids are crystalline solids which must be purified by repeated crystallization before partial reduction with Lindlar’s catalyst. Silver ion chromatography is used to remove over-reduced and/or trans-olefinic isomers to purify the polyenes. On the other hand, in the Wittig reaction, an alkyl halide and its phosphonium salt reacts with a base to produce ylid that is subsequently condensed with an aldehyde:

Figure 3326. Wittig Reaction

The product may be a mixture of cis- or trans-isomers which can be later optimized. Cis is favored at low temperature, high dilution and absence of Li+, while sodium bistrimethyl-silylamide is recommended as the base. Acids can be incorporated with isotopic hydrogen or carbon. Deuterium label is usually incorporated by exchange or reduction (hydrogenation). Hydrogen on carbon atom adjacent to an aldehyde or ketone function is replaced by deuterium from 2H2O with mildly acidic or basic catalyst which promotes enolization.

Figure 27. Incorporation of isotopic hydrogen by hydrogenation for aldehydes

34    

Carboxylic acids need more vigorous conditions.

Figure 285. Incorporation of isotopic hydrogen by hydrogenation for carboxylic acids

On a more complex note, one of the classic syntheses of fatty acids involves the synthesis of prostaglandins. The first main contributions in this field are those from Corey in 1869. Although many precursors have been used for the synthesis of PGs, the bicycloheptane precursur (Corey’s lactone) is one of the most successful building blocks which is also used for the industrial production of PG analogs. The commercially available lactone can be converted into the dialdehyde and in the classical approach side chains introduced by Wittig and Wittig-Horner reactions. His synthesis of PGF2α and PGE are shown in the following figure:

Figure 36. Corey synthesis of prostaglandins

35    

Figure 3729. (continued) Corey synthesis of prostaglandins

36    

Figure 3830. (continued) Corey synthesis of prostaglandins

37    

POLYKETIDES

The aims of organic synthesis was to develop new synthetic routes to polyketide aromatic systems (i.e. benzenes, naphthalenes, anthracenes) and to carry out mechanistic studies on the chemical factors that govern the inherent preference for one cyclisation mode over another and the role of enzymes in controlling ring formation in vivo. One notably successful approach developed by Tom Harris in the 1970s was the condensation of poly-anions of polyketones with acylating agents. A striking example is the condensation of acetate with the acetoacetate anion. The resulting polyketone had precisely the type of structure predicted by Collie.

Figure 3931. Harris poly-anion condensation

Several important conclusions emerged from the Harris investigations. Most importantly, it was clear that the folding of a polyketone chain formed in vivo must be under enzymatic control, so that unwanted cyclisation modes are suppressed. In particular, it is necessary to prevent premature cyclisation of the chain in the early stages of elongation. Several other groups used pyrone chemistry to generate transient intermediates with polyketone chains. The researchers were motivated by the idea that enzymatic control in vivo might be achieved by selection of the folding point of the chain so that the first carbocyclic ring is formed at the appropriate site. The preferred mode for subsequent cyclisations would then be progressive formation of extra six-membered rings until a linear polycyclic system is generated. A striking example of this principle in the biosynthesis of the fungal metabolites rubrofusarin and alternariol is the folding of a heptaketide chain to give these phenolic natural products. Control of cyclisation in both cases is achieved by formation of the first carbocyclic ring at the appropriate position in the chain.

A common technique in formulating the total synthesis of polyketides is through retrosynthesis,

that is, the product (the polyketide itself) is methodologically broken down in order to obtain an idea on what starting material and processes to use and follow. The following are two examples of polyketide synthesis.

Pteridic acids A and B are spiroacetal-containing polyketides, isolated by Igarashi et al. from a

fermentation broth of Streptomyces hygroscopicus TP-A0451 (obtained from the stems of the bracken Pteridium aquilinium). Preliminary biological testing indicated that the pteridic acids have potent plant growth promoter properties, inducing the formation of adventitious roots in kidney beans at exceptionally low concentrations.

In the synthetic plan shown below, the (Z)-enone (7) was thought to function as a suitable acyclic

precursor to the pteridic acids and might arise, in turn, from coupling of the C1-C11 aldehyde (8) with the C12-C16 vinyl bromide (9). These two subunits should be accessible with a high level of stereocontrol using aldol methodology developed in the laboratory.

38    

Figure 32. Retrosynthesis of pteridic acid

Below shows the formation of the C1-C11 aldehyde (8) that began with a boron-mediated anti aldol reaction to rapidly establish the required C6-C10 stereopentad. Formation of the (E)-enolate of the Roche ester derived ethyl ketone (S)-10,9 upon treatment with c-Hex2BCl/Et3N, was followed by addition of the a-chiral aldehyde (R)-(11). The expected anti,-anti aldol adduct (12) was generated in 87% yield with excellent selectivity (>99:1 dr) in this matched situation. The high level of enolate p-facial selectivity is governed by the formation of a stabilizing formyl hydrogen bond with the oxygen of the PMB ether in the bicyclic aldol transition state shown, acting in unison with the minimization of allylic strain between the a-stereocentre of the enolate and the methyl substituent. This stereoinduction from the boron enolate is further enhanced by the conformational preference of the aldehyde a-stereocentre, which in the case of (E)-boron enolate aldol reactions is matched/reinforced for the formal FelkineAnh adduct, which avoids syn-pentane interactions between the aldehyde substituents and the enolate methyl group. Installation of the remaining C7 stereocentre was achieved using an EvanseSaksena hydroxyl-directed reduction. Thus, treatment of (12) with Me4NBH(OAc)3 in MeCN/AcOH (3:1) provided the corresponding 1,3-anti diol 13 in 87% yield and 95:5 dr, which was then converted into the acetonide 14 by (MeO)2CMe2 in the presence of catalytic acid (PPTS). Hydrogenolysis of the PMB ether (H2, Pd(OH)2), followed by oxidation of the resulting alcohol (15) using DesseMartin periodinane (DMP), (15) led smoothly to the corresponding aldehyde (16). With the correctly configured aldehyde (16) in hand, elaboration of the diene side chain of the pteridic acids was undertaken via a Hornere-Wadsworthe-Emmons olefination. A high degree of selectivity for the desired (E,E)-diene 17 (>95:5) was achieved using triethyl- 4-phosphonocrotonate with LiHMDS as base. Finally, TBAF-mediated cleavage of the TBS ether, followed by DesseMartin oxidation of the resulting alcohol (18) to give the aldehyde 8 (79%, 3 steps), completed the efficient stereocontrolled construction of the C1-C11 subunit.

 Figure 41. Formation of C1-C11 Aldehyde

39    

Synthesis of the C12-C16 vinyl bromide coupling partner (9) commenced with preparation of the ketone (19) via a second antiselective boron aldol reaction, this time using our lactate-based methodology as shown below. Based on our standard procedure, the Weinreb amide (20) derived from ethyl (S)-lactate was treated with n-PrMgCl followed by benzoylation to give the propyl ketone (S)-21. Enolization of (21) with c-Hex2BCl/Me2NEt, followed by addition of acetaldehyde at -85°C, generated the desired anti adduct (22) (90:10 dr). Again the enolate π-facial selectivity is presumably governed by the formation of a stabilizing formyl hydrogen bond, now involving the carbonyl oxygen of the benzoate, in the bicyclic aldol transition state shown, along with minimization of allylic strain between the a-stereocentre of the enolate and the ethyl substituent. From here, TBS ether formation to give the b-siloxy ketone (19) was followed by a one-pot NaBH4 reduction/ester solvolysis, then cleavage of the resulting glycol with NaIO4, to generate aldehyde (23) in 92% overall yield. Treatment of (23) with CBr4 and PPh3 converted the aldehyde into the corresponding vinyl dibromide (24), (18) the less hindered bromide of which was selectively reduced (Pd(PPh3)4, Bu3SnH)19 to provide the necessary (Z)-alkenyl bromide (9) and complete the efficient preparation of the C12-C16 subunit.

 Figure 42. Formation of C12-C16 vinyl bromide

At this point, the development of suitable reaction conditions for the coupling of (8) and (9) was investigated as shown below. Gratifyingly, generation of the corresponding (Z)-alkenyl lithium species, via treatment of the bromide (9) with tert-butyl lithium in Et2O at -78°C, followed by addition of the aldehyde (8), generated two epimeric alcohol adducts, which were subjected directly to DesseMartin oxidation to give the (Z)-enone (7) in 53% overall yield. With the desired linear precursor (7) in hand, the desired end product required a suitable acid treatment to cleave the TBS ether and acetonide groups, thereby inducing spiroacetalization of the resulting triol with the C11 ketone. Concerns over the intervention of competing undesired reaction pathways proved to be justified, as exposure of (7) to a range of acidic conditions (e.g.,CSA, HCl, PPTS, TFA, H2SiF6, etc.) led to intractable complex mixtures of products. At this stage, all that remained was saponification of the esters to produce the corresponding acids, avoiding any potential equilibration at the C11-acetal centre. Separate treatment of the epimeric ethyl esters with KOH in EtOH generated pteridic acids A (25/5, 76%) and B (26/6, 73%).

40    

Figure 42. Coupling of aldehyde and vinyl bromide

Graphislactone A was identified as reduction product of the fungal metabolite botrallin in 1968, but it was first isolated as a natural product from the lichen Graphis scripta var. pulverulenta in the late nineties together with graphislactones B–D. Graphislactone A is an antioxidant and a scavenger of free radicals as well as a moderate inhibitor of AChE.

Retrosynthesis of the target compound suggested a Suzuki coupling as key step for the

construction of the central biaryl bond. Boronate (4) suitable for this transformation could be prepared according to a published procedure starting with commercially available acetal-protected phloroglucinic acid (2) as shown below or, with one more step necessary, from phloroglucinic acid (1). The other component required for a Suzuki coupling was prepared essentially according to a known protocol starting with commercially available orsellinic acid (6), which can be easily prepared in two steps from orcinol (5) with 65% yield.

Figure 43. Synthesis of boronate for Suzuki coupling

The methyl ester (7) was obtained with methyl iodide and potassium carbonate (91%). Chlorination with freshly distilled sulfuryl chloride afforded chlorinated methyl orsellinate (8) in 95% yield, while the dichlorinated substrate (9) was obtained in traces. Separation was simply achieved by recrystallization as shown below.

41    

Figure 44. Synthesis of aryl bromide for Suzuki coupling

Selective methylation of the 4-hydroxyl group with subsequent saponification (à10) and decarboxylation using 2,2-bipyridyl in the presence of CuO2 at elevated temperature (160°C) furnished the known compound (11) with 58% yield which is shown below. The aryl bromide (12) suitable for a Suzuki coupling was obtained by bromination with N-bromosuccinimide. The correct constitution of compound (12) was unambiguously proven by analysis of proton NOESY spectra. Suzuki coupling of aryl bromide (12) and boronate (4) with concomitant lactonization was achieved with a protocol, which previously proved itself suitable for the synthesis of similar compounds, thus yielding the natural product graphislactone G in 72%. The reaction occurred with complete chemoselectivity. No Suzuki coupling of the chloro function was observed.

Figure 45. Suzuki coupling yielding graphilactone G

BIOSYNTHESIS Excess glucose from energy production, and glycogen synthesis is converted to fatty acids via acetyl-CoA, and to reducing equivalents (NADPH) required for fatty acid synthesis. It involves sequential

42    

addition of two carbon units from acetyl-CoA to the activated carboxyl end of a growing chain by fatty acid synthase, which is a complex of several proteins. Initially, a single acetyl-CoA is activated to malonyl-CoA by a carboxylase reaction, which is catalyzed by acetyl-CoA carboxylase. This is the commitment step of fatty acid synthesis. The reaction proceeds by reacting acetyl-CoA with bicarbonate (HCO3-), as the source of CO2, and ATP, which is hydrolyze to drive the CO2 transfer. The commitment step is a biotin dependent carboxylase reaction, which consists of three polypeptide chains that constitutes the carboxylase enzyme: a biotin carboxylase (BC) activity that covalently attached a carboxyl group to the ureido ring of the biotin cofactor at the N-1 position, a carboxyltransferase activity that transfers the carboxyl group to the acceptor molecule, and a biotin carboxyl carrier protein that covalently carries the biotin group through an amide linkage between its valeric side chain and ε-nitrogen of a lysine residue. Biotin carboxylases vary by its carboxyltransferase component. It can differ greatly in sequence and structure, depending on the identity of the acceptor molecule for the carboxyl group. For example, there is no similarity between the carboxyltransferase components of acetyl-CoA carboxylase, and pyruvate carboxylase. But their BC activity is shared among these enzymes. The BC enzyme, which is Mg2+ ATP dependent, exists as a dimer consisting of two monomers, each with three domains i.e. A, B, and C, of almost similar conformation. It belongs to a larger family of proteins characterized by an ATP-grasp fold. Active site is located at the interface between the B domain and the two other domains (Figure 46).

Figure 46. Overall structure of E. coli BC dimer in complex with the substrates: biotin, bicarbonate and ADP (pink, green, black)

Crystallographic analyses show that two of the oxygen atoms (O1 and O2) of bicarbonate are recognized by bidentate ion pair interactions with the side chain guanidinium group of Arg292. While the third oxygen of bicarbonate (O3) is located near the side chain carboxylate group of Glu296, which states that this oxygen most likely carries the proton of bicarbonate and Glu296 acts as a base during catalysis to extract the proton from bicarbonate. The carbonyl oxygen in the ureido ring of biotin has hydrogen bonding interactions with the side chain of Arg338, which also exhibits interaction with bicarbonate. The N1 atom of biotin was observed to be located near the O1 atom of bicarbonate. This close approach probably helps to position the biotin group in the active site, and also has implications for the catalysis of BC. However, the valeric side chain is less ordered and has less electron density. One of its carboxylate

43    

oxygen atoms is hydrogen bonded to the main-chain amide of residue Asp382. The valeric side chain normally forms an amide bond with BCCP. Hence, substrate for BC activity is usually in the form of BCCP-biotin. During binding of ATP, the B domain undergoes a large movement to close over the active site upon ATP binding. Hence, the B domain is placed even closer to the active site. Experimental studies show that if free biotin is introduced rather than BCCP-biotin, the active site is still exposed to the solvent. It was observed that free biotin is a poorer substrate, and BCCP may help cover up this active site.Further closure of the B domain brings the phosphate group of the adenine nucleotide to the bound position of the bicarbonate and make the BC enzyme competent for catalysis.

Figure 47. Active site of BC

Binding of ATP generates a BC and ATP complex which has been previously studied by changing mutation studies to inhibit catalysis. i.e. glutamate was changed to lysine, or binding of ADP-Mg2+ rather than ATP. Consensus of these studies show that the magnesium ion is coordinated by a terminal oxygen from the alpha and gamma phosphates. A second magnesium ion coordinates the beta and gamma phosphates. After the gamma phosphate is transferred to bicarbonate, the two metal ions are likely to remain with ADP, one between the alpha and beta phosphate and one with the beta-phosphate. Moreover, the O1 atom of bicarbonate is located directly below of the gamma phosphorus atom. This suggests that this oxygen is in the appropriate location for initiating a nucleophilic attack on the phosphorus atom. It is also possible that an additional conformational rearrangement of the B domain can further reduce the distance between the ATP and bicarbonate and thereby facilitate catalysis. Previous studies state that a carboxyphosphate intermediate occur, though this has not been observed experimentally for the BC reaction. Structural studies show that O1 atom of the bicarbonate initiaites the

44    

nucelophilic attack on the gamma-phosphate of ATP which could lead to the carboxyphosphate intermediate. Deprotonation of bicarbonate occur with Glu296 as base. The pKa of bicarbonate is 10.3, which make deprotonation very unlikely. However, bicarbonate interacts with positively charged side chains of BC, i.e. arginine residues. This would reduce its apparent pKa. Hence, the combined effects of the local environment, i.e. closure and positive charges, make the pKa values of Glu296 and bicarbonate closer, which facilitates deprotonation of bicarbonate. Afterwards, the carboxyphosphate intermediate undergoes decomposition, producing carbon dioxide and phosphate ion. One of the oxygen atoms of the phosphate group is expected to be near the N1 atom of biotin. Phosphate group acts as a base which deprotonates the N1 atom leading to enolization of the ring. The resulting oxyanion is stabilized by the presence of arginine residues. Lastly the N1 atom attacks the carbon dioxide which leads to the formation of carboxybiotin.

Figure 48. Molecular mechanism for the catalysis by BC

Acceptor molecule, i.e. acetyl-CoA, is then carboxylated releasing biotin again for another carboxylatin process. The activated acetyl-CoA, which is malonyl-CoA, is attached to a 4’-phosphopantetheine moiety on acyl carrier protein (ACP) of fatty acid synthase. Another molecule, termed as primer that is either acetyl-CoA or butyryl coA, is attached to another ACP of the fatty acid

45    

synthase. Subsequently, it is transferred and covalently bounded to a cysteine group of the enzyme. Cysteine bound acetyl-CoA reacts with ACP bound malonyl-CoA by nucleophilic substitution, releasing CO2 in the process. Resulting product is a β-ketoacyl. The remaining steps involve reduction of β-ketoacyl. Reducing equivalents (NADPH) transform the carbonyl compound to an alcohol, which is dehydroxylated by releasing water, using a dehydratase, Reduction of the double bond occur by another NADPH and the final product is released, i.e. a saturated fatty acid. The process repeats by transfer of acetyl coA to ACP, then to cysteine group of enzyme, and nucleophilic substitution to the growing chain. Hence, fatty acid synthase sequentially adds two to four carbon units in the form of acetyl-CoA to the chain.

Figure 49. Reaction sequence for de novo BC of FA

Biosynthesis of polyketides is catalyzed by a collection of multienzyme proteins, which consists of coordinated group of active sites, called polyketide synthases (PKSs). These enzymes are evolutionarily related enzyme assemblies. Genes encoding these enzymes are organized in clusters to facilitate coordinated regulation of these enzymes. Associated with these genes are resistance genes and export

46    

genes to prevent the organism from self-intoxication of its products. Biosynthesis begins by activating two or more carbon compounds, such as acetyl-CoA, propionyl-CoA, and butyryl-CoA. Activated counterparts of preceding compounds are malonyl-CoA, methylmalonyl-CoA, and ethylmalonyl-CoA. This process resembles and closely related that of fatty acid synthases (FASs). Beta-keto intermediates are generated by decarboxylative Claisen thioester condensation of an activated unit, e.g. malonyl-CoA. The process requires Beta-ketoacyl synthase (KS), acyl-carrier protein (ACP), and an optional malonyl/acyl transferase (MAT/AT). In contrast to fatty acid synthesis, where ketoreductase (KR), dehydratase (DH), and enoyl reductase (ER) follows after the aforementioned processes, polyketide chain extension cycle is followed by full, partial, or no reduction steps, which results to a complex pattern of functionalization (Fig 49). Also, PKSs use several types of building blocks. Hence, it generates various products of different chain lengths. A final comparison between FASs and PKSs, both products are cleave off from the thioester-bound substrates of the enzyme complexes. PKSs are divided into three major categories, i.e. type I, type II, and type III, based on their mode of actions and architecture. Type I PKSs consist of a large multifunctional enzyme complexes where each enzyme has a catalytic domain and a carrier protein domain organized like beads on a string. These assembly lines are further divided to two different systems, e.g. modular and iterative. In modular system, each module consists of KS, AT, and ACP domains, and optional Beta-keto processing domains (KR, DH, ER). The number of modules reflects the number of cycles the process undergoes, and the presence of KR, DH, and ER domains determine the degree of Beta-keto reduction in each module. Hence, it is possible to detemine the structure or architecture of the product from the number of catalytic domains. A good example of a proto-typical PKS is erythromycin biosynthesis (Fig 51).

Figure 50. FA (A) and PK (B) biosynthesis (route X and Y represent BC of unreduced and

fully/partially reduced polyketides

47    

Figure 51. Erythromycin A biosynthesis, 6-DEB; 6-deoxyerythronolide B

The gene cluster for the PKS in erythromycin biosynthesis consists of three open reading frames, namely DEBSI (eryAI), DEBS2 (eryAII), and DEBS3 (eryAIII) with a total of six modules and a loading domain. The biosynthesis starts with the loading of a starter unit propionyl-CoA to the holo-ACP domain through long ‘swinging arims’. The KS domain of module 1 self acylates with the extender unit, which is methylmalonyl-CoA. Decarboxylative condensation between the ACP-tethered starter unit and the extender unit results in the formation of a diketide. Reduction of the keto group results in the presence of a hydroxyl group in the C3 position of the diketide. The AT domain of module 1 catalyzes transfer of the growing chain from the loading ACP to the ACP of module 1. A detailed scheme for the process of how type I PKSs catalyse one round of polyketide extension is shown (Fig 52). The process then cycles until the desired length of the chain is achieved. The final product is cleave off from the ACP domain by thioesterase domain of eryAIII, which catalyzes intramolecular lactone formation. Hydroxyl group of the C3 domain serves as the nucleophile to the carbonyl carbon of the thioester resulting in the formation of erythronolide B. Other type I PKSs, where a module is used more than once, skips domains or lacks AT domains, where ATs load starter or extender units to ACPs in trans fashion . An example of an AT deficient iterative type I PKSs, which are usually found in fungi, is lovastatin biosynthesis (Fig. 53). In this PKS, KR, DH, and ER domains are optionally used for every elongation cycle. Regulation of when and how the domains were used is still not known

48    

Figure 52. Sequential reactions showing how the modules move for a typical type I PKS

Figure 53. Lovastatin biosynthesis. The intermediate undergoes intramolecular Diels-Alder reaction to establish a decalin moiety of lovastatin.

49    

Table 1. Characteristics of domains

A more common type of PKS that is found in most prokaryotes are type II PKSs. A minimal set of enzymes are used iteratively to generate poly beta-keto thioester. Ketoacyl synthases,which includes a KS domain (KSalpha) and chain length factor (KSbeta), ACP, and malonyl-CoA:ACP transacetylase (MAT), which is borrowed from the endogenous fattya acid synthase, constitute the minimal set of a type II PKS (Fig 55). Other domains, i.e. ketoreductases (KRs), cyclases (CYCs), and aromatases (AROs), work cooperatively with the minimal PKS to control folding, cyclization, and aromatization pattern of the nascent polyketide chain. Type II PKS system is responsible for the biosynthesis of some known antibiotics, i.e. actinorhodin, which was first reported from Streptomyces coelicolor, doxorubicin (Fig 57), gilvocarin V, landomycin A, and tetracycline (Fig 56). Acetyl-CoA or propionyl-CoA serves as starter units, and malonyl-CoA for chain elongation. To exemplify further, actinorhodin is synthesized with actinorhodin PKS and malonyl-CoA as the sole substrate. The parent polyketides, which is a two 16 carbon product called SEK4 and SEK4b, are primed by decarboxylation of malonyl-S-ACP, followed by an interthiol transfer of the resulting acetyl group onto the active site of the KS. This is followed by seven rounds of chain elongation. Each round consists of decarboxylative C-C bond formation between the nucleophilic malonyl-S-ACP and the growing polyketide chain bound to the KS via an electrophilic thioester linkage. The full length poly-Beta-ketoacyl chain is released from the ACP yielding the 16 carbon products (Fig 54). X-ray crystallography of actinorhodin KSalpha and KSbeta (CLF) proteins form a heterodimer with an elaborate pocket at their interface into which the growing polyketide is extruded. Changing selected KSbeta residues allow longer products to be synthesize. In contrast to tight association of KS and CLF, KS-ACP interactions are weak but specific. This prevents interchange of substrates between ACPs present in the same cell. On the other hand, consequence of weak KS-ACP interactions is that multiple KS-ACP exchanges occur during chain elongation. Type II PKSs are usually found in actinomycetes and most gram negative bacteria. Lastly, iterative type III PKSs do not require ACP-bound extender units for chain elongation. Rather, it utilizes acyl-CoA substrates directly. Biosynthesis of naringenin chalcone (Fig 13), a plant product, occurs by sequential addition of three molecules of malonyl-CoA to a p-coumaryl-CoA starter unit. This is a typical type III PKS catalysis, commonly referred as stilbene/chalcone synthases found in plants, bacteria and some fungi. In addition to type I, II, and III PKSs, there are mixed type polyketide synthases, i.e. type I-type II, type I-type II, type I-type III, FAS-PKS hybrid or PKS-NRPS (non-ribosomal peptide synthases).

50    

Figure 54. Biosynthesis of SEK4 and SEK4b by minimal PKS from actinorhodin biosynthesis pathway

Figure 55. Polyketide chain formation/elongation by type II PKSs

51    

Figure 56. Some examples of type II PKS derived antibiotics

Figure 57. Type II PKS-catalyzed biosynthesis of doxorubicin

Figure 58. Type II PKS-mediated biosynthesis of naringenin chalcone

Products from type I, II, and III PKSs undergo further reactions referred as post-PKS tailoring reactions. Such reactions are glycosylation, methylation, oxidation, reduction, prenylation, and

52    

halogenaitons to generate biologically active compounds. These tailoring reactions cause drastic changes to the pharmacological property of the parent polyketides, such as solubility or binding ability of the drug to a receptor. A typical example would be eythromycin A, where the PKS product (6-DEB) has no activity. Oxygenation and glycosylation reactions change it into a powerful antibacterial agent. Post-PKS tailoring reactions provide a unique prospect for designing natural product derivatives, i.e. combinatorial biosynthesis. One of the most important post-PKS tailoring reaction is glycosylation, which modulates biological activity of the molecule. Some examples are activation of 6-DEB by glycosylation, and inactivation of oleandomycin or erythromycin by addition of glucose to their desosamine moieties. Glycosylation usually occurs in the later stages of a biosynthetic pathway and the reaction involves the glycosyltransferases (GTs) enzyme. GTs use NDP-activated sugars as donor substrates and an acceptor substrate. Transfer yields O-, N-, S-, or C- glycosidic bonds. Glycosylation chemistry involves a GT-mediated nucleophilic attack of the acceptor substrate at the anomeric carbon of the donor substrate, with the simultaneous release of the nucleotide diphosphate, either in an SN2-fashion or via an oxocarbenium ion intermediate (Fig 59). C-glycoside formation from microbial secondary metabolites attached the sugar moiety ortho or para to an electron rich functional groups, such as the hydroxyl group (-OH), of the aromatic moiety. Attack at the anomeric center of the sugar was mad possible because of the high electron density at the ortho or para by such functional groups (Fig 60). Exact mechanisms for C- or N- glycosylations are still obscure. GTs can be categorized further to inverting GTs and retaining GTs based on the stereochemical change at the anomeric carbon of the sugar .

Figure 58. Proposed mechanism for inverting type O-GTs

53    

Figure 59. A hypothetical mechanism for C-glycosylation

Most sugar moieties of microbial natural products are often deoxygenated to various extents prior to attachment to aglycones. Deoxygenation of hexose sugars begins by conversion of hexose-1-phosphate into its respective nucleotide diphosphate (NDP) derivative by NDP-hexose-nucleotidyl transferases. Sources of NDP are usually NTPs, e.g. ATP, CTP, GTP, UTP, or TTP. Deoxysugars of nystatin and amphotericin B are derived from GDP mannose. Deoxygenation of C6 carbon and oxidation of C4 carbon of the activated sugar is catalyzed by NDP-hexose-4.6-dehydratase to generate NDP-4-keto-6-deoxy-D-hexose. The product is a common intermediate of most 6-deoxysugar pathways (Fig 60)

Figure 60. Biosynthesis of 6-deoxysugars through the common intermediate

54    

Moreover, products of PKSs also undergo a series of oxidations such as hydroxylations, epoxidation, anthrone oxidations, desaturations, Baeyer-Villiger monooxygenation, which are usually needed for biological activities. These reactions are catalyzed by a large family of enzymes called oxygenases. This group of enzymes are further classified as monooxygenases, which extracts one O-atom from an O2-molecule, and dioxygenases, which utilizes both O-atoms of an O2-molecule. Most monooxygenases utilize NADPH as reducing equivalents and catalyze hydroxylations, epoxidations, or Baeyer-Villiger oxidations. Dioxygenases, on the other hand, result in the production of peroxides or dioxetanes, which further rearrange or react to generate diols (Fig 61).

Figure 61. Structural diversity introduced by oxygenase

Cytochrome P-450 (CYP450s) monooxygenases are family of hemoproteins, which catalyzes stereospecific hydroxylation and epoxidation observed in post-PKS modifications. It can be either one-component, i.e. reductase activity is associated with the enzyme, or two -component, i.e. it requires another enzyme called ferredoxin reductase. PikC-mediated hydroxylation of narbomycin in the pikromycin biosynthetic pathway is a good example of a CYP450 stereospecific hydroxylation (Fig 62). Several analogues of pikromycin and erythromycin can be generated by the broad substrate specificity of CYP450s.

Figure 62. Hydroxylation of narbomycin by PikC

55    

Baeyer-Villiger monooxygenases (BVMOs) are also involved in several secondary metabolite biosynthetic pathways, e.g. MtmOIV from mithramycin, CmmOIV from chromomycin, JadH from jadomycin, and GilOI from gilvocarin. BVMOs are flavin-dependent and catalyze the Baeyer-VIlliger oxidation reaction, i.e. insertion of oxygen in cyclic ketone to form a lactone. The reaction requires NADPH as a reducing molecular oxygen. MtmOIV catalyzes conversion of premithramycin to premithramycin B-lactone, which eventually opens up to yield mithramycin DK, which is the penultimate intermediate of the mithramycin biosynthetic pathway (Fig 63). The mechanism occurs by transfer of two electrons, and two protons from NADPH to FAD generating FADH transiently. The reduced flavin immediately receives an oxygen molecule to generate a reactive peroxyflavin ion. Nucleophilic attack of ion on the ketone carbon followed by the migration of the C-C bond generates a lactone (Fig 64).

Figure 63. MtmOIV-catalyzed Baeyer-Villiger oxidation of premithramycin B

56    

Figure 64. Mechanism of BVMO catalysis

Lastly, methylation reactions are also common among post-PKS tailoring reactions. A variety of bioactive microbial secondary metabolites are N-, C-, or O-methylated that occur by activity of methyltransferases (MTs) in a regioselective manner. These enzymes require S-adenosyl methionine (SAM) as the methyl donor source. MTs are widely used for combinatorial biosynthetic approaches.

57    

RECENT STUDIES The study, "The First Total Synthesis of (±)-4-methoxydecanoic acid: A Novel Antifungal Fatty Acid," by N. Caballeria, C. Miranda, and K. Parang demonstrated that mid-chain methoxylated fatty acids are valuable compounds that can be used to develop more potent antifungal agents. The synthesis of (±) -4-methoxydecanoic acid (1) was done in six steps with a 25% overall yield from the commercially available 4-penten-1-ol. The synthesized compound demonstrated 17-fold higher antifungal activity against Candida albicans ATCC 60193 and Cryptococcus neoformas ATCC 66031 when compared to an unsubstituted n-decanoic acid. The isolation of (±)-4-hydroxydecanoic acid is particularly difficult because it is able to cyclize easily to y-decalactone. (±)-4-hydroxydecanoic acid is known to have antibacterial activity, however, the presence of a hydroxyl group in the acyl chain greatly decrease toxicity towards fungi. The y-decalactone has been shown to have higher antibacterial activity than (±)-4-hydroxydecanoic acid since it inhibits the growth of bacteria such as Bacillus subtilis and some fungi such as Fusarium oxysporum and Trichothecium roseum at concentrations between 0.1 and 0.7 mM. Isolated from Lactobacillus planatarum, 3-(R)-hydroxydecanoic acid has antifungal activity against different molds and yeasts. Studies have shown that α-methoxylation increases the antifungal activity of fatty acids, but in terms of the effect of mid-chain methoxylation, no reports have been published on the antifungal activity of such compounds. In this study, it is hypothesized that (±)-4-methoxydecanoic acid could be a good starting template to such an effect. The methoxylated fatty acid should be more fungitoxic than the compound previously described. In addition, the choice of a 10-carbon chain length is investigated because of the fact that capric acid kills Candida albicans at 10 mM by the postulated mechanism of the fungal plasma membrane disintegration. Another advantage of using (±)-4-methoxydecanoic acid is that it will not cyclize to a y-lactone. Moreover, this study compares the efficiency of mid-chain methoxylation to alpha-methoxylation in terms of the antifungal activity of fatty acids. Previous antifungal studies against C. albicans showed that 2-OMe-10:0 acid is more fungitoxic than capric acid, but for Cryptococcus neoformans, capric reigned as the better antifungal agent. The synthesis involves six steps. First, the primary alcohol, 4-penten-1-ol (2), is protected by dihydropyran (DHP) and catalytic amounts of p-toluenesulfonic acid (PTSA) in CHCl3 at room temperature for 5 hours to afford the 1-[(tetrahydropyran-2-yl)oxy]-2-pentene (3) in an 88% yield. The double bond was epoxidized in the presence of magnesium monoperoxyphthalate (MMPP) in EtOH as solvent for 48 hours. An 89% yield of the products, 4,5-epoxy-1-[(tetrahydropyran-2-yl)oxy]pentane, was observed after purification of the crude product by using silica gel column chromatography and elution with hexane or diethyl ether. MMPP was observed to be more efficient than the m-chloroperoxybenzoic acid (m-CPBA) in epoxidizing these alkenes. The tetrahydropyran (THP) protected epoxide 3 was then opened with 1-pentylmagnesium bromide assisted by catalytic amounts of copper (I) chloride in THF at the temperature range of (-78C to -30C), which afforded the desired 4-hydroxy-1[(tetrahydropyran-2-yl)oxy]decane (4) in an 81% yield after silica gel column chromatographic purification. The free hydroxyl group in 4 was readily methylated with methyl iodide in the presence of sodium hydride in THP. This afforded the 4-methoxy-1[(tetrahydropyran-2-yl)oxy]decane (5) in an 88% yield. Deprotection of the primary alcohol was achieved by treating the alcohol with PTSA in CHCl3 at 45C for two hours. This afforded the (±)-4-methoxydecan-1-ol . Final oxidation of the acid was done by reacting the alcohol with pyridinium dichromate in DMF for 24 hours. This resulted in a 63% yield of 1. The overall yield was observed to be 25%.

58    

Scheme 7 The synthesis of (±)-4-methoxydecanocic acid. Reagents and conditions: (i) DHP/PTSA, CHCl3, room temperature, 5 hours, 88%; (ii) MMPP/EtOH, 48 hours, 89%; (iii) CH3(CH2)3CH2MgBr, Cu(I)/THF, -78C to -30C, 81%; (iv) NaH/CH3I, THF. OC to room temperature, 2 hours, 88%; (v) PTSA,

CHCl3, 45C, 2 hourse, 73%; (vi) PDC/DMF, 24 hours, 63% The NMR spectrum of 1 showed significant absorption for carbons and hydrogens bearing the methoxy functional group. The methoxy protons resonated at δ 3.32 ppm, the methoxy carbon at δ 56.5 ppm, while the methine (CHOCH3) hydrogen and methine carbon resonated at δ 3.20 ppm and δ 79.9 ppm, respectively. The resulting NMR absorbances seem to be characteristic for saturated mid-chain methoxylated fatty acids. In the 70 eV electron impact (EI) mass spectrum of 1, the typical McLafferty rearrangement of fatty acids at m/z=60 was greatly reduced by the presence of the methoxy functional group. In the mass spectrum of 1 the a-fragmentation at both sides of the methoxylated carbon predominated, but the fragments containing the carboxyl end (at m/z = 117 corresponding to C5H9O3+ and at m/z = 85 corresponding to C4H5O2+) were the most abundant. Table 2. Antifungal activity (MIC values, uM) against Candida albicans (SDB) and Crytococcus neoformans (SDB) at 35-37C after 24-48 hours

The antifungal activity of 1 was determined against a fluconazole-resistant strain of C. albicans (ATCC 60193) ad C. neoformans (ATCC 66031). n-Decanoic acid was used as a control. It was observed that (±)-4-methoxydecanoic acid (1) gave a minimum inhibitory concentration of 1457 um, while that of the n-decanoic acid gave an MIC of 25,478. This implied that the synthesized fatty acids was approximately 17-fold more antifungal against both fungal strains as compare to n-decanoic acid. Thus, C-4 methoxylation increased the antifungal activity of the parent n-decanoic acid. The factors that affect the

59    

improved antifungal activity are suspected to include its solubility, efficiency of disrupting fungal membranes, and inhibition activity. The addition of the C-4 methoxy functional group makes it possible to be soluble than n-decanoic acids. This facilitated its interaction with the target sites. With this the efficiency of disrupting fungal membranes increases. The compound 1 may also inhibit fatty acid biosynthesis within the fungi and interact with some key enzymes.

This study is significant to the scientific world because it is an advancement or a breakthrough in fungal infection treatment. Despite the arrival of new effective drugs, therapeutic problems due to new pathogenic fungal species, slow microbiological diagnosis, variable drug availability, toxicity, lack of oral or intravenous preparations, and significant drug interactions for some agents, and development of resistance or breakthrough infections. The results of this study could use fatty acid amide analogues to find new ways to cure fungal infections. The study, "A Flexible Synthesis of C33-C39 Polyketide Region of Apratoxin: Synthesis of Natural and Unnatural Analogues," describes the syntheses of the polyketide moiety of apratoxins A and C and two other polyketide analogues and demonstrates the versatility of such synthesis in introducing different alkyl groups at C39 position. This makes it easier to synthesize a number new polyketide analogues, or more specifically - new apratoxin analogues. Cytotoxic apratoxin A and analogues were recently discovered class of cyclodepsipeptides of marine origin. They are known for their antitumoral activity and uncommon mechanism of action.

Figure 64. Apratoxin and analogues

Figure 65. The four synthesized analogues

60    

An efficient synthetic pathway was developed to provide the polyketide moiety of apratoxin A (1) and C (2) analogues. The parent molecules used were pivalaldehyde and butyraldehyde for apratoxin A and apratoxin C, respectively. The parent molecules underwent List's aldolisation reaction with proline as an organocatalyst to produce their corresponding hydroxyketones.

Scheme 8 Synthesis of apratoxin A and C

It was observed that butyraldehyde gave a lower yield. The diasteroselective reduction of the hydroxyketones 6 and 6’ under Prasad reaction conditions furnished the diols 7 and 7', with a better diastereoselectivity for the pivalaldehyde due to steric hindrance. The diastereomer products were then separated by means of silica gel chromatography. The cyclic sulfate was obtained by cyclic sulfite formation (8 and 8') with thionyl chloride. This was followed by oxidation with RuCL4 and NaIO4 as cooxidant to produce 9 and 9'. The sulfate was subjected to a regioselective nucleophilic substitution with allylmagnesium bromide. When R=iPr for 9', the regioselectivity of the reaction was not complete due to less steric hindrance with the yield of the formation of its regioisomer only being 30%. The substitution reaction induced complete inversion of configuration by SN2 for both cases. Final hydrolysis of the newly formed sulfate furnished the free alcohol. The parent molecule for the molecule 3 is a racemic epoxide in the molecule 11. Hydrolytic kinetic resolution furnished the pure (R)-epoxide 13, which was subjected to regioselective nucleophilic substitution with homoallyl magnesium bromide to yield 14.

61    

Scheme 9

Hex-5-en-1-ol was the parent molecule used in the synthesis for the less hindered analogue 4 without alkyl group. From this, the rest of the synthesis was identical for all the four analogues and was performed in multiple steps. First, the free alcohol was silylated with TESCI (16) and the double bond was subjected to oxidative cleavage with ozone to give the corresponding aldehyde (17). Then, a non-Evans anti-aldolisation type reaction was employed to obtain the last two stereocenters. However, various attempts to furnish the aldol adduct were unsuccessful. The synthesis then proceeded with a Brown's cotylation reaction. This was a more efficient option in yielding the corresponding homoallylic alcohol. (±)-α-methoxydiisopinocampheyl borane was used as chiral auxiliary to obtain the desired enantioselectivity at the C35 chiral center, and the trans-butene was used as adduct to give the anticompound with a good diastereoselectivity. The syn diastereomers were identified by silica gel chromatography. The compounds 17c and 17d did not possess a methyl group in the beta position from the aldehyde have less diastereoselectivity due to a loss of steric hindrance. The latter was then protected as TBDMS silyl ether, followed by ozonolysis to give an aldehyde. This was then oxidized with NaClO2 into its corresponding polyketide moieties.

62    

Scheme 10

This study is significant because it is a stepping-stone for future structure-activity relationship (SAR) studies of the potent antitumoral compound, apratoxin, by means of proposing a strategy to synthesize the oxazoline analogues of the said compounds.