Information to Users

INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. U M I films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send U M I a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact U M I directly to order.

University Microfilms international A Bell & Howell Information Company 300 North Zeeb Road. Ann Arbor. Ml 48106-1346 USA 313-761-4700 000.521-0600 Order Number 9412014

Arachidonic acida ct-reductone strategies: Asymmetric syntheses of 2-hydroxy tetronic acid antimetabolites

Mantri, Padmaja, Ph.D.

The Ohio State University, 1993

UMI 300 N. Zeeb Rd. Ann Arbor. M l 48106 ARACHIDONIC ACID ac/-REDUCTONE STRATEGIES:

ASYMMETRIC SYNTHESES OF 2-HYDROXYTETRONIC ACID

ANTIMETABOLITES

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate

School of the Ohio State University

Padmaja Mantri, B.S.

The Ohio State University

1993 Dissertation Committee: Approved by

Dr. Donald T. Witiak

Dr. Robert W. Brueggemeier

Dr. Dennis R. Feller Advisor

Dr. Michael H. Zehfus College of Pharmacy To

Ayn Rand

ii ACKNOWLEDGEMENTS

To the following, I express my gratitude and appreciation for their help and support.

Professor and Dean Donald T. Witiak, my advisor for his guidance, enthusiasm, advice, persistence, patience, and never ending support.

Neelam Gupta for the initial financial support that made graduate studies possible.

Prabhakar and Shabana for intellectually stimulating discussions about chemistry and philosophy and for their much valued friendship.

Dr. Dennis R. Feller and Dr. Karl J. Romstedt for providing the associated pharmacological support.

Dr. Cottrell, Dr. David Chang and Dr. Kurt Loening for their help in procurring NMR, mass spectral data and chemical nomenclature, respectively.

The faculty, staff, my dissertation committee and colleagues, Pat and Al for help and suggestions.

My parents, Madhukar Mantri and Vijaya Mantri for their love, support and patience.

Pineabrim McCullough Jr., whose love, understanding and encouragement made this time period enjoyable.

The Ohio State University, University of Wisconsin-Madison and American Cancer Society (Ohio Division) for financial support. VITA

August 20, 1967 ...... Bom- Bombay, India

1989...... B.S. Pharmacy, C.U.Shah College of Pharmacy, S.N.D.T. University, Bombay, India

1989-199 0...... Graduate Fellow, The Ohio State University, Columbus, Ohio

1990-199 2...... Graduate Teaching Assistant, The Ohio State University, Columbus, Ohio

1992-1993...... Graduate Research Associate, The Ohio State University, Columbus, Ohio

1989-1993...... Academic Challenge Fellow, The Ohio State University, Columbus, Ohio

PRESENTATIONS

Mantri, P.M. and Witiak, D.T. Approaches to the asymmetric syntheses of polyalkenyl-2-hydroxytetronic acids: Potential promoters of IL-2-induced lymphokine activated killer activity. American Cancer Society International symposium, September 9-12,1992, Columbus, OH.

Mantri, P.M. and Witiak, D.T. ac/-Reductone chemistry and biology: Asymmetric syntheses for 4-substituted-2-hydroxytetronic acids. 205th ACS National Meeting, March 28-April 2,1993, Denver, CO.

iv Witiak, D.T., Triozzi, P.L., Feller, D.R., Romstedt, K.J., Tehim, A.K., Hopper, A.T. and Mantri, P.M. 4-Substituted-2-hydroxytetronic acid aci- reduct ones improve lymphokine-activated killer cell activity in vitro. 84th Annual Meeting of the American Association for Cancer Research, May 19-22,1993, Orlando, FL.

FIELD OF STUDY

Major Field: Pharmacy

Studies in : Medicinal Chemistry

v TABLE OF CONTENTS

DEDICATION...... ii

ACKNOWLEDGEMENTS...... iii VITA...... iv

LIST OF TABLES...... viii

LIST OF FIGURES...... ix

CHAPTER Page

I INTRODUCTION AND HISTORICAL BACKGROUND 1

Prostaglandin H synthase ...... 1 Inflammation ...... 8 Cyclooxygenase inhibitors ...... 9 Stereochemical preferences of cyclooxygenase ...... 20 Metabolism of asymmetric propionic acids ...... 23 5-Lipoxygenase...... 27 Substrate and product analogues ...... 33 FLAP inhibitors...... 37 Iron chelators ...... 45 Antioxidants, which scavenge free radicals ...... 60 Stereochemical preferences of 5-lipoxygenase ...... 62 II aci-REDUCTONES...... 64

Introduction ...... 64 Drug Design...... 66 Biological properties...... 67 Synthetic chemistry ...... 69

vi III STATEMENT OF THE PROBLEM...... 75 IV RESULTS AND DISCUSSION...... 81

Method A Unsuccessful Organocuprous condensation approach ...... 83 Method B Successful Wittig reaction approach ...... 94

V EXPERIMENTAL SECTION...... 116

BIBLIOGRAPHY...... 158

vii LIST OF TABLES

TABLE PAGE

1. Classification of nonsteroidal antiinflammatory drugs ...... 9

2. Inhibitory constants of NSAIDs derived hydroxamates ...... 47

3. Comparison of in vitro and in vivo inhibitory constants of type A and type B hydroxamate ...... 53

4. Comparison of inhibitory constants and pharmacokinetic parameters of type A, type B hydroxamates and N-hydroxyureas ...... 59 LIST OF FIGURES

FIGURE PAGE

1. Interaction of 5-lipoxygenase and 5-lipoxygenase activating protein ...... 30

2. 5-Lipoxygenase inactivation by antioxidants ...... 61 CHAPTER I

INTRODUCTION AND HISTORICAL BACKGROUND

All mammalian cells, except red blood cells, produce potent inflammatory mediators; among these are the prostaglandins (PG), prostacyclins, thromboxanes (TX) and leukotrienes (LT). Physiologically, these autacoids produce inflammation, pain and fever, regulate blood pressure and induce blood clotting. However, abnormal eicosanoid concentrations are observed in pathological conditions such as rheumatoid arthritis, psoriasis, asthma, anaphylactic shock and ulcerative colitis. This has provided the impetus for developing inhibitors of autacoid biosynthetic enzymes, PGH synthase and 5-lipoxygenase (5-LO).

Prostaglandin H Synthase: Prostaglandin H synthase (PGH synthase) catalyzes the committed step in the biochemical synthesis of autocoids, PGs, prostacyclins and TXs. PG biosynthesis utilizes the substrate arachidonic acid (AA, 1), present as a triglyceride ester in the cell membranes. The rate-limiting step in AA metabolism involves hydrolysis of triglycerides or phospholipids by phospholipases to generate free eicosanoid AA. PGH synthase is ubiquitous in all cell types; however, the generation of various PGs, prostacyclins and TXs is tissue- specific. This specificity resides in the selective presence of enzymes that convert PGH 2 to different autacoids. For example, only TXs are 1 2 generated in blood platelets, whereas only prostacylins are found in the smooth muscle cells.1

PHOSPHOLIPIDS

^ Phospholipases

COOH t. AA

Cyclooxygenase

2. PGG2 COOH

Peroxidase COOH COOH

OH Prostacyclin P HQ 8. PGI synthase b

HO OH OH 4. PGD. PGE r Thromboxane Synthase synthase COOH COOH 7 ' TXA

HO OH PGFa OH ySynthase HO COOH

Qh 6. PGH 2a Metabolic pathway for AA using PGH synthase 3

PGH synthase is bound to the endoplasmic reticulum and has two catalytic activities associated with this single protein.1 (1) The bis- oxygenase or cyclooxygenase (CO) activity oxygenates AA to endo- hydroperoxide PGG 2 (2), and (2) the peroxidase (PO) activity reduces hydroperoxide PGG 2 to the corresponding secondary alcohol PGH 2 (3).

These two subunits are connected by a prosthetic heme. Site directed mutagenesis (SDM) studies reveal that Tyr 385 and His 309 are important amino acids for enzyme activity. Mutation of His 309 results in total loss of

CO and PO activities. Tyr 385 acts as a conduit for electron transport between the heme and the CO active site. Tyr 385 forms a radical which abstracts the p ro-S hydrogen at C-13 of AA, the first step in the metabolism of this substrate. His 309 is one of the ligands for the prosthetic heme.1

There are two isozyme forms for PGH synthase: PGH synthase-1 which is the constitutively expressed form of the enzyme, and PGH synthase-2 which is the isozyme induced by phorbol esters, liposaccharides and serum proteins.2 Approximately, 62% homology exists between the amino acid residues of murine PGH synthase-1 and murine PGH synthase-2;3*5 the amino acids necessary for catalytic activity are highly conserved.3*11 The two isozymes differ in their hydrophobic signal peptide sequence. In PGH synthase-2 the signal peptide is a short cleaved sequence compared to one in PGH synthase-1.4*7 Furthermore, there are differences in N-glycosylation between the two isozymes.3>4'6*

912 To date, neither tissue specificity nor differential binding of substrate or inhibitors to these two isozyme forms have been studied. 4

The molecular weight of PGH-synthase-1, based on the amino acid sequence obtained from cloned sheep, mouse and human cDNAs, is

65,500 daltons.6*8 However, the enzyme is isolated as a homodimer with Mr of 72,000 by SDS-PAGE.13*15 It is not known whether ovine PGH synthase functions as a monomer or a dimer in membranes. The difference between the isolated DNA and the cDNA is N-glycosylation at three Asn residues; Asn 68, Asn 144 (located in the NH 2 -te rm in a l peptide), and Asn 410 (located in the COOH-terminal peptide). SDM at each of the four possible Asn residues (68, 104, 144 and 410) and at Ser or Thr flanking highly conserved (Asn-X-Ser/Thr) N-glycosylation sequences indicates glycosylation at Asn 68, 144 and 410.16-17 Mutant PGH synthases lacking N-glycosylation at either Asn 68 or Asn 144 exhibit decreased (CO/ PO; 18%/11.2% and 6.8%/3.5% of control, respectively) activity following expression in COS-1 cells. However, mutants lacking both Asn 68 and Asn 144 and/or Asn 410 N-glycosylation have no CO or PO activity. Partial deglycosylation with endoglycosidase

H, subsequent to formation of conformationally active enzyme, does not affect enzyme activity. This suggests a role for N-glycosylation in procurring and maintaining the active conformation of PGH synthase.16

Murine PGH synthase-2 exhibits an Mr of 72 and 74 kDa for the native forms, and an Mr of 65 kDa the unglycosylated form.18 Using an experimental strategy similar to the one described for PGH synthase-1 the extent and location of N-glycosylation has been determined for PGH synthase-2. Both forms of PGH synthase-2 are N-glycosylated at Asn 53,

Asn 130 and Asn 396. The two native forms of murine PGH synthase-2 5 differ wherein, the 72 kDa form facks N-glycosylation at Asn 580.

However, such differential N-glycosylation does not result in differential activity. PGH synthase-2 mutants also lack enzyme activity when expressed unglycosylated in tunicamycin-treated COS-1 cells. This implicates a similar role (as seen for PGH synthase-1) for N-glycosylation in achieving the proper conformation.

A proposed model for the orientation of the enzymes involves two transmembrane, hydrophobic regions. This allows the enzymes to cross the membrane twice and position both the NH 2 and the COOH terminii on the lumen side. The catalytic site is in the cytoplasm.16 Hydrophobicity plots predict only one membrane-spanning domain,7 and this model does not explain the origin of the second membrane spanning domain.

Very little is known about inhibitors of PGH synthase-2. This isozyme has only recently been discovered, and its availability is limited since this protein is inducible rather than constitutive. The inhibitory work described is primarily on PGH synthase-1. As pointed out, PGH synthase-

1 possesses spatially distinct interactive sites for CO and PO. The cyclooxygenase subunit binds the substrate AA as well as both competitive and non-competitive, non-steroidal anti-inflammatory drugs.5-9-15 The PO binding site accomodates secondary alkyl hydroperoxides. The prosthetic heme is flanked by the highly conserved

His 309 on the proximal side and His 388 on the distal side.20 The heme has an interactive location and is shared by the CO and PO subunits.1*21 6

The highly conserved Tyr 385 serves as the channel for electron transfer

between the peroxidase and the CO interactive sites.22

The sequence for the catalytic conversion of the substrate A A to

PGH 2 likely occurs via the following steps: A hydroperoxide activator

binds the PO site of the enzyme and oxidizes the heme group to a species

possessing a higher oxidation state.23"25 This species, in turn,

homolytically abstracts a hydrogen radical thereby generating the Tyr 385

radical.27 Once formed, the tyrosyl radical initiates the CO reaction. Based on the model proposed by Hamberg and Samuelsson, AA is held by the enzyme PGH synthase-1 in a conformation resulting from rotation

about the C-9/C-10 single bond.28 The resultant orientation produces a

kink as shown in AA (1 ). The Tyr 385 radical abstracts the AA 13-proS

hydrogen, which following double bond migration results in the

generation of C-11 radical 9. This radical is trapped by molecular oxygen from the side opposite to the initial radical generated by the Tyr 385 species to produce intermediate 1 0 . The resulting peroxide radical

undergoes intramolecular attack at C-9 of 10 providing endoperoxide 11 which possesses a radical at C-8. Sequential intramolecular attack at C-

12 results in the generation of a trans-bicyclie endoperoxide radical 12.

This intermediate is quenched by a second oxygen molecule affording

15-hydroperoxide PGG 2 (2) wherein the second oxygen attacks from the same side as the first.29*31 The PO subunit of PGH synthase reduces the hydroperoxide PGG 2 (2) to the corresponding secondary alcohol PGH 2

(3 ). 7

PGH SYNTHASE PGH SYNTHASE

TYR 385 TYR 385 c HO

COOH COOH 1 9

►

COOHCOOH 10 11 PGH SYNTHASE PGH SYNTHASE

// TYR 385 TYR 385

'OOH

HOOC HOOC

The hydroperoxide-initiated free-radical chain reaction is the rate limiting step for both PGH synthase and 5-LO. Continued presence of 8

lipid peroxide is essential, and in the presence of a peroxide scavenger

such as glutathione peroxidase chain termination takes place. The

enzyme is catalytically inert until the level of lipid peroxide increases.32 In

normal tissue, the peroxide is continually removed by cellular peroxidase

thus limiting autocoid biosynthesis. In the presence of inflammatory signals which attract phagocytes, levels of lipid peroxides increase thereby increasing the levels of the autacoids generated.33

All three oxygens in PGEi are derived from molecular oxygen. The

oxygens at C-9 and C-11 on the five-membered ring of PGEi originate

from the same molecule of oxygen. This led to the proposal invoking

endoperoxide and hydroperoxide intermediates in the mechanism of PG biosynthesis.28-34

Inflammation: Inflammation is a disorder involving local

increases in the number of leuckocytes and a variety of complex mediator

molecules. Since inflammation involves many factors such as

neutrophils, macrophages, lymphocytes and mast cells, and since these are involved in the release of histamine, eosinophil chemotactic factor, PGs, LTs, platelet activating factor (PAF), hexosaminidase, interleukins

(IL) , interferons, inhibition of many rate limiting steps could be important in the successful treatment of inflammatory disorders.35 Even though there are so many mediators, inhibitors of PGH synthase serve as effective antiinflammatory agents, especially in the treatment of acute inflammatory disorders. Thus, generation of PGs is one of the important rate limiting steps in the expression of inflammatory symptoms.35-36 Non- 9 steroidal antiinflammatory drugs (NSAIDs) inhibit PGH synthase and are classified chemically as shown in Table 1. NSAIDs have been extensively reviewed ,37-39 ancj on|y unique features of these drugs and newer PGH synthase inhibitors are described.

Table 1. Classification of Non Steroidal Antiinflammatory Drugs Non Steroidal Antiinflammatory Drugs

Carboxylic acids Pyrazoles Oxlcam

Salicylates Acetic acids Propionic Fenamates

Aspirin Indomethacin Ibuprofen Mefanimic Phenybuta- Piroxicam Diflunisal Acetmetacin Flurbiprofen acid zone Isoxicam Fendosal Sulindac Naproxen Flufenamic Apazone Tenoxicam Tolmetin Ketoprofen acid T rimethazone Diclofenac Fenoprofen Meclofena- Kebuzone Fendonac Benoxaprofen mete Suxibuzone Isoxepac Indoprofen Niflumic acid Feprazone Clidanac Pirprofen Tolfenamic Oxepinac Carprofen acid Fenclorac Oxaprozin Metiazinic acid Pranoprofen Benzofenac Miroprofen Etodolac Tioxaprofen Sumadione Cicloprofen Clamidoxic Furaprofen acid Butibufen Fenbufen Bucloxic acid Pratizinic acid

Cyclooxygenase Inhibitors: CO can be inhibited by three different mechanisms : (1) Reversible competitive inhibition; (2) mechanism based or time-dependent inactivation; and (3) reversible noncompetitive inhibition. 10

CO utilizes a wide variety of fatty acids as substrates, and these are oxygenated at different rates. Such substrates possess at least one 1,4 skipped double bond as found in 8,11,14-eicosatrienoic acid (1 4 ) and AA (Km = 2-10 pM, Vmax = 1400/nmol enzyme).40 When the oxygenation

rate is low, these fatty acids serve as competitive inhibitors by occupying the active site of CO. For example, eicosapentaenoic acid [15; 20 : 5 (n-

3)], present in fish oils, behaves as a substrate only in the presence of elevated peroxide concentrations. This substrate is oxygenated by the enzyme at one half the rate for the normal substrate AA.1 In the presence of normal peroxide levels competitive inhibition (Kj = 2.5 pM) of sheep vesicular PGH synthase is observed. Several additional fatty acid analogues such as docosahexaenoic acid (16; Ki = 5.2 pM), 3,6,9- eicosatrienoic acid (17; Kj = 6.0 pM) and 9,12,15-eicosatrienoic acid (18;

Kj = 20.0 pM) are competitive inhibitors which bind the active site of CO in ram PGH synthase without forming products.26-30 Thus C-18

(octadecenoic) to C-22 (docosaenoic) acids possessing one to six skipped double bonds inhibit CO presumably because they mimick substrate AA. 11

‘COOH 14

-COOH

‘COOH 17

‘COOH 18

Competitive inhibition is also observed for NSAID aryl propionic acids such as ibuprofen 19 (Kj = 5 pM; PGH synthase in vitro), pirprofen 20 (Kj = 1.3 pM; sheep seminal vesicle PGH synthase) and naproxen 21

(IC50 = 100 pM; bovine seminal vesicle PGH synthase). Such inhibition is related to hydrophobicity.42-43 12

COOH COOH

J l X j C O O H

Time-dependent inactivation is observed using 5,8,11,14- eicosatetraynoic acid (22, ETYA) in the presence of oxygen and a hydroperoxide activator. Such suicide inhibition is seen for both CO and

5-LO.44-45 ETYA is metabolized by the enzyme and converted to a reactive allene intermediate 23. Such metabolic activation alkylates catalytic sites in both CO and 5-LO.46

COOH

COOH 13

Aspirin (24) also is a time-dependent inhibitor of CO. This drug irreversibly acetylates Ser 530 on PGH synthase-1. Initially, aspirin competes with the substrate AA for binding to the CO site (Kj = 20m M).5-42

Once AA has been competitively displaced, aspirin acetylates Ser 530 thus blocking the enzyme.5-47 Such acetylation has no effect on PO activity. Ser 530 does not participate in catalytic mechanisms of the enzyme, and replacement of this amino acid by Ala 530 using SDM results in a mutant possessing a similar kinetic profile. However, mutation of Ser 530 to Asn 530, a change presenting steric bulk similar to that of acetylated Ser 530, results in a mutant with no CO activity. Nonetheless,

PO activity is retained 20 Seemingly, upon acetylation of serine, aspirin inhibits CO through steric hinderance at the active site. The reaction prevents formation of the appropriate conformation required for AA binding.5-20-4849

Other NSAIDs such as indomethacin (25; IC 50 = 0.4 pM, PGH synthase in vitro) and flurbiprofen (26; ED 50 = 25 mg/kg; carragenan foot edema) inhibit PGH synthase without covalent reaction. These become tightly associated with the protein resulting in a time-dependent, conformational modification of PGH synthase.50 The carboxylic acid and halogen function found in these NSAIDs play a cooperative role in the time-dependent inactivation of the enzyme. This time dependent inactivation may also occur by chemical modification involving halogenation or arylation, and/or physical binding resulting in an allosteric intramolecular rearrangement.42 14

Recently, indomethacin was found to potentiate lnterleukin-2 (IL-2)-

induced lymphokine activated killer (LAK) activity.51 This is proposed to

be a function of inhibition of the formation of PGE 2 and reactive oxygen

species (ROS).51 Paradoxically, both PGE 2 and ROS are known to

abrogate IL-2-induced LAK activity.

COOH COOH

COOH F 26

Using MODBUILDER and classical and quantum mechanical calculations, a model of the CO active site was proposed based upon indomethacin SAR. Accordingly, the active site of CO consists of two non- coplanar hydrophobic regions and a cationic center.52-53 CPK space filling models of AA, based on the mechanistic steps involved in the formation of the cyclic endo-peroxide PGG 2 , is in agreement with this proposal. The carboxy binding center is adjacent to a broad hydrophobic binding region and ir-electron acceptor regions bind the A8, A11 and/or 15

A 14 double bonds of AA from the underside. The A 5 double bond undergoes hydrophobic interaction with the enzyme active site.

The p-chloro substituent on the aryl amide, is thought to co ordinate with the active site heme. The carboxylic acid of NSAIDs like indomethacin bind the cationic site through coulombic or hydrogen bonding interactions, and the indole nitrogen interacts with the A® double bond electron accepting group of AA. Both the indole ring and the aryl amide substituents of indomethacin interact with a hydrophobic groove on the enzyme normally responsible for interacting with AA double bonds. This model provides a rationale for the 80 % decrease in potency observed for bioisosteric E-indane inhibitor 27 wherein the E-geometry precludes interaction of the aryl substituent with the second hydrophobic domain.52 Similarly, conformationally restricted tetracycle 2 8 does not d inhibit PGH synthase because this flat molecule cannot interact with the two non-coplanar hydrophobic regions in the active site.53 Whereas, this model qualitatively predicts and rationalizes the inhibitory activity of

NSAIDs derived from acetic and propionic acids, the model does not explain the binding of potent PGH synthase inhibitors of the salicylic, anthranilic and fenamic acids types 54 16

COOH COOH CH3

27 28

Bicyctic endoperoxide PGH 2 mimic (+) 29 is a time-dependent

PGH synthase inhibitor [IC 50 = 0.3 pM; AA-induced platelet aggregation

(AAIPA)]. Comparisons of bicyclic ether (+) 29 and the bioactive AA conformation using CPK models indicates the same spatial location for pro-HR-C-14 a to the ether function and AA-13proHs abstracted by the Tyr

385 radical.55 Like indomethacin, (+) 29 binds the apoenzyme of PGH

synthase resulting in a protein conformational change. This

conformational change is detectable; the change prevents trypsin-

catalyzed hydrolysis at Arg 277. Interestingly, whereas (+) 29 inhibits PGH synthase, the (-) enantiomer is a thromboxane antagonist (AAIPA

IC50 = 38.5 pM) 55

Heptenoic acid (+) 29 affords complete protection against AA-

induced sudden death in the mouse (10 mg/kg). Unfortunately, catabolic p-oxidation provides for a short duration of action, a,a-Dimethyl carboxylic acid 30 (AAIPA IC 50 = 0.08 pM) is resistant to this metabolism, and thus, this compound possesses a longer but unspecified, duration of action. SAR description for these novel transition-state mimics utilizes PG 17 nomenclature. Translocation of oxygen from position '15* of the hexyl ether 29 (AAIPA IC 50 for the racemate = 0.5 pM) to either C-'14' or C-'16' provides heptyl ether 31 (AAIPA IC 5 0 = 340 pM) and pentyl ether 32

(AAIPA IC50 = 105 pM), respectively, with a 200-fold decrease AAIPA potency.55 During PG biosynthesis radicals are formed at C-13 and C-15. Formation of analogous radical species is electronically disfavoured for heptyl ether 31 and pentyl ether 32, and this may contribute to the 200- fold decrease in potency observed for the bicyclic ethers. COOH

H (+) 29

COOH

(-) 29

COOH

H COOH

Replacement of the hexyl ether group with arylalkoxy or cycloalkoxy functions also decreases potency; note phenylpropyl ether

33 (AAIPA IC50 = 26 pM) and cyclohexylmethylether 34 (AAIPA IC 50 =

165 pM). Changing the degree of unsaturation does not affect the potency of these time-dependent inhibitors; i.e. note (E)-2'-hexenyl ether 19

35 (AAIPA IC50 = 0.3 pM), (Z)-2'-hexenyl ether 36 (AAIPA IC 50 = 3.3 pM), acetylenic ether 37 (AAIPA IC 50 = 0.2 pM) and saturated ether 38

(AAIPA IC50 = 0.2 pM). However, conformationally restricted tricycle 39 does not inhibit AAIPA even at 1mM.55*67 Presumably, potency is a function of the ability of bicyclic compounds 29 and 35-38 to mimic the transition state of endoperoxide PGG 2 . Tricycle 39, likely does not possess the transition-state configuration of frans-endoperoxide PGG 2 and is therefore inactive.

COOH COOH

O 37 O 38

COOH

° 39 20

Reversible noncompetitive inhibition is exemplified by radical

trapping antioxidants. These substances are involved in scavenging of

PG and LT biosynthetic radical intermediates. As mentioned previously, phagocytic leukocytes are inflammatory mediators which generate peroxides at the site of injury. For these reasons, and because PGH synthase is an oxidase, antioxidants are anticipated to serve as anti

inflammatory agents. However, in the presence of abnormally high levels

of lipid peroxides constraints are posed wherein the antioxidant may be

quenched prior to complete reduction of the lipid hydroperoxide.33-58 Interestingly, stereochemical preferences have been observed for

antiinflammatory radical scavenging phenolic amines 40 and 41 .60 Only

(+) f-butyl 40 (ED50 = 6.5 mg/kg) and (+) dibromo 41 (ED50 = 4.3 mg/kg) phenols inhibit carragenan induced paw edema in rat, although both

isomers of amines 40 and 41 scavenge free radicals to the same extent.

40 41

Stereochemical preferences of Cyclooxygenase : CO is an asymmetric target which preferentially binds the S(+) isomer of certain a- arylpropionic acids.61 When enantiomeric differences are observed the 21 more active isomer is called the eutomer and the less active the distomer.

The ratio of activities of the two enantiomers is known as the eudismic ratio. In vitro, a wide range of eudismic ratios are observed for arylpropionic acids: ratios of 160 (ibuprofen; 19), 878 (flurbiprofen; 26),

100 (indoprofen; 42), 133 (naproxen; 21), 6.4 (pirprofen; 20), 23 (carprofen; 43) and 35 (fenprofen; 44) are observed. The eudismic ratio is species dependent (naproxen = 133 and 70, respectively, for sheep and bovine PGH synthase).62-63 These stereochemical preferences have been rationalized using the model previously described for interaction of indomethacin with CO. Accordingly, the likely carboxyl binding site lies below the plane of the "ring binding region"52. The lower potency associated with fl(+)-propionic acids may result from unfavourable interactions of the methyl substituent with the edge of the CO active site.

CXKK-. W X COOH O M I 42 43

COOH 44 22

Another model developed based upon the mechanism of PG biosynthesis also provides a rationale for the stereochemical preference of CO interaction with (S)-a-substituted propionic acids. This model presupposes a very specific conformation for the peroxyl radical intermediate 45.64 This intermediate ultimately serves as precursor for five chiral centers and a frans-double bond during the biosynthetic process. Conjugated double bonds are formed at C-12 and C-14 in radical intermediates 9, 10, and 11. The backbone from C-11 to C-16 is coplanar, and this results in maximum jc-orbital overlap during the production of frans-A-^ PGG2 (2). The C-1 to C-7 loop is essential; there is minimal change in side-chain orientation during cyclization. This model provides a rationale for the binding of the carboxyl oxygen of a-aryl substituted propionic acids 46 (S-3-chloro-4-cyclohexylphenylpropionic acid; ID 50 = 0.05 pg/mL with PGH synthase of sheep seminal vesicles) to the same enzyme site accepting the peroxyl oxygen at C-11 of AA. The backbone which mimicks the substrate-enzyme interaction is highlighted for p-cyclohexyl-substituted 46. The phenyl ring of this inhibitor mimics the planar Tt-system of the C-13 to C-15 substrate backbone. The (S)- methyl group falls in the groove which normally accepts the C-7 intermediate segment. Since this groove is on the backside of the enzyme and there is no corresponding hydrophobic groove on the front side, the (Ff)-methyl group is poorly accomodated. Thus, stereochemical preferences of biosynthetic intermediates provide a rationale for increased CO inhibitory potencies of {S)-propionic acids. 23

45 46

Metabolism of asymmetric propionic acids: Interestingly, the (R)- and (S)-enantiomers of certain asymmetric propionic acids undergo stereospecific metabolic conversions. Chiral inversion of certain less potent or inactive fl-(-)-propionic acids to more potent S-(+) isomers occurs without other alterations. This process is substrate-dependent and takes place at different rates in various species. For example, in the case of benoxaprofen 47, inversion half-lives in man are 40 times that observed in rat.62 Metabolic stereoconversion of racemic hydratropic acid (48) to the (+)-glycine conjugate was observed as early as 1922.

Within the NSAIDs, ibuprofen was the first substituted 2-arylpropionic acid reported to undergo chiral inversion; such chiral inversion subsequently has been demonstrated for cicloprofen (49), benoxprofen (4 7 ), ketoprofen (50), clidanac (51), naproxen (21), fenoprofen (44) and tioxaprofen (52).65 24

COOH

47 48

C q O ^ ooo" COOH

49 50

COOH

Cl 51 52

Ibuprofen is metabolized to tertiary alcohol 53, secondary alcohol

54 and primary alcohol 55. The primary alcohol may be further oxidized to the corresponding acid 56, which in turn undergoes p-oxidation to produce 57. However, only {+) isomers of tertiary alcohol 53 and diacid

56 are detected in man when either {+)- or (-)-ibuprofen are administered.66 These observations provided the first information concerning metabolic inversion of configuration; metabolism to produce 25

tertiary alcohol 53 does not introduce a chiral center. Therefore, inversion of configuration must have taken place a to the carbonyl group.

COOH

COOH HO. COOH 54 55 OH HO

COOH

HO COOH 57

Administration of /efra-deuterated ibuprofen fl{-)-58 generates

three metabolites (S)-59, (S)-60, and (R)-61. Furthermore, the S(+)-

metabolites are b/s-deuterated.67 Chemically, inversion may result by sequential a-hydroxylation, dehydration and stereospecific reduction.

Seemingly, this does not represent the mechanism involved in metabolic chiral inversion since hydroxyacids, atrolactic acid (62) and tropic acid

(63), are excreted unchanged.62 Rather, R-ibuprofen is thought to form 26 activated thioester 64 with coenzyme A. Stereospecific reduction of conjugated thioester 65, following FAD catalyzed dehydrogenation results in S-ibuprofen.67 Such an inversion is observed with methylmalonyl Coenzyme A metabolism. Although there is no evidence, this enzyme is proposed to catalyze chiral inversion of (fl)-profens.67 Only (Ff)-2-aryl propionic acids form thioesters with coenzyme A. Thus, unidirectional chiral inversion of (R)-profens results in their metabolic activation. Consequently, even though a number of chiral propionic acids are available as NSAIDs, all except naproxen are administered as racemates.68

COOH

(S)-59 (fl)-61 27

OH c h 6 h r COOH j^jp^COOH

62 63

ATP AMP

H COOHI ( SCoA CoASH (R)-19

FAD FADH2

► SCoA 65

TPNH TPN

SCoX H COOH

(SJ-19

5-Lipoxygenase : 5-LO catalyzes the first committed step in the biosynthesis of LTs by oxygenating AA (1) to produce 5- hydroperoxyeicosatetraenoic acid (HPETE; 6 6 ). Dehydration yields LTA 4

(6 7), the precursor to LTB 4 (6 8 ) and a wide variety of peptidoleukotrienes .69 This, in addition to reactions catalyzed by PGH 28 synthase, is a second metabolic pathway for AA. The cDNA for human leukocyte 5-LO reveals an open reading frame of 674 amino acids. No signal sequence or membrane spanning domains are obvious from this sequence.70 5-LO contains 1.1 mol of iron per mol of the protein 71 Iron is thought to play a role in hydrogen abstraction and peroxide formation via radical intermediates 71 Inference for histidines as ligands in 5-LO is drawn from soybean LO. In this enzyme, the active site iron is hexacoordinated (4+1 to the imidazole nitrogen and 2+1 to the oxygen ligands) and does not form heme or iron-sulfur clusters.73 Comparison of the amino acid sequence of 5-LO with other LOs indicates 6 histidines to be conserved.74-75 Mutations at His 372 and His 550 result in proteins which do not bind iron and are catalytically inactive. Whereas the His 367 mutant is inactive, this protein binds varying amounts of iron (0.2-0.5 mol iron/mol enzyme) dependent on the mutated amino acid. His 362, His 390 and His 399 mutants are partially active and have the same iron content as the native enzyme.76-77 Based on these results His 372 and His 550 are the proposed iron ligands for human 5-LO. Work with His 367 mutants suggests that either this amino acid is located at the active site and plays a role in substrate binding, or that this amino acid is a loose ligand for iron and holds iron in place in the active site.77 His 362, 390 and 399 are not involved in active site iron coordination. 29

.COOH I OOH .COOH I 6 6 . HPETE COOH

67. LTA 4 I .COOH

68 . LTB4

In its pure form human 5-LO is a highly unstable (half-life = 45 min at 37 °C), 78-kDa cytosolic protein requiring Ca2+, ATP and other unidentified proteins and cofactors for activation.78-79 Activation of 5-LO to the ferric form is followed by translocation to the cell membrane where

5-LO binds a transmembrane 5-LO activating protein (FLAP) and catalyzes two steps in the biosynthesis of leukotrienes (Figure 1).41-80 5. LO first abstracts the pro-S hydrogen from the C-7 position of A A (Km =

11.9 nM; Vmax = 8.5 mmol/min for human PMN 5-LO); this is followed by the migration of a double bond to form a (Z,E)-1,3-diene radical 69.

Oxygen adds from the side opposite to proton abstraction and results in 30 the formation of intermediate 5-HPETE ( 6 6 ) .81 This first step is essentially diffusion controlled with an activation energy of 5.2 kcal/mol 82 In the second step 5-LO catalyzes the dehydration of 5-HPETE. This

involves the stereospecific removal of the C 10 pro-fl hydrogen to produce the epoxide LTA 4 (67).81

FLAP

Phopholipasesho lipases

1 ------• Active 5-LO leukotrienes

Inactive 5-LO Fe2+

Figure 1. Interaction of 5-lipoxygenase and 5-lipoxygenase activating protein ,COOH 66

COOH 67

5-LO recognises the lipophilic character of AA and the ir-character of the double bond at position C-584. Like CO, 5-LO is a ‘suicide enzyme* and is irreversibly inactivated by hydroperoxide metabolites produced during the metabolic conversion of AA85. LTA 4 is subsequently hydrated by a specific LTA4 hydrolase to produce a chemotactic factor LTB 4 (68).

LTB4 stimulates leukocyte migration and aggregation of polymorphonuclear leukocytes. Alternatively, LTA 4 conjugates with glutathione to generate peptidoleukotriene LTC 4 (72). LTC 4 is 32

metabolized both by loss of the glutamyl side chain, and by hydrolysis of

glycine. This generates LTD4 (73) and LTE 4 (74), respectively.

Together these LTs are referred to as slow reacting substance of

anaphylaxis (SRS-A ).86

COOH

.COOH

72. LTC4 X = glutathione 73. LTD4 X = cysteine-glycine 74. LTE4 X - cysteine

Even though other LO such as 12-LO (platelets) and 15-LO are

known, inhibition of 5-LO is the focus of most antiinflammatory research

e ffo rts .64 Inhibition of 5-LO suppresses the synthesis of all LTs. Therapeutically, 5-LO inhibitors are useful in the treatment of diseases

such as psoriasis, arthritis, asthma, inflammation, rhinitis, conjunctivitis,

myocardial ischaemia, ulcerative colitis and Crohn's disease .84 Classical

NSAIDs are not effective LT biosynthetic inhibitors and thus considerable

effort is focussed on the development of selective or dual CO and 5-LO 33

inhibitors. New entities are screened utilizing the LT inducer A23187 in

stimulated broken RBL cells and leukocytes. In such studies the release

of 5-LO metabolites is monitored using radioimmunoassays. Broadly, 5- LO inhibitors are classified as substrate and product analogues, FLAP inhibitors, iron chelators, and antioxidants.

Substrate and product analogues : Analogues of AA

substrate or products, including any of the LTs, competitively bind the active site of 5-LO. Although, a number of aromatic 5-LO antimetabolites have been designed to mimic the AA backbone, only those analogues with an obvious structural resemblance to the substrate or product are discussed in this section. 5-LO recognizes the 5,8-Z,Z-1,4-diene in AA and stereospecifically transforms the substrate to 6,8-Z,E-1,3-diene-5- hydroperoxide 66. 5-LO interacts with appropriately substituted acetylenes in a similar manner. Thus, whereas, acetylenic 5,6- dehydroarachidonic acid (DHA; 75) is an irreversible and time-dependent inhibitor of 5-LO (IC50 = 10 nM; RBL-1 5-LO); 8,9-DHA (76) has no effect on this enzyme.46-80-89

Possibly, 5-LO metabolizes 5,6-DHA in a manner similar to substrate AA by abstracting the pro-S hydrogen at C-7. Generation of a- acetylenic radical 77 is followed by triple bond migration to provide reactive allenic radical 78. This 4,5-DHA radical may be responsible for the irreversible, time-dependent inactivation of the enzyme or may generate other radical inactivating species by trapping oxygen and 34 forming allene hydroperoxide 79. Rearrangement of 79 furnishes enone radical 80.129

Radiolabelled allene 4,5-DHA inhibits 5-LO in a manner similar to acetylene 5,6-DHA. This provides credibility to the proposed reactive allene intermediate during inactivation.90'92 Eicosatetraynoic acid

(ETYA; 22) is the all acetylenic analogue of AA. This compound inhibits both 5- and 12-LO [RBL-1 cell 5-LO, IC50 = 28 pM. Human PMN 5-LO,

IC 50 - 100 pM. Human platelet 12-LO, IC 5 0 = 1 pM] by generating allenic reactive species not unlike those formed from 5,6-DHA.88-94-95

COOH COOH

75 76

77 78

OOH

O 79 80 35

Replacement of the methylene group at C-7 of AA with sulfur provides sulfide 81, a mechanism based inhibitor. The intermediate radical sulfonium cation 82 alkylates 5-LO before or after oxygenation.

This produces the corresponding hydroperoxide which form enzyme adduct 83.97

OOH Enz

Nu '

7,7-Disubstituted analogues block C-7 hydrogen abstraction during

LT generation. However 7,7-dimethylarachidonic acid (84) is only a weak inhibitor of 5-LO (IC 50 = 100 pM in RBL-1).98-99 Another structural modification leading to enzyme inhibition consists of substituting the 5,6- double bond with an isoelectronic group incapable of undergoing oxidation. Although, substitution of the 5,6-bond with a phenyl ring generates substrate analogue 85 possessing some 5-LO inhibitory activity (94 % inhibition at 100 pM in RBL-1 supernatent), cyclopropyl substitution generates inactive triene 8 6 ."-1oo 36

COOH COOH

.COOH

AA hydroxamate 87 was the first hydroxamate synthesized as substrate-based 5-LO iron-chelator inhibitor. This compound blocks RBL- 1 5-LO with a Kj of 0.13 pM. N-methyl hydroxamide 88 has a three-fold potency increase (Kj = 0.04 pM ).96-102

Lipoxygenase products 12- and 15- HETE and HPETE, serve as feedback inhibitors for LO. Interestingly, LTA 4 cyclopropyl analogues 89 and 90 inhibit 5-LO100. The methyl ester 90 (IC50 = 18 pM) is 2.5 times as potent as the corresponding carboxylic acid 89 (IC50 - 44 pM).100

Thus analogues resembling substrate and products of 5-LO serve as inhibitors of this enzyme. 37

O O

NHOH

87 88

/ (CH2)4COOR

\ = / \ / \ / \

89 R = H 90 R = CH3

FLAP inhibitors : FLAP is a 18 kDa hydrophobic membrane

protein possessing three 20-30 amino acid hydrophobic regions.104 The three a-helica! hydrophobic transmembrane domains are connected by two hydrophilic loops and the amino and carboxy terminii lie on opposite sides of the cell membrane.41-80 FLAP mediates translocation of 5-LO either by both modifying and activating the enzyme, or by serving as a receptor for 5-LO. Because, neither phosphorylation nor significant molecular weight changes are detectable for activated 5-LO, FLAP likely serves as its receptor.104

FLAP-mediated translocation of 5-LO is required for AA accessibility. Majority of the substrate AA exists in the cell membrane and is inaccessible to cytosolic 5-LO in the absence of FLAP.105 Human

FLAP binds radiolabelled AA photoaffinity analogue 95. Such photoaffinity labelling of FLAP is competitively inhibited by AA (IC 50 = 10- 38

20 nM). The Km of AA for human 5-LO is similar to the IC 50 of AA when

inhibiting photoaffinity labelling by azide 95. Thus, AA concentrations

required for half-maximal 5-LO velocity during metabolic conversion to LTs is similar to the concentration required for inhibiting 50 % FLAP photoaffinity labelling. This indicates that FLAP specifically binds and

transfers AA to 5-LO.106

,COOH

125j 95

LT biosynthesis is co-operatively catalyzed by FLAP and 5-LO and LTs are generated only after 5-LO binds FLAP. In vivo, both these enzymes are required for LT formation. Consequently, human

osteosarcoma 143 cell lines expressing either 5-LO or FLAP do not

generate LTs when stimulated by Ca2+ ionophore A23187. Expression of

both 5-LO and FLAP is required for significant LT production.104

FLAP antimetabolites inhibit LT biosynthesis in vivo by preventing translocation of 5-LO. These compounds act by either directly binding and blocking the 5-LO docking site, or by rendering the binding domain

inaccessible due to allosteric conformational change. FLAP 39 antimetabolites do not inhibit 5-LO directly. Rather, these compounds prevent 5-LO from binding FLAP and accessing substrate AA. Therefore, the translocation inhibitors are inactive against isolated 5-LO.107

FLAP antimetabolite, thioindole 91 (MK 886) inhibits rat and human polymorphoneutrophils in vivo (IC50 = 2.7 and 3.5 nM respectively) and expectedly has no effect on purified 5-LO in vitro.'107

Interestingly, photoaffinity labelling of this translocation inhibitor led to the discovery of FLAP. 125l radiolabelled azide 92 undergoes photochemical reaction (UV radiation) and generates FLAP alkylating nitrene 93 Furthermore, immobilization of MK 886 on Sepharose 12 generates affinity column gel 94, useful for purifying FLAP.41 40

^CgcL,N ^ COOH

b f ^ a N:

COOH COOH

H H

GEL

Activity of N-benzyl indole type translocation inhibitors is structurally associated with a) hydrophobic N-benzyl-/>substitution; b) a 2\2‘-dimethyl-substituted propanoic or butanoic acid at C-2; c) a lipophilic sulfide at C-3; and d) a non-polar sterically hindered group at C-5. The 41

minimally active indole is qualitatively represented in 96.107 Although,

the significance of these functional groups is currently unknown, indole 96 may act by mimicking the AA-FLAP binding conformation.

Lipophilic - SR Lipophilic thio group steric group

N (CH2)nCOOH

C 3 -C 4 chain lengths 2,2-dimethyl substituted x f

Only lipophilic monosubstituent

Quinoline 97 is also a nanomolar FLAP inhibitor (human PMN IC 50 = 6 nM). Interestingly, quindole 99 possessing both indole and quinoline

rings is 3-fold more potent (human PMN IC50 = 2 nM) when compared to

indole 91 and quinoline 97. Furthermore, all three types of FLAP

inhibitors, indole 91, quinoline 97 and quindole 99 inhibit photoaffinity

labelling by azides 92 and 98. This inhibition occurs with IC50S ranging from 30 nM to 100 nM.108 COOH

COOH

Conformationally restricted tricyclic quindole 100 is 10-fold less potent (FLAP IC 50 = 19 nM) than bicyclic analogue 99. Replacing the quinoline ring in tricyclic analogue 100 with a 5-methoxypyridine function 43

affords 101, a 5-LO inhibitor with an IC 50 of 396 nM (rat 5-LO). Inhibition

of 5-LO is accompanied by a 2-fold drop in FLAP potency (IC 50 = 40 nM).

Replacing the 5-methoxypyridyl group with a p-phenylpyridine function

increases 5-LO potency 6-fold. Thus, thiopyran 102 inhibits rat 5-LO with

an IC 50 of 60 nM. This compound is no longer a translocation (FLAP) inhibitor. Biosteric replacement of carboxylic acid group in quindole 102 homologue by a tetrazole function provides 103 having a 3-fold greater

potency (rat 5-LO IC 50 = 17 nM; human PMN IC 50 = 8 nM). This compound is in clinical development.109-110 Thus, replacing the quinoline functionality by a 5-phenylpyridine group transforms FLAP inhibitor 100 to a 5-LO inhibitor 1 0 2 . The differential specificity of the two heterocyclic inhibitors may reflect the preferred AA binding conformation for FLAP and 5-LO. 44

100

CH.

COOH

XT 101

Cl JO*

102 X = COOH 103 X = NN V'V H 45

Iron chelators : Hydroxamic acids form strong complexes with transition metals and possibly inhibit 5-LO by chelating active site Fe+3.

During metabolic conversion of substrate AA 104 to LTA4, the ferric form of 5-LO homolytically abstracts a hydrogen radical at C-7 generating radical intermediate 105. This radical is oxygenated by 5-LO to provide

5-HPETE (106). A specific binding pocket on the enzyme accepts the peroxyl- functionality of 5-HPETE and transforms this intermediate to

LTA4 . Hydroxamates (107) are proposed to bind this 5-hydroperoxy- pocket of 5-HPETE. Following binding, iron is chelated by such hydroxamates resulting in a dead end 5-LO complex .111-112

OOH

104 105 106

Fe~

107 46

N-Hydroxamates of PGH synthase inhibitors such as meclofenamic acid (108), indomethacin (25) , sulindac (109) and ibuprofen (19) are dual CO and 5-LO inhibitors. For all 4 analogues, N-methylation results in the generation of 5-LO selective antimetabolites. O-methylation provides

CO selective compounds. These hydroxamates have been evaluated in intact RBL-1 cell lines for both CO and 5-LO inhibitory properties. The data is summarized in Table 2. The prodrug sulindac (109) is reduced in vivo to sulfide metabolite. While a PGH synthase inhibitor in vivo, sulindac is inactive in v/'fro.115 Systemic hydrolysis of N- and O- substituted hydroxamates to corresponding carboxylic acids limits their use.

O O

sII o 108 109

A X = NHOCH3 B X = NHOH C X = N(CH3)OH 47

Table 2. Inhibitory constants of NSAID derived hydroxamates

IC50 pM

A X = NHOCH3 B X - NHOH C X = N(CH3)OH

CO 5-LO CO 5-LO CO 5-LO

108 0.55 16 1.1 3.3 15 1.5

25 <0.2 24 1.1 7.5 5.2 1.4

109 NSa 26 NSa 13 NSa 1.0

19 5.8 24 % (32P 10 20 20 1.2 a- No significant inhibition at 32 pM b-24 % inhibition at 32 pM

Hydrophobicity is an important parameter determining potency for hydroxamates having structure Ar-X(R 2 )-CO-N(Ri)-OH 110. These compounds are referred to as type A hydroxamates. Type A hydroxamates possess a large group [ArXfFte)-] bonded to the hydroxamate carbonyl and small group (Ri) bonded to the hydroxamate nitrogen. A highly significant correlation between potency, and substituted hydrophobicity and electronic constants for the p-substituent is observed for benzohydroxamic acids 111 (eq 1).117 Thus, for eq 1, n = 10, s = 0.220, r = 0.945, p < 0.0001, n is the hydrophobicity of the entire group bonded to the hydroxamate, and ap is the Hammett electronic constant for the aryl substituent. Positive coefficients for it and op indicate 48 that hydrophobic and electron withdrawing substituents positively correlate with inhibitory potency. The hydrophobic descriptor alone explains greater than 76 % of the variance; 13 % is a function of the electronic descriptor. Accordingly, the electron donating hydrophilic p- hydroxybenzylhydroxamate 112 inhibits RBL-1 5-LO with an IC 50 of 190 pM and the electron-withdrawing hydrophobic p-trifluoromethyl- hydroxamate 113 has an IC 5 0 of 27 pM. N-Methylation of the hydroxamic acid results in an 8-fold increase in potency relative to the unsubstituted hydroxamate; see hydroxamate 114 {IC50 =110 pM) and

N-methyl hydroxamate 115 (IC50 = 14 pM).117 log (I/IC50) = 0.49 n + 0.45

-OH

< r S 111

. O ’ • j X i 112 113

OH r * y V O H H L I CH '

114 115

The potency against 5-LO in RBL-1 of 111 related aromatic and heteroaromatic hydroxamates bonded using different spacer units such as alkyl (methyl, ethyl, propyl) and ethers (aryloxymethyl) correlate with tc according to equation 2: log (I/IC50) = 0.57 n’ - 1.16 INH - 0.69 I Big 2 - 0.064 h + 4.30 eq 2 n = 111, s = 0.323, r = 0.94, p < 0.0001 where n‘ is the hydrophobicity of the first three carbons of the side chain; l|MH is an indicator variable and equals 1 or 0 if R 1 is hydrogen or any other substituent respectively; I Big 2 is a second indicator variable with a value of 0 {R 2 - H, CH3 ) or 1 (R2 larger than methyl); h is the third 50 indicator variable for conjugation (equals 0 when the aryl group is either directly bonded or bonded via a conjugated system, and 1 otherwise ).117

Large a-substituents are disfavoured. Thus, a-p-bromophenyl substituted styryl hydroxamate 117 (IC50 = 1.8 pM RBL in vitro) is 10-fold less potent when compared to parent hydroxamate 116 (IC50 = 0.1 pM

RBL in vitro). Steric hinderance of the bulky a-substitutent may decrease potency by obstructing iron-hydroxamate chelation. Quantitatively, the steric effect is accounted for using indicator variable Isig 2- A second indicator variable Inh quantifies 8-15 fold potency increases of N- methylated hydroxamates 115 (IC50 = 14 pM) when compared to unsubstituted hydroxamate 114 (IC50 = 110 pM). Directly bonded aryl hydroxamate 1 1 1 (no spacers), and conjugated hydroxamate 116 are both 4 fold more potent than hydroxamates using non-conjugated or heteroatom spacers. Conjugation increases hydroxamate electron density resulting in tighter Fe+3-inhibitor interaction. QSAR equation 2 utilizes indicator variable h to explain this effect.117

The most important physical property associated with the hydroxamate series is hydrophobicity. Limiting the hydrophobicity to the first 3 carbons provides an improved correlation constant of 0.94 when compared to 0.82 for the entire substituent. The 3 carbons utilized in the ji'-calculation limit a 12 A° hydrophobic binding boundary. Thus, increased hydrophobic substitution at distances greater than 12 A0 from the hydroxamic acid group does not affect potency, in vitro. Additionally, a 51 hydrophobicity coefficient of 0.57 indicates these inhibitors to be 57 % desolvated in the enzyme active site.118

*OH -OH N I 0 0 ^

116

As predicted using QSAR equation 2, styryl-N-methyl hydroxamate

116 is a potent 5-LO inhibitor in vitro (IC50 = 0.1 pM; RBL cells).

Unfortunately, 116 exhibits only 66 % inhibition in the rat peritoneal anaphylaxis model (100 mg/kg; oral administration). Other conjugated and directly bonded hydroxamates also are either inactive or exhibit a significant potency decrease in vivo. Furthermore, low potency in vivo for these hydroxamates correlates to greater than 50 % plasma hydrolysis within 1 min of iv. administration (20 mg/kg). Interestingly, introduction of a substituted methyl spacer increases hydrolytic resistance and methyl hydroxamate 118 exhibits plasma concentrations of 8.4 and 7.9 pM, 20 and 60 min following oral administration (100 mg/kg). This compound inhibits rat peritoneal anaphylaxis with an IC 50 of 0.28 pM.119 52

118

As pointed earlier, the above compounds are type A hydroxamates.

Reversing the substituent pattern by appending large bulky groups to the hydroxamate N and small sustituents to the carbonyl function provides type B hydroxamates 120. The type B sustitution pattern confers increased plasma hydrolytic resistance. Improved plasma levels thus obtained translate to upto 5 times increased potencies in rat peritoneal anaphylaxis model when compared to type A compounds. Selected examples (121-124 A and B) of potencies in vitro and in vivo are shown in Table 3. Thus, whereas plasma levels of only 8 pM of type A hydroxamate 124A are detected 1h following oral administration (100 mg/kg), the corresponding type B analogue 124B exhibits concentrations of 110

O OH

119 TYPE A 120 TYPE B 53

Table 3. Comparison of in vitro and in vivo inhibitory constants of Type A and Type B hydroxamates.

Compd RBL-cells IC50 Rat peritoneal anaphylaxis

Type A Type B Type A Type B 121 0.78 0.79 62 17 1 22 0.29 0.42 65 24 1 2 3 0.59 0.54 54 19

124 0.28 0.37 40 8

121 122

O C ^

123 124

A X s CON(OH)CH3 B X = N(OH)COCH3 54

The methyl group of acetylated type B hydroxamates provides

optimal steric bulk. Decreased potencies are observed with smaller or

larger functionalities. Thus, whereas acetylated 121B (ED 50 = 28 pM )

and 124B (ED 50 = 68 pM) inhibit rat peritoneal anaphylaxis, their formyl

(R = H) counterparts 126 and 125, respectively, are ineffective at 200 pM/kg. Interestingly, the potencies in vitro of these compounds are comparable [125 (IC 50 = 0.94 pM); 124B (IC50 = 0.37 pM); 126 (IC 50 =

1.9 pM); 121B (IC50 = 0.79 pM)]. Large substituents are also detrimental to activity and benzyl 127 and isobutyl 128 substituted type B hydroxamates are also inactive at 200 pM/kg (rat anaphylaxis model).

The bulky substituents are thought to decrease oral absorption and bioavailabity; neither the parent compounds 127 and 128 nor their metabolites have been detected in the plasma.121 55

O HCL H (X A N X R, r R

j V "

124B Rt = CH3 121B R, = CH3 125 Rt = H 126 R, = H 127 R1 = Ph 128 R, =CH2CH(CH3)2

h q - A

( X ° 1;129

Propenyl type B hydroxamate 130 (plasma T 1/2 = 2 h; IC50 = 0.04 pM; A23187 stimulated leukocytes) forms 3 metabolites in humans.

Hydrolysis following a-oxidation results the major carboxylic acid metabolite 131. This process likely occurs following reduction of the propenyl group since propyl hydroxamates are hydrolytically unstable.122 The propanoic acid 131 may undergo p-oxidation to benzoic acid 132.

Reduction of hydroxamate 130 provides acetamide 1 3 3 .

Glucuronidation results O-conjugate 134.123 Whereas, type A hydroxamates are metabolized by simple amide hydrolysis, type B hydroxamates utilize a metabolic pathway involving a-oxidation prior to hydrolysis. This may explain why type B hydroxamates have higher plasma levels and longer durations of action. 56

I c x j x ^ v 131 134 O

132

Besides iron chelation, current evidence indicates an additional redox-based inactivation mechanism for Type A and Type B hydroxamates. The EPR spectrum of ferric soybean LO treated with 2 eq of N-hydroxy-N-methyltetradecylamide (142) does not show the characteristic ferric signal at g = 6. A similar effect on the ESR spectrum for soybean LO is seen using hydroxamic acids 124A and 124B.

Therefore, hydroxamic acids may inhibit LO by reducing the ferric active site in this enzyme. This generates an EPR invisible ferrous form.127

Alkaline ferricyanide oxidation of hydroxamate 135 involves a one- electron transfer to generate nitroxyl radical 136. This radical couples to 57

provide dimer 137. This dimer rearranges to nitroso compound 139 and an unstable b/s-acylated nitroso intermediate 138. Hydrolysis of intermediate 138 provides acid 141 and hydroxamate 140. Thus, chemically 1 mole of hydroxamate 135 is converted to 0.5 mol of carboxylic acid 141 and 0.5 mol of hydroxamate 140.114 Analogous products have not been isolated during enzymatic inactivation of soybean

LO by hydroxamates 124A and 124B. Therefore, even though the EPR spectrum of soybean LO indicates loss of the ferric oxidation state in the enzyme, redox-based inactivation mechanisms for such iron chelators has not been substantiated. The mechanism of 5-LO inhibition by hydroxamates remains controversial.

Ph^l Ph"l i RV N 'o - V N- o ’ o o o 135 136 137

Ph>, o Ph-,

V N-OH + R_COOH 0 + 1 3 8 O ~ 140 141 Ph‘ N1, 139 58

142

Iron chelating N-hydroxyureas (143) obtained by replacing small alkyl substituent in type B hydroxamates with an amine, exhibit further improvement in oral bioavailability and potency in vivo. A direct comparison of various factors affecting bioavailability including plasma concentration levels, plasma half-life, c-RBL IC 50 and ED50 (rat peritoneal anaphylaxis) for type A hydroxamate 124A, type B hydroxamates 124B and 144 and hydroxy urea 143 are shown in Table 4 .121-124 Clearly, high potency in vivo, reasonable plasma half-life and high plasma concentration levels provide the superior bioavailability profile for N- hydroxyurea 145.

OH 1

143

144 145 59

Table 4. Comparison of inhibitory constants and pharmacokinetic

parameters of Type A, Type B hydroxamates and N-hydroxyureas.

cRBL ICgg EDjjjj (R.P.A.)a Plasma Tj /? Plasma levelb

Type A (rat) iv. 1 hb 12 4 A 0.28 pM 40 mg/kg 0.4 h 8 pM

Type B 12 4 B 0.37 pM 8 mg/kg 1.1 h 110 pM

144 0.5 pM 21.8 mg/kg 6.0 h 126 pM

Hydroxy urea 145 0.62 uM 9.3 mg/kg 5.6 h 259 uM a- rat peritoneal anaphylaxis b- Plasma levels in rats 1 h after administrating 100 mg/kg compound

Benzothiophene hydroxyurea 146 (zileuton; IC 50 = 0.9 pM human

PMNL), is currently in Phase III trias as a LO inhibitor indicated for the treatment of asthma and ulcerative colitis.125 Although, (+) zileuton is available, comparitive biological properties for the two enantiomers are not reported.134 This compound is excreted as a urea and diastereomeric glucuronide.126 Thus, unlike type A and type B hydroxamates, N-hydroxy ureas are not hydrolyzed in humans. This may explain the superior bioavaibility profile associated with this class of iron chelators. 60

OH NH;

O h V

146

Antioxidants, which scavenge free radicals : A variety of reducing agents such as phenols, catechols, hydroquinones, naphthols and pyrrazoiones are inhibitors of 5-LO. 5-LO antioxidant inhibitors have been recently reviewed.128-129 Antioxidants interact non-specifically with 5-LO by scavenging radical intermediates and/or reducing the active heme site.105 Potency for these inhibitors is directly proportional to hydrophobicity.128

The ESR spectrum of soybean LO incubated with phenolic antioxidant nordihydroguaretic acid (147; NDGA) and phenindone (148) shows a hyperfine splitting pattern. This indicates formation of free radical intermediates. Additionally, the characteristic ferric signal at g = 6 is lost and the enzyme is reduced to an ESR silent ferrous form 149.

Hydroperoxides oxidize the inhibited ferrous enzyme and regenerate catalytically active protein.130 61

HO, OH < Y % X ) HO' 147 148

Enz-Fe (III) L(RH)

LOOH LO- Enz-Fe (II) + L (R*) LOOH I 149

Enz-Fe (II) + LOO-

Figure 2. 5-Lipoxygenase inactivation by antioxidants

Therapeutic use of 5-LO antioxidant inhibitors is limited; NDGA

147 and phenindone 148 induce methaemoglobinaemia. However, this side-effect is not universal for all antioxidants and topically used phenol

149 has no reported side effects.105 Stereochemical preferences of 5-lipoxygenase Compared to PGH synthase very little information about 5-LO active site

stereospecificity is available. Redox (R)- and (S)-indazolines 150 are

dual CO and 5-LO inhibitors. Interestingly, eudismic ratios of 8 are

observed for both CO and 5-LO. In human blood, R-150 is 285 more times selective for 5-LO (IC50 = 1.4 pM) than CO (IC 5 0 = 400 pM).

However, the S-enantiomer exhibits only a 4-fold selectivity for 5-LO

(IC 50 - 11 pM) over CO (IC 5 0 = 48 pM).131 Stereoselective 5-LO inhibition has also been observed with thiazole 151; eudismic ratios of 80 and 150 are seen in RBL 5-LO and human whole blood. The absolute stereochemisty of the dextrorotary eutomer [IC 50 = 0.13 pM (RBL); IC50 =

0.5 pM (human whole blood)] is not known.132-133 Thus, although enantioselective 5-LO antimetabolites are known these compounds have not served to elucidate active site conformational preferences. 63

c c r ^ o

150 151

In conclusion, PGH synthase inhibitors and 5-LO inhibitors are useful for treating acute and chronic inflammatory disorders. Major advances in CO inhibition have provided clinically effective NSAIOs. In comparison, 5-LO inhibitors only recently have been extensively studied.

Inhibitors of 5-LO are useful in the treatement of ulcerative colitis and asthma. Due to the variety of inhibitory mechanisms available, substances possessing more than one mechanism of action may be synergestically superior antimetabolites. Often substrate shunting results in LT side effects when CO inhibitors are administered and vice-versa. In this respect dual CO and 5-LO inhibitors are of interest. CHAPTER II

ac/-REDUCTONES

Introduction: aci-Reductones serve as non classical carboxylic

acid bioisosters, possessing a biologically relevant redox potential.

Whereas, other non-classical bioisosteres such as tetrazole 152, sulfonic

acids 153, sulfonamides 154, phosphonic acids 155 and 5- hydroxyisoxazole 156 are available, only the 3-hydroxypyran-4-one 157 and the 2-hydroxytetronic acids 158 systems have received attention as

acidic reductones applicable to drug design.140 Introduction of a hydroxyl group in vinylogous acid 160 (pKa = 3.76) generates a readily oxidizable reductone, an ene-diol 161, with the pKa of 4.37. These aci- red uctones are referred to as Carboxylic Acid Mimics designed to deliver a biologically relevant Redox functionality (CAM-R analogues). Such functionalities are found in ascorbic acid analogues 163 and are derivatized in the chlorothricin macrolide antibiotic 164.135

64 65

n n -SO 3H r R - - s o 2n h r , R-P-OH 1 nu OH 152 153 154 155

C r & HO OH 156 157 158

HO OH

HO OH 163 COOH

164 66

: 0 6|

HO HO HO OH O O 159 160 161 162

Drug Design: In drug design, the vinylogous -OH of a ci- reductone is proposed to mimic the carboxylic -OH whereas the C=C mimics the 0=0 of the carboxylic acid.140 Additionally, replacement of a carboxylate in a substrate or inhibitor with an ac/-reductone function introduces a stereocenter at a site a- to the original carbonyl group.

Appropriately constructed acAreductones possess a range of biologically relevant redox potentials (Ei = 0.027 - 0.209 V at pH 7.4) and are proposed to reduce the active site of redox enzymes.

CO, an enzyme with oxidizing heme at active site may be inhibited by such aci- red uctones owing to hydrophobic binding, covalent bonding and/or subsequent one or two electron transfer from the aci-reductone to the active site. The oxidized trione 162, thus formed, is proposed to undergo reaction with the nucleophilic amino acid residues generating 165 in the hydrophobic active site. Such reactivity following reduction of the enzyme site, is a function of the highly electrophilic, electropositive central carbonyl functionality found in trione 162. Appropriately constructed CAM-R analogues might serve as both enantioselective and potent enzyme inhibitors working by both hydrophobic and redox based 67 mechanisms. Additionally, with the discovery of the two isozymes for CO,

CO-1 and CO-2, use of selective inhibitors might serve as tools for the elucidation of isozyme roles in biology, including mechanisms related to complementarity and/or pathogenicity differences, tissue specificity and isozyme topography requirements.

162 165

Biological Properties: ac/-Reductones synthesized in our laboratory possess numerous biological properties and have potentially many therapeutic applications. CAM-R analogues exhibit antilipidemic and antithrombotic activities and promote lymphokine-activated killer

(LAK) activity in lnterleukin-2 (IL-2)-treated human peripheral blood mononuclear cells (PBMC).

ac/-Reductones inhibit CO-dependent, AA-induced platelet aggregation.137-139-141 A positive linear free energy relationship is observed between enzyme inhibition and calculated hydrophobicity (tc) pa ram e te r . 142 Thus, 4-biphenyl- (166) and 4-(4'-chlorobiphenyl)-2- hydroxytetronic acids (167) possess an estimated V of 1.96 and 2.67 66 and inhibit AA-induced platelet aggregation with IC 50S of 135 and 44 p,M142. Selected CAM-R analogues also inhibit purified PGH synthase-1.

HO OH HO OH

166 167

4-Aryl-2-hydroxytetronic acids 168 potentiate iL-2-induced LAK activity. This activity is related in part to CO inhibition. Highly tumoricidal lymphocytes induced by IL -2 have therapeutic potential in the treatment of cancers for which conventional antineoplastic therapy is not useful.

However, the by products of IL-2 activation, PGE 2 and ROS such as superoxide anion radical abrogate LAK activity, aci- Red uctones such as 4-aryl-p 4-alkyl- and 4,4-spiroalkyl-2-hydroxytetronic acids inhibit CO and the production of PGE 2 as well as scavenge ROS, thus improving IL- 2 - induced LAK activity. In standard 4-hour 51 Cr release assays, the improvement in LAK activity observed is comparable to the combined synergy obtained using CO inhibitor indomethacin and ROS scavenging enzymes superoxide dismutase (SOD) and catalase .51-143 Thus, aci- reductones may be useful in IL-2 therapy of cancer. 69

HO OH

168

acAReductones are antilipidemic and lower total serum cholesterol, triglycerides, VLDL, and LDL in cholesterol/cholic acid-fed rats.

Additionally, these compounds decrease apoB in VLDL in vivo and inhibit copper-catalyzed LDL oxidation in wfro.137-139

Free-radicals play a significant role in UV-, drug- and xenobiotic- induced toxicities and activate molecular oxygen to superoxide and other

ROS including hydrogen peroxide and hydroxyl radical. Defense mechanisms, including enzymes such as SOD and catalase and radical scavengers such as glutathione, retinoic acid and ascorbic acid protect proteins and nucleic acids from free-radical toxicities by quenching ROS. Inadequate protection from ROS results in myocardial ischemia, photosensitivity, radiation sensitization, red cell hemolysis and atherosclerosis. 4-Aryl-2-hydroxytetronic acids 168 have antioxidant efficiencies similar to probucol and about twice that of a-tocopherol .140

Synthetic Chemistry: Early synthesis of 4-aryl-2-hydroxytetronic acids involve condensation of aromatic aldehyde 169 with glyoxal sodium bisulfite 170 and potassium cyanide under alkaline conditions to provide enamine 171. Following oxidation to dehydro species 172 70

(acidic sodium nitrite), reduction with ascorbic acid furnishes tetronic acid 173.144

i^ v ' c h o - I J CHtOHJSOgNa^^rx 3 v\ v / ( nh2NaNO2 10% KCN, Na2C03 * O OH H2S04 169 171

R^G-rV°\ / Ascorbic acid R'O ^ w \ / °

O o MeOH*reflux HO OH 172 173

Another approach utilizes intramolecular Claisen condensation of methoxyacetylmethyl ester 174 with a sterically hindered base (LDA, 2.2 eq). Acetylation (AC 2O, Pyr) of methoxytetronic acid 175, followed by deprotection (BBr 3, CH2CI2) furnishes 2 -hydroxytetronic acid 176.136 71

OCH3 HO OCH3 174 175

1. AcgO, Pyr

2. BBr3, OH2CI2 HO OH

176

The third method developed in our laboratories utilizes intramolecular cyclization of benzyloxyacetoxymethyl ester 177 with a sterically hindered base (lithium hexamethydisilazide or lithium dicyclohexylamide; -78°C and -100 °C, respectively) . Deprotection of benzyloxytetronic acid 178 affords the corresponding 2-hydroxytetronic acid 179. 145 Whereas, this methodology is applicable for the synthesis of asymmetric 4-alkyl-2-hydroxytetronic acids with greater than 99 %e.e., stereogenically labile 4-aryl-2-hydroxytetronic acids are generated in less than 35 % e.e. 72

O (X ^ P h CH3V ° V O LiHMDA \ = f A 70 Of"* / \ COOCH3 0 HO OBn

177 178

10% Pd/C \ __ / ► ) — \ cyclohexene, EtOH h o OH reflux 179

Other methods for generating aci- red uctones involve oxygenation of protected tetronic acid. Thus, lithiated methoxytetronic acid 180 on treatment with trimethoxyborane provides boronate ester. In situ oxidative hydrolysis of this intermediate provides protected ac/'-reductone 181. Demethylation (48 % HBr) provides spiro-2-hydroxytetronic acid 182.183 73

1. LDA, -78 °C

C C r0*” "- > O CT 3. AcOH, H20 2 P o h c h 3o c h 3o 180 181

48 % HBr

'-OK.OH c h 3o 182

D-Erythroascorbic acid 187 is prepared in four steps from D- glucose 183. Oxidation of D-glucose( 0 2 ; KOH) affords D-arabinonate 184 which undergoes cyclization (H+) to provide lactone 185. Vanadium pentoxide-catalyzed oxidation generates y-hydroxyester 186. This ester undergoes cyclization to provide D-erythroascorbic acid (187;

NaOAc/MeOH ).196 74

CHO H— — OH COOH — H HO' -H HO— 0 2, KOH H+ HOH2C, H—-OH H- -OH ■OH H— -OH H- f r CH2OH CH2OH OH

183 184 185

COOMe o=H VnO* NaOAc H- -OH HOH2C>y Q ^ O MeOH ] H- -OH CH2OH HOOH

186 187

In conclusion, numerous methods involving Claisen condensation of a-hydroxymethylester precursors, oxygenation of protected tetronic acids and lactonization of y-hydroxy esters are available for the synthesis of aci- red uctones. CHAPTER III

STATEMENT OF THE PROBLEM

Previously, certain 4-aryl-substituted aoreductones of the 2- hydroxytetronic acid type were shown to inhibit the enzyme CO. Early

CAM-R compounds prepared may be viewed as analogues of the

arylpropanoic or aryloxyacetic acids antimetabolites, whereas this

research focuses on modification of the substrate AA. The major emphasis is to design asymmetric syntheses for (fl)-200 and (S)-AA- CAM-R (201) species. Of interest is whether such compounds will serve as enantioselective substrates and be recognized and processed by CO to give AA-related metabolites, and/or whether they will be stereoselective antimetabolites. Whether they are metabolites and/or enzyme inhibitors, a further question relates to mechanism. Will the CAM-R system produce a reversible competitive, non-competitive, irreversible or redox mechanism-based inhibition?

75 76

HO OH

(R)’ 200

HO OH (S)-201

The proximity of the prosthetic heme to the two active sites (CO and

PO) may also influence structure-controlled redox interactions, ff-200, S-

201 and related systems (eicosanoate analogues R-202, S-203, R-204 and S-205) should provide the initial information concerning such interactions and the stereoselectivity involved. Additionally, mechanisms reflecting enzyme inhibition vs. radical scavenging in tissue cultures may be analyzed using optically pure enantiomers. If the eudismic ratio is > 1 specific interactions are likely involved; if the eudismic ratio = 1 possibly enzyme inhibition occurs by simple radical scavenging.

AA-CAM-R (200 and 201) and related eicosanoids (202-205) are also anticipated to find therapeutic use in the treatment of free radical toxicities. Free radicals and ROS such as superoxide anion radical, hydroxyl radical and hydrogen peroxide are produced continuously in biomembranes of living cells. Various enzymatic defense systems consisting of SOD, catalase, glutathione peroxidase scavenge free 77

radicals and ROS and prevent tissue, DNA and protein damage. Additional protection against ROS is offered by natural antioxidants such as ascorbic acid, urate and glutathione. Among these natural antioxidants, ascorbic acid possesses a redox potential of 0.162 V (pH = 7.4) and is the most efficient and is consumed first. Unfortunately, in numerous pathological conditions including reperfusion damage, stroke, cerebral ischaemia, convulsive seizures, Alzheimer's disease and inflammation large amounts of generated free radicals overwhelm these defense systems and result in oxidative stress .209-210 Furthermore, hydrophilicity prevents ascorbic acid penetration in lipid cell membrane, the primary site of free radical and ROS production. Hydrophobic arachidonate [{/?)-200 and (S)-201], eicosaenoate

[(R)-202, (SJ-203J, and eicosanoate [(fl)-204 (S)-205] ac/-reductones are anticipated to circumvent this problem, cross lipid membranes and scavenge ROS at their site of origin. Interestingly, removal of 3-4 conformationally rigid Z-double bonds provides analogues 202-205 possessing a greater degree of freedom and conformational flexibility. HO OH (S)-205

Asymmetric ac/-reductones of the general structure 206 may be synthesized from a-hydroxymethyl esters 207 using Claisen cyclization with a stericaliy hindered base. Retrosynthetically, olefinic ester 208 serves as synthon for CAM-R systems 200, 201, 202 and 203. 79

c h 3o OH HO OH

207 206

Although, a number of methods are available for the synthesis of a-

hydroxymethyl esters, no general method for the synthesis of asymmetric alkenyl, polyalkenyl and long chain a-hydroxy fatty acids (greater than 12

C) are available. Conventional methodologies for the synthesis of chiral a-hydroxy acids and esters involve asymmetric induction of anions in the

presence of strong oxidizing agents such as dioxirane. This route is not useful for generation of a-hydroxy ester synthon 208 because the

skipped tetraene in AA makes all allylic carbons labile to oxygenation.

R ,COOCH3

208

Retrosynthetically, two chirons can be used to generate intermediate 208, an electrophilic condensation precursor 210 (Method

A) and a five carbon aldehyde 211 (Method B ).146 Organocuprous, organolithium and Grignard reagents are useful as nucleophiles in

Method A, whereas Wittig reactions are applicable to Method B. Use of the chiral pool is one possibility for introducing asymmetry. In addition to

AA-CAM-R analogues (R)~ and (fS>4-benzyl-2-hydroxytetronic acids (214 60 and 215) and (R)- and (S,M-hexyl-2-hydroxytetromc acids (216 and

217) were synthesized in order to explore synthetic possibilities.

R /= \ COOCH3

& Method A Method B

*COOCH3 COOCH 3

x ^ c o o c h 3 + „

(,V = V ,+ OH R A HO Or COOCHg \ v JpuLi \S P+(Ph)3 B r "

209 210 212 211

HO OH HO OH (fl)-214 (S)-215

HO OH HO OH

(R )-216 (S)-217 CHAPTER IV

RESULTS AND DISCUSSION

Optically pure (R)~ and ('S>4-benzyl-2-hydroxyt©tronic acids, (214

and 215), can be prepared using Claisen condensation methodology

previously developed in our laboratories .145 Whereas, 4-aryl-substituted

a ci- red uctones are stereogenically labile and undergo racemization

under basic conditions required for cyclization, this methodology provides

chiral 4-alkyl-substituted acAreductones with greater than 99 % e.e.. (S)- Benzyltetronic acid is synthesized starting from commercially available (S)-phenyl lactate (220). Methyl ester 221 (97 %) obtained by refluxing

lactate 220 in acidic methanol is esterified with benzyloxyacetyl chloride to produce benzyloxyacetate 222 (1 eq pyridine, 88 %).

O O

220 BnOCH2COCI Pyr - COCH2OBn

Benzyloxyacetate 222 cyclizes in the presence of a sterically hindered non-nucleophilic base at -78 °C, to afford benzyloxytetronic acid 225 (85 %). In Claisen cyclizations, secondary carbons are preferentially

81 82 deprotonated (when compared to tertiary carbons) by use of sterically hindered bases. The anion formed reacts with the methyl ester carbonyl effecting cyclization to a five-membered ring possessing a 1,3-dicarbonyl system. This system preferentially exists as a conjugated lactone (1H

NMR analysis). The second equivalent of base deprotonates the vinylogous acid generating an anion. Therefore, use of one equivalent of base provides only 42 % yields.

'OBn S'' °^^OBn

COLo c h 3 222

V OCH3 OBn 224

Benzyl group deprotection using either transfer catalytic hydrogenation [10 % Pd/C; cyclohexene (100 mmol); EtOH] or hydrogenation over 10 % Pd/C generates optically pure (S)-benzyl aci- reductone 215.147171 In this case, a cleaner reaction and good yields

(90 %) are obtained using the latter conditions. Transfer hydrogenation results in only 70 % yields. The (fl)-enantiomer 214 is prepared from ( R)- lactate 226 using methodologies employed for the (S)-isomer. 83

, 10 % Pd/C , EtOH 75 % ------B. 1 0 % Pd/C, H 2 , MeOH 90% 225 215

Optical purity is determined by 1H NMR analysis of the

diastereomeric salt of benzyloxytetronic acids (S)-215 and (R)-214 with (Rj-methylbenzylamine. The asymmetric proton resonance signal

observed at 500 MHz for each of the pure (R)/(R) and (S)/{R)

diastereomeric salts, appears as a doublet of a doublets. The racemate

(R,S)/(fl) shows a multiplet resulting from the overlap of these resonance signals. Conventional methodologies for the synthesis of chiral a- hydroxyacids and esters involve asymmetric induction of anions in the presence of strong oxidizing agents such as molecular oxygen and dioxirane .148 This route is not useful for preparation of a-hydroxyester synthon 208 because the skipped tetraene in AA renders all allylic carbons labile to oxygenation. Both the organocuprous and Wittig reactions described utilize a preformed chiral center present in malic acid 229.

Method A: Unsuccessful Organocuprous condensation approach : This approach consists of using organocuprous condensation reactions to obtain the ( R)~ and (S)-a-hydroxymethyl ester precursor to the Claisen cyclization . Thus, commercially available (S)- malic acid (229) is protected as the dioxolane acid 230 using either 84

excess dimethoxypropane and p-toluenesulfonic acid (PTSA, 70 %, 30

min) or pyridinium p-toluenesulfonate (PPTS, 85%, 48 h) as a

ca ta lyst. 149'150 Acid 230 is reduced to the corresponding primary

alcohol 231 with borane-tetrahydrofuran complex at -30 °C for 2 h, 4 °C

for 11 h, and room temperature for 9 h .149-151-152 This unstable primary

alcohol (90 %) is immediately converted to the stable tosylate. Such

conversion is critical since primary alcohol 231 undergoes 1H NMR

detectable intramolecular lactonization within 3-4 h.

ch3o o c h 3 a Dv ___ OH OH ------9 v - o r PTSA, 70 % h o V t 6 h o X ° 229 230

BH3-THF, 90 %

OTs TsCI, 0-4 0 C ^ - OH Pyridine, 95 %

232 231

Unfortunately tosylate 232 is a suitable electrophile in only one instance. In other instances, bromide 233, triflate 234 and iodide 235 have also been explored as precursors to organocuprate reagents. Three electrophiles, tosylate 232, primary bromide 233 and triflate 234 are generated from primary alcohol 231. Tosylate formation (95 %) requires slow addition of p-toluenesulfonyl chloride in order to control the exotherm of the reaction and prevent formation of a black decomposition 85

mixture. Triflate 234 (72 %) is generated using trifluorosulfonyl chloride.

B rom ide 233 (70 %) is formed by treating alcohol 231 with

dibromotriphenylphosphorane. Additionally, refluxing tosylate 232 with sodium iodide in acetone furnishes iodide 235 (74 %).

231 233. X = - Br 234. X = - OSO2CF3 235. X = - 1

The fragment corresponding to the polyalkenyl side chain 209 is obtained by monochlorinating commercially available 2-butyn-1,4-diol

(236) with thionyl chloride in pyridine .153 Chloro-2-butyn-1-ol (237; 64

%) is subjected to a series of acetylenic condensation reactions to furnish desired tetradecenyl complementary fragment 2 4 5 .154 Treatment of chloride 237 with 1-heptynylmagnesium bromide in the presence of 1.5 mol % Cu(l)CI in refluxing tetrahydrofuran furnishes 2,5-undecadiynol

238 (85 %). Alcohol 238 is converted in 3 h to bromide 239 (80 %) using phosphorous tribromide in refluxing ether. This bromide is purified by distillation under reduced pressure. Generally, such purification of acetylenic compounds containing up to 12 carbons is possible without significant decomposition to side products such as allenes. However, contaminating acidic impurities have to be washed from bromide 239 prior to distillation. Failure to do so, results in explosive allenes / radicals at temperatures of approx. 100 °C. 86

-.OH SOCI2 -Cl / = = ^ ^ HO pyri 64 % HO 236 237

EtMgBr, CuCI THF, 85 % X 238. X = - OH

239. X = - Br

2,5,8-Tetradectriyn-1-ol (240; 76 %) is obtained by employing a second acetylenic condensation of bromide 239 with 3 eq of the Grignard derived from propargyl alcohol. This alkylation is facilitated using 6.5 mol

% Cu(l)l in refluxing tetrahydrofuran (22 h). In cases where the carbon backbone exceeds 12 carbons improved yields for acetylenic condensation reactions are observed using Cu(l)l instead of Cu(l)CI c a ta ly s t. 155 The unstable 1,4-skipped triyne 240 is purified by crystallization at -20 °C in petroleum ether and is immediately reduced to the all Z -trie n e 241 (92 % ).113-155 This partial reduction is facilitated with Lindlar catalyst by monitoring the uptake of a calculated amount of hydrogen in the presence of quinoline poison .156 87

OH . 3 eq r EtMgBr 6 eq

Cul, THF , 22 h

239

Lindtar catalyst. Pd/5 % BaSO

H2, Quinoline, 92 % -OH

240

PPh3.Br2 , 84 %

Treatment of primary alcohol 241 with dibromotriphenyl

phosphorane in acetonitrile furnishes the allylic bromide 242 (64 % )157.

When more than two skipped double bonds are present, mild brominating

conditions provide better yields. Treatment of bromide 242 with ethynyl

magnesium bromide in the presence of 2 mol % Cu(l)l gives 4,7,10-

hexadecatrien-1-yne (243; 72 %).1 se-ieo The lithium acetylide of 243 (n-

BuLi; -78 °C) is quenched with iodine at -25 °C to provide ethylenic iodide 244 (90 %). This acetylene is reduced to all-c/s-tetraene iodide

245 (78 %) using dicyclohexyl borane .161-162 Iodide 245 undergoes

lithiation (-30 °C, n-BuLi) and is converted to the diorganocopper reagent

194 with Cu(l)l at -20 °C. This organocopper reagent is alkylated with dioxolane iodide 235.163 Under these conditions and at temperatures of -30 °C, -40 °C and -65 °C complex decomposition products are observed. 88

R =-[ CH2

.___ Rr MgBr Cu(l)l BuLi, l2l - 25 °C r— V ------► ------J T H F ,1 0 h ether

242 243

BH3 .DMS Cyclohexene

244 245

235 CuLi— 0.5 M 245 X = l 194 BuLi 193 X = Li

195

Additionally, condensations with simple alkenyl organocuprous reagents failed. Although, 1-bromotetradecane (246) serves as a three step precursor to c/s-hexadec-l-enyl iodide (248), and lithiation of iodide

248 (BuLi; -35 °C) and formation of organocuprous species (Cul; 0.5 eq) is successful ,164 reaction of cuprate 250 with tosylate 232 at -25 °C in the synthesis of 251 fails. Unlike condensations involving iodide electrophiles the tosylate does not decompose and is observed in the 1H 89

NMR spectrum. Hexadecyne (247; 85 %) is obtained by nucleophilic

displacement of bromide 246 using a lithium acetylide ethylene diamine

complex in DMSO (10 °C ).165 The lithium acetylide of 247 is quenched

(I2, BuLi, -78 °C) and the resultant ethylynic iodide (88 %) is reduced to

the c/s-hexadec-1-enyl iodide (248; 69 %) with dicyclohexylborane .161

-Li EDA Br 11 DMSO, 10°C 11 246 247

1. I2 , BuLi

2. BH3 .DMS 11 cyclohexene 248

/ f V1 0 N / = N * 0.5 M 1 0 248 X = i 250 BuLi \ z 249 X = U

251

Another strategy involves use of higher order copper reagents. Iodide 248 is lithiated (2 eq; f-BuLi; -78 °C) and may be converted to a heteroorganocuprous species using 3,3-dimethylbutyne-copper complex at -35 °C. However, the organocuprous species generated on treatment with copper bromide-dimethyl sulfide ,166-167 when undergoing reaction with iodide 235 results in decomposition. 90

I 246 X - 1 252 249 X = Li O.

251

Propenyl iodide 254, derived from propargyl alcohol (253) serves

as a three carbon source amenable to further elongation. f-BDMS

protected propenyl iodide 254 is prepared utilizing a sequence similar to

one employed for the synthesis of hexadecenyl iodide 248. The iodide

254 should undergo facile reaction owing to less steric hinderance associated with proposed hydrophobic collapse anticipated for long chain

iodides 245 and 248. Subsequent to lithiation (f-BuLi, 2 eq, -78 °C), alkene 191 is converted to a organocopper species. Even in this case, attempted alkylation with dioxolane iodide 235 did not lead to isolable product although the reactant was likely consumed as indicated by analysis of the 1H NMR spectrum.

1 . TBDMS-CI Imd, DMF 2. I2 .BuU HO*

253 cyclohexene & ►

290

A model reaction involving alkylation of tosylate 2 3 2 with dibutylcuprous lithium in ether at -78 °C provides protected a-

hydroxyoctanoic acid 257 (95 % ).168 However, use of tosylate 232, brom ide 2 3 3 , triflate 234 or iodide 235 as electrophiles proved

unsuccessful for the preparation of synthon 208.

Cu(l)l, - 78°C ether, 95 % 232 257

Hexyl dioxolane 2 5 7 is readily converted to (,SJM-hexyl-2-

hydroxytetronic acid 217 in four high yield steps. Hydrolytic deprotection

with acidic methanol at reflux ( 6 h) affords methyl ester 258 (95 %). Esterification of secondary alcohol 258 with benzyloxyacetyl chloride in the presence of pyridine generates benzyloxyacetate 259 (89 %), the

reactant for the Claisen cyclization. Non-nucleophilic sterically hindered 92 base, lithium hexamethyldisilazide ( 2.1 eq), cyclizes benzyloxydiester 259 in THF (-78 °C) and furnishes benzyloxytetronic acid 26 0. Following cyclization benzyloxytetronic acid 260 undergoes deprotection using catalytic transfer hydrogenation [10 %Pd/C; cyclohexene (100 eq);

EtOH; 1 h]147 to furnish fS>4-hexyl-2-hydroxytetronic acid (217; 75 %).

The enantiomer (/?>4-hexyl-2-hydroxytetronic acid (216) is obtained from dioxolane protected ('RJ-malic acid (261) utilizing methodology developed for the (S)-enantiomer. Optical purity is determined by 1H

NMR analysis of the diastereomeric salts of benzyoxytetronic acids (S)- 217 and (R)-216 with ('R^methylbenzylamine. The asymmetric proton resonance signal observed at 500 MHz, for each of the pure (R)/(R) and

(S)/(R) diastereomeric salts appears as a doublet of a doublets. The racemate (R,S)/(R) provides the anticipated multiple! resonance signal.

BnOCH2COCI 2 5 7 95% c h 3o 6 h Pyr 89 % 258 93

Pd 10% C , EtOH 2 6 2 o. 75%

Unfortunately, reaction conditions similar to those employed for the

construction of hexyl ac/-reductone precursor 257 fail in the attempted synthesis of the a-hydroxyundecenoic acid 269, the precursor to the

desired nonene acAreductone. Thus, the organocuprous reagent derived

from model sp 2 iodide 268 does not undergo reaction with electrophilic

dioxolane tosylate.

CuLi 0.5 M 268 X = l 197 BuLi 196 X = Li

269

Triflates are known to solvolyze 10 5 to 106 times faster than tosylates due to the electron withdrawing effects of three fluorine atom s.169 Unfortunately, although triflate 232 was anticipated to be a good electrophile ,169 treatment with organocuprous derivative 234 does not provide desired dioxolane 269 as indicated by 1H NMR analysis. The ease of transfer of the organic group in organocuprous reagents

R-1R 2CUU is dependent upon the s-character of the functionalities 94 attached to copper. Decreasing transferability is observed with increasing

s-character. Thus, sp 3 carbons transfer faster than sp 2 carbons.

Acetylenic sp carbons are known not to transfer. This may explain why polyalkenyl transfer fails in the synthesis of intermediates 251, 269 and

290.

192

269

Method B: Successful Wlttlg reaction approach : S in ce s p 2 carbons are less nucleophilic and therefore are less easily transferred as organocuprous reagents, the Wittig approach represents an attractive alternative. The Wittig reaction involves use of an unknown five carbon aldehyde equivalent. Both isomers of 5-oxotetrahydrofuran-2- carboxylic acid (270) are commercially available and can serve as precursors for a-hydroxyglutaric acid semialdehyde 291. However reduction of acid 270 using either DIBAL-H (-78 °C) or UAIH 4 does not provide the target hemiacetal .170-171 Possibly, carboxylate anion generated during DIBAL-H and LiA!H 4 reduction interferes with lactone reduction.

A. DIBAL-H, -78 °C — ^

OO 270 291 95

Alternatively, reduction of butyl ester 271 using DIBAL-H (1.5

eq)171 furnishes desired hemiacetal 272. DCC-mediated esterification of

acid 270 with f-butyl alcohol in the presence of 4-pyrrolidinopyridine provides precursor ester 271 (85 %}172. The aldehydes 272 and 273 are obtained in a 2 : 1 ratio and a combined yield of 72 %. Even though f-

butyl esters are relatively more resistant to DIBAL-H reduction171, in case

of lactone 271 reduction to generate aldehyde 273 competes favourably with lactone ring reduction.

/-BuOH, 4-PP OfBu DCC, CHaCfe

270 ° 271 °

DIBAL-H + o ^ ? S r 0 2 : 1 O 272 273

Acid 270 is an expensive chiron; yields of only 48 % (for hemiacetal 272) early in a multi-step synthetic scheme are undesirable.

For these reasons relatively inexpensive enantiomers of malic acid were of interest. Anticipatedly, these commercially available chirons are convertible to the desired 5-carbon aldehyde equivalents 278 and 283.

Dioxolane aldehyde 274 is obtained by oxidation of alcohol 231 using either Collins reagent (54 %)174, pyridinium dichromate (62 %)175 or Ratcliffe reagent (81 %)176. Although all three methods furnish 96

aldehyde 274, Ratcliffe reagent provides a better yield on a 1-2 g scale,

but only poor yields at the 10 g level.

A. Collins oxidation B. Pyridinium dichromate

C. Ratcliffe oxidation

231 274

Alternatively, Rosenmund reduction of acid chloride 275 serves as a high yield source of aldehyde 274. DMF reacts with thionyl chloride forming a highly reactive chlorinating Vilsmeyer species. Thus, the

protected acid chloride 275 (~ 100 % crude yield) is obtained at room temperature by treating acid-labile dioxolane 230 with excess thionyl chloride and catalytic amount of DMF. Anhydrous conditions are important; traces of thionyl chloride are removed prior to reduction.

Hydrogenation of acid chloride 275 in refluxing xylenes using 5%

Pd/BaS 0 4 and thioquinanthrene poison furnishes aldehyde 274 (90 %).

Thioquinanthrene prevents overreduction to the corresponding alcohol.201 The reaction is monitored by trapping gaseous HCI in water and titrating with 5M sodium hydroxide solution. The Rosenmund reduction is clean and high yielding (90 %). Even though high temperatures (refluxing xylenes; 135 °C) are employed racemization of the stereocenter is not observed. Optical rotations obtained for aldehyde

274 synthesized using Rosenmund methodology or low temperature

Collins oxidation are the same. Aldehyde 274 requires homologation as the synthon 211 required for Wittig reaction has a 5-carbon backbone. 97

Homologation of aldehyde 274 with methoxymethyltriphenyl phosphine chloride (K-OfBu; THF; 0 °C) provides enol ether 276 {78 %).177-178

Pd/5% BaS04 XX 275 274

OCH3 CH3OCH(PPh)+3 Br

K-0©u. THF, 0°C

274 276

Enol ethers undergo hydrolysis with a variety of reagents including

saturated perchloric acid in ether, trimethylsilyl iodide in acetone, and acetic acid in water.171-179 These conditions fail for enol ether 276.

Rather, enol ether 276 is readily converted to the hemiacetal 278 using a two step hydrolytic protocol. Transketalization in acidic methanol at reflux produces hemiketal 277 (86 %) and hydrolysis of the hemiketal in 25 % aqueous acetic acid (10 h) generates hemiacetal 278 (90 %).

o ^ OCH3 CH3O H , H

277. R = - CH 3 276 AcOH : 5:1 278. R = - H 98

Preparation of a second 5-carbon aldehyde synthon 283 utilizes

Arndt-Eistert methodology. Reaction of acid chloride 275 with 2 eq of diazomethane produces diazomethylketone 279 (85 %). Arndt-Eistert rearrangement in the presence of benzyl alcohol and freshly prepared silver oxide furnishes the homologated benzyl ester 280 (72 %). This ester undergoes facile debenzylation to protected a-hydroxyglutaric acid

281 (95 %). Arndt-Eister rearrangement of diazom ethyl ketones in water provide the corresponding acids.180 Unfortunately, attempted formation of acid 281 in water only leads to decomposition. Rosenmund reduction of acid chloride 282 obtained from protected a-hydroxyglutaric acid 281 furnishes aldehyde 283 (85 %; 2 steps).

CHgNg, 2 op 275 ° x ° 0 279 O

PhCH2 OH, AgNOa X

dioxane

280. X = -CH2Ph Pd/ 1 0 % C. H:

One convergent approach for the synthesis of AA-CAM-R analogues involves construction of ac/-reductone aldehyde 288 to be 99 used with a variety of Wittig synthons. Aldehyde 268 is obtained in five steps from protected a-hydroxyglutaric acid semialdehyde 283. Chain elongation using [2-(1,3-dioxolan-2-ylmethyl)triphenylphosphonium bromide (THF; K-/OBu) furnishes alkene 284 (67 %).181 Deprotective transketalization (methanol; H 2 SO4 ) affords methyl ester 285 (95 %).

Esterification (benzyloxyacetyl chloride; Pyr; 1.5 eq) provides aci- reductone precursor benzyloxyacetate 286 (87 %). Benzyoxyacetate

286 undergoes cyclization (LiHMDA; 2.1 eq; -78 °C) affording vinylogous acid 287 (86 %). Hydrolysis in 25 % aqueous acetic acid generates aldehyde 288 (88 %). Unfortunately, this aldehyde is unstable; treatement with phosphorane 289, produced with either n-BuLi or K- fOBu, results in decomposition.

2 8 3 K-OlBu, THF

284 O OCH3

285. R = -H BnOCH2 COCI, Pyr 286. R = -COCH2OBn 100

o c h 3 LiHMDA O 286 HO OBn O 287

AcOH : HgO H O 3:1 HO OBn 288

289 O

CH2 P(Ph)3 + Br' HO OBn 289 = 306

Hemiacetal 278, synthesized in excellent yield, is a preferred source of 5-carbon aldehyde 211. Additionally, utilization of hemiacetal

278 circumvents use of highly sensitizing diazomethane in the synthesis of intermediate 304. The 15 carbon counterpart to Wittig precursor 278 is obtained by acetylenic condensation of bromide 239 with 3 eq of the

Grignard derived from 3-butyn-1-ol 294 in refluxing THF (Cul, 75 %).

Unstable triyne 300 is partially reduced to the all Z-triene 301 (92 %) using Lindlar catalyst and a calculated amount of hydrogen .156 Primary alcohol 301 is converted under mild conditions to homoallylic bromide

302 (90 %) using dibromotriphenylphosphorane (CH 3CN).

Triphenylphosphine salt 303 is prepared by heating bromide 302 and triphenylphosphine in the presence of a small amount of acetonitrile .182 The resulting phosphine salt 303 has an Rf of zero in 1:5 ethyl acetate : hexanes, whereas bromide 302 migrates with an Rf of 1.

The rate of salt formation is directly proportional to the reaction 101 temperature, but temperatures greater than 70 °C result in decomposition of the triene. At 70 °C, phosphine salt formation is complete within 72 h.

*OH. 3 ®q . EtMgBr 6 eq OH 239 Cul, THF , 2 2 h * * 300

Lindlar catalyst, Pd/5 % BaS0 4

H2 , Quinoline. 92 %

301. X = -OH P(Ph)3 .Br2l CH3CN ^ 302. X = -Br P(Ph>3 303. X = *P+(Ph)3 Br'

Numerous examples of Wittig Z-olefinations are available in the synthetic prostaglandin literature .93-184*194 Accordingly, all-(Z)- tetraalkenyl-a-hydroxymethyl ester 304 is obtained in 85 % yield by Wittig olefination of hemiacetal 278 in HMPA. The triphenylphosphorane is derived from triene 303 using n-BuLi. Two equivalents of phosphorane are required in order to prevent allylic deprotonation. Methyl a- hydroxyarachidonate 304 is the required Claisen precursor necessary for synthesis of AA-CAM-R 201. Esterification (benzyloxyacetyl chloride; pyridine) generates benzyloxyacetate 305 (85 %). Cyclization (lithium hexamethyldisilazide; 2.1 eq; -78 °C) provides benzyioxydiester 306. 102

303 , nBuLi 278 CH30 -60 °C, HMPA OR

304. R = - H BnOCHaCOCI, Pyr 305. R = - C O CH^Bn

Boron-trifluoride etherate-ethylmercaptan and lithium/ammonia have been reported to deprotect benzyloxyalkenes .171-195 Unfortunately, tetraalkene 306 decomposes on treatment with either reagent. The 1H

NMR spectrum for the BF 3 etherate/ EtSH deprotection product shows a partial reduction of downfield unsaturation accompanied by extraneous upfield peaks. This partial reduction likely occurs because of nucleophilic attack of the mercaptan on the skipped double bonds. The resulting mixture undergoes further undefined decomposition.

o 305

I LiHMDA [ 2 .1 eq

306 HO OBn

A. BF3 .OEt2, EtSH B. U/NH3 1 r

DECOMPOSITION 103

Whereas benzyl protected alkyl- and aryl-ac/-reductones are easily deprotected to the corresponding CAM-R analogues and preparation of hexyl ac/-reductone 217 and benzyl ac/-reductone 215 is successful, this protecting group is not easily removed in polyalkenyl systems such as 306. Methyl eicosaenoate 309, possessing one double bond, was viewed as a model compound needed to determine the appropriate protecting group required for assembling alkenyl ac/-reductones.

Accordingly, Wittig reaction of hemiacetal 278 with pentadecyltriphenyl phosphorane furnishes methyl eicosaenoate 309 (80 %).

Deprotection of p-methoxybenzyl (PMB) ethers is carried out under mild conditions, which do not decompose the unsaturated side-chain .171 Diester 310 (96 %), obtained by coupling secondary alcohol 309 with /> methoxybenzylglycolic acid (292), readily cyclizes to p- m ethoxy benzyloxytetronic acid 311 (84 %). Unfortunately, PMB ether

311 does not undergo deprotection using either dicyanodichloroquinone (DDQ) or eerie ammonium nitrate (CAN). a-Hydroxy-substituted PMB ethers are known to form hemiacetal intermediates such as 312 under DDQ deprotection conditions. Although, hemiacetal 312 is detected in the 1H NMR spectrum less than 30 % yields are obtained. Anticipatedly, hydroxyl group in tetronic acid 311, if protected, cannot interfere during deprotection by attacking the benzylcarbocation intermediate.

Accordingly, acetylation (AC 2O, Pyr) of ac/'-reductone 311 furnishes protected tetronic acid 313 (82 %). Unfortunately, even acetylated-PMB- aci-reductone 313 does not undergo deprotection with either DDQ or

CAN. 104

308 1. BuLi, THF , - 35 °C 2. HMPA 3. r \ 0.5 eq HO O COOCH3 278

c h 3o 3 OR o c h 3 309. R = - H 292 ,X~0.XT' HO *-► 310. R = - COCHj-OPhp-OCHa 4-PP. DCC

LiHMDA, 2.1 eq I'

HO OCH2 PhpOCH3 311 o c h 3

o c h 3

In our laboratories, methyl ethers of the type 316 previously have been used as sources of the corresponding ac/-reductones . 197-198

Accordingly, sequential esterification (methoxyacetyl chloride; 85 %) of methyl eicosaenoate (309) and cyclization (LiHMDA; -78 °C) affords methoxytetronic acid 315 (82 %). Acetylation furnishes P/'s-protected

316 (89 %). Desired 2-hydroxytetronic acid 203 is obtained in 45 % yield following deprotection with 5 mol BBr 3 in CH2 CI2 . Thus, less than satisfactory yields are obtained using this methodology for the synthesis of act-reductone 203, and when employed for the synthesis of AA-CAM-R

201 only decomposition occurred. 106

c h 3o '

£>r 309. R = - H CH-jOO-feCOCI, Pyr 314. R = *C0CH20CH3

^LiHMDA

R2O OR1

315. R, = -OCH3 , R2 = -H CH3COCIp Pyr 316. Rt = -OCH3, R 2 = -COCH3

| BBr3

HO OH 203

Cyclization of methoxyacetate 317 using 2.1 eq of LiHMDA provides tetronic acid 318 (82 %). Acetylation furnishes protected aci- reductone 319 (88 %). 1H NMR analysis reveals bromination of the side chain to take place during attempted deprotection of acetate 319 using 5 mol BBr3 in CH 2CI2- Whereas neither deprotection nor decompostion is observed with 2 mol BBr3 solutions, use of greater than 2 mol BBr3 solutions invariably results in simultaeous deprotection and bromination of the skipped tetraene 319. Trimethylsilyl iodide, another methyl ether 107

deprotecting reagent also iodinates tetraene 319.171 Clearly, use of a

methyl ether protecting group does not furnish polyalkenyl aci-

reductones.

c h 3o Or 304. R = - H CH3OCH2COCI, Pyr|~^~ 317. R =-C0CH20CH 3

JtJHMDA

R20 OR1

318. R, = -OCHs, R2 = -H CH3COCI, Pyr[~^~ 319. R1 = -OCH3i R2 = -COCH3

| BBr3

DECOMPOSITION

A third protecting group investigated, the f-BDMS group,

undergoes facile deprotection in the presence of mild reagents such as tetrabutylammonium flouride and acetic acid .171 f-Butyldimethyl- silyloxyacetate 320 is obtained in 75 % yield by esterifying methyl arachidonate (304) with f-butyldimethylsilyloxyacetyl chloride . 199

Unfortunately, attempts to cyclize diester 320 failed. Conceivably, steric hinderance between the f-BDMS group and bulky base (LiHMDA; LDA) 108 serves as explaination for the observed unreactivity, but surprisingly, a sterically non-hindered base (n-BuLi) also fails in the cycli 2ation reaction.

A fourth protecting group investigated is acid-labile trifluoroacetate.

Trifluorotriester 321 is obtained in good yields (no starting material detected by 1H and 13C NMR) but is extremely labile. During attempted purification, this compound readily undergoes deprotection to glycolate

324 on silica g e l. Acid labile methoxymethyl (MOM) and tetrahydropyran

(THP) ethers were also investigated as protecting groups in the synthesis of AA-CAM-R. MOM 322 (75 %) and THP ethers 323 (72 %) are obtained from methyl arachidonate (304) in the presence of DCC. However, like f-BDMS diester 320, these derivatives do not undergo cyclization to ac/-reductones 326 and 327, respectively, using either

LiHMDA or LDA.

o 320. R = - TBDMS 321. R = -COCF3 322. R = - MOM 323. R = - THP 324. R = - H

HO OR 325. R =-TBDMS 326. R = -MOM 327. R = -THP 109

Subsequently, the pivaloyl protecting group was studied. Pivaloylacetoxy triester 328 is obtained in 74 % yield from methyl

arachidonate (328), but cyclization to ac/-reductone 329 takes place in only 56 % yield. A common feature associated with successful cyclization is a methyl or methylene x-system bonded to the hydroxyl group of glycolate 324. Bonding the gylcolate oxygen to silicon (f-BDMS), a carbonyl group (pivaloyl) or methoxy functionality (MOM, THP) results in derivatives which either do not undergo cyclization or undergo such chemistry in low yields. These observations provided the impetus to investigate use of the allyl protecting function. In fact, ally! ethers are widely used protecting groups in carbohydrate chemistry and are hydrolytically deprotected following isomerization to the 1-propenyl ether .202' 206

c h 3o

o 328. R=-COC(CH3)3

LiHMDA. 2.1 eq

HO OR

,----- 329. R = -COC(CH3)3 AcOH : HzO [ ^ 2 0 1 . R = -H 110

Accordingly, allyoxyacetic acid (330), prepared in 75 % yield from chloroacetic acid and sodium allyloxide ,200 undergoes coupling with methyl arachidonate 3 0 4 in the presence of DCC and 4- pyrrolidinopyridine in CH 2CI2. Allyloxyacetate 331, formed in 92 % yield, readily cyclizes in the presence of 2.1 eq of LiHMDA (-78 °C) to furnish acAreductone 332 in 89 % yield. Although no precedence exists for the use of allyl ether in non-aromatic polyolefinic systems, and a I lyoxy tetronic acid 332 decomposes in the presence of Wilkinson's catalyst (3h reflux ) , 171 isomerization to 1-propenyl ether 333 is effected in 79 % yields under mild conditions employing an iridium catalyst. This catalyst is prepared by first treating 20 mol % [Bis(methyldiphenylphosphine)](1,5- cyclooctadiene)iridium(l) hexafluorophosphate 207 with hydrogen (5 min) in freshly distilled peroxide free tetrahydrofuran .208 After 5 min the red catalyst turns colorless and the hydrogen is evacuated .208 Allyl ether

332 is introduced and isomerization is complete in 2-3 h.

The cyclooctadiene ligand co-ordinates with two iridium cAorbitals in complex 293. Based on existing knowledge, hydrogenation of this iridium catalyst, reduces cyclooctadiene and the cAorbitals of iridium no longer complex the ligand. These free d-orbitals now form an oxygen- directed n-allyl complex selectively with the allyl ether of tetronic acid

332. Following formation of the n-allyl complex, iridium catalyzes a 1,3- hydride shift to generate propenyloxytetronic acid 333. Hydrolysis of this enol ether using 50 % aqueous acetic acid (15 min) furnishes AA-CAM-R

201 (95 %) in 99 % e.e.. Optical purity can be assessed by 1H NMR 111 analysis (500 MHz) of the diastereomeric allyloxytetronic acid salts ($)•

201 and (fl )-200 with ('R^methylbenzylamine.

6 h o 304

DCC , 4 - PP, CH2CI2

92% O

O LiHMDA, 2.1 eq ,-78°C 89 %

332 HO O ------112

AcOH : H 2O 1:1 ,15 min 95% f

HO OH

The allyloxy Claisen condensation also represents a marked improvement for the construction of alkene-CAM-R analogue 203, when compared to the one previously described route using a methyl protecting group. Allyoxydiester 334 (93 %) is obtained from secondary alcohol

309 using DCC coupling. Cyclization using 2.1 eq of LiHMDA furnishes ac/-reductone 335 (89 %). Isomerization with [Bis(methyldiphenylphosphine)](1,5-cyclooctadiene)iridium(l) hexafluorophosphate (20 mol %) followed by hydrolysis produces 2-hydroxytetronic acid 203 in 74 % overall yields ( 2 steps). 309. FU-H HO' 4-PP, DCC 334. R = - COCH2OCH2CH=CH2

LiHMDA. 2.1eq, -78 °C

HO OR

335. R=-CH2CH=CH2 (COD) Ir (P(CH3) Phg + PF6 ' 336. R = * CH=CHCH3 AcOH : hfeO 1 : 1 203. R = - H

For the completely saturated acAreductone 205 use of the allyloxy

protecting group is not required. The saturated acAreductone is obtained from benzyloxyacetate diester 33 7 prepared in 87 % yield by esterification of eicosaenoate 309. Cyclization (2.1 eq of LiHMDA) provides tetronic acid 338 (84 %). One pot reduction and deprotection using hydrogen over 10 % Pd/C furnishes saturated CAM-R 205 (72 %) in > 99 %e.e. (fl)-AA-CAM-R 200, (F?)-octadecenyl-CAM-R 202 and (ff)- octadecanyl-CAM-R 204 have also been prepared in > 99 % e.e. from

(R)-malic acid (339) utilizing the reactions developed for the respective

(S)-enantiomers. 114

c h 3o OR 309. R = - H BnOCHaCOCI, Pyr £ 337. R = - COCH2OBn

^LiHMDA, 2.1 eq

HO OBn 336 P d ,1 0 %C,H2 , MeOH

HO OH 205

In conclusion, a general method for the asymmetric synthesis of optically pure 4-alkenyl- and 4-polyalkenyl-2-hydroxytetronic acids using allyloxy Claisen methodology has been developed. This high yielding synthesis (62 %; 4 steps; > 99 % e.e.) is applicable for the preparation of skipped and isolated double bond-containing ac/-reduct ones. It is likely that this method will be useful for the synthesis of conjugated polyalkenyl systems as well since these are anticipatedly more stable when compared to skipped tetraenes. Additionally, two Wittig synthons 274 and 278 possessing either a four carbon or a five carbon backbone provide a convergent synthetic route for numerous alkenyl-, polyalkenyl- and alkyl-oc-hydroxymethyl ester Claisen condensation precursors to the 115

desired ac/-reductone targets. In all cases, greater than 99 % e.e. is obtained using this methodology. CHAPTER V

EXPERIMENTAL SECTION

Melting points were determined in open capillaries with a Thomas- Hoover Uni-Melt Apparatus and are uncorrected. Infrared spectra were

recorded by a Laser Precision Analytical RFX-FTIR spectrometer (model TSI-400). Nuclear magnetic resonance spectra were obtained with either an IBM-Bruker model NR-250, 270 or 500 FT NMR spectrometer. TMS

(CDCI3, acetone-afe, CD 3OD) was used as internal standard. Chemical shifts were reported on the 5 scale with peak multiplicities: s, singlet; d, doublet; dd, doublet of doublets; ddd, doublet of doublet of doublets; t, triplet; dt, doublet of triplets; q, quartet; m, multiplet. THF was distilled from

Na/Benzophenone ketyl and CH 2CI2 was dried over P 2O 5 . Optical rotations were taken on a Perkin-Elmer model 241 polarimeter using a 10 cm long 1 mL cell. Mass spectra were aquired with either a Kratos MS25RFA or a VG 70-250S mass spectrometer. Elemental analyses were performed by Galbraith Laboratories Inc. Knoxville, TN.

116 117

Methyl (S)-3-phenyllactate (221). In a 250 mL round bottom flask fitted with a reflux condenser, 3.0 g (18.07 mmol) of S-phenyllactic acid (220) in 225 mL of methanol containing 4 drops of conc. sulfuric acid was heated at reflux for 7 h. The reaction mixture was cooled, 0.6 g of sodium bicarbonate was added, and the methanol was evaporated under reduced pressure. The residue was taken up in 200 mL of ether, and the ether layer was washed with 2 x 75 mL of water, 2 x 100 mL of saturated sodium bicarbonate solution and 2 x 75 mL of brine, dried

(Na2 S0 4 ) and evaporated under reduced pressure to provide 3.12 g of crude product which was purified over silica gel using ethyl acetate : hexanes (1 : 5) to furnish 2.92 g (90 %) of white crystalline solid: mp 46-47

°C; IR (KBr, cm '1) 3473 (br), 3029, 2954, 1739, 1496, 1454; 1H NMR

(CDCI3 ) 5 (ppm) 7.34-7.19 (m, 5H), 4.50-4.43 (m, 1H), 3.77 (s, 3H), 3.14

(dd, J = 4.4, 13.9, 1H), 2.96 (dd, J= 6 .8, 13.9, 1H), 2.72 (d, J= 6.2, 1H) ;

HRMS calcd for C 10H 12O3 (M+) 180.0786, found 180.0786

Methyl (R)-3-phenyllactate (226) was prepared in sim ilar yield as for the S-isomer 221 from R-220 : mp 47 °C; IR (KBr, cm*1) 3479

(br), 3029, 2954, 1739, 1496, 1454; 1H NMR (CDCI 3 ) 5 (ppm) 7.34-7.19

(m, 5H), 4.50-4.43 (m, 1H), 3.78 (s, 3H), 3.14 (dd, 4.3, 13.9, 1H), 2.97

(dd, 6 .8, 13.9, 1H), 2.72 (d, J = 6 .2 , 1H) ; HRMS calcd fo r C io H l 2 0 3

(M+) 180.0786, found 180.0795 118

Methyl (S)-3-phenyllactate, (benzyloxyacetate) (222).T o a two-necked, flame-dried 250 mL round bottom flask under argon was added 2.92 g (16.2 mmol) of methyl (S)-3-phenyllactate (221) in 80 mL of anhydrous CH 2CI2- Benzyloxyacetyl chloride (4.492 g, 24.33 mmol) was added with stirring and the solution was cooled to 0 °C in an ice-salt bath. Pyridine (1.93 g, 24.33 mmol) was added dropwise and reaction contents were stirred for 30 min at 0 °C, warmed to room temperature, and stirred for an additional 3 h. The reaction was quenched with 40 mL of ice-water and 75 mL of CH 2CI2 was added. Following stirring overnight the CH 2CI2 layer was washed with 3 x 75 mL of 10% aqueous HCI solution, 3 x 100 mL of saturated sodium bicarbonate solution and 2 x 100 mL of brine, dried (Na 2 S0 4 ) and concentrated under reduced pressure. The crude product was purified over silica gel using ethyl acetate : hexanes (1:5) to yield 4.68 g (88 %) of white crystalline solid : mp 51-52 °C; [a]o 25 -14.3°

(C = 2 .4 .CH3 0 H): IR (KBr, cm '1) 2948, 2886, 1766, 1745, 1455, 1434 ; 1H

NMR (CDCI3 ) 6 7.36-7.19 (m, 10H), 5.36 (dd, J * 4.5, 8.7, 1H), 4.57 (s,

2H), 4.19 (d, J= 16.7, 1H), 4.09 (d, J= 16.7, 1H), 3.74 (s, 3H), 3.23 (dd, J =

4.5, 14.3, 1H), 3.11 (dd, J= 8.7, 14.3, 1H) ; HRMS calcd for C 19H2 0 O5

(M+) 328.1311, found 328.1269

Methyl (/?)-3-phenyllactate, (benzyloxyacetate) (227) was prepared in similar yield as for the S-isomer 222 from R-226 : mp 49-50

°C; [a]D25 14.7° (c = 0 .4 .CH3OH); IR (KBr, cm*1) 2948, 2886, 1766, 1745, 119

1455, 1434 ; 1H NMR (CDCI3 ) S (ppm) 7.35-7.19 (m, 10H), 5.36 (dd, J =

4.5, 8.7, 1H), 4.57 (s, 2 H), 4.19 (d, J = 16.7, 1H), 4.09 (d. J= 16.7, 1H),

3.74 (s, 3H), 3.23 (dd, J = 4.5, 14.3, 1H), 3.11 (dd, J = 8.7, 14.3, 1H);

HRMS calcd for C 19H20O5 (M+) 328.1311, found 328.1303

(S)-5-Benzyl-3-(benzyloxy)-4-hydroxy-2(5H)-furanone (225). To a flame-dried three-necked argon purged 100 mL flask equipped with a low temperature thermometer was added 2.38 g (14.73 mmol) of hexamethyldisilazide in 35 mL of anhydrous tetrahydrofuran.

The solution was cooled to -25 °C, and 5.9 mL of a 2.5 M n-BuLi (14.73 mmol) solution in hexanes was added dropwise while maintaining the temperature below -15 °C. Following the addition, the reaction was stirred and held between -3 and -5 °C for 45 min and cooled to -78 °C. Benzyloxyacetyl diester 222 (2.3 g, 7.01 mmol) in 10 mL of anhydrous tetrahydrofuran was added dropwise. The reaction was stirred for 75 min and quenched at -78 °C with 25 mL of precooled 10% aqueous HCI solution. After warming to room temperature, the aqueous layer was extracted with 3 x 80 mL of ether and the combined organic layers washed with 3 x 100 mL of brine and dried (Na 2 SC>4 ). The ether was evaporated under reduced pressure to provide 1.98 g of crude white solid which was recrystallized using ether/petroleum ether to give 1.82 g (85%) of white crystalline solid : mp 181-182 °C; [a]o 25 -57.1° (c = 0 .9 .CH3OH);

IR (KBr, cm-1) 3031 (br), 2719, 1743, 1662, 1454 ; 1H NMR (CD 3COCD3 )

6 (ppm) 7.34-7.29 (m, 5H), 7.29-7.21 (m, 5H), 4.96 (dd, J = 3.7, 6.5, 1H), 120

4.87 (d, 18.2, 1H), 4.79 (d, J= 18.2, 1H), 3.26 (dd, J= 3.7, 14.5, 1H), 2.88

(dd, J - 6.5, 14.5, 1H); 13C NMR (CD3 COCD 3 ) 5 169.0, 160.2, 138.2,

136.4, 130.6 ( 2C), 129.2 (2 C), 129.1 (2C), 129.0 (2C), 128.8, 127.6, 121.9,

76.0, 73.6, 38.6; HRMS calcd for C 1 8H 1 6 O 4 (M+) 296.1048, found

296.1045; Anal. Calcd for C 18H 16 O4 : C, 72.96; H 5.44. Found : C, 72.59;

H 5.53.

(ff)-5-Benzyl-3-(benzyloxy)-4-hydroxy-2(5W)-furanone (228) was prepared in similar yield as for S-225 from ft-227.: mp 182-

183 °C; [a]D25 57.8° (c = 0.4, CH 3OH); IR (KBr, cm‘ 1) 3029(br), 2717,

1743, 1660, 1454 ; 1H NMR (CD 3 COCD3 ) S (ppm) 7.34-7.29 (m, 5H),

7.29-7.21 (m, 5H), 4.96 (dd, J= 3.7, 6.5, 1H), 4.86 (d, 18.4, 1H), 4.79 (d, J

= 18.4, 1H), 3.26 (dd, 3.7, 14.4, 1H), 2.88 (dd, J= 6.5, 14.4, 1H); HRMS calcd for C 18H 16 O4 (M+) 296.1048. found 296.1045

(S)-5-Benzyl-3,4-dlhydroxy-2(5H)-furanone(215). In a 250 mL argon-flushed hydrogenation bottle was suspended 0.2 g of palladium on 10% carbon in 10 mL methanol. To this suspension was added 2.0 g

(6.76 mmol) of tetronic acid 215 and 25 mL methanol. The mixture was shaken at room temperature under hydrogen (35 psi) and monitored by

TLC (about 5-6h). After filtration (Celite pad) the filtrate was evaporated under reduced pressure, and the residue was purified by recrystallization from acetone/hexanes to furnish 1.25 g (90%) of a white crystalline solid : 121 mp 142-144 °C; [a ]D25 -40.2 o (c = 2 .1.CH3 0 H); IR (KBr, c m '1) 3334(br),

1762, 1681, 1455, 1319 ; 1H NMR (CD 3 COCD3 ) 5 (ppm) 7.28-7.21 (m,

5H), 4.93 (dd, J = 3.5, 6.7, 1H), 3.29 (dd, J = 3.5, 14.5, 1H), 2.88 (dd, J =

6.7, 14.5, 1H); HRMS calcd for C 1 1 H 1 0 O 4 (M+) 206.0579, found

206.0583. Anal. Calcd for C 11H 10O 4 ; C, 64.08, H, 4.82. Found : C,

63.99; H, 4.89.

(ff)-5-8enzyl-3,4»dihydroxy-2(5H)>furanone (214) was prepared in similar yield as for S-215 from ft-228 : mp 142-144 ° C ; [a]o25

40.8 o (c=0 .8 ,CH3OH); IR (KBr, cm-t) 3336(br), 1762, 1679, 1455, 1319;

1H NMR (CD3COCD3 ) 6 (ppm) 7.28-7.22 (m, 5H), 4.93 (dd, J= 3.6, 6.7,

1H), 3.29 (dd, J = 3 .6 , 14.5, 1H), 2.88 (dd, J = 6 .7 , 14.5, 1H); HRMS calcd for C 1 1 H 1 0 O 4 (M+) 206.0579, found 206.0575; Anal. Calcd for

C 11H10O4 : C, 64.08; H, 4.82. Found: C, 63.93 H. 4.86.

(S)-5-(2-Hydroxyethyl)-2,2-dlmethyl-1,3-dioxolan-4-one

(231)149-151-152. In a 500 mL flame-dried round bottom flask fitted with a septum and under argon was placed 11.0 g (63.2 mmol) of dioxolane

230 dissolved in 200 mL of anhydrous tetrahydrofuran. The reaction was cooled (-20 °C to -30 °C) and 70 mL of 1M borane-tetrahydrofuran complex was added dropwise over 2h. Following addition the reaction vessel was placed in a refrigerator at 4 °C for 11 h, warmed to room temperature, stirred at room temperature for 9 h and chromatographed 122

(silica gel) using acetone as eluant. Following evaporation under

reduced pressure, the residue was chromatographed as before to

generate 9.1 g (90%) of the alcohol as a colorless unstable liquid which

was dried under reduced pressure and utilized as such in the next

reaction. : IR (neat, cm '1) 3480(br), 2994, 2940, 2888, 1791, 1220 ; 1H

NMR (CDCI3 ) 5 (ppm) 4.58 (dd, J= 5.1, 7.0, 1H), 3.91-3.79 (m, 2 H), 2.28-

2.13 (m, 1H), 2.13-1.97 (m, 1H), 1.64 (s, 3H), 1.57 (s, 3H)

(R)-5-(2-Hydroxyethyl)-2,2-dlmethyl-1 t3-dioxolan-4-one

(340) was prepared in similar yield as for S-231 from ft-339 .: IR (neat,

cm-1) 3453(br), 2994, 2940, 2888, 1791, 1220 ; 1H NMR (CDCI3 ) 8 4.58

(dd, J= 5.1, 7.0, 1H), 3.91-3.79 (m, 2H), 2.28-2.13 (m, 1H), 2.12-1.97 (m,

1H), 1.64 (s, 3H), 1.57 (s, 3H)

(S)-5-(2-Hydroxyethyl)-2,2-dimethyl-1,3-dioxolan-4-one

p-toluenesulfonate (232). Dried alcohol 231 (9.0 g, 59.2 mmol) was dissolved in lOOmL of anhydrous pyridine under argon and cooled to -4

°C. To this solution maintained at approx. 0 °C was added dropwise

11.3g (59.2 mmol) of p-toluenesulfonyl chloride dissolved in 100 mL of

pyridine. Following addition, the mixture was placed in the refrigerator (0-

4 °C) overnight. Water (150 mL) was added, and the aqueous mixture extracted with 4 x 200 mL of ether. The ether layers were combined, washed with 3 x 150 mL of water, 3 x 150 mL of saturated copper sulfate 123 solution (until no dark blue color remained), 2 x 100 mL of water, 3 x 150 mL of brine, dried (Na2S04), filtered, and evaporated under reduced pressure. The crude tosylate 232 was purified by column chromatography using ethyl acetate : hexanes (1 : 1) to provide 16.7 g (95

%) of white solid : mp 48-49 °C; [a ]D25 -3.7 o (c = 0.3, CH 3 OH); IR (KBr, cm '1) 2989, 1785, 1390, 1357, 1278 ; 1H NMR (CDCI 3) 5 (ppm) 7.80 (d, J

= 8.3, 2 H), 7.36 (d, J= 8.3, 2H), 4.43 (dd, 4.4, 8.1, 1H), 4.27-4.14 (m, 2H), 2.46 (s, 3H), 2.30-2.17 (m, 1H), 2.09-1.92 (m, 1H), 1.58 (s, 3H), 1.51

(s, 3H); HRMS calcd for C 14H 18O6 S (M+) 314.0824, found 314.0817

(R)-5-(2-Hydroxyethyl)-2,2-dimethyM,3-dioxoIan-4-one p-toluenesulfonate (341) was prepared in similar yield as for S-232 from ff-340.: mp 52-53 °C; [a ]D25 3.8 0 (c = 1.1, CH3OH); IR (KBr, cm -1)

2992, 1785, 1388, 1357, 1278 ; 1H NMR (CDCI 3 ) 5 7.80 (d, J = 8.3, 2H),

7.36 (d, J= 8.3, 2H), 4.43 (dd, J= 4.4, 8.1, 1H), 4.27-4.14 (m, 2H), 2.46 (s,

3H), 2.30-2.18 (m, 1H), 2.09-1.93 (m, 1H), 1.58 (s, 3H), 1.51 (s, 3H);

HRMS calcd for C 14H 18O6 S (M+) 314.0824, found 314.0830

(S)-5-Hexyl-2,2-dimethyl-1,3-dioxolan-4-one (257). To a flame-dried 500 mL three-necked round bottom flask under nitrogen containing a suspension of 4.85 g (25.48 mmol) of cuprous iodide in 200 mL of anhydrous ether and held at -30 °C was added dropwise 31.85 mL of 1.6 M n-BuLi (50.96 mmol) in hexanes. The dark red-brown solution 124

was stirred at -30 °C to -40 °C for 2 h and cooled to -78 o q .168

Dioxolane tosylate 232 (4.0 g, 12.74 mmol), dissolved in 30 mL of anhydrous ether and 10 mL of anhydrous tetrahydrofuran, was added dropwise while maintaining the temperature below -70 °C. The reaction

mixture was stirred for 18 h at -78 °C. Following completion (TLC

monitoring), the reaction was warmed to -10 °C and quenched by the

addition of 125 mL of precooled saturated ammonium chloride solution.

Ether (100 mL) was added and the mixture filtered over Ceiite. The aqueous phase was extracted with 3 x 175 mL ether. The combined ether extract was washed with 2 x 125 mL of saturated ammonium chloride solution, 1 x 75 mL of water and 2 x 100 mL of brine, dried (Na2S04), filtered and concentrated under reduced pressure. The product was purified (silica gel) using ethyl acetate : hexanes (1 : 5) as eluant to give 2.44 g (95 %) of colorless o il: [a]o 25 0.6 ° (c = 1.9, CH 3OH);

IR (neat, cm-1) 2958, 2933, 2861, 1797 ; 1H NMR (CDCI 3 ) S (ppm) 4.39

(dd, J - 4.4, 7.1, 1H), 1.95-1.81 (m, 1H), 1.81-1.62 (m, 1H), 1.61 (s, 3H),

1.54 (s, 3H), 1.49-1.38 (m, 2H), 1.35-1.29 (m, 6 H), 0.89 (t, J = 6 .6 , 3H) ; 13

C NMR (CDCI3) 8 (ppm) 173.3, 110.2 , 74.2, 31.7, 31.6, 28.8, 25.8, 24.8,

22.5, 13.9; HRMS calcd for C 11H20O3 (M+) 200.1412, found 200.1422

(/?)-5-Hexyl-2,2-dimethyM,3-dioxolan-4-one (342) w a s prepared in similar yield as for S -257 from R -341.: [a ]o 25 2.9 0

(c = 1.3 ,CH 3 0 H); IR (neat, cm-1) 2958, 2931, 2859, 1793 ; 1H NMR

(CDCI3 ) (ppm) 8 4.39 (dd, J = 4.4, 7.1, 1H), 1.95-1.81 (m, 1H), 1.81-1.62 125

(m, 1H), 1.61 (s. 3H), 1.54 (s, 3H), 1.49-1.38 (m, 2H), 1.236-1.28 (m, 6 H),

0.89 (t, J = 6 .6 , 3H) ; HRMS calcd for C 11H20 O3 (M+) 200.1412, found

200.1431

Methyl (S)-2-hydroxyoctanoate (258). In a 250 mL round bottom flask fitted with a reflux condenser was placed 2.4 g (12 mmol) of hexyl dioxolane 257 in 150 mL of methanol containing 2 drops of conc. sulfuric acid. Following heating at reflux for 6 h, the reaction mixture was cooled, and 0.5 g of sodium bicarbonate was added. The solvent was evaporated under reduced pressure and the residue dissolved in 200 mL of CH2CI2- The solution was washed with 2 x 75 mL of water.2 x 100 mL of saturated sodium bicarbonate solution, and 2 x 75 mL of brine, and dried (Na 2 S0 4 ). The CH 2CI2 solvent was evaporated under reduced pressure to provide 2.07 g of crude product which was purified over silica gel using ethyl acetate : hexanes (1 : 5) to furnish 2.0 g (96 %) of pale yellow oil : [a]D25 -2.9 0 (c = 0.8, CH3OH); IR (neat, cm’ 1) 3475 (br), 2925,

2857, 1739; 1H NMR (CDCI 3 ) 6 (ppm) 4.19 (m, 1H), 3.79 (s, 3H), 2.73 (d,

J= 4.8, 1H), 1.89-1.75 (m, 1H), 1.72-1.56 (m, 1H), 1.55-1.39 (m, 2H), 1.39-

1.28 (m, 6 H), 0.88 (t, J * 6 .6 , 3H) ; HRMS calcd for C 9 H 1 8O 3 (M+)

174.1255, found 174.1255.

Methyl (/?)-2-hydroxyoctanoate(343) was prepared in similar yield as for S-258 from R-342.: [a]D25 2.7 0 (c = 0.3, CH3 OH); IR (neat, 126 cm-1) 3496(br), 2929, 2859, 1749 ; 1H NMR (CDCI 3 ) 5 (ppm) 4.19 (m,

1H), 3.79 (s, 3H), 2.73 (d, J= 4.8, 1H), 1.89-1.75 (m, 1H), 1.72-1.57 (m,

1H), 1.55-1.39 (m, 2 H), 1.39-1.28 (m, 6 H), 0.88 (t, J= 6 .6 , 3H) ; HRMS calcd for C 9H 18O3 (M+) 174.1255, found 174.1256.

Methyl ($)>2-hydroxyoctanoate, (benzyloxy)acetate

(259). In a two-necked, flame-dried 250 mL round bottom flask under argon was added 4.28 g (25.0 mmol) of methyl 2-hydroxyoctanoate 258 in 100 mL of anhydrous CH 2CI2. Benzyloxyacetyl chloride (6.82g; 36.9 mmol) was added and the mixture was cooled to 0 °C in an ice-salt bath. Pyridine (2.92 g, 36.9 mmol) was added dropwise. The mixture was stirred for 30 min at 0 °C, warmed to room temperature, stirred for an additional 3 h, and quenched with 30 mL of ice-water. An additional 50 mL of CH2CI2 was added. Following stirring overnight, the CH 2CI2 layer was separated and extracted with 3 x 50 mL of 10% aqueous HCI solution, 3 x 75 mL of saturated sodium bicarbonate solution, and 2 x 100 mL of brine, dried (Na 2SC>4) and concentrated under reduced pressure.

The residue was purified over silica gel ethyl acetate : hexanes (1 : 6 ) as eluant to give 6.7 g (88 %) of pale yellow liquid : [

CH 3 OH); IR (neat, cm-1) 2927, 2859, 1735, 1455, 1438 ; 1H NMR

(CDCI3 ) 6 (ppm) 7.36-7.29 (m, 5H), 5.11 (t, J= 6.5, 1H), 4.66 (s, 2H), 4.23

(d, J= 16.7, 1H), 4.17 (d, J= 16.7, 1H). 3.75 (s, 3H). 1.87-1.72 (m, 2H),

1.42-1.15 (m, 8H), 0.87 (t, J = 6.6, 3H) ; HRMS calcd for C 18H2 6 O5 (M+)

322.1780, found 322.1760. 127

Methyl (R)-2-hydroxyoctanoate, (benzyloxy)acetate (344)

was prepared in similar yield as for 5-259 from fl-343.: [o i] d 2 5 10.8 0

(CsI.I.CHaO H); IR (KBr, cm-1) 2925, 2857, 1733, 1455, 1436 ; 1H NMR

(CDCI3 ) 8 (ppm) 7.36-7.29 (m, 5H), 5.10 (t, J s 6.5, 1H), 4.65 (s, 2H), 4.23

(d, J = 16.7, 1H), 4.17 (d, J = 16.7, 1H), 3.74 (s, 3H), 1.87-1.72 (m, 2H),

1.42-1.15 (m, 8H), 0.87 (t, J = 6.6, 3H) ; HRMS calcd forCi8H26C>5 (M+)

322.1780, found 322.1782.

(S)-3-(Benzyloxy)-5-hexyl-4-hydroxy-2(5H)-furanone (260). In a flame-dried three-necked argon purged 100 mL flask equipped with a low temperature thermometer was added 2.1 g (13.04 mmol) of hexamethyldisilazide in 35 mL of anhydrous tetrahydrofuran . Following cooling to -25 °C, 5.3 mL of a 2.5 M n-BuLi in hexanes (13.04 mmol) was added dropwise and with stirring while maintaining the temperature below -15 °C. The mixture was stirred and held between -3 °C and -5 °C for 45 min and cooled to -78 °C. Benzyloxyacetyldiester 259 (2.0 g; 6.21 mmol) in 10 mL of anhydrous tetrahydrofuran was added dropwise. The reaction was stirred for 75 min and quenched at -78 °C with 30 mL of precooled 10% aqueous HCI solution. After warming to room temperature the product was extracted with 3 x 80 mL of ether, and the organic layer was washed with 3 x 100 mL brine, dried (Na 2 S0 4 ) and evaporated under reduced pressure to give 1.65 g of crude white product which was recrystallized using ether/petroleum ether to furnish 1.53 g 128

(85%) of white crystals : mp 74-75 °C; M o 25 -18.9 o (c = 2 .0 , CH3OH); IR

(KBr, cm '1) 3035(br), 2950, 2921, 1735, 1646, 1465; 1H NMR

(CD3COCD3 ) 5 (ppm) 7.43-7.29 (m, 5H), 5.06 (d, 16.1, 1H), 5.01 (d, J =

16.1, 1H), 4.69 (dd, J= 3.7, 7.0, 1H), 1.90-1.82 (m, 1H), 1.61-1.45 (m, 1H),

1.38-1.15 (m, 8H), 0.87 (t, J = 6 .8, 3H); 13C NMR (CD3 COCD3 ) 5 (ppm)

171.6, 169.4, 161.2, 138.3, 129.2 (2C), 129.1 (2C), 128.9, 120.9, 75.9,

73.4, 32.6, 32.4, 24.4, 23.2, 14.3 ; HRMS calcd for C 1 7 H2 2 O 4 (M+)

290.1518, found 290.1538. Anal calcd for C 17 H2 2 O 4 C 70.36, H 7.50; found C 70.32, H 7.64.

(/?)-3-(Benzyloxy)-5-hexyl-4-hydroxy-2(5H)-furanone

(345) was prepared in similar yield as for the S-isomer 260 from R-344

: mp 86-87 °C; [a ]D25 18.8 ° (c = 0.9, CH3 OH); IR (KBr, cm’ 1) 3035(br),

2950, 2921, 1735, 1646, 1465 ;1H NMR (CD 3 COCD3 ) S (ppm) 7.43-7.29

(m, 5H), 5.07 (d, 16.2, 1H), 5.01 (d, J = 16.2, 1H), 4.69 (dd, J = 3.7, 7.0,

1H), 1.90-1.82 (m, 1H), 1.61-1.45 (m, 1H), 1.38-1.15 (m, 8 H), 0.87 (t, J = 6.7, 3H); HRMS calcd forCi7H2204 (M+) 290.1518, found 290.1505

(S)-5-Hexyl-3,4-dlhydroxy-2(5H)-furanone (217). In a two-necked 100 mL round bottom flask was added under argon 0.7g

(2.41 mmol) of benzyloxy-2-hydroxytetronic acid 260, 0.7 g of 10% Pd/C, and 4.96 g (60.35 mmol) of cyclohexene in 50 mL of absolute ethanol.

The mixture was stirred and heated at reflux, filtered (Celite) and the 129 solvent removed under reduced pressure. The residue was recrystallized from acetone/hexanes to provide 0.362g (75%) of white solid : mp 100-

101 °C; [a ]D25 -14.1 0 (c = 0.4, CH3OH); IR (KBr, cm‘ 1) 3426 (br), 2921,

1766, 1662 ;1H NMR (CD 3 COCD3) 6 (ppm) 4.56 (dd, J = 3.4, 7.0, 1H),

1.98-1.84 (m, 1H), 1.57-1.43 (m, 1H), 1.42-1.22 (m, 8 H), 0.90 (t, J= 6.7,

3H); HRMS calcd for C 10H 16 O4 (M+) 200.1049, found 200.1049; Anal.

Calcd for C 10H16 O4 : C, 59.98; H, 8.05: Found: C, 60.05; H, 8.05.

(/?)-5-Hexyl-3,4-dihydroxy-2(5H)-furanone (216) w a s prepared in similar yield as for the S-isomer 217 from Ff-345: mp 98-99

°C; [a]D25 14.2 o (c = 1.9 , CH3 OH); IR (KBr, cm‘ 1) 3423 (br), 2921, 1768,

1660 ;1H NMR (CD 3COCD3) 5 (ppm) 4.57 (dd, J= 3.4, 7.0, 1H), 1.98-1.84

(m, 1H), 1.57-1.43 (m, 1H), 1.42-1.22 (m, 8H), 0.90 (t, J= 6.7, 3H); HRMS calcd for C 10H 16 O4 (M+) 200.1049, found 200.1049.

(S)-2,2-Dimethyl-5-oxo-1,3-dioxolane-4-acetyl chloride

(275). To 20g ( 114.9 mmol) of dioxolane acid 228 in a dry 250 mL round bottom flask was added under argon and at room temperature 75 g

(630 mmol) of thionyl chloride and 2 drops of DMF. The reaction mixture was stirred until evolution of gaseous HCI ceased (oil bubbler; approx. 2 hours). The excess thionyl chloride was distilled in vacuo and remaining traces were removed under reduced pressure (9 h). The acid chloride

(275, 22.1 g) thus obtained was utilized without further purification in the 130

next step.: IR (neat, cm-1) 2998, 1793, 1751, 989, 958 ;1H NMR (CDCI 3 ) 8

(ppm) 4.69 (dd, J= 3.6, 6.4, 1H), 3.56 (dd, J= 3.6, 18.1, 1H), 3.36 (dd, J =

6.4, 18.1, 1H), 1.65 (s, 3H), 1.58 (s, 3H).

(fl)-2,2-Dlmethyl-5-oxo-1,3-dloxolane-4-acetyl chloride

(346) was prepared in similar yield as for the S-275 from fl-339 : IR

(neat, cm-1) 2996, 1793, 1751, 989, 958 ;1H NMR (CDCI 3 ) 5 4.69 (dd, J =

3.6, 6.4, 1H), 3.53 (dd, .7=3.6, 18.1, 1H), 3.35 (dd, J= 6.4, 18.1, 1H), 1.65

(S, 3H), 1.58 (S, 3H).

(S)-2,2-Dimethyl-5-oxo-1,3-dioxolane-4-acetaldehyde

(274). To a 500 mL three-necked flask equipped a mechanical stirrer,

reflux condenser and gas inlet dispersion tube was added 22.1 g (115

mmol) of crude dioxolane acid chloride 275 dissolved in 250 mL of anhydrous xylenes. To this solution was added 2.0 g of 5% palladium on barium sulfate and 0.2 mL of stock quinoline-sulfur poison solution

(prepared by refluxing 1g of sulfur with 5 mL of quinoline for 6 h and

diluting to a final volume of 70 mL with anhydrous xylenes ).201 Hydrogen

gas was bubbled through the stirred reaction mixture and the hydrogen chloride gas generated was trapped in 175 mL of water containing a few drops of phenolphthalein indicator. The mixture was heated at 135 °C and monitored by titration of the hydrogen chloride solution with 5M sodium hydroxide solution. On completion (approx. 3 h) the reaction 131

mixture was cooled to room temperature and 1.5 g of Norit was added. The mixture was filtered (Celite) and the filtrate concentrated under

reduced pressure. The residue was purified over silica gel using ethyl

acetate : hexanes (1 : 3) to furnish 16.2 g (89 %) of white solid : mp 37-38

°C ; [a]o25 1.4 o (c = 4.54, CH3 OH); IR (neat, cm*1) 2994, 2744, 1793,

1725, 1386; NMR (CDCI 3 ) 5 (ppm) 9.78 (s, 1H), 4.80 (dd, J= 3.6, 6 .8 ,

1H), 3.10 (dd, 3.6, 18.3, 1H), 2.92 (dd, J= 6.9, 18.3, 1H), 1.63 (s, 3H),

1.58 (s, 3H) ; HRMS calcd for C 7 H 10O4 (M+) 158.0579, found 158.0572.

(R)-2l2-Dimethyl-5-oxo-1,3-dioxolane-4-acetaldehyde

(347) was prepared in similar yield as for the S-274 from fl-346. : mp

37-38 °C; [oc ]D25 3.7 0 (c - 5.12, CH3OH); IR (neat, cm’ 1) 2996, 2746,

1791, 1727, 1388; 1H NMR (CDCI 3 ) 8 (ppm) 9.78 (s, 1H), 4.80 (dd, J =

3.6, 6 .8 , 1H), 3.11 (dd, J= 3.6, 18.3, 1H), 2.93 (dd, J = 6 .9 , 18.3, 1H), 1.63

(5, 3H), 1.58 (s, 3H) ; HRMS calcd for C 7 H 10O 4 (M+) 158.0579, found

158.0574.

(S)-5-(3-Methoxyallyl)-2,2-dimethyM ,3-dioxolan-4-one

(276). In a dry 500 mL three-necked round bottom flask under argon was dissolved 14.2 g (126.58 mmol) of potassium f-butoxide in 300 mL of anhydrous tetrahydrofuran. The solution was cooled to 0 °C and 44 g

(126.6 mmol) of methoxymethyltriphenylphosphine chloride was added slowly and with stirring (20 min). The resulting orange-red solution was 132 stirred at 0 °C for 45 min, and 10 g (63.3 mmol) of dioxolane aldehyde 274 in 50 mL of anhydrous tetrahydrofuran was added dropwise (15 m in ).178 The mixture was allowed to stir at ambient temperature for 1 h and quenched by addition of 100 mL of brine. Following stirring for 1 h the mixture was extracted with 3 x 250 mL of ether. The combined ether extract was washed with 2 x 150 ml of brine, dried (Na 2 S0 4 ) and filtered.

The filtrate was evaporated under reduced pressure to yield 18 g of crude brown colored liquid (contaminated with the triphenylphosphorane). This residue was purified over silica gel using ethyl acetate : petroleum ether

(1 : 9) to give 9.2 g (78 %) of a mixture of inseparable E : Z enol ethers as a colorless liquid: [a]o 25 -3.2 ° (c = 2.8, CH3 OH); IR (neat, cm-1) 2994,

2938, 1793, 1658; 1H NMR (CDCI 3 ) 6 (ppm) for the E-enol ether (75 %)

6.41 (d, 12.7, 1H), 4.49-4.34 (m, 2H), 3.53 (s, 3H), 2.72-2.32 (m, 2H),

1.60 (s, 3H), 1.54 (s, 3H), for the Z-enol ether (25%) 6.04 (d, J = 6.1, 1H), 4.78-4.61 (m, 2H), 3.61 (s, 3H), 2.72-2.32 (m, 2H), 1.60 (s, 3H), 1.54 (s,

3H); HRMS calcd for C 9H14O4 (M+) 186.0892, found 186.0894.

(R)-5-(3-Methoxyallyl)-2,2-dimethyl-1,3-dioxolan-4-one

(348) was prepared in similar yield as for the S-276 from Ft-347: [ cc] d 2 5

3.1 0 (c = 3.07, CH 3 OH); IR (neat, cm '1) 2992, 2740, 1793, 1656 ; 1H

NMR (CDCI3 ) S (ppm) for the E-enol ether (72 %) 6.42 (d, J= 12.7, 1H),

4.49-4.34 (m, 2H), 3.53 (s, 3H), 2.72-2.32 (m, 2H), 1.61 (s, 3H), 1.54 (s,

3H); for the Z-enol ether (28 %) 6.05 (d, J = 6.1, 1H), 4.78-4.61 (m, 2H), 133

3.61 (s, 3H), 2.72-2.32 (m, 2H), 1.61 (s, 3H), 1.54 (s, 3H); HRMS calcd for

C 9H 14O4 (M+) 186.0892, found 186.0891.

Methyl (2$)-tetrahydro*5-methoxy*2*furoate (277). To the

enol ethers (276, 4.0 g , 21.5 mmol) dissolved in 150 mL of anhydrous

methanol was added 5-6 drops of conc. H2 SO4 . The resultant solution

was heated for 6 h and cooled to room temperature. Sodium bicarbonate

(0.5 g) was added, and the methanol was removed in vacuo. The residue was dissolved in 250 mL of CH 2CI2 and washed with 2 x 100 mL of saturated sodium bicarbonate solution and 2 x 125 mL of brine. The organic extract was dried (Na 2 SC>4 ) and the solvent removed under reduced pressure to produce a colorless liquid which was purified over silica gel using ethyl acetate : hexanes (1:1) as eluant to provide 2.93 g

(86 %) of colorless liquid : [a]o 25 27.3 0 (c = 1.2, CH3 OH); IR (neat, cm-1)

2958, 1739, 1213, 1105: 1H NMR (CDCI 3 ) 5 (ppm) for diastereomer A

(66 %) 5.21 (m, 1H), 4.64-4.52 (m, 1H), 3.77 (s, 3H), 3.37 (s, 3H), 2.44-1.83

(m, 4H); for diastereomer B (33%) 5.08 (m, 1H), 4.64-4.52 (m, 1H), 3.77 (s,

3H), 3.42 (s, 3H), 2.44-1.83 (m, 4H); HRMS calcd for C 7 H 12O 4 (M+)

160.0735, found 160.0718.

Methyl (2/?)-tetrahydro-5-methoxy-2-furoate (349) w a s prepared in similar yield as for the S-277 from fl-348 : [a]o 25 -28.33 0 (c

-0 .1 , CH3 OH); IR (neat, cm’ 1) 2956, 1754, 1209, 1105; 1H NMR (CDCI 3 ) 134

8 (ppm) diastereomer A ( 66 %) 5.21 (m, 1H), 4.64-4.52 (m, 1H), 3.77 (s,

3H), 3.38 (s, 3H), 2.42-1.84 (m, 4H) for diastereomer B (33%) 5.08 (m,

1H), 4.64-4.52 (m, 1H), 3.77 (s, 3H), 3.42 (s, 3H), 2.42-1.84 (m, 4H) ;

HRMS calcd for C 7 H 12O4 (M+) 160.0735, found 160.0750.

Methyl (2S)-tetrahydro-5-hydroxy-2>furoate (278).

Hemiketal 277 (2.93 g, 18.3 mmol ) was stirred with 500 mL of 25 %

aqueous acetic acid for approx. 10 h (monitor by TLC). Upon reaction

completion the aqueous acetic acid was removed in vacuo and the

residue was purified over silica gel using ethyl acetate : hexanes ( 1:1) to

give 2.4 g (90 %) of a colorless liquid : [c i] d 25 9-3 0 (c = 2.7, CH 3 OH); IR

(neat, cnrr1) 3457 (br), 2956, 1735, 1062, 1010; 1H NMR (CDCI 3 ) 8 (ppm)

for diastereomer A (58%) 5.62 (m, 1H), 4.60 (dd, J= 6.5, 8.1, 1H), 3.78 (s,

3H), 2.46-1.93 (m, 4H); for diastereomer B (42%) 5.75 (m, 1H), 4.73 (dd, J

= 3.8, 8.5, 1H), 3.76 (s, 3H, ), 2.46-1.93 (m, 4H) ; HRMS calcd for

C6 H10O4 (M+) 146.0579, found 146.0574.

Methyl (2/?)-tetrahydro-5-hydroxy-2-furoate (350) w a s

prepared in similar yield as for the S-278 from fl-349 : [a]o25 -9.2 0 (c =

1.8, CH3 OH); IR (neat, cm '1) 3543 (br), 2956, 1741, 1068, 1010; 1H NMR

(CDCI3 ) 8 (ppm) for diastereomer A (58%) 5.62 (m, 1H), 4.67 (dd, J= 6.5,

8.1, 1H), 3.78 (s, 3H), 2.46-1.93 (m, 4H); for diastereomer B (42%) 5.75 135

(m, 1H), 4.73 (dd, J= 3.8, 8.5, 1H), 3.76 (s, 3H), 2.46-1.93 (m, 4H) ; HRMS

calcd for C 6 H10O4 (M+) ,146.0579, found 146.0577.

4-Chloro-2-butyn-1-ol (237) was prepared according to Bailey and Fujiwara 153. Thus, to 10g (116.3 mmol) of 2-butyne-1,4-diol dissolved in 10 g of pyridine and 12 mL of benzene in a flame-dried 250

mL three-necked round bottom flask was added dropwise and over 6h

under argon 15.2 g (127.8 mmol) of thionyl chloride while maintaining the temperature between 10 °C and 20 °C. The mixture was warmed to room temperature, stirred overnight, and poured into 30 mL of ice-water. The benzene layer was separated and the aqueous layer was extracted with 4 x 50 mL of ether. The combined organic layers were washed with 5 x 250 mL of saturated sodium bicarbonate solution, 100 ml of water and 2 x 100 ml of brine, and dried (Na 2 SC>4 ). The solvent was evaporated in vacuo and the yellow residual oil was purified by distillation under reduced pressure (55°C, 0.4 mm of Hg pressure) to furnish 7.6g (64 %) of colorless liquid. HRMS calcd for C 4 Hs35CIO (M+), 104.0029, found

104.0025; C4 H537CIO (M+) , 105.9999, found 105.9997.

2,5-Undecadiyn-1-ol (238) was prepared according to Ege154.

Thus, in a 1 L three-necked flame-dried round bottom flask fitted with a reflux condenser and a rubber septum was placed 4.5 g (370 mmol) of magnesium in 200 mL of anhydrous tetrahydrofuran. Bromoethane (23 g; 136

210 mmol) in 200 mL of anhydrous tetrahydrofuran was added dropwise under argon; the reflux rate was controlled by cooling in ice-water. The mixture was heated to reflux and stirred for 1 h. 1-Heptyne {18 g , 187.5

mmol), dissolved in 150 mL of anhydrous tetrahydrofuran, was slowly

added dropwise in order to control the rate of liberated ethane gas.

Following addition the mixture was heated to reflux, and after 45 min 0.3 g

( 3.03 mmol) of cuprous (I) chloride was added. After heating for 15 min 7.5 g (75 mmol) of 4-chloro-2-butyn-1-ol (237) dissolved in 150 mL of anhydrous tetrahydrofuran was added dropwise. The reaction mixture

was heated at reflux for 3 h, cooled, and quenched by addition of 250 mL of ice-water saturated with ammonium chloride. The resulting mixture was filtered (celite) and the filtrate extracted with 3 x 300 mL of ether. The organic layers were washed with 2 x 250 mL of saturated ammonium chloride solution, 150 mL of brine, dried (Na 2 SC>4 ) and concentrated in

vacuo to provide 10.5 g of yellowish brown residue which was purified by distillation (105 °C; 0.4 mm of Hg) to furnish 9.9 g (85 %) of colorless oil.

HRMS calcd for C 11H 16 O (M+) ,164.1201, found 164.1194.

1>Bromo-2,5-undecadiyne (239) was prepared according to Sprecher158. Thus, to a 250 mL flame-dried round bottom three-necked flask, fitted with a reflux condenser, was added under argon 7.0 g (42.68 mmol) of 2,5-undecadiynol (238) in 100 mL of anhydrous ether. The reaction vessel was placed in an ice-salt bath and phosphorous tribromide (5.2 g, 19.4 mmol) was added dropwise with stirring over 30 137

min. The reaction mixture was heated at reflux for 3 h, cooled, and

quenched by addition of 75 mL of ice-water. The mixture was extracted with 3 x 150 mL of ether and the ether layers were washed several times

with saturated sodium bicarbonate solution, 100 mL of water, 2 x 125 mL

of brine, and dried (Na 2 SC>4 ). The solvent was removed under reduced

pressure and purified by distillation (110 °C; 0.4 mm of Hg) to provide 7.7

g (80 %) of a colorless oil.

3,6,9-pentadecatriyn-i-ol (300). In a 1 L three-necked flame- dried round bottom flask fitted with a reflux condenser and a rubber

septum was placed 3.86 g (159 mmol) of magnesium in 250 mL of anhydrous tetrahydrofuran. Bromoethane (17.3 g; 158.6 mmol) in 250 mL anhydrous tetrahydrofuran was added dropwise under argon, and the reflux rate was controlled with the aid of an ice-water bath. The mixture was heated to reflux and stirred for 1 h. 3-Butyn-1-ol (5.56g, 79.3 mmol), dissolved in 150 mL of anhydrous tetrahydrofuran, was slowly added dropwise with stirring (2 h). Following addition, the reaction was heated to reflux. After stirring for 90 min, 0.5 g (2.63 mmol) of cuprous (I) iodide was added. After 75 min 9.0 g (39.65 mmol) of bromo-2,5-undecadiyne (239) dissolved in 150 mL of anhydrous tetrahydrofuran was added. The mixture was heated at reflux for 12 h and an additional 0.25 g (1.32 mmol) of cuprous (I) iodide was added. The mixture was heated at reflux for 7 h, cooled and quenched by addition of 400 mL of ice-water saturated with ammonium chloride. After filtration (Celite), the filtrate was extracted with 138

3 x 400 mL of ether. The ether layers were washed with 2 x 300 mL of saturated ammonium chloride solution, 3 x 200 mL of water, 250 mL of brine, dried (Na 2 SC>4 ), and concentrated in vacuo to yield 6.2 g of yellow- brown oil. The residue was partially purified by crystallization (petroleum ether) at -20 °C to produce 5.75 g (75 %) of an unstable yellow oil (room temperature) which was utilized immediately in the next reaction. : IR

(neat, cm '1) 3365, 2956, 2933, 2225; 1H NMR (CDCI 3 ) 8 (ppm) 3.70 (t, J

= 6.2, 2H), 3.17-3.13 (m, 4H), 2.45 (dt, 1.5, 5.9, 12.0, 2H), 2.15 (dt, J =

2.1, 6.9, 13.9, 2H), 1.68 (br, 1H), 1.55-1.43 (m, 2H), 1.43-1.26 (m, 4H),

0.89 (t. J = 6.9, 3H).HRMS calcd for C 15H2 0 O (M+) 216. 1514, found

216.1519.

(3Z,6Z,9Z)-3,6,9-pentadecatrien-1-ol (301). 3,6,9-

Pentadecatriynol (300, 5.75 g, 26.6 mmol), 5 % palladium on barium sulfate (0.5 g) and 5 drops of 3 % quinoline in methanol was added to a

500 mL hydrogenation flask .156 Hydrogen, at an initial pressure of 72 psi, was taken up over 30 min. The mixture was filtered (Celite) and the filtrate was evaporated in vacuo to produce 5.7 g of crude triene which was purified over silica gel using ethyl acetate : hexanes (1:5) yielding 5.5 g

(93 %) of light yellow oil. IR (neat, cm’ 1) 3336(br), 3012, 2958, 2927,

1652, 719; 1H NMR (CDCI 3 ) 8 (ppm) 5.60-5.28 (m, 6 H), 3.67 (t, J = 6.4,

2H), 2.88-2.79 (m, 2H), 2.43-2.33 (m, 2H), 2.12-1.98 (m, 2H), 1.68-1.53

(m, 2H), 1.53-1.23 (m, 6 H), 0.89 (t, J= 6.7, 3H); HRMS calcd for C 15H26 O

(M+) 222.1984, found 222.1990. 139

1-Bromo-(3Z,6Z,9Z)-3,6,9-pentadecatriene (302). To a

250 mL three-necked round bottom flask was added under nitrogen 9.46 g (36.1 mmol) of trjphenylphosphine dissolved in 150 mL of anhydrous acetonitrile. After cooling to 0 °C (ice-salt bath), bromine (5.77 g, 36.1 mmol) was added dropwise with stirring. The mixture was warmed to room temperature and stirred for 30 min .157 3,6,9-Pentadecatrienol (301,

6.16 g, 27.75 mmol), dissolved in 50 mL of anhydrous acetonitrile, was added dropwise (15 min) and stirred for approx. 4 h. Upon reaction completion acetonitrile was removed in vacuo and the residue was dissolved in 75 mL of ether. Hexanes were utilized to precipitate the triphenylphosphorane side product which was removed by filtration. The crude residue, obtained after concentration of the filtrate in vacuo, was purified over silica gel using ethyl acetate:hexanes (1:9) as eluant. The product 7.0 g (90 %) was obtained as a light yellow oil. IR (neat, cm-1)

2958, 2927, 1652, 1267, 723 ;1H NMR (CDCI 3 ) 8 (ppm) 5.69-5.28 (m,

6 H), 3.38 (t, J= 7.1, 2H), 2.89-2.75 (m, 4H), 2.75-2.52 (m, 2H), 2.12-1.93

(m, 2H), 1.49-1.31 (m ,6 H), 0.89 (t, J = 6 .8, 3H); 13C NMR (CDCI3) 8 (ppm)

131.0, 130.6, 128.9, 127.4 (2C), 126.3, 32.2, 31.5, 30.9, 29.3, 27.3, 25.8,

25.7, 22.6, 14.0; HRMS calcd for CisH2 5 Br (M+) 284. 1139, found

284.1101.

[(3Z,6Z,9Z)-3,6,9-Pentadecatrienyl]triphenylphosphonium bromide (303). Bromo-3,6,9-pentadecatriene (302, 140

7.0 g, 24.6 mmol) was treated with 7.5 g (28.6 mmol) of triphenylphosphine in 25 mL of acetonitrile. The mixture was heated to 70 °C under nitrogen atmosphere. Following reaction completion (monitor salt formation by TLC; 72 h) the mixture was dried for 36 h under reduced pressure to ensure removal of traces of acetonitrile. The yellow residue

13.45 g was used in the next reaction without further purification : IR (neat, c m '1) 3010, 2958. 1652, 1191, 723; 1H NMR (CDCI 3 ) 5 (ppm) 7.93-7.61

(m, 15 H). 5.69-5.12 (m, 6 H). 4.02-3.91 (m, 2H), 2.69-2.48 (m, 4H), 2.09-

1.92 (m, 2H), 1.89-1.72 (m, 2H), 1.46-1.25 (m, 6 H), 0.88 (t, J = 6.9, 3H).

Methyl (25)-2-hydroxyarachldonate (304). To a flame-dried 500 mL three-necked flask, fitted with a low temperature thermometer and a rubber septum, was added under argon 12.35 g (22.57 mmol) of 3,6,9- pentadecatrienetriphenylphosphine bromide (303) and 350 mL of anhydrous tetrahydrofuran. The solution was cooled to -35 °C and 14.1 mL of 1.6 M n-BuLi in hexanes (22.56 mmol) was added dropwise with stirring. The dark red solution was warmed to room temperature, stirred for an additional 30 min and cooled to -35 °C. Hexamethyl- phosphoramide (20.6 g; 115 mmol) was slowly added, and stirring was continued for an additional 45 min at -35 °C. The solution was cooled to

-60 °C, and hemiacetal 278 (1.65 g, 11.28 mmol) dissolved in 25 mL of anhydrous tetrahydrofuran was added dropwise (approx. 15 min). The mixture was stirred at -60 °C for 2 h and warmed to room temperature. Upon completion (TLC monitoring, approx. 8-9 h) the reaction was 141 quenched by the addition of 100 mL of 10 % aqueous HCI solution and extracted with 3 x 300 mL of ethyl acetate. The organic layers were washed with 3 x 250 mL of water, 2 x 200 mL of brine, dried (Na 2 SC>4 ), filtered and concentrated in vacuo. The residue contaminated with the triphenylphosphorane, was purified over silica gel using ethyl acetate : hexanes (1 : 5) as eluant to provide 3.1 g (82 %) of yellow oil : [ajo 25 10.2

° (c = 5 .4 .CH3 0 H); IR (neat, cm-1) 3477 (br), 3012, 2956, 1739, 1652,

721; 1H NMR (CDCI 3 ) 6 (ppm) 5.45-5.25 (m, 8H), 4.20 (dd, J = 4.0, 7.6,

1H), 3.79 (s, 3H), 2.92-2.74 (m, 4H), 2.35-1.92 (m, 4H), 1.91-1.62 (m, 2H),

1.49-1.24 (m, 6 H), 0.89 (t, J= 6.4, 3H); HRMS calcd for C 2 1 H34O 3 (M+)

334.2507, found 334.2509. Anal. Calcd for C 2 1 H 3 4 O 3 : C, 75.41; H,

10.25. Found C, 75.36; H, 10.13.

Methyl (2/?)-2-hydroxyarachidonate (351) was prepared in similar yield as for the S-304 from R-350 : [a]o 25 -10.5 0 (c = 0.9,

CH3OH); IR (neat, cm '1) 3504 (br), 3012, 2956, 1739, 1652, 723 ;1H NMR

(CDCI3 ) 6 (ppm) 5.45-5.25 (m, 8H), 4.20 (dd, J = 4.0, 7.7, 1H), 3.79 (s,

3H), 2.92-2.74 (m, 4H), 2.35-1.92 (m, 4H), 1.90-1.62 (m, 2 H), 1.49-1.24

(m, 6 H), 0.89 (t, J= 6.4, 3H); HRMS calcd for C 2 1 H34O 3 (M+) 334.2507, found 334.2506.

Methyl (2S)-2-hydroxyarachidonate, (allyoxy)acetate

(331). To a dry 250 mL two-necked round bottom flask , fitted with a 142

rubber septum was added under argon 1.5 g (4.49 mmol) of methyl

arachidonate 304 dissolved in 125 mL of anhydrous CH 2CI2. The

solution was cooled to 10 °C (ice bath) and 1.30 g of allyoxyacetic

acid 200 (330; 11.23 mmol) dissolved in 15 mL of anhydrous CH 2CI2 and 0.133 g (0.90 mmol) of 4-pyrrolidinopyridine dissolved in 2 mL of

anhydrous CH 2CI2 was added. A solution of 2.32 g (11.23 mmol) of DCC

in 25 mL of CH 2CI2 was added dropwise with stirring, warmed to room

temperature, and stirred overnight. CH 2CI2 was removed by distillation under reduced pressure and the residue was chromatographed on silica gel using ethyl acetate : hexanes (1 : 5) as eluant to furnish 1.79 g (92 %)

of yellow oil : [ah 25 -8.1 0 (c = 0.1, CH3 OH); IR (neat, cm"1) 3012, 2956,

1751, 1652 ;1H NMR (CDCI 3 ) 5 (ppm) 5.94-5.82 (m, 1H), 5.41-5.22 (m,

10H), 5.11 (t, J = 6.3, 1H), 4.23 (d, 16.6, 1H), 4.22 (d, J = 16.6, 1H),

4.12 (dt, J = 1.3, 5.7, 2H), 3.75 (s, 3H), 2.89-2.68 (m, 4H), 2.32-1.89 (m,

6 H), 1.61-1.49 (m, 2H), 1.42-1.31 (m, 6 H), 0.89 (t, J = 6.5, 3H); HRMS

calcd for C 2 6 H4 0 O 5 (M+) 432.2875, found 432.2856; Anal. Calcd for

C2 6 H40O5 : C, 72.19; H, 9.32: Found: C, 71.90; H, 9.11.

Methyl (2fl)-2-hydroxyarachidonate, (allyoxy)acetate

(352) was prepared in similar yield as for the S-331 from Ff-351 : [a]o25

7.0 ° (c = 0.1, CH 3OH); IR (neat, cm"1) 3012, 2956, 1756, 1648; 1H NMR

(CDCI3 ) 5 (ppm) 5.94-5,82 (m, 1H), 5.45-5.22 (m, 10H), 5.11 (t, J = 6.3,

1H), 4.23 (d, 16.6, 1H), 4.22 (d, J = 16.6, 1H), 4.12 (dt, J= 1.3, 5.7, 2H), 143

3.75 (s. 3H), 2.89-2.68 (m, 4H), 2.28-1.89 (m, 6H), 1.61-1.49 (m, 2H), 1.42-

1.31 {m, 6H), 0.89 (t, J = 6.5, 3H); HRMS calcd for C 2 6 H 4 0 O 5 (M+)

432.2875, found 432.2858.

(S)-3-(Allyoxy)-4-hydroxy-5-[(a//-Z)-3f6,9,12-octadecatetraenyl]-2(5H)-furanone (332). To a flame-dried three-necked 250

mL round bottom flask under argon fitted with a low temperature thermometer and a septum was added 1.23 g (7.59 mmol) of hexamethyldisilazane in 100 mL of anhydrous tetrahydrofuran. The contents were cooled to -25 °C (dry-ice/CCU) and 4.75 mL of 1.6 M (7.59 mmol) n-BuLi in hexanes was added dropwise with stirring while maintaining the temperature below -15 °C. The stirred reaction mixture was warmed to -5

°C, the contents maintained between -5 °C and 0 °C for 45 min, and cooled to -78 °C (dry ice/acetone). Allyloxydiester 331 (1.56 g, 3.61 mmol) in 30 mL of anhydrous tetrahydrofuran was added dropwise with stirring while maintaining the temperature below - 68 °C. Following addition the mixture was stirred at - 78 °C for 75 min and quenched by addition of 40 mL of 10 % aqueous HCI solution. Ether (125 mL) was added, the mixture warmed to room temperature, and extracted with 3 x

100 mL of ether. The ether extract was washed with 2 x 75 mL of brine, dried (Na 2S 0 4 ) and concentrated in vacuo to yield 1.34 g of crude product which was purified over silica gel using 10 % methanol in chloroform as eluant to provide 1.28 g (89 %) of yellow oil : [a]p25 -9.7 0 (c

= 0.2, CH3OH); IR (neat, cm '1) 3081 (br), 3012, 2956, 1747, 1670, 723; 1H 144

NMR (CD3 COCD3 ) 8 (ppm) 6.05-5.89 (m, 1H), 5.44-5.14 (m, 10H), 4.72

(dd, 3.5, 7.7, 1H), 4.48 (dt, J = 1.2, 5.6, 2H), 2.89-2.69 (m, 6 H), 2.38-

1.88 (m, 4H), 1.71-1.21 (m, 8H), 0.87 (t, J = 6 .6 , 3H); MS(FAB) (M+1)+

401; HRMS calcd for C 2 5 H36 O4 (M+) 400.2614, found 400.2606; Anal.

Calcd for C 2 5 H36 O4 + H2O: C, 71.74; H, 9.15: Found: C, 71.90; H, 9.11.

(/?)-3-(Allyoxy)-4-hydroxy-5-[(a//-Z)-3,6,9,12- octadecatetraenyl]-2(5H)-furanone (353) was prepared in similar yield as for the S-332 from fl-352 : [a ]D25 9 4 <> (c = 0.3, CH3 OH); IR

(neat, cnH ) 3081 (br), 3012, 2956, 1749, 1670, 723; 1H NMR

(CD3COCD3 ) 8 (ppm) 6.05-5.89 (m, 1H), 5.44-5.14 (m, 10H), 4.74 (dd, J =

3.5, 7.7, 1H), 4.48 (dt, J = 1.1, 5.7, 2H), 2.87-2.69 (m, 6 H), 2.42-1.88 (m,

4H), 1.71-1.21 (m, 8 H), 0.87 (t, J= 6 .6 , 3H); MS(FAB) (M+1)+ 401; HRMS calcd for C 2 5 H36 O4 (M+) 400.2614, found 400.2607.

(S)-4>Hydroxy-5-[(a//-Z)-3l6t9l12-octadecatetraenyl]-3- [(£)-propenyloxy]-2(5H)-furanone (333). To a flame-dried three necked 250 mL round bottom flask was added under argon 0.254 g (0.30 mmol) of [Bis(methyldiphenylphosphine)] ( 1,5-cyclooctadiene)iridium(l) hexafluorophosphate suspended in 50 mL of freshly distilled peroxide free anhydrous tetrahydrofuran. The flask was evacuated and the argon displaced with hydrogen. The red colored suspension turned to a colorless solution and after 5 min the flask was evacuated and replaced 145 with argon. The allyloxytetronic acid (332, 0.6 g, 1.5 mmol)> dissolved in 25 mL of peroxide free tetrahydrofuran, was added and reaction completion monitored using TLC (approx. 3 h). The solvent was evaporated under reduced pressure and the residue was purified over silica gel using 10% methanol in chloroform as eluant to furnish 0.47 g

(79 %) of dark yellow o il: [a]o 25 -11.2 0 (c = 0.2, CH3OH); IR (neat, cm-1)

3081 (br), 3012, 2956, 1745, 1662, 721; 1H NMR (CD 3 COCD 3 ) 8 (ppm)

6.42*6.35 (m, 1H), 5.49-5.23 (m, 8H), 5.05-4.89 (m, 1H), 4.77 (dd, J= 3.5,

7.9, 1H), 2.92-2.74 (m, 4H), 2.31-1.57 (m, 6 H), 1.51 (dd, 7 = 1.6, 6.9, 3H),

1.45-1.28 (m, 8 H), 0.87 (t, J = 6.4, 3H); MS(FAB) (M+1)+ 401, (M+Na)+

423 ; HRMS calcd for C 2 5 H36 O4 (M+) 400.2614, found 400.2616; Anal.

Calcd for C 2 5 H36 O4 + H2O: C, 71.74; H, 9.15: Found: C, 71.70; H, 9.06.

(/?)-4-Hydroxy-5-[(0//-Z)-3,6,9,12-octadecatetraenyl]-3-

[(E)-propenyloxy]*2(5H)-furanone (354) was prepared in similar yield as for the S-333 from R-353 : [a ]D25 11.7 0 (C = 0.2, CH3 OH); IR

(neat, cm-1) 3081(br), 3012, 2956, 1749, 1664, 696; 1H NMR

(CD3COCD3 ) 8 (ppm) 6.42-6.35 (m, 1H), 5.49-5.23 (m, 8H), 5.05-4.89 (m,

1H), 4.77 (dd, J = 3.5, 7.9, 1H), 2.92-2.74 (m, 4H), 2.31-1.57 (m, 6 H), 1.51

(dd, J = 1.6, 6.9, 3H), 1.45-1.28 (m, 8H), 0.87 (t, J = 6.4, 3H); MS(FAB)

(M+1)+ 401, (M+Na)+ 423 ; HRMS calcd for C 2 5 H3 6 O4 (M+) 400.2614, found 400.2615. 146

(S)-3,4-Dihydroxy-5*[(a//-Z)-3,6,9,12-octadecatetraenyl]-

2(5H)-furanone (201). To a 100 mL round bottom flask was added under nitrogen 0.3 g (0.74 mmol) of 1-propenyl ether 333 dissolved in 60 mL of 50% aqueous acetic acid. The stirred solution was heated at reflux

(oil bath) for 15 min. cooled, and concentrated in vacuo. The residue was

chromatographed on silica gel using 12 % methanol in chloroform as

eluant to provide 0.26 g (95%) of yellow oil : [a)o 25 -13.5 0 (c = 0.2,

CH 3OH); IR (neat, cn r1) 3220 (br), 3012, 2956, 1751, 1670, 1652, 723,

694; 1H NMR (CD 3COCD3 ) 6 (ppm) 5.48-5.34 (m, 8H), 4.68 (dd, J= 3.4,

7.9, 1H), 2.85-2.72 (m, 4H), 2.35-2.19 (m. 2H), 2.18-1.97 (m, 4H), 1.65-

1.52 (m, 1H), 1.48-1.28 (m, 7H), 0.87 (t, J = 6.7, 3H); HRMS calcd for

C2 2 H32O4 (M+) 360.2301, found 360.2308; Anal. Calcd for C 2 2 H32 O4 +

0.33 H2O: C, 72.1; H, 8.98: Found: C, 72.18; H, 8.93.

(/?)-3,4-Dihydroxy-5-[(a//-Z)-3,6,9,12-octadecatetraenyl]-

2(5H)-furanone (200) was prepared in similar yield as for the S-201 from fl-354 : [a]D25 13.2 o (c = 0.2, CH3 OH); IR (neat, cm‘ 1) 3241 (br),

3012, 2956, 1751, 1675, 1652, 723, 694; 1H NMR (CD 3COCD3 ) S (ppm)

5.48-5.34 (m, 8H), 4.68 (dd, J= 3.4, 7.9, 1H), 2.85-2.72 (m, 4H), 2.35-2.19 (m, 2H), 2.18-1.97 (m, 4H), 1.65-1.52 (m, 1H), 1.48-1.28 (m, 7H), 0.87 (t, J

= 6.7, 3H); HRMS calcd for C 2 2 H32O4 (M+) 360.2301, found 360.2305;

Anal. Calcd for C 22H32O4 + 0.5 H2 O: C, 71.51; H, 9.00: Found: C, 71.52;

H, 9.14. 147

Pentadecyltrlphenylphosphonium bromide (308). Under nitrogen atmosphere 1-bromopentadecane (307, 6.5 g, 22.4 mmol) was treated with 5.87 g (22.4 mmol) of triphenylphosphine in 15 mL of acetonitrile and heated to 135 °C. The reaction was monitored by TLC (salt formation, about 18 h), and the mixture was dried under reduced pressure (24 h) to ensure removal of traces of acetonitrile and provide 12.3 g of a colorless solid which was utilized in the next reaction without further purification.

Methyl (2S,5Z)-2-hydroxy-5-eicosenoate (309). To a flame-dried 500 mL three-necked flask fitted with a low temperature thermometer and a rubber septum, was added under argon 12.2 g (22.06 mmol) of pentadecyltrlphenylphosphonium bromide (308) and 300 mL of anhydrous tetrahydrofuran. Following cooling to -35 °C, 13.8 mL of 1.6 M n-BuLi in hexanes (22.06 mmol) was added dropwise and with stirring. The orange solution was warmed to room temperature and stirred for an additional 30 min. The mixture was cooled to -35 °C, and 18.5 g (103.3 mmol) of hexamethylphosphoramide was added slowly. The reaction mixture was stirred for 45 min and cooled to -60 °C. Hemiacetal 278 (1.6 g, 11.03 mmol) dissolved in 25 mL of anhydrous tetrahydrofuran was added dropwise, and stirring was continued for 1 h at -60 °C. The mixture was warmed to room temperature; reaction completion was monitored using TLC. The reaction was quenched by addition of 100 mL of 10 % aq.

HCI soln. and extracted with 3 x 300 mL of ethyl acetate. The organic 148

layers were washed with 3 x 250 mL of water, 2 x 200 mL of brine, dried

(Na2 S0 4 ), filtered and evaporated in vacuo. The residue, contaminated

with triphenylphosphorane, was purified over silica gel using ethyl acetate

: hexanes (1 : 5) to yield 2.98 g (80%) of colorless oil. : [a]o 25 8.9 0 (c =

1.4, CH3 OH); IR (neat, cm-1) 3475(br), 3006, 2925, 1739, 721; 1H NMR

(CD3COCD3 ) S (ppm) 5.48-5.24 (m, 2H). 4.20 (m, 1H), 3.79 (s, 3H), 2.72

(d, J= 5.3, 1H), 2.32-2.15 (m, 2H), 2.15-1.98 (m, 2H), 1.92-1.64 (m, 2H),

1.48-1.21 (m, 24H), 0.88 (t, J = 6.3, 3H); 13q NMR (CD 3COCD3 ) 5 (ppm)

175.7, 131.5, 127.9, 70.0, 52.4, 34.4, 31.9, 29.7 (5C), 29.6 ( 2 C), 29.3 (3C),

27.2, 22.7 (2C), 14.0 ; HRMS calcd for C 2 1 H4 0 O3 (M+) 340.2977, found

340.2977; Anal. Calcd for C 2 1 H4 0 O 3 : C, 74.07; H, 11.84: Found: C,

74.13; H, 11.91.

Methyl (2R,5Z)-2-hydroxy-5-eicosenoate (355) was

prepared in similar yield as for the S-309 from R-350 : [a]o 25 -8.8 0 (c =

2.1, CH3 OH); IR (neat, cm"1) 3482 (br), 3006, 2925, 1739, 721; 1H NMR

(CD3COCD3 ) 5 (ppm) 5.48-5.24 (m, 2 H). 4.20 (m, 1H), 3.79 (s, 3H), 2.72

(d, J= 5.3, 1H), 2.32-2.15 (m, 2H), 2.15-1.98 (m, 2H), 1.92-1.64 (m, 2H),

1.48-1.21 (m, 24H), 0.88 (t, J= 6.3, 3H); 13c NMR (CD 3COCD3 ) 5 175.7,

131.5, 127.9, 70.0, 52.3, 34.4, 31.9, 29.7 (5C), 29.6 (2C), 29.3 (3C), 27.2,

22.7 ( 2 C), 14.0 ; HRMS calcd for C 2 1 H 4 0 O 3 (M+) 340.2977, found

340.2970. 149

Methyl (25,5Z)-2-hydroxy-5-eicosenoate, (allyloxy)- acetate (334). To a dry 100 mL three-necked round bottom flask fitted

with a rubber septum was added under argon 0.65 g (1.91 mmol) of

methyl eicosaenoate 309 and 50 mL of anhydrous CH 2CI2. Following

cooling to 10 °C (ice bath), 0.45 g (3.82 mmol) of allyloxyacetic acid

(330) dissolved in 15 mL of anhydrous CH 2CI2 and 0.028 g (0.191 mmol)

of 4-pyrrolidinopyridine dissolved in 2 mL of anhydrous CH 2CI2 were added. A solution containing 0.79 g (3.824 mmol) of DCC in 15 mL of

CH2CI2 was added dropwise and the stirred mixture was warmed to room

temperature and stirred for an additional 8 h. CH2CI2 was evaporated in

vacuo and the product was purified over silica gel using ethyl acetate :

hexanes (1 : 5) as eluant to furnish 0.78 g (93 %) of colorless oil : [a]o 25

-6.5 0 (c = 1.6, CH 3OH); IR (neat, cm-1) 3006, 2923, 1758, 721; 1H NMR

(CD3COCD3 ) 6 (ppm) 6.01-5.86 (m, 1H), 5.51-5.24 (m, 4H), 5.10 (t, J =

6.3, 1H), 4.22 (d, J= 16.6, 1H), 4.23 (d, J= 16.6, 1H), 4.13 (dt, J= 1.1, 5.7,

2H), 3.75 (S, 3H), 2.26-2.12 (m, 2H), 2.07-1.87 (m, 4H), 1.42-1.28 (m,

24H), 0.88 (t, J = 6.3, 3H); HRMS calcd for C 2 6 H 4 6 O 5 (M+), 438.3345

found 438.3330.

Methyl (2/?,5Z)-2-hydroxy-5-eicosenoate, (allyloxy)- acetate (356) was prepared in similar yield as for the S-334 from ft-

355 : [a )D25 6.4 ° ( c s 1.4, CH3 OH); IR (neat, cm"1) 3006, 2921, 1756,

721; 1H NMR (CD 3COCD3 ) S (ppm) 6.01-5.86 (m, 1H), 5.51-5.24 (m, 4H),

5.10 (t, J= 6.3, 1H), 4.22 (d, 16.6, 1H), 4.23 (d, J= 16.6, 1H), 4.13 (dt, J 150

= 1.1, 5.7, 2H), 3.75 (s, 3H), 2.26-2.12 (m, 2H), 2.07-1.87 (m, 4H), 1.42-

1.28 (m, 24H), 0.88 (t, J = 6.3, 3H); HRMS calcd for C 2 6 H4 6 O 5 (M+)

438.3345, found 438.3347.

(S)-3-(Allyoxy)-4-hydroxy-5-[(Z)-3-octadecenyl]-2(5H)-

furanone (335). To a flame-dried three-necked 250 mL round bottom flask fitted with a low temperature thermometer and a septum was added under argon 0.648 g (4.02 mmol) of hexamethyldisilazane in 50 mL of

anhydrous tetrahydrofuran. The contents were cooled to -25 °C (dry-

ice/CCl4 ), and 2.51 mL of 1.6 M (4.02 mmol) of n-BuLi in hexanes was

added dropwise with stirring while maintaining the temperature below -15

°C. The mixture was warmed to -5 °C and the contents maintained between -5 °C and 0 °C for 45 min. The mixture was cooled to -78 °C (dry ice/acetone) and 0.84 g (1.91 mmol) of allyloxydiester 334 in 20 mL of anhydrous tetrahydrofuran was added dropwise with stirring while

maintaining the temperature below - 68 °C. Following addition the

mixture was stirred at - 78 °C for 75 min and quenched with 10 mL of 10

% aqueous HCI solution. Ether (75 mL) was added, and the reaction mixture was warmed to room temperature and extracted with 3 x 75 mL of ether. The ether extracts were washed with 2 x 50 mL of brine, dried

(Na2 S0 4 ) and concentrated in vacuo to provide 0.75 g (97 %) of residue.

Purification over silica gel using 10 % methanol in chloroform as eluant provided 0.69 g (89 %) of white solid : mp 51-54 °C; [a]o 25 -9.0 0 (c = 0 .6 ,

CH3 OH); IR (neat, c m '1) 3079 (br), 3002, 2915, 1741,1654, 734, 719; 1H 151

NMR (CD3 COCD3 ) 8 (ppm) 6.04-5.89 (m, 1H), 5.47-5.15 (m, 4H), 4.72

(dd, J - 3.5, 7.6, 1H), 4.48 (dt, 1.1, 5.9, 2H), 2.21-1.54 (m, 6 H), 1.45-

1.28 (m, 24H), 0.87 (t. J = 6 .8 , 3H); HRMS calcd for C 2 5 H 4 2 O 4 (M+)

406.3083, found 406.3084; Anal. Calcd for C 2 5 H4 2 O 4 : C, 73.85; H,

10.41. Found: C, 73.52; H, 10.27 .

(ff)-3-(Allyoxy)-4-hydroxy-5-[(Z)-3-octadecenyl]-2(5H)- furanone (357) was prepared in similar yield as for the $-335 from R-

356 : mp 49-53 °C; [a]D25 8.9 0 (c = 0.8, CH3OH); IR (neat, cm-1) 3079

(br), 3002, 2915, 1741,1654, 734, 719; 1H NMR (CD 3COCD 3 ) 5 (ppm)

6.04-5.89 (m, 1H), 5.47-5.15 (m, 4H), 4.72 (dd, J= 3.5, 7.6, 1H), 4.48 (dt, J

« 1.1, 5.9, 2H), 2.21-1.54 (m, 6 H), 1.45-1.28 (m, 24H), 0.87 (t, J = 6 .8, 3H);

HRMS calcd for C 2 5 H42O4 (M+), 406.3083 found 406.3094.

(S)-4-Hydroxy-5-[(Z)-3-octadecenyl]-3-[(E)-propenyloxy]-

2(5H)-furanone (336). To a flame-dried three-necked 100 mL round bottom flask under argon was added 0.172 g (0.203 mmol) of

[bis(methyldiphenylphosphine)] (1,5-cyclooctadiene)iridium(l) hexafluorophosphate suspended in 35 mL of freshly distilled peroxide free anhydrous tetrahydrofuran. The flask was evacuated and the argon displaced with hydrogen. The red suspension turned to a colorless solution after 5 min. The flask was evacuated, and the hydrogen replaced with argon. The allyloxy tetronic acid (335, 0.415 g, 1.022 mmol) was 152 dissolved in 25 mL of peroxide free tetrahydrofuran and added to the activated catalyst. On completion (TLC; approx. 3 h) the solvent was evaporated in vacuo and purified over silica gel using 10 % methanol in chloroform as eluant to provide 0.33g (79 %) of white waxy solid : mp 42-

45 °C ; [a ]D25 -13.0 0 (c = 0.3, CH3 OH); IR (neat, cm-1) 3079 (br), 3004,

2919, 1739, 1681, 1658, 738, 721; 1H NMR (CD 3 COCD3 ) 8 (ppm) 6.43-

6.36 (m, 1H), 5.49-5.31 (m, 2H), 5.05-4.91 (m, 1H), 4.77 (dd, J= 3.5, 7.8,

1H), 2.24-1.62 (m, 6 H), 1.51 (dd. J= 1.7, 6.9, 3H), 1.44-1.24 (m, 24H),

0.87 (t, J = 6 .8, 3H); HRMS calcd for C 2 5 H4 2O 4 (M+) 406.3083, found

406.3083.

(ft)-4-Hydroxy-5-[(Z)-3-octadecenyl]-3-[(£)-propenyloxy]-

2(5H)-furanone (358) was prepared in similar yield as for the S-336 from H-357 : mp 42-45 °C; [«]d 25 12.8 ° (c = 0.3, CH3OH); IR (neat, cm '1)

3079 (br), 3004, 2919, 1739, 1681, 1658, 738, 721 ;1H NMR

(CD3COCD3 ) 8 (ppm) 6.43-6.36 (m, 1H), 5.49-5.31 (m, 2H), 5.05-4.91 (m,

1H), 4.77 (dd, J= 3.5, 7.8, 1H), 2.24-1.62 (m, 6 H), 1.51 (dd, J * 1.7, 6.9,

3H), 1.44-1.24 (m, 24H), 0.87 (t, J - 6 .8, 3H); HRMS calcd for C 2 5 H4 2 O4

(M+) 406.3083, found 406.3076.

(S)-3,4-Dihydroxy-5-[(Z)-3-octadecenyl]-2(5H)-furanone (203). To a 100 mL round bottom flask was added under nitrogen 0.2 g

(0.49 mmol) 1-propenyl ether 336 dissolved in 60 mL of 50% aqueous 153

acetic acid. The solution was heated at reflux (oil bath) for 15 min, cooled,

and the aqueous acetic acid removed in vacuo. The residue was purified

over silica gel using 12 % methanol in chloroform as eluant to give 0.17 g

(95%) of white waxy solid : mp 64-66 °C; [a]o 25 -9.9 0 (c = 0.5, CH3OH);

IR (neat, cm*1) 3421 (br), 2917, 2850, 1754, 1668, 719; 1H NMR

(CD3COCD3 ) 8 (ppm) 5.49-5.28 (m, 2H), 4.67 (dd, 3.4, 7.8, 1H), 2.32-

1.88 (m, 4H), 1.69-1.51 (m, 2H), 1.44-1.25 (m, 24H), 0.87 (t, J= 6.7, 3H);

HRMS calcd for C 2 2 H38O4 (M+) 366.2770, found 366.2780; Anal. Calcd for C2 2 H38O4 + 0.33 H2 O: C, 70.93; H, 10.46: Found: C, 70.73; H, 10.31.

(ft)-3,4-Dihydroxy-5-[(Z)-3-octadecenyl]-2(5H)-furanone (202) was prepared in similar yield as for the S-203 from fl-358 : mp

65-67 °C; [a ]D25 9.5 0 (c = 0.5, CH3OH); IR (neat, cm*1) 3421 (br), 2917,

2850, 1754, 1666, 734, 719; 1H NMR {CD 3COCD3) 5 (ppm) 5.49-5.28 (m,

2H), 4.67 (dd, J = 3.4, 7.8, 1H), 2.32-1.88 (m, 4H), 1.69-1.51 (m, 2H), 1.44-

1.25 (m, 24H), 0.87 (t, J = 6.7, 3H); HRMS calcd for C 2 2 H 3 8 O 4 (M+),

366.2770 found 366.2780; Anal. Calcd for C 2 2 H3 8 O 4 + 0.33 H2 O: C,

70.93; H, 10.46: Found: C, 70.53; H, 10.25.

Methyl (2S,5Z)-2-hydroxy-5-eico$enoate, (benzyloxy)- acetate (337). To a two-necked, flame-dried 100 mL round bottom flask was added under argon 0.275 g (0.81 mmol) of methyl-2- hydroxyeicosaenoate 309 in 40 mL of anhydrous CH 2CI2 and 0.20 g 154

(1.09 mmol) of benzyloxyacetyl chloride. The solution was cooled to 0 °C (ice-salt bath), and pyridine (0.086 g, 1.09 mmol) was added dropwise.

The mixture was stirred for 30 min at 0 °C, warmed to room temperature, and stirred for an additional 8 h. The reaction was quenched with 10 mL of ice-water. CH 2CI2 (20 mL) was added, and the mixture stirred for 6 h.

The CH 2CI2 layer was washed with 3 x 20 mL 10% aqueous HCI solution, 3x15 mL of saturated sodium bicarbonate solution, 2 x 25 mL of brine, dried (Na 2 S0 4 ) and concentrated under reduced pressure. The residue was purified over silica gel using ethyl acetate : hexanes (1:5) as eluant to yield 0.34 g (87 %) of white solid : [cc] d 25 -5.4 0 (c = 0.5, CH3 OH); IR (neat, cm "1) 3006, 2923, 1756, 1455, 734, 698; 1H NMR (CD 3COCD3 ) 5 (ppm)

7.40-7.26 (m, 5H), 5.48-5.22 (m, 2H), 5.12 (t, J= 6.4, 1H), 4.66 (s, 2H),

4.25 (d, J= 16.7, 1H), 4.17 (d, J= 16.7, 1H), 3.76 (s, 3H), 2.20-1.87 (m,

6 H), 1.49-1.21 (m, 24H), 0.88 (t, J= 6.4, 3H); HRMS calcd for C 3 0 H48O5

(M+) 488.3501, found 488.3501.

M e th y l (2R,5Z)-2-hydroxy-5-elcosenoate, (benzyloxy)acetate (359) was prepared in similar yield as for the S-

337 from R-355 : [ct]D25 5.2 0 (c = 0.5, CH3 OH); IR (neat, cm '1) 3006,

2923, 1756, 1455, 734, 698; 1H NMR (CD 3COCD3 ) 5 (ppm) 7.40-7.26 (m,

5H), 5.48-5.22 (m, 2H), 5.12 (t, J= 6.4, 1H), 4.66 (s, 2H), 4.25 (d, J= 16.7,

1H), 4.17 (d, J= 16.7, 1H), 3.76 (s, 3H), 2.20-1.87 (m, 6 H), 1.49-1.21 (m,

24H), 0.88 (t, J = 6.4, 3H); HRMS calcd for C 3 0 H4 8 O 5 (M+) 488.3501, found 488.3501. 155

(S)-3-(Benzyloxy)-4-hydroxy-5-[(Z)-3-octadecenyl)- 2(5H)-furanone (338). To a flame-dried three-necked 100 mL flask

equipped with a low temperature thermometer was added under argon

0.302 g (1.87 mmol) of hexamethyldisilazide in 25 mL of anhydrous

tetrahydrofuran. The solution was cooled to -25 °C and 1.17 mL of a 1.6

M n-BuLi (1.87 mmol) in hexanes was added dropwise with stirring while maintaining the temperature below -15 °C. The reaction was held between -3 °C and -5 °C for 45 min, cooled to -78 °C, and 0.434 g (0.89 mmol) of benzyloxyacetyldiester 3 3 7 in 8 mL of anhydrous tetrahydrofuran was added dropwise. The reaction was stirred for 2 h and quenched at -78 °C with 10 mL of cooled 10% aqueous HCI solution.

Ether (15 mL) was added, and the mixture was warmed to room temperature and extracted with 3 x 50 mL of ether. The organic layers were washed with 3 x 25 mL of brine, dried (Na 2 S0 4 ) and concentrated under reduced pressure The residue was purified over silica gel using

10% methanol in chloroform as eluant to furnish 0.34 g (84%) of white solid : mp 76-79 °C; (a ]D25 -13.8 o (c = 0.3, CH3OH); IR (KBr, cm-1) 3033

(br), 3004, 2917, 1739, 1660, 1402, 1342, 738, 728, 696; 1H NMR

(CD3COCD3 ) 6 (ppm) 7.43-7.29 (m, 5H), 5.48-5.27 (m, 2H), 5.06 (d, J =

11.3, 1H), 5.01 (d, J= 11.3, 1H), 4.69 (dd, J= 3.5, 7.6, 1H), 2.28-1.51 (m,

6 H), 1.42-1.27 (m, 24H), 0.86 (t, J= 6.7, 3H); HRMS calcd for C 2 9 H44O4

(M+) 456.3239, found 456.3239; Anal. Calcd for C 2 9H44O4 + 0.2 H2 O: C,

75.68; H, 9.72: Found: C, 75.57; H, 9.68. 156

(ft)-3-(Benzyloxy)-4-hydroxy-5-[(Z)-3-octadecenyl]- 2(5H)-furanone (360) was prepared in similar yield as for the S-338

from R-359 : mp 76-79 °C; [a ]D25 13.5 ° (c = 0.2, CH3OH); IR (KBr, cm’ 1)

3033 (br), 3004, 2917, 1739, 1660, 1400, 1342, 738, 730, 696; 1H NMR

(CD3COCD3 ) 6 (ppm) 7.43-7.29 (m, 5H), 5.48-5.27 (m, 2H), 5.06 (d, J =

11.3, 1H), 5.01 (d, J - 11.3, 1H), 4.69 (dd, J= 3.5, 7.6, 1H), 2.28-1.51 (m,

6 H), 1.42-1.27 (m, 24H), 0.86 (t, J = 6.7, 3H); HRMS calcd for C 2 9 H44O4

(M+) 456.3239, found 456.3243.

(5)-5-Octadecyl-3,4-dlhydroxy-2(5H)-furanone (205). T 0 a 250 mL hydrogenation bottle was added 0.02 g of 10 % palladium on carbon in 5 mL methanol. To this suspension was added 0.1 g (0.219 mmol) of benzyloxytetronic acid 338 dissolved in 15 mL methanol. Hydrogenation was initiated at 40 psi and at room temperature. The reaction was monitored for completion by TLC (approx. 5-6h), filtered

(Celite) and evaporated under reduced pressure. The residue was purified over silica gel using 10% methanol in chloroform as eluant to generate 58 mg (72 %) of white solid : mp 110-112 °C; [cc] d 25 -6.8 0 (c =

0.3, CH3OH); IR (KBr, cm '1) 3380 (br), 2917, 2848, 1741, 1668; 1H NMR

(CD3COCD3 ) 5 (ppm) 4.67 (dd, J = 3.5, 7.3, 1H), 2.01-1.88 (m, 1H), 1.57-

1.47 (m, 1H), 1.42-1.23 (m, 32H), 0.87 (t, J * 6 .8 , 3H); HRMS calcd for

C2 2 H40O4 (M+) 368.2927, found 368.2928; Anal. Calcd for C 2 2 H4 0 O4 +

0.5 H2O: C, 69.99; H, 10.95: Found: C, 70.27; H, 11.22. 157

(fl)-3,4-Dihydroxy-5-3-octadecyl]-2(5H)-furanone (204) w a s

prepared in similar yield as for the S-205 from R-360 : mp 114-117 °C;

[a]025 6.9 ° (c = 0.2, CH 3OH); IR (KBr, cm '1) 3411 (br), 2917, 2848, 1754,

1668; 1H NMR (CD 3 COCD3) 8 (ppm) 4.67 (dd, J = 3.5, 7.3, 1H), 2.01-1.88

(m, 1H), 1.57-1.47 (m, 1H), 1.42-1.23 (m, 32H), 0.87 (t, .7=6.8, 3H); HRMS calcd for C 2 2 H4 0 O 4 (M+) 368.2927, found 368.2927; Anal. Calcd for

C2 2 H40O4 + 0.5 H2O: C, 69.99; H, 10.95: Found: C, 69.71; H, 11.09. BIBLIOGRAPHY

1. Smith, W. L.; Mamett, L. J. Prostaglandin endoperoxide synthase: structure and catalysis . Biochim. Biophys. Acta 1991, 1083, 1-17.

2. Simmons, D. L.; Xie, W.; Chipman, J. G.; Evett, G. E. Multiple ccyclooxygenases: cloning of a mitogen-inducible form In Prostaglandins, Leukotrienes, Lipoxins and P, Bailey, A F J. M., Ed.; Plenum Press: New York, 1991, pp 67-78.

3. Xie, W.; Chipman, J. G.; Robertson, D. L.; Erikson, R. L.; Simmons, D. L. Expression of a mitogen-responsive gene encoding prostaglandin synthase is regulated by mRNA splicing. Proc. Natl. Acad. Sci. 1991, 88, 2692-2696.

4. Kujubu, D. A.; Fletcher, B. S.; Varnum, B. C.; Lim, R. W.; Herschman, H. R. TIS10, a phorbol ester tumor promoter-inducible m-RNA from Swiss 3T3 cells, encodes a novel prostaglandin synthase/cyclooxygenase homologue. J. Biol. Chem. 1991, 266, 12866-12072.

5. Dewitt, D. L.; El-Harith, E. A.; Kraemer, S. A.; Andrews, M. J.; Yao, E. F.; Armstrong, R. L.; Smith, W. L. The aspirin and heme-binding sites of ovine and murine prostaglandin endoperoxide synthases. J. Biol. Chem. 1990, 265, 5192-5198.

6. DeWitt, D. L.; Smith, W. L. Primary structure of prostaglandin G/H synthase from sheep vesicular gland determined from the complementary DNA sequences. Proc. Natl. Acad. Sci. USA 1988, 85, 1412-1416.

7. Merlie, J. P.; Fagan, D.; Mudd, J.; Needleman, P. Isolation and characterization of the cDNA for sheep seminal vesicle prostaglandin endoperoxide synthase (cyclooxygenase). J. Biol. Chem. 1988, 263, 3550-3553.

8. Yokoyama, C.; Takai, T.; Tanabe, R. Primary structure of sheep prostaglandin endoperoxide synthase deduced from cDNA sequence. FEBS Lett 1988, 231, 347-351. 158 159

9. Hla, T.; Nielson, K. Human cyclooxygenase-2 cDNA. Proc. Natl. Acad. Sci. USA 1992, 89, 7384-7388.

10. Shimokawa, T.; Smith, W. L. Essential histidines of prostaglandin endoperoxide synthase. J. Biol. Chem. 1991, 266, 6168-6173.

11. Shimokawa, T.; Kulmacz, R. J.; Dewitt, D. L.; Smith, W. L. Tyr 385 of prostaglandin endoperoxide synthase is required for cyclooxygenase catalysis. J. Biol. Chem. 1990, 265, 20073-20076.

12. Yokoyama, C.; Tanabe, R. Cloning of human gene encoding prostaglandin endoperoxide synthase and primary structure of the enzyme. Biochem. Biophys. Res. Common. 1989, 165, 888-894.

13. Miyamoto, T.; Ogino, N.; Yamamoto, S.; Hayashi, O. Purification of prostaglandin endoperoxide synthetase from bovine vesicular gland microsomes . J. Biol. Chem. 1976, 251, 2629-2636.

14. Van der Ouderaa, F. J.; Buytenhek, M.; Nugteren, D. H.; Van Dorp, D. A. Purification and characterization of prostaglandin endoperoxide synthetase from sheep vesicular glands. Biochim. Biophys. Acta 1977, 487, 315-331.

15. Van der Ouderaa, F. J.; Buytenhek, M.; Nugteren, D. H.; Van Dorp, D. A. Acetylation of prostaglandin endoperoxide synthase with acetylsalicylic acid . Eur. J. Biochem. 1980, 109, 1-8.

16. Otto, J. C.; DeWitt, D. L.; Smith, W. L. /V-Glycosylation of prostaglandin endoperoxide synthases-1 and -2 and their orientations in the endoplasmic reticulum . J. Biol. Chem. 1993, 268, 18234-18242.

17. Mutsaers, J. H. G. M.; van Halbeek, H.; Kamerling, J. P.; Vliegenthart, J. F. G. Determination of the structure of the carbohydrate chains of prostaglandin endoperoxide synthase from sheep. Eur. J. Biochem. 1985, 147, 569-574.

18. Regier, M. K.; DeWitt, D. L.; Schindler, M. S.; Smith, W. L. Subcellular localization of prostaglandin endoperoxide synthase-2 in murine 3T3 cells. Arch. Biochem. Biophys. 1993, 301, 439-444.

19. Mizuno, K.; Yamamoto, S.; Lands, W. E. M. Effects of non-steroidal antiinflammatory drugs on fatty acid cyclooxygenase and prostaglandin hydroperoxide activities. Prostaglandins 1982,2 3 , 743-757. 160

20. Smith, W. L.; Shimokawa, T.; Otto, J. C. Structure/function relationships of PGH synthase: heme binding and carbohydrate attachment sites In Prostaglandins, Leukotrienes, Lipoxins and PAF, Xlth Washington International Spring Symposium Bailey, J. M., Ed.; 1991.

21. Kulmacz, R. J.; Tsai, A. L.; Palmer, G. Heme spin states and peroxide-induced radical species in prostaglandin H synthase. J. Biol. Chem. 1987 ,262, 10524-10531.

22. Karthein, R.; Dietz, R.; Nastainczysk, W.; Ruf, H. H. Higher oxidation states of prostaglandin H synthase. EPR study of a transient tyrsoyl radical in the enzyme during the peroxidase. Eur. J. Biochem. 1988, 171, 313-320.

23. Hemler, M. E.; Cook, H. W.; Lands, W. E. M. Prostaglandin biosynthesis can be triggered by lipid peroxides. Arch. Biochem. Biophys. 1979, 193, 340-345.

24. Hemler, M. E.; Lands, W. E. M. Evidence for a peroxide-initiated free-radical mechanism of prostaglandin biosynthesis. J. Biol. Chem. 1980, 255, 6253-6261.

25. Kulmacz, R. J.; Lands, W. E. M. Requirements for hydroperoxide by the cyclooxygenase and peroxidase activities of prostaglandin H synthase. Prostglandins 1983, 25, 531-540.

26. Smith, W. L.; Lands, W. E. M. Oxygenation of unsaturated fatty acids of soybean lipoxygenase. J. Biol. Chem. 1972, 247, 1038- 1047.

27. Picot, D.; Garavito, R. M. In Cytochrome P450: Biochemistry and Biophysics Schuster, I., Ed.; Taylor and Francis: Philadelphia, 1989, pp 29-36.

28. Hamberg, M.; Samuelsson, B. On the mechanism of the biosynthesis of prostaglandin Et and F ia. J. Biol. Chem. 1967, 242, 5336-5343.

29. Marnett, L. J.; Maddipati, K. R. Prostaglandin H synthase In Peroxidases: Chemistry and Biology Everse, J., Everse, K., Grisham, M., Ed.; CRC Press: Boca Raton, 1991, pp 293-334.

30. Hamberg, M.; Samuelsson, B. Oxygenation of unsaturated fatty acids by the vesicular gland of sheep. J. Biol. Chem. 1967, 242, 5344-5354. 161

31. Hamberg, M.; Samuelsson, B. Prostaglandin endoperoxides. Novel transformations of arachidonic acid in human platelets . Proc. Natl. Acad. Sci. USA 1974, 71, 3400-3404.

32. Lands, W. E. M.; Cook, H. W.; Rome, L. H. Prostaglandin biosynthesis: consequence of oxygenase mechanism upon in vitro asays of drug effectiveness. Adv. Prostaglandin, Thromboxane Leukotriene Res. 1976, 1, 7-17.

33. Cook, H. W.; Lands, W. E. M. Mechanisms for suppression of cellular biosynthesis of prostaglandins. Nature 1976, 260, 630- 632.

34. Samuelsson, B. On the incorporation of oxygen in the conversion of 8,11,14-eicosatrienoic acid to prostaglandin Ei. J. Am. Chem. Soc. 1965, 87, 3011.

35. Lands, W. E. M. Mechanisms of action of antiinflammatory Drugs . Adv. Drug Res. 1985, 14, 147-167.

36. Robert, A. Antisecretory, antiulcer, cytoprotective and diarrhoegenic properties of prostaglandins. Adv. Prostaglandin, Thromboxane Leukotriene Res. 1976, 2, 507-520.

37. Shen, T. Y. Perspectives in nonsteroidal anti-inflammatory agents . Angew. Chem. Int. Ed. 1972, 11, 460-472.

38. Shen, T. Y.; Ham, E. A.; Cirillo, V. J.; Zanetti, M. Structure-activity relationship of certain prostaglandin synthetase inhibitors. In Prostaglandin Synthetase Inhibitors Robinson, H. J., Vane, J. R., Ed.; Raven Press: New York, 1974, pp 19-31.

39. Lombardino, J. G. Medicinal chemistry of acidic nonsteroidal antiinflammatory drugs. In Nonsteroidal Antiinflammatory Drugs Lombardino, J. F., Ed.; Wiley-lnterscience: 1905, pp 253-431.

40. Marshall, P. J.; Kulmacz, R. J.; Lands, W. E. M. Constraints on prostaglandin biosynthesis in tissues. J. Biol. Chem. 1987, 262, 3510-3517.

41. Miller, D. K.; Gillard, J. W.; Vickens, P. J.; Sadowski, S.; Leveitle, C.; Mancini, J. A.; Charleson, P.; Dixon, R. A. F.; Ford-Hutchinson, A. W.; Fortin, R.; Gauthier, J. Y.; Rodkey, J.; Rouzer, C. A.; Sigal, I. S.; Strader, C. D.; Evans, J. F. Identification and isolation of a membrane protein necessary for leukotriene production . Nature 1990, 3 43, 278-281. 162

42. Rome, L. H.; Lands, W. E. M. Structural requirements for time dependent inhibition of prostaglandin biosynthesis by antiinflammatory drugs. Proc. Natl. Acad. Sci. U.S.A. 1975, 72, 4863-4865.

43. Higgs, G. A.; Vane, J. R. Inhibition of cyclooxygenase and lipoxygenase . British Medical Bulletin 1983, 39, 265-270.

44. Vanderhoek, J. Y.; Lands, W. E. M. Acetylenic inhibitors of sheep vesicular gland oxygenase. Biochim. Biophys. Acta. 1973, 296, 374-381.

45. Kuhn, H.; Holzhutter, H. G.; Schewe, T.; Hiebsch, C.; Rapoport, S. The mechanism of inactivation of lipoxygenases by acetylenic fatty acids. Eur. J. Biochem. 1984, 139, 577-583.

46. Corey, E. J.; Park, H. Irreversible inhibition of the enzymic oxidation of arachidonic acid to 15-(hydroperoxy)-5,8,11(Z), 13(E)- eicosatetraenoic acid (15-HPETE) by 14,15-dehydroarachidonic acid. J. Am. Chem. Soc. 1982, 104, 1750-1752.

47. Roth, G. J.; Stanford, N. S.; Majerus, P. W. Acetylation of prostaglandin synthetase by aspirin . Proc. Natl. Acad. Sci. USA 1975, 72, 3073-.

48. Smith, W. L.; DeWitt, D. L ; Shimokawa, T.; Kremer, S. A.; Meade, E. A. molecular basis for the inhibition of prostanoid biosynthesis by nonsteroidal antiinflammatory agents. Stroke 1990, 21, IV24-IV28.

49. Roth, G. J.; Siok, C. J.; Ozols, J. Structural characteristics of prostaglandin synthetase from sheep vesicular gland. J. Biol. Chem. 1980,255, 1301-1304.

50. Kulmacz, R. J. Topography of prostaglandin H synthase. J. Biol. Chem. 1989,264, 14136-14144.

51. Triozzi, P. L.; Ailabouni, J.; Rinehart, J. J.; Witiak, D. T. aci- Reductones enhance interleukin-2-induced lymphocyte cytotoxicity . Int. J. Immunopharmac. 1993, 15, 47-54.

52. Gund, P.; Shen, T. Y. A model for the prostaglandin synthetase cyclooxygenase site and its inhibition by antiinflammatory arylacetic acids . J. Med. Chem. 1977, 20, 1146-1152.

53. Shen, T. Y.; Winter, C. A. Chemical and biological studies on indomethacin, sulindac and their analogs . Adv. Drug Res. 1977, 12, 89-245. 163

54. Cohen, N. C. Drug design in three dimensions . Adv. Drug Res. 1985, 14, 41-145.

55. Hall, S. E.; Han, W. C.; Haslanger, M. F.; Harris, D. N.; Ogletree, M. L. 9,11 -Epoxy-9-homo-14-oxaprosta-5-enoic acid derivatives. Novel inhibitors of fatty acid cyclooxygenase . J. Med. Chem. 1986, 29, 2335-2347.

56. Pal, I.; Odenwalter, R.; Marnett, L. J. 7-Oxabicycloheptylprostanoic acids: potent time-dependent cyclooxygenase inhibitors that induce a conformational change in the prostaglandin endoperoxide synthase protein . J. Med. Chem. 1992, 35, 2340- 2342.

57. Harris, D. N.; Phillips, M. B.; Michel, I. M.; Goldenberg, H. J.; Steinbacher, T. E.; Ogletree, M. L.; Hall, S. E. Inhibition of prostaglandin biosynthesis by SQ 28,852, a 7- oxabicyclo[2.2.1]heptane analog . Prostaglandins 1986, 31, 651- 667.

58. Lands, W. E. M.; LeTellier, P. R.; Rome, L. H.; Vanderhoek, J. Y. Inhibition of prostaglandin biosynthesis In Adv. Biosci. ed.; Bergstrom, S., Ed.; Permagon Press: Vieweg, 1973; Vol. 9, pp 15- 28.

59. Lelievre, V.; Martin, B.; Bure, J.; Junien, J. L. Direct inhibition of cyclooxygenase pathway in platelets by tixocortisol 21-pivalate: comparison with related structures, steroidal and non-steroidal antiinflammatory drugs . Agents and Actions 1988, 24, 179-186.

60. Payne, T. G.; Dewald, B.; Siegel, H.; Gubler, H. U.; Ott, H.; Baggiolini, M. Radical scavenging and stimulation of prostaglandin synthesis not anti-inflammatory? . Nature 1982, 296, 160-162.

61. Juby, P. F.; Goodwin, W. R.; Hudyma, T. W.; Partyka, R. A. Antiinflammatory activity of some indan-1-carboxylic acids and related compounds. J. Med. Chem. 1972, 15, 1297-1306.

62. Caldwell, J.; Hutt, A. J.; Fournel-Gigleux, S. The metabolic chiral inversion and dispositional enantioselectivity of the 2-arylpropionic acids and their biological consequenses . Biochemical Pharmacology 1988, 37, 105-114.

63. Ku, E. C.; Wasvary, J. M. Inhibition of prostaglandin synthase by pirprofen studies with sheep seminal vesicle enzyme. Biochim. Biophys. Acta 1975, 384, 360-368. 164

64. Appleton, R. A.; Brown, K. Conformational requirements at the prostaglandin cyclooxygenase receptor site: a template for designing non-steroidal anti-inflammatory drugs . Prostaglandins 1979, 18, 29-34.

65. Hutt, A. J.; Caldwell, J. The metabolic inversion of 2-arylpropionic acids- a novel route with pharmacological consequences. . J. Pharm. Pharmacol. 1983, 35, 693-704.

66. Mills, R. F. N.; Adams, S. S.; Cliffe, E. E.; Dickinson, W.; Nicholson, J. S. The metabolism of ibuprofen. . Xenobiotica 1973, 3, 589-595.

67. Wechter, W. J.; Loughhead, D. J.; Reischer, R. J.; Van Giessen, G. J.; Kaiser, D. J. Enzymatic inversion at saturated carbon: nature and mechanism of the inversion of R(-)p-isobutylhydratropic acid . Biochem. Biiophys. Res. Comm. 1974, 61, 833-837.

68. Lennard, M. S.; Tucker, G. T.; Woods, H. F. In Biopharmaceutics Taylor, J. B., Ed.; Permagon Press: Oxford, 1992; Vol. 5, pp 196- 197.

69. Samuelsson, B.; Funk, C. D. Enzymes involved in the biosynthesis of leukotriene B4. J. Biol. Chem. 1989, 264, 19469-19472.

70. Rouzer, C. A. Regulation and structure of human leuckocyte 5- lipoxygenase In Lipoxygenases and their Products ed.; Crook, S. T.. Wong, A Academic Press Inc.: San Diego, 1991; pp 51-65.

71. Percival, M. D. Human 5-lipoxygenase contains an essential iron. J. Biol. Chem. 1991, 266, 10058-10061.

72. Falgueyret, J. P.; Hutchinson, J. H.; Riendeau, D. Criteria for the identification of non-redox inhibitors of 5-lipoxygenase . Biochemical Pharmacology 1993, 45, 978-981.

73. Navaratnam, S.; Feiters, M. L.; Al-Hakim, M.; Allen, G. A.; Veldink, G. A.; Vliegenthart, J. F. G. Iron environment in soybean lipoxygenase. Biochim. Biophys. Acta 1988, 956, 70-76.

74. Steczo, J.; Donoho, G. P.; Clemens, J. C.; Dixons, J. E.; Axelrod, B. Conserved histidine residues in soybean lipoxygenase: functional consequences of their replacement. Biochemistry 1992, 31, 4053- 4057.

75. Zhang, Y. Y.; Radmark, O.; Samuelsson, B. Mutagenesis of some conserved residues in human 5-lipoxygenase: Effects on enzyme activity. Proc. Natl. Acad. Sci. USA. 1992, 89, 485-489. 165

76. Percival, M. D.; Ouellet, M. The characterization of 5 histidine- serine mutants of human 5-lipoxygenase . Biochem. Biophys. Res. Commun. 1992, 186, 1265-1270.

77. Zhang, Y. Y.; Lind, B.; Radmark, O.; Samuelsson, B. Iron content of human 5-lipoxygenase, effects of mutations regarding conserved histidine residues . J. Biol. Chem. 1993, 268, 2535-2541.

78. Rouzer, C. A.; Samuelsson, B. On the nature of the 5-lipoxygenase reaction in human leuckocytes: enzyme purification and requirement for multiple stimulatory factors . Proc. Natl. Acad. Sci. USA. 1985, 82, 6040-6044.

79. Rouzer, C. A.; Shimizu, T.; Samuelsson, B. On the nature of the 5- lipoxygenase reaction in human leuckocytes: Characterization of a membrane associated stimulatory factor . Proc. Natl. Acad. Sci. USA. 1985, 82, 7505-7509.

80. Dixon, R. A. F.; Diehl, R. E.; Opas, E.; Rands, E.; Vickens, P. J.: Evans, J. F.; Gillard, J. W.; Miller, D. K. Requirement of a 5- lipoxygenase-activating protein for leukotriene synthesis . Nature 1990, 343 , 282-284.

81. Corey, E. J.; Lansbury, P. T. Stereochemical course of 5- lipoxygenation of arachidonate by rat basophil leukemic cell (RBL- 1) and potato enzymes. J. Am. Chem. Soc. 1983, 105, 4093.

82. Lagutina I. O.; Sudina, G. F.; Kobelkor, V. S. D. Regulation of the activity of 5-lipoxygenase from potato tubers under the action of temperature. Biochemistry USSR 1990, 55, 520-525.

83. Ford-Hutchinson, A. W. Potential and therapeutic value of development of novel 5-lipoxygenase inhibitors. In Lipoxygenases and their Products Crook, S. T., Wong, A., Ed.; Academic Press Inc.: San Diego, 1991, pp 137-160.

84. Rokach, J.; Young, R. N. The development of new anti leukotriene drugs: specific leukotriene D4 antagonists and 5-lipoxygenase inhibitors in renal eicosanoids In Adv. Expt. Med. Biol. Dunn, M.J. Patrono, C.Cinotti, G.A. Ed.; Plenum Press: New York, 1989; Vol. 259.

85. Ueda, N.; Kaneko, S.; Yoshimoto, T.; Yamamoto, S. Purification of arachidonate 5-lipoxygenase from porcine leuckocytes and its reactivity with hydroperoxyeicosatetraenoic acids . J. Biol. Chem. 1988,261, 7982-7988. 166

86. Samuelson, B. Leukotrienes: mediators of immediate hypersensitivity reactions and inflammation . Science 1983, 220, 568-575.

87. Hammarstrom, S. Selective inhibition of platelet n-8 lipoxygenase by 5,8,11-eicosatriynoic acid. Biochim. Biophys. Acta. 1977, 487, 517-519.

88. Sok, D. E.; Han, C. Q.; Pai, J. K.; Sih, C. J. Inhibition of leukotriene biosynthesis by acetylenic analogues. Biochem. Biophys. Res. Commun. 1982, 107, 101-108.

89. Corey, E. J.; Munroe, J. E. Irreversible inhibition of prostaglandin and leukotriene biosynthesis from arachidonic acid by 11,12- dehydro and 5,6-dehydroarachidonic acids, respectively. J. Am. Chem. Soc. 1982, 104, 1752-1754.

90. Corey, E. J.; Lansbury, P. T., Jr.; Cashman, J. R.; Kantner, S. S. Mechanism of the irreversible deactivation of arachidonate 5- lipoxygenase by 5,6-dehydroarachidonate . J. Am. Chem. Soc. 1984, 106, 1501-1503.

91. Corey, E. J.; Kantner, S. S.; Lansbury, P. T., Jr. Irreversible inhibition of the leukotriene pathway by 4,5-dehydroarachidonic acid. Tett. Lett. 1983, 24, 265-268.

92. Patterson, J. W., Jr.; Pfister, J. R.; Wagner, P. J.; Murthy, D. V. K. Synthesis of 4,5,8-eicosatrienoic acids. J. Org. Chem. 1983, 48, 2572-2576.

93. Schlosser, M. The stereochemistry of the Wittig reaction. In Topics in stereochemistry Eliel, E. L., Allinger, N. L., Ed.; Wiley- Interscience: 1970; Vol. 5. 1-30.

94. Sun, F. F.; McGuire, J. C.; Morton, D. R.; Pike, J. E.; Sprecher, H.; Kunau, W. H. Inhibition of platelet arachidonic acid 12- lipoxygenase by acetylenic acid compounds. Prostaglandins 1981, 21, 333-343.

95. Corey, E. J.; Cashman, J. R.; Eckrich, T. M.; Corey, D. R. A new class of irreversible inhibitors of leukotriene biosynthesis. J. Am. Chem. Soc. 1985, 107, 713-715.

96. Corey, E. J.; Cashman, J. R.; Kantner, S. S.; Wright, S. W. Rationally designed potent, competitive inhibitors of leukotriene biosynthesis . J. Am. Chem. Soc. 1984, 106, 1503-1504. 167

97. Corey, E. J.; d'Alarcao, M.; Kyler, K. S. Synthesis of new lipoxygenase inhibitors 13-thia and 10-thiaarachidonic acids. Tett. Lett. 1985, 26, 3919-3922.

98. Perchonock, C. D.; Finkelstein, J. A.; Uzinskas, I.; Gleason, J. G.; Sarau, H. M.; Cieslinski, L. B. Dimethyleicosatrienoic acids: inhibitors of the 5-lipoxygenase enzyme. Tett. Lett. 1983, 24, 2457- 2460.

99. Pfister, J. R.; Murthy, D. V. K. Synthesis of three potential inhibitors of leukotriene biosynthesis. J. Med. Chem. 1983, 26, 1099-1103.

100. Arai, Y.; Shimoji, K.; Konno, M.; Konishi, Y.; Okuyama, S.; Iguchi, S.; Hayashi, M.; Miyamoto, T.; Toda, M. Synthesis and 5- lipoxygenase inhibitory activities of eicosanoid compounds. J. Med. Chem. 1983, 26, 72-78.

101. Kolasa, T ; Brooks, D. W.; Bell, R. L.; Lanni, C. Substrate-derived N- hydroxyurea inhibitors of 5-lipoxygenase . Med. Chem. Res. 1993, 3, 96-108.

102. Kerdesky, F. A. J.; Holms, J. H.; Schmidt, S. P.; Dyer, R. D.; Carter, G. W. Eicosatetraenehydroxamates: inhibitors of 5-lipoxygenase. Tett. Lett. 1985, 26, 2143-2146.

103. Kreft, A. F.; Failli, A. A.; Musser, J. H.; Kubark, D. M.; Banker, A. L.; Steffan, R.; Demerson, C. A.; Nelson, J. A.; Shah, U. S.; Gray, W.; Marshall, L. A.; Holloway, D.; Sturm, R.; Carlson, R. P.; Berkenkopf, J.; Grimes, D.; Weichman, B. M.; Chang, J. Y. Conversion of a cyclooxygenase (CO) inhibitor into a 5-lipoxygenase (LO) inhibitor: A general route to novel orally active anti-inflammatory and anti allergy drugs . Drugs Exptl. Clin. Res. 1991, XVII, 381-387.

104. Rouzer, C. A.; Ford-Hutchinson, A. W.; Morton, H. E.; Gillard, J. W. MK-886, a potent and specific leukotriene biosynthesis inhibitor blocks and reverses the membrane association of 5-lipoxygenase in ionophore challenged leukocytes . J. Biol. Chem. 1990, 265, 1436-1442.

105. Musser, J. H.; Kreft, A. F. 5-Lipoxygenase: properties, pharmacology and the quinolinyl(bridged)aryl class of inhibitors . J. Med. Chem. 1992, 35, 2501-2524.

106. Mancini, J. A.; Abramovitz, M.; Cox, M. E.; Wong. E.; Charleson, S.; Perrier, H.; Wang, Z.; Prasit, P.; Vickers, P. J. 5-Lipoxygenase activating protein is an arachidonate binding protein . FEBS-Lett. 1993, 318, 277-281. 168

107. Gilliard, J. W.; Dixon, R.; Ethier, D.; Evans, J.; Ford-Hutch in son, A. W.; Fortin, R.; Girard, Y.; Guindon, Y.; Hamel, P.; Jones, T.; Leveille, C.; Lord, A.; Miller, D.; Morton, H.; Rouzer, C.; Yoakim, C. Therapeutic potential of 5-lipoxygenase inhibitors: the discovery and development of MK-886, a novel mechanism leukotrine inhibitor In Inflammation and Drug Therapy Series', Rainsford, K.D. Velo, G.P. Ed. Kluwer Academic Publishers: Boston, 1992; pp 275- 286.

108. Mancini, J. A.; Prasit, P.; Coppolino, M. G.; Charleson, P.; Leger, S.; Evans, J. F.; Gillard, J. W.; Vickers, P. J. 5-Lipoxygenase-activating protein is the target of novel hybrid of two classes of leukotriene biosynthesis inhibitors . Molecular Pharmacology 1992, 41, 267- 272.

109. Hutchinson, J. H.; Prasit, P.; Choo, L. Y.; Riendeau, D.; Charleson, S.; Evans, J. F.; Piechuta, H.; Ball, R. G. Development of L-689,065 - the prototype of a new class of potent 5-lipoxygenase inhibitors . Biomed. Chem. Lett. 1992, 2, 1699-1702.

110. Hutchinson, J. H.; Riendeau, D.; Brideau, C.; Chan, C.; Delorme, D.; Denis, D.; Falgueyret, J. P : Fortin, R.; Guay, J.; Hamel, P.; Jones, T. R.; Macdonald, D.; McFailane, C. S.; Piechuta, H.; Scheigetz, J.; Tagari, P.; Therien, M.; Girard, Y. Substituted thiopyrano[2,3,4- c,d] in doles as potent, selective and orally active inhibitors of 5- lipoxygenase. synthesis and biological evaluation of L-691,816 . J. Med. Chem. 1993, 36, 2771-2787.

111. Cashman, J. R. Leukotriene biosynthesis inhibitors . Pharm. Res. 1985, 6, 253-261.

112. Summers, J. B.; Mazdiyasni, H.; Holms, J. H.; Ratajczyk, J. D.; Dyer, R. D.; Carter, G. W. Hydroxamic acid inhibitors of 5-lipoxygenase . J. Med. Chem. 1987, 30, 574-580.

113. Nigam, S. S.; Weedon, B. C. L. Unsaturated fatty acids. Part II. The synthesis of linolenic acid . J. Chem. Soc. 1956, 4049-4054.

114. Oliver, T. R.; Waters, W. A. The mechanism of oxidation of hydroxamic acids by alkaline ferricyanide . J. Chem. Soc. B. 1971, 677-681.

115. Flynn, D. L.; Capiris, T.; Cetenko, W. J.; Connor, D. T.; Dyer, R. D.; Kostlan, C. R.; Nies, D. E.; Schrier, D. J.; Sircar, J. C. Non-steroidal anti-inflammatory drug hydroxamic acids. Dual inhibitors of both cyclooxygenase and lipoxygenase . J. Med. Chem. 1990, 33, 2070-2072. 169

116. Flynn, D. L.; Belliolti, T. R.; Boctor, A. M.; Connor, D. T.; Kostlan, C. R.; Neis, D. E.; Ortwine, D. F.; Schrier, D. J.; Sircar, J. C. Styrylpyrazoies, styrylisoxazoles and styrylisothiazoles. Novel 5- lipoxygenase and cyclooxygenase inhibitors . J. Med. Chem. 1991, 34, 518-525.

117. Summers, J. B.; Kim, K. H.; Mazdiyasni, H.; Holms, J. H.; Ratajczyk, J. D.; Stewart, A. O.; Dyer, R. D.; Carter, G. W. Hydroxamic acid inhibitors of 5-lipoxygenase: quantitative structure-activity relationships . J. Med. Chem. 1990, 33, 992-998.

118. Hansch, C.; Klein, T. E. Molecular graphics and QSAR in the study enzyme-ligand interactions. On the definition of bioreceptors. Acc. Chem. Res. 1986, 19, 392.

119. Summers, J. B.; Gunn, B. P.; Mazdiyasni, H.; Goetze, A. M.; Young, P. R.; Bouska, J. B.; Dyer, R. D.; Brooks, D. W.; Carter, G. W. in Vivo characterization of hydroxamic acid inhibitors of 5-lipoxygenase . J. Med. Chem. 1987, 30, 2121-2126.

120. Summers, J. B.; Gunn, B. P.; Martin, J. G.; Mazdiyasni, H.; Stewart, A. O.; Young, P. R.; Goetze, A. M.; Bouska, J. B.; Dyer, R. D.; Brooks, D. W.; Carter, G. W. Orally active hydroxamic acid inhibitors of leukotriene biosynthesis . J. Med. Chem. 1988, 31, 5-9.

121. Summers, J. B.; Gunn, B. P.; Jonathan, G. M.; Martin, M. B.; Hormoz, M.; Stewart, A. O.; Young, P. R.; Bouska, J. B.; Goetze, A. M.; Dyer, R. D.; Brooks, D. W.; Carter, G. W. Structure-activity analysis of a class of orally active hydroxamic acid inhibitors of leukotriene biosynthesis . J. Med. Chem. 1988, 31, 1960-1964.

122. Bhattacherjee, P.; Boughton-Smith, K.; Follenfant, R. L.; Garland, L. G.; Higgs, A.; Hodson, H. F.; Islip, P. J.; Jackson, W. P.; Moncada, S.; Payne, A. N.; Randall, R. W.; Reynolds, C. H,; Salmon, J. A.; Tateson, J. E.; Whittle, B. J. R. The effects of a novel series of selective inhibitors of arachidonate 5-lipoxygenase on anaphylactic and inflammatory responses In B io lo g y o f Leukotrienes. Annals of the New York Academy of Sciences; Levi, R. Krell, R.D. Ed. 1988; Vol. 524, pp 307-320.

123. Garland, L. G.; Salmon, J. A. Hydroxamic acids and hydroxyureas as inhibitors of arachidonate 5-lipoxygenase . Drugs Future 1991, 16, 547-558.

124. Summers, J. B. European Patent Application Patent 0279281, 1988; Chem. Abstr. 03-02-1988, 170

125. Bell, R. L.; Young, P. R.; Lanni, A. C.; Summers, J. B.; Brooks, D. W.; Rubin, P.; Carter, G. W. The discovery and development of zileuton: an orally active 5-lipoxygenase inhibitor , Int. J. Immunopharmac, 1992, 14, 505-510.

126. Braeckman, R. A.; Granneman, G. R.; Rubin, P. R.; Kesterson, J. W. Pharmacokinetics and metabolism of new 5-lipoxygenase inhibitor A-64077 after single oral administration in man . J. Clin. Pharmacol. 1989, 29, 837.

127. Nelson, M. J.; Batt, D. G.; Thompson, J. S.; Wright, S. W. Reduction of active-site iron by potent inhibitors of lipoxygenase . J. Biol.Chem. 1991, 266, 8225-8229.

128. Batt, D. G. 5-Lipoxygenase inhibitors and their anti-inflammatroy activities . Prog. Med. Chem. 1992, 29, 1-63.

129. Fitzsimmons, B. J.; Rokach, J. Enzyme inhibitors and leukotriene receptor antagonists In Leukotrienes and Lipoxygenases Rokach, J., Ed.; Elsevier: Amsterdam, 1989, pp 427-502.

130. Van der zee, J.; Eling, T. E.; Mason, R. P. Formation of free radical metabolites in the reaction between soybean lipoxygenase and its inhibitors . Biochemistry 1989, 28, 8363-8367.

131. Bruneau, P.; Delvare, C. Indazolinones, a new series of redox- active 5-lipoxygenase inhibitors with built-in selectivity and oral activity . J. Med. Chem. 1991, 34, 1028-1036.

132. Bird, T. G. C.; Bruneau. P.; Crawley, G. C.; Edwards, M. P.; Foster, S. J.; Girdeau, J. M.; Kingston, J. F.; McMillan, R. M. (Methoxyalkyl) thiazoles: a new series of potent, selective and orally active 5- lipoxygenase inhibitors displaying high enantioselectivity . J. Med. Chem. 1991, 34, 2176-2186.

133. Crawley, G. C.; Briggs, M. T.; Dowell, R. I.; Edwards, P. N.; Hamilton, P. M.; Kingston, J. F.; Oldham, K.; Waterson, D.; Whalley, D. P. 4-Methoxy-2-methyltetrahydropyrans: chiral leukotriene biosynthesis inhibitors, related to ICI D138, which display enantioselectivity . J. Med. Chem . 1993, 36, 295-296.

134. Hsiao, C. N.; Kolasa, T. Synthesis of chiral zileuton, a potent and selective inhibitor of 5-lipoxygenase . Tett. Lett. 1992, 33, 2629- 2632.

135. Ireland, R. E.; Varney, M. D. Approach to the total synthesis of chlorothricolide: synthesis of (+)-19,20-dihydro-24-O- 171

methylchlorothricolide, methyl ester, ethyl carbonate . J. Org. Chem. 1986, 51, 635-648.

136 Ireland, R. E.; Thompson, W. J. An approach to the total synthesis of ch loroth ricolide: the synthesis of the top half. J. Org. Chem. 1979, 44, 3041-3052.

137. Witiak, D. T.; Kodrady, S. S.; Patel, S. T.; Huzoor-Akbar; Feller, D. R. Hypocholesterolemic and antiaggregatory properties of 2- hydroxytetronic acid redox analogues and their relationship to clofibric acid . J. Med. Chem. 1982, 25, 90-93.

138. Witiak, D. T.; Feller, D. R.; Newman, H. A. I.; Kim, S. U.; Esbenshade, T. A. Medicinal chemistry aspects of antilipidemic drugs: ac/-reductone antlipidemic and antiaggregatory agents . Actual. Chem. Ther. 1988, 15, 41-62.

139. Witiak, D. T.; Kim, S. K.; Tehim, A. K.; Sternitzke, K. D.; McCreey, R. L.; Kim, S. U.; Feller, D. R.; Romstedt, K. J.; Kamanna, V. S.; Newman, H. A. I. Synthetic ac/-reductones: 3,4-dihydroxy-2H-1- benzopyran-2-ones and their cis- and frans-4a,5,6,7,8,8a- hexahydro diastereomers. antiaggregatory, antilipidemic, and redox properties compared to those of the 4-substituted 2 - hydroxytetronic acids . J. Med. Chem. 1988, 31, 1437-1445.

140. Witiak, D. T.; Feller, D. R.; Newman, H. A. I. Chemical approaches to the treatment of atherosclerosis In Trends in Medicinal Chemistry '90 Sarel, S., Mechoulam, R., Agranat, I., Ed.; Blackwell Scientific Publications: Oxford, 1990, pp 243-256.

141. Witiak, D. T.; Kim, S. K.; Romstedt, K.; Newman, H. A. I.; Feller, D. R. Comparitive antiaggregatory activity in human platelets of a benzopyranone ac/'-reductone, clofibric acid, and a 2,3- dihydrobenzofuran analogue . J. Med. Chem. 1986, 29, 2170- 2174.

142. Romstendt, K.; Feller, D. R. ; Witiak, D. T. Unpublished results.

143. Witiak, D. T.; Triozzi, P. L.; Feller, D. R.; Romstedt, K. J.; Tehim, A. K.; Hopper, A. T.; Mantri, P. “4-Substituted-2-hydroxytetronic acid aci-reductones improve lymphokine-activate killer cell activity in vitro," American Association for Cancer Research meeting, Florida, 1993.

144 Danh, H.; Lawendel, J.S.; Hoegger, E.F.; Fischer, R. Schenker, E. A new preparation of aromatically substituted reductones Experientia 1954, 10, 245-246. 172

145. Witiak, D. T.; Tehim, A. K. Efficient synthesis of optically pure stereogenically labile 4-substituted-2-hycfroxytetronic acids . J. Org. Chem. 1990, 55, 1112-1114.

146. Warren, S. In Organic synthesis: The disconnection approach. John Wiley and Sons, Chichester 1989.

147. Anantharamaiah, G. M.; Sivanandaiah, K. M. Transfer hydrogenation: a convenient way for removal of some commonly used protecting groups in peptide synthesis . J. Chem. Soc. Perkin I Trans. 1977, 490-491.

148. Wasserman, H. H.; Lipshutz, B. H. Reactions of lithium enolates with molecular oxygen a-hydroxylation of amides and other carboxylate derivatives . Tett. Lett. 1975, 1731-17734.

149. Collum, D. B. Total synthesis of monensin. Ph.D. thesis , Columbia University, 1980.

150. Sterling, J.; Slovin, E.; Barasch, D. Use of malic acid as a chiral synthon: 24,25-dihydroxycholecalciferol . Tett. Lett. 1987, 28, 1685-1688.

151. Still, W. C., Columbia University, personal communication, 1990.

152. Collum, D. B.; McDonald, I., J.H.; Still, W. C. Synthesis of the polyether antibiotic monensin. 2. preparation of intermediates . J. Am. Chem. Soc. 1980, 102, 2118-2120.

153. Bailey, W. F.; Fujiwara, E. Acetylenes I. Mixed dihalides and halohydrins from butynediol. J. Am. Chem. Soc. 1955, 77, 1955.

154. Ege, S. N.; Wolovsky, R.; Gensler, W. J. Synthesis of methyl 5,8,11,14-eicosatetraenoate (methyl arachidonate) . J. Am. Chem. Soc. 1961, 83. 3080-3085.

155. Osbond, J. M.; Philpott, P. G.; Wickens, J. C. Essential fatty acids. Part I. Synthesis of linoleic, y-linolenic, arachidonic and docosa- 4,7,10,13,16-pentaenoic acid . J. Chem. Soc Chem Comm. 1961, 2779-2787.

156. Lindlar, H.; Dubuis, R. Palladium catalyst for partial reduction of acetylenes . Organic Synthesis 46, 89-92.

157. Tai, H. H.; Hsu, C. T.; Tai, C. L.; Sih, C. J. Synthesis of [5,6-3H]- arachidonic acid and its use in development of a sensitive assay for prostacyclin synthetase . Biochemistry 1980, 19, 1989-1993. 173

158. Sprecher, H. The total synthesis and metabolism of 7,10,13,16- docosatetraenoate in the ra t. Biochim. Biophys. Acta. 1967, 144, 296-304.

159. Sprecher, H. The synthesis of 1- 14C-arachidonate and 3- 1 4 C- docosa-7,10,13,16-tetraenoate . Lipids 1971 , 6, 889-894.

160. Sprecher, H. The organic synthesis of unsaturated fatty acids . Prog. Chem. Fats other Lipids 1978, 15, 219-254.

161. Brown, H. C.; Campbell, J. B. Synthesis and applications of vinylic organoboranes . Aldrichimica Acta. 1981, 14, 3-11.

162. Ravid, U.; Silverstein, R. M.; Smith, L. R. Synthesis of the enantiomers of 4-substituted y-lactones with known absolute configuration . Tetrahedron 1978, 34, 1449-1452.

163. Posner, G. H.; Ting, T. S.; Lentz, C. M. A mechanistic and synthetic study of organocopper substitution reactions with some homoallylic and cyclopropylcarbinyl substrate . Tetrahedron 1976, 32, 2281- 2287.

164. Guindon, Y.; Zamboni, R.; Lau, C. K.; Rokach, J. Stereospecific synthesis of leukotriene B 4 . Tett. Lett. 1982, 23, 739-742.

165. Novis Smith, W.; Beumel, O. F. Preparation of alkynes and dialkynes by reaction of monohalo- and dihaloalkanes with lithium acetylide-ethylenediamine complex . Synthesis 1974, 441-442.

166. Lipshutz, B. H.; Kozlowski, J.; Wilhelm, R. S. Chemistry of higher order mixed organocuprates. 2. Reactions of epoxides . J. Am. Chem. Soc. 1982, 104, 2305-2307.

167. Lipshutz, B. H.; Wilhelm, R. S.; Kozlowski, J. A. The chemistry of higher order organocuprates . Tetrahedron 1984, 40, 5005-5038.

168. Johnson, C. R.; Dutra, G. A. Reactions of lithium diorganocuprates with tosylates. I. Synthetic aspects. . J. Am. Chem. Soc. 1973, 95, 7777-7782.

169. Kotsuki, H.; Kadota, I.; Ochi, M. A Novel carbon-carbon bond forming reation of triflates with copper(l)-catalyzed Grignard reagents. A new concise and enantiospecific synthesis of (+)-axo- brevicomin, (5R6S)-{-)-6-acetoxy-5-hexadecanolide, and L-factor . J. Org. Chem. 1990, 55, 4417-4422. 174

170. Winterfeldt, E. Applications of diisobutylaluminium hydride (DIBAH) and triisobutylaluminium (TIBA) as reducing agents in organic synthesis . Synthesis 1975, 617-630.

171. Greene, T. W.; Wuts, P. G. M. In Protective groups in organic synthesis 2nd ed.; John Wiley and sons: New York, 1991;.

172. Hassner, A.; Alexanian, V. Direct room temperature esterification of carboxylic acids . Tett. Lett. 1978, 46, 4475-4478.

173. Hamada, Y.; Kawai, A.; Matsui, T.; Hara, O.; Shioiri, T. 4- Alkoxycarbonyloxazoles as p-hydroxy-a-amino acid synthons: efficient, stereoselective synthesis of 3-amino-2,3,6- trideoxyhexoses and a hydroxyamino acid moiety of AI-77-B . Tetrahedron 1990, 46, 4823-4846.

174. Collins, J. C.; Hess, W. W.; Frank, F. J. Dipyridine-chromium(VI) oxide oxidation of alcohols in dichloromethane . Tett. Lett. 1968, 3363-3366.

175. Corey, E. J.; Schmidt, G. Useful procedures for the oxidation of alcohols involving pyridinium dichromate in aprotic media . Tett. Lett. 1979, 399-402.

176. Ratcliffe, R.; Rodehorst, R. Improved procedure for oxidations with the chromium trioxide-pyridine complex . J. Org. Chem. 1970, 35, 4000-4002.

177. Levine, S. G. A new aldehyde synthesis . J. Am. Chem. Soc. 1958, 80, 6150-6151.

178. Newton, R. F.; Wadsworth, A. H. Synthesis of stable prostacyclin analogues from 2,3-disubstituted bicyclo[3.2.0]heptan-6-ones . J. Chem. Soc. Perkin Trans. 11982, 823-830.

179. Zhdanov, Y. A.; Alexeeva, V. G. A novel synthesis of 2- deoxyhexoses . Carbohyd. Res. 1969, 10, 184-185.

180. Bachmann, W. E.; Struve, W. S. The Arndt-Eistert synthesis . Organic Reactions 1942, /. 38-55.

181. S tow ell, J. C.; Keith, D. R. 2-(1,3-Dioxan-2-yl)- ethylidenetriphenylphosphorane, a new three-carbon homologating agent. Synthesis 1979, 132-133. 175

182. Dauben, W. G.; Gerdes, J. M.; Bunce, R. A. Organic reactions at high pressure. Preparation of Wittig phosphonium salts at ambient temperature . J. Org. Cham. 1984, 49, 4293-4295.

183 Witiak, D.T.; Tehim, A.K. Synthetic approaches to 4-spiro-2- hydroxytetronic acids J. Org. Chem. 1987, 52, 2324-2327.

184. Bergelson, L. D.; Shemyakin, M. M. Synthesis of naturally occuring unsaturated fatty acids by sterically controlled carbonyl olefination . Angew. Chem. Int. Ed. 1964, 3, 250-260.

185. Cohen, N.; Banner, B. L ; Lopresti, R. J.; Wong, F.; Rosenberg, M.; Liu, Y. Y.; E., T.; Liebman, A. A. Enantiospecific syntheses of leukotrienes C 4, D4, and [14,15-3H2] leukotriene E 4 dimethyl ester. J. Am. Chem. Soc. 1983, 105, 3661-3672.

186. Corey, E. J.; Schaaf, T. K.; Huber, W.; Koelliker, U.; Weinshenker, N. M. Total synthesis of prostaglandins F2a and E 2 as the naturally occuring forms. J. Am. Chem. Soc. 1970, 92, 397-398.

187. Corey, E. J.; Shirahama, H.; Yamamoto, H.; Terashima, S.; Venkateswarlu, A.; Schaaf, T. Stereospecific total synthesis of prostaglandins E 3 and F3or J. Am. Chem. Soc. 1971, 93, 1490- 1491.

188. Jain, S. C.; Dussourd, D. E.; Conner, W. E.; Eisner, T.; Guerrero, A.; Meinwald, J. Polyene pheromone components from an arctid moth {Utetheisa omatrix): characterization and synthesis . J. Org. Chem. 1983, 48, 2266-2270.

189. Johnson, W. S.; Semmelhack, M. F.; Sultanbawa, M. U. S.; Dolak, L. A. A new approach to steroid total synthesis. A nonenzymic biogenetic-like olefinic cyclization involving the stereospecific formation of five asymmetric centers . J. Am. Chem. Soc. 1968, 90, 2994-2996.

190. Maercker, A. The Wittig reaction . Organic reactions 1965, 14, 270- 401.

191. Kwok, P. Y.; Muellner, F. W.; Chen, C. K.; Fried, J. Total synthesis of 7,7- 10,10- and 13,13-difluoroarachidonic acids . J. Am. Chem. Soc. 1987, 109, 3684-3692.

192. Marshall, D.; Whiting, M. C. Researches on polyenes. Part IV. The (o-p-methoxyphenyl-polyene aldehydes and their absorption spectra . J. Chem. Soc. 1956, 4082-4088. 176

193. Mori, K.; Ebata, T. Synthesis of optically active pheromones with an epoxy ring, (+)-disparlure and both the enantiomers of (3Z,6Z)-c/s- 9,10-epoxy-3,6-hexeicosadiene . Tetrahedron 1986, 42, 3471- 3478.

194. Rokach, J.; Zamboni, R.; Lau, C. K.; Guindon, Y. The stereospecific synthesis of leukotriene A 4 (LTA4), 5-epi-LTA4, 6 -epi-LTA4 and 5- epi, 6 -epi-LTA4 . Tett. Lett. 1981, 22, 2759-2752.

195. Kuji, K.; Ichikawa, K.; Node, M.; Fujita, E. Hard acid and soft nucleophile system. New efficient method for removal of benzyl protecting group . J. Org. Chem. 1979, 44, 1661-1664.

196. Liang, Y.T.S.; Liu, X.S.; Seib, P.A. Synthesis of D-erythroascorbic acid J. Carb. Chem. 1990, 9, 75-84.

197. Witiak, D. T.; Tehim, A. K. Synthetic approaches to 4-spiro-2- hydroxytetronic acids . J. Org. Chem. 1987, 52, 2324-2327.

198. McOmie, J. F. W.; Watts, M. L.; West, D. E. Demethylation of aryl methyl ethers by boron tribromide . Tetrahedron 1968, 24, 2289- 2292.

199. Wissner, A.; Grudzinskas, C. V. Reaction of te/t-butyldimethylsilyl esters with oxalyl chloride-dimethylformamide: preparation of carboxylic acid chlorides under neutral conditions . J. Org. Chem. 1978, 43, 3972-3974.

200. Evans, R. M.; Owen, L. N. Dithiols. Part II. Dimercaptopropyl ethers of glycerol and glycolic acid, with some further observations on “BAL-lntrav“ . J. Chem. Soc. 1949, 244-248.

201 Hershberg; Cason .The Rosenmund reduction . Org. Synth. 1941, 21, 84.

2 0 2 . Gent, P. A.; Gigg, R. The allyl ether as a protecting group in carbohydrate chemistry, isomerizations with fr/s-[triphenylphosphine]rhodium (I) chloride . J. Chem. Soc. Chem. Comm. 1974, 277-278.

203. Gigg, J.; Gigg, R. The allyl ether as a protecting group in carbohydrate chemistry . J. Chem. Soc. C. 1966, 82-86.

204. Gigg, R.; Warren, C. D. The allyl ether as a protecting group in carbohydrate chemistry . J. Chem. Soc. C1968, 1903-1911. 177

205. Boss, R.; Scheffold, R. Cleavage of allyl ethers with Pd/C . Angew. Chem. Int. Ed. 1976, 15, 558-559. 206. Boullanger, P.; Chatelard, P.; Descotes, G.; Klloosterman, M.; van Boom, J. H. Use of the allyloxycarbonyl protective group in carbohydrate chemistry . J. Cart). Chem. 1986, 5, 541-559.

207. Haines, L. M.; Singleton, E. Cationic complexes of iridium(l) containing phosphites, phosphines, or arsines as ligands . J. Chem. Soc. Dalton Trans.1972, 1891-1896.

208. Oltvoort, J. J.; van Boeckel, C. A. A.; de Koning, J. H.; van Boom, J. H. Use of the cationic iridium complex 1.5-cyclooctadiene- bis[methyldiphenylphosphine]-iridium hexafluorophosphate in carbohydrate chemistry: smooth isomerization of allyl ethers to 1- propenyl ethers . Synthesis 1981, 305-308.

209. Bondy, S.C.; LeBel, C.P.; The relationship between excitotoxicity and oxidative stress in the central nervous system. Free Radical Biology and Medicine 1993, 14 633-642.

210. van Acker, S.A.B.E.; Koymans, L.M.H.; Bast, A. Molecular pharmacology of vitamin E: structural aspects of antioxidant activity Free Radical Biology and Medicine 1993, 15 311-328.