INFORMATION TO USERS

This material was produced from a microfilm copy of the original document. While the moat advanced technological means to photograph and reproduce this document have bean used, the quality is heavily dependent upon the quality of the original submitted.

The following explanation of techniques is provided to help you understand markings or patterns which may appear on this reproduction.

1. The sign or "target" for pages apparently lacking from the document photographed is "Missing Paga(s)". If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting thru an image and duplicating adjacent pages to insure you complete continuity.

2. When an image on the film is obliterated with a large round black mark, it is an indication that the photographer suspected that the copy may have moved during exposure and thus cause a blurred image. You will find a good image of the page in the adjacent frame.

3. Whan a map, drawing or chart, etc., was part of the material being photographed the photographer followed a definite method in "sectioning" the material. It is customary to begin photoing at the upper left hand comer of a large sheet and to continue photoing from left to right in equal sections with a small overlap. If necessary, sectioning is continued again - beginning below the first row and continuing on until complete.

4. The majority of users indicate that the textual content is of greetest value, however, a somewhat higher quality reproduction could be made from "photographs" if essential to the understanding of the dissertation. Silver prints of "photographs" may be ordered at additional charge by writing the Order Department, giving the catalog number, title, author and specific pages you wish reproduced.

5. PLEASE NOTE: Some pages may have indistinct print. Filmed as received.

Xerox University Microfilms 300 North Z««b Road Am* Artor, Michigan 48106 75-26,549 BQKFLMAN, Gordon Herman, 1948- THE CHEMISTRY OF A MOLD METABOLITE. The Ohio State University. Ph.D., 1975 Chemistry, organic

XSTOX University Microfilms , Ann Arbor, Michigan 4«10fl

THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED. THE CHEMISTRY OF A MOLD METABOLITE

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

By

Gordon Herman Bokelman, A.B.

A A A A A

The Ohio State University

1975

Reading Committee: Approved by

Prof. J. L. Beal Prof. R. W. Doskotch Prof. L. A. Mitscher Adviser College of Pharmacy To Ann and Jennifer

i i ACKNOWLEDGMENTS

I would like to express my1 gratitude to my adviser,

Professor Lester A. Mitscher, for the opportunity to work on the challenging research projects which he envisioned.

Professor Mitscher has been a gifted teacher of the princi­ ples and philosophy of chemistry.

I am also indebted to Professors Jack L. Beal and

Raymond W. Doskotch for their contributions to my under­ standing of the field of natural products chemistry.

The friendship, inspiration, and contribution of the graduate students and postdoctoral fellows I have known are greatly appreciated.

Financial assistance from the Chemistry Department of The Ohio State University, Abbott Laboratories, N.I.H., and The American Foundation for Pharmaceutical Education is gratefully acknowledged. VITA

April 15, 1948 ...... Born-Bloomington, Indiana

1970 ...... A.B., Chemistry, Indiana University, Bloomington, Indiana.

1970- 1971...... Teaching Assistant, Chemistry Department, The Ohio State University, Columbus, Ohio.

1971-1975 ...... Research Associate, College of Pharmacy, The Ohio State University, Columbus, Ohio.

FIELDS OF STUDY

Major Field: Natural Products Chemistry

iv TABLE OF CONTENTS

Page

DEDICATION ...... ii

ACKNOWLEDGMENTS...... iii

VITA ...... iv

LIST OF FIGURES...... vii

INTRODUCTION ...... 1

The purpose of the research...... 31 Previous studies on terrein ...... 34

EXPERIMENTAL ...... 35

Methodology...... 35 Purity of reagents and solvents ...... 36 1-Iodo-7-octene (34i) ...... 36 7- Iodoheptanoic acid (.39)...... 3 8 7-Iodo-1-heptanol (4 0) First procedure ...... 38 Second procedure...... 39 THP-ether of 7-iodo-1-heptanol ( £ 1 ) ...... 40 Tetrakis [iodo (tri-n-butylphosphine) copper (I)] (£2) . 41 Attempted conjugate addition of THP-ether £1 to trans- 2-methyl- 2-pentenal...... 42 Attempted addition of 1-iodo-7-octene (_38) to benzophenone...... 43 7-Bromoheptanoic acid ( £ £ ) ...... 46 7-Bromo-1-heptanol (45) First procedure...... 47 Second procedure...... 47 THP-ether of 7-bromo-1 - heptanol (£6)...... 48 Attempted addition of THP-ether £6 to carbon dio x i d e ...... 49 Attempted addition of 2-chloro-1,1- diethoxyethane (48) to benzophenone ...... 51

v Page

Attempted addition of 2-bromo-l,l- diethoxyethane (4j)) to benzophenone...... 53 THP-ether of 2-bromoethanol (5_1)...... 55 Attempted reaction of THP-ether _51_ with lithium...... 56 Attempted conjugate addition of THP-ether 51 to trans-2-methyl-2-pentenal ...... 57 Trans-5-methyl-1,5-octadien-4-ol (53) ...... 59 Methyl 5-hexenoate (54) ...... 62 3-Carbomethoxy- 4 (S) ,1>TR) -dibenzoxy-2- cyclopenten-1 - one etFylenethioketal (56)...... 62 5 (R)-Benzoxv-3-carbomethoxy-2-(2-propenyT)- 7-cyclopcnten-1-one ethy lenethioket al (5^7). . . . 64 Terrein diacetate (58) First procedure...... 66 Second procedure...... 67 4 (S) ,5(R) -Diacetoxy-3-(trans-1-propenal)- 2- cyclopenten-1-o n e ...... 67 4 (S),5(R)-Diacetoxy-3-(3-hydroxy-trans-1- octenyl) - 2-cyclopenten-1-one (6TT) T ...... 69 Attempted thioketalization of ketone (60) ...... 71 4(S),5(R)-Diacetoxy-3-(3-[2-tetrahydro- pyranyloxy]-trans-1-octenyl)-2-cyclopenten- 1-one (6_1)...... 72 4 (S) ,5(S) - Diacetoxy-1-(3-[2-tetrahydropyranyloxy]- trans-1-octenyl) -cyclopenten- 3rQl (62)...... 74 3- (4 (S)", 5 (S) - Diacetoxy -1 - [ 5-hydroxy - trans -1- octenyl]-cyclopentenyl) ethyl malonate...... 76 Attempted manganese dioxide oxidation of alcohol (63)...... 77 Attempted biphasic chromic acid oxidation of alcohol (^3...... 78 4 (S),5(S)-Dibenzoxy-1-formyl-3(R)- Kydroxycyclopentene (65)...... 79 4 (S) ,5 (S) - Dibenzoxy - 3 (R^hydroxy-1- j3-oxo-trans-1-octenyl)-cyclopentene (_66).... 80 3(R)-(4(S),5(S)-Dibenzoxy-1-(3-oxo-trans- 1 - octenyl ]-cyclopentenyl) ethyl maTonate (67^) . • 81 Attempted cyclization of compound 6jf with disodium salicylate ...... 83 Attempted cyclization of compound §]_ with sodium methoxide...... 83 1 (R) , 5(R),8(R)-Benzoxy-4(R)-carbethoxy-6- X3-OXO-trans-1-octenyl)-3-oxo-2-oxabicyclo [3.3.0] oct-6-ene (6fJ)...... ' ...... 84

DISCUSSION...... 87

BIBLIOGRAPHY ...... 107

vi LIST OF FIGURES

Figure Page

1 Some representative prostaglandins ...... 6

2 Examples of stereochemical variations of the natural prostaglandins ...... 8

3 Prostaglandin biosyntheses ...... 11

4 Synthetic routes to prostaglandins from a cyclopentane nucleus ...... 13

5 Potential cyclization routes to prostaglandins ...... 15

6 The Just synthesis of dl-PGF^a (fO and dl-PGE^ methyl ester~T9) - Part I. . . . 16

7 The Just synthesis of dl-PGFia (8) and dl-PGEj methy ester * Part II. . . . 17

8 Upjohn modification of the Just synthesis ...... , 20

9' The Corey bicycloheptene route to natural prostaglandins - Part I...... 22

10 The Corey bicycloheptene route to natural prostaglandins - Part II . . . . 24

11 The Corey bicycloheptene route to natural prostaglandins - Part III. . . . 26

12 The Miyano synthesis of dl-PGE. and dl-PGF. - Part I...... 1 ...... 28 — la 13 The Miyano synthesis of dl-PGE. and dl-PGF. - Part I I ...... 29 — la 14 Conversion of terrein to one or more series of prostaglandins ...... 32

vii Figure Page

15 Laboratory glassware prepared for organolithium reactions ...... 45

16 Synthetic routes to the THP-ether of 7-iodo-1-heptanol ( 4 1 ) ...... 87

17 Attempted formation of organolithium compounds from primary alkyl iodides .... 89

18 Synthetic routes to the THP-ether of 7-bromo- 1-heptanol ( 4 j > ) ...... 90

19 Reactions between a 99:1 1ithium-sodium alloy and three primary alkylhalides. . . . 92

20 Proposed reaction between a 99:1 lithium- sodium alloy and 2-bromo or chloro-1,1- diethoxyethane ...... 94

21 Attempted formation of an organolithium reagent from the THP-ether of 2-bromoethanol (_51)...... 96

22 Reactions of diallyl lithium cuprate with a ,8-unsaturated carbonyl compounds . . 97

23 The synthesis of 4 (S)»5(R)-dibenzoxy- 3-formyl-2-cyclopenten-1-one ethylene- thioketal (jij>)...... 98

24 Conversion of compound 5j> to 5(R)-benzoxy- 3-carbomethoxy-2-(2-propenyl)-3-cyclo- penten-l-one ethylenethioketal (57^) .... 99

25 The synthesis of 3-(4(S),5(S)-diacetoxy-1- [3-hydroxy-trans-1-octenyT]-cyclopentenyl) ethyl malonate ( 6 3 ) ...... 101-2

26 The synthesis of 4 (S)p5(S)-dibenzoxy-1- (1f 2-dihydroxypropyl)- 3 (R)- hydroxy cyclopentene ( 6 4 ^ ) ...... 103

27 Conversion of compound 6£ to 1(R)P5(R),8(R)- benzoxy-4(R)-carbethoxy-6-(5-oxo-trans- 1-octenyl)- 3-oxo-2 -oxabicyclo [3.3.0]oct- 6-ene (68)...... 104

vi i i INTRODUCTION

In 1930 two gynecologists in New York, Raphael

Kurzrok and Charles C. Lieb (1), found that fresh human semen could produce contraction or relaxation of the human uterus. A few years later, physicians Maurice W.

Goldblatt (2) in England and Ulf S. von Euler (3) in Sweden independently conducted experiments with human semen and with extracts from the seminal vesicular gland of sheep.

They reported that these fluids contain a potent lipid substance which could stimulate smooth muscle tissue to contract or, upon injection, produce a sharp lowering of blood pressure in an animal. In the mistaken belief that the primary source of the active material was the prostate gland, von Euler named the substance "prostaglandin."

The next major development in this field occurred in 1957 when Bergstrom and his associates in Sweden iso­ lated and crystallized the first prostaglandins, PGE^ and

PGF. . These same workers determined the structures of la their prostaglandins by methods that included classical degradation and X-ray crystallographic studies on suitable derivatives (4-6).

The prostaglandins are now known to constitute a large series of unsaturated hydroxy fatty acids . These

1 2 hormone-like compounds display a wide diversity of bio­ logical effects (7-17). Prostaglandins affect the activity of smooth muscles, secretion (including some endocrine gland secretions), and blood flow. These individual effects, however, may be quite specific. For example, one prostaglandin (PGF- ) raises blood pressure, while another b Ct closely related prostaglandin (PGE^) lowers blood pressure.

The effects of the prostaglandins on the female

reproductive system are particularly striking. A very low dose of either PGE2 or PGF2a given intravenously stimulates uterine contractions, and thus may be used to facilitate

labor. Oral administration of these compounds is also

effective in inducing labor. pt^F2a *s clinicallT indicated

as an abortifacient for termination of pregnancy in humans.

In experiments with female monkeys, PGF2a has been

shown to greatly reduce the secretion of progesterone by

the corpus luteum of the ovary. Since an adequate supply

of progesterone is needed to ensure implantation of a

fertilized ovum in the wall of the uterus, it appears that

prostaglandins might become important agents in controlling

population growth.

The prostaglandins may also have the potential to

prevent peptic ulcers. It has been shown that PGE^ or PGE2

can inhibit gastric secretions in dogs. Furthermore, these

same two prostaglandins can prevent gastric and duodenal 3 ulcers in rats.

A number of other possible pharmacological uses of the prostaglandins are being investigated, including:

1) opening the airways to the lungs, 2) regulating blood pressure, 3) clearing the nasal passages, and 4) regulating metabolism. With regard to regulating metabolism, PGE^ has been shown to strongly counteract the effects of many hor­ mones in the stimulation of lipolysis (18). This action may take place through altered levels of adipose tissue cyclic

AMP (19).

A number of intriguing hypotheses have been sug­ gested concerning other physiological roles of the prosta­ glandins. Endogenous prostaglandins of the F, series have been postulated to operate on sympathetic neurotransmission by a negative feedback mechanism (20) . It has also been suggested that the anti-inflammatory action of aspirin and certain other drugs may be attributed to their blocking the synthesis of prostaglandins (21). If this latter hypothesis is correct, then the prostaglandins may play a fundamental role not only in normal physiological functions but also in certain pathological conditions.

Nomenclature

The prostaglandins are a family of closely related

fatty acids containing a cyclopentane ring with two adjacent side chains, one of which has the carboxyl group at the 4 terminal position. The naturally occurring prostaglandins may be regarded as being derivatives of prostanoic acid (_1) » whose carbon atoms are numbered as shown. Although they may be named by their relationship to prostanoic acid, the natural prostaglandins are usually divided into five groups and referred to by the letters A,B,C,E and F.

All five groups feature a hydroxyl group at and a trans double bond at the 13,14 position. The E and F series are frequently referred to as the primary prosta­ glandins. They both possess an additional hydroxyl group at Cj j . The E series contains a carbonyl group at Cg, whereas the F series has another hydroxyl group at that position. The A,B and C series may be regarded as dehy­ dration products of the E series. These three series lack a hydroxyl group at but they possess a double bond in

the ring.

The A prostaglandins, in which the additional double bond is at C^0 are the principal products of the dehy­

dration of the E series. Compounds of the A series will

readily rearrange under alkaline conditions to the B prosta­

glandins, where the ring double bond is at Cg (i*e * be­

tween the two side chains). The enzyme prostaglandin

isomerase causes a single shift of the 10,11 double bond of

the A series to yield the C prostaglandins (22) , which have

the ring double bond at C ^ The c prostaglandins are

unstable and also isomerize under mild alkaline conditions 5 to the B series.

Each of these five series is subdivided on the basis of how many additional side chain double bonds are present.

The total number of double bonds is indicated by a subscript numeral after the letter. Therefore, prostaglandin (PGE^) contains only the ^ double bond. has an addi­ tional (cis) double bond at Cj. ^ and PGE^ also possesses a third (cis) double bond at ^g . These additional side chain double bonds occur in the same positions and with the same configuration for all five groups, although prosta­ glandins PGAj, PGBj and PGC^ have never been isolated.

Members of the A and B series are known which have an additional hydroxyl group at C^g. For prostaglandins isolated from natural sources, hydroxyl groups located at

Cg, and Clg positions all have the a configuration. The hydroxyl group at C^g has the 6 configuration. Since this stereochemistry is so far invariant, reference to it is usually omitted in the abbreviated forms used for the natural prostaglandins. However, in the F series tho*-^ additional term a or 3 is always incorporated (e.g., PG F ^ or PGF-j^) to denote the configuration of the Cg hydroxyl group. Although the PGF^ compounds do not occur naturally, they have been known for a long time as reduction products of the E prostaglandins.

Some of the naturally occurring prostaglandins are shown in Figure 1. 6

OH OH OH OH PGE, PGE*

OH OH CO-H

* OH OH OH OH PGF, PGF 3«C

* OH OH PGC, PGA.

« * OH OH OH PGBi 19-O H -P G B 2

FIG. 1. Some representative prostaglandins. 7

Three types of stereochemical variations of the natural prostaglandins can be encountered (see Figure 2).

First, the side chains may have a cis relationship to one another; these are termed 8-isoprostaglandins (e.g., 8-iso-

PGE^). Secondly, one or more of the hydroxyl groups may have the 8 configuration. Hydroxyl groups at C.^ and with the 6 configuration may also be referred to by the term epi. In addition, the and hydroxyl groups may be named after the Cahn-Ingold-Prelog Convention, i.e. S for the a and R for the £} form. Finally, all pros­ taglandins are capable of existing in two optically active forms. Natural prostaglandins are levorotatory. Their enantiomers, which are dextrorotatory, are referred to as the ent forms.

A method based on the structure of prostanoic acid is adopted to give the full systematic nomenclature of a prostaglandin (23).' Thus the complete systematic name of

PGE^ is (-)-lla,15(S)-dihydroxy-9-oxo- 15-trans-prostenoic acid.

Sources of Prostaglandins

Prostaglandins may be obtained from four sources:

1) direct extraction of tissues, 2) fermentation, 3) labo­ ratory biosynthesis, and 4) total chemical synthesis.

Although the first major source of natural prosta­ glandins for research was sheep seminal vesicles, it has 8

CO-H

OH OH OH OH

8 - is o - POE 1 11-opi-PG E 1

OH .COuH

OH OH OH OH

•nt-PGE, PGF 1/?

FIG. 2. Hxamples of stereochemical variations of the natural prostaglandins. 9 been well established that prostaglandins are found in low concentrations in nearly all mammalian tissues (24). The concentration of the natural prostaglandins in most tissues is less than 1 yg/g, (One exception is human seminal fluid, where the concentration is 50 to 60 yg/ml.) Thus, extrac­ tion of mammalian tissues will not provide quantities of prostaglandins sufficient for pharmacological uses.

One prostaglandin, lS-epi-PGA^, has been discovered in relatively large quantities in a horny coral, the

Gorgonian (Plexaura homoinalla), found in the Caribbean region (25). This prostaglandin and the 15-acetyl derivative of its methyl ester are present in the cortex of that species to the extent of 0.2 and 1.3 percent respectively.

These compounds are isolated by solvent extraction and then purified by chromatography. Although these 15-epi prosta­ glandins have relatively low biological activity, they can be chemically converted to the highly active prostaglandins

PGF2a and PGE2 methyl ester (26).

It has recently been reported that the fatty acid extract of yellow onions may be an important natural source of PGA^ (27).

Prostaglandin products have also been produced by culturing Pseudomonas aeruginosa aerobically in a medium free of fatty acids (28) . The highest yield of total prostaglandin material so far reported from this source has been 2.80 pg/ml. 10

It was demonstrated in 1964 by Bergstrom and assoc­

iates (29,30) in Stockholm and van Dorp and associates (31,

32) in Holland that prostaglandins are biosynthesized from

C2 Q straight chain carboxylic acids. The biosyntheses have been demonstrated in a number of different animal tissues.

Of even greater significance, however, is the fact that enzyme extracts can be used to convert the substrate acids

into prostaglandins on a laboratory scale. Until total

chemical syntheses became available this was the only prac­

tical means of obtaining gram quantities of the principal prostaglandins for biological investigations.

The enzyme system most commonly used in these bio­

syntheses is derived from homogenates of the vesicular

gland of the sheep. By the use of vesicular gland homog­

enates, the prostaglandins PGE^, PGE^ and PGE^ are derived

respectively from 8 ,11,14-eicosatrienoic acid, 5,8,11,14-

eicosatetraenoic (or arachidonic) acid and 5,8,11,14,17-

eicosapentaenoic acid (see Figure 3). However, by using

guinea pig lung homogenates for the enzyme system, signifi­

cant quantities of PGF compounds are also formed (33),

The prostaglandin products are purified by solvent

extraction, followed by chromatography, to give yields up

to sixty percent (34).

Several isomers of the natural prostaglandins have

been isolated from these incubations (35,36). Furthermore,

novel prostaglandins with additional oxygen functions have 11

OH i«nolc OH acid PGE,

CO^H

OH 5,8,11,14-Eico*at«tra«no

OH 5,8,11,14,17-Eicosap«nta- OH •noic acid PGE-

FIG. 3. Prostaglandin biosyntheses. 12 been obtained using different enzyme systems (37,38).

Studies on the mechanism of these biosyntheses have established the involvement of an intermediate endoper- oxide (39).

In summary, biosynthesis has been useful as a method of producing limited quantities of prostaglandins in a single conversion step from starting materials which are either commercially available or can be readily synthesized.

However, this approach is dependent upon enzyme systems which are available only in limited amounts and at consider­ able cost. Biosynthesis also lacks the inherent flexibility of total chemical synthesis for the preparation of unnatural stereoisomers and novel structural variants.

Total Chemical Syntheses

Synthetic routes to the prostaglandins can be divided into two categories--those in which the ring is present in the starting material, and those in which it is formed by a cyclization step within the synthesis.

The molecular dissections shown in Figure 4 suggest possible ways in which side chains might be connected to or elaborated on a cyclopentane nucleus. Pathway A represents a-alkylation of a carbonyl compound, the opening of an epoxide by an organometallie reagent, or the conjugate addition of an organometallic reagent. Pathway B stands for a Wittig or similar condensation reaction. Pathway C 13

CCL.H

OH OH

FIG. 4 Synthetic routes to prostaglandins from a cyclopentane nucleus. 14 indicates the nucleophilic addition of an organometallic reagent to an aldehyde.

Figure 5 illustrates the required intermediates for five potential cyclization reactions.

The first synthesis of natural prostaglandins was reported by Just and Simonovitch in 1967 (40) . This syn­ thesis is outlined in Figures 6 and 7.

The tetrahydropyranyl derivative of cyclopenten-4 - ol (2 ) was used as the starting material. Compound 2_ was converted to the bicyclic adduct (3), in which the carbo- ethoxy group had exclusively the exo configuration. Re­ duction of the ester (3^) with ethereal lithium aluminum hydride, followed by oxidation with dilute Jones reagent, yielded the aldehyde 4. Wittig condensation of 4 with n-hexyltriphenyl phosphonium bromide produced a mixture of geometrical isomers from which the cis isomer could be isolated in sixty to eighty percent yield. Subsequent hydrolysis of the acetal group and oxidation resulted in the formation of a ketone (_5) . Alkylation of ketone _5 with methyl 7-iodoheptanoate could be achieved in a yield of only ten to twenty percent. The resulting ketone (6) provided entrance to both the F and F series of prosta­ glandins .

Reduction of the ketone carbonyl group in 6, fol­ lowed by hydrolysis, yielded the olefin _7, which had the 15

FIG. 5. Potential cyclization routes to prostaglandins. 16

0 thp I. N^CHCOjEt, Cu, 60-100° THPQ Z. NaOMt, M«OH, A H 2

THPO I.LAH, Et^O 2 2.CrQ3/ HjO, M.^CO, ' O c ° " H© -15° H

4 2.M.OH, (CO^H)^ H^O, A H 3.Cr03f H20,M*2C0,

5 H |4

6

FIG. 6. The Just synthesis of dl-PGF. (8) anxl dl-PGE methyl ester (9)-Part TT 17

12 13

2.M.OH, H 0,Na0H

7

H C O ^ H , H C O ^ H ,

N«2C0 3 No^COg , " Z ° t H Z°Z

CO^H

OH OH

HO..

OH OH 8

FIG. 7. The Just synthesis of tH-PCF^ (8) and dl-PGE methylester (9)-Part ITT 18 correct stereochemistry for the critical epoxidation- solvolysis step. Treatment of 7^ with 97$ formic acid

(buffered with sodium carbonate) and 30% hydrogen peroxide was claimed to produce a mixture from which a thirty per­ cent yield of amorphous dl.-PC>Fja (8) could be isolated.

Similar treatment of ketone 6 was claimed to produce impure dJ-PGE^ methyl ester (9). No yield was specified for the latter reaction.

For the two epoxidation-solvolysis steps it was anticipated that nucleophilic attack by water on the proton- ated epoxide species should proceed at since attack at

C ^ 2 is sterically blocked by a substituent at Cg. The trans configuration of the double bond (C^ in the products is consistent with the most favorable conformation for the double ring opening sequence. However, a mechanism involv­ ing initial hydrolysis of each epoxide to a diol followed by a cyclopropylcarbinyl rearrangement cannot be excluded.

The results of the Just synthesis were quickly chal­ lenged by investigators at Smith Kline and French. They reported that Ma) no detectable amount of cll-PGE^ is formed in the synthesis described and b) only a low yield of a separable mixture of dl-PGF. isomers was obtained even r — la under improved conditions. None of these correspond to authentic PGFla." (41)

The Just synthesis was further investigated by chemists at The Upjohn Company (42,43) and by Just and 19 associates (44). As shown in Figure 8, the intermediate 6 was treated with osmium tetroxide to yield two erythro-vic- glycol racemates. After separation of the mixture, each racemate was converted to its b ismethanesul fonate (1_0) and solvolyzed in 2:1 acetone-water at 25° C. Solvolysis pro­ duced d_l-PGE^ methyl ester (1_1) in yields of five to ten percent (depending on the starting racemate) along with equal amounts of dl-15-epi-PGE^ methyl ester (12). Using conditions slightly modified from the original Just pro­ cedure, it was confirmed that cfL-PGF^a methyl ester could be synthesized in low yield (along with an equal amount of the 15-epimer).

In all of the reactions involving solvolysis of an epoxide it was found that at least seventy-five percent of the products resulted from attack at the cyclopropylcarbiny1 carbon without participation of the ring (44). This result was attributed to the epoxide oxygen atom which is able to stabilize the developing cyclopropylcarbinyl cation, thus decreasing the extent of delocalization of that charge into the ring. The net result is a decrease in nucleophilic attack on the cyclopropyl ring. This problem is a major obstacle to the Just synthesis of prostaglandins.

A new approach (45) was developed which gave inter­ mediates identical to those of the Just synthesis, except that the olefinic side chain had an endo configuration. 20

H h p o

SO2 SOj I I M « M .

10

2:1, (M * )2C 0 :H 20 ;

25

+ OH OH

OH OH

11 12

FIG. 8. Upjohn modification of the Just synthesis 21

Hydrolysis of a vic-glycol in the final step produced a

seventeen to nineteen percent yield of dl-PGE^ methyl ester (along with an equal amount of the 15-epimer). Modi­

fications of this approach provided syntheses of (H-PGE2 methyl ester (46) and dl_-PGEj methyl ester (47).

The first practical synthetic route to all of the natural prostaglandins was the bicycloheptene approach developed by E. J. Corey's research group (48). This ap­ proach features excellent control of stereochemistry, good overall yields, and an opportunity for optical resolution at an early stage. However, the synthesis of each prosta­ glandin requires approximately twenty steps (see Figures 9-

U).

Conversion of cyclopentadiene (13) to its thallous salt followed by treatment with chloromethyl methyl ether at -20° C produced the 5-substituted cyclopentadiene derivative (14), which was uncontaminated with 1- and 2-

substituted isomers (49). The Diels-Alder condensation of compound with a-chloroacryl chloride afforded the adduct

15 (50). It should be noted that this one reaction provided

stereochemical control for all of the resultant asymmetrical centers in the ring.

Treatment of intermediate 1_5 with sodium azide in dimethoxyethane produced the corresponding acyl azide (16), which upon heating underwent Curtius rearrangement to the

isocyanate (17). Hydrolysis of the isocyanate (17) with 22 OM»

i .-t i (so 4 )2 , k o h , n 2 o 2.C1CH2OM*, Et2Of -20 13

M*

14

15

M*

NaN. 15 m o c h -N ( « 2 )2 3 Cl

16 17

AcO H, H a O 17 ------< “ - C 0 0 Cl 18

,c o 2 h c h 2c i 2 , n o h c o 3

18 2. NaOH, H20 OM« 3. H2 C O j

OH

19

FIG. 9. The Corey bicycloheptene route to natural prosta­ glandins - Part I . 23 aqueous acetic acid led finally to the bicyclic ketone (1_8) .

This ketone (1J1) could be synthesized in eighty percent overall yield from thallous cyclopentadienide with less than two percent contamination with other isomers.

The syn relationship between the double bond and the methoxymethy1 group in ketone had an important influence on the next reaction. That is, the Baeyer-Villiger rearrangement proceeded smoothly on compound using m-chloroperbenzoic acid, with no competing epoxidation of

the double bond. The resultant lactone was readily hydro­

lyzed to the hydroxy acid (1_9) . This acid (19) was resolved using the base ephedrine. The ( + )-enantiomer (as shown by structure JJ3) was shown to be of the proper absolute config­ uration for synthesis of the natural prostaglandins.

Iodolactonization of the resolved hydroxy acid (19)

stereospecifically placed the fourth asymmetrical center on

the cyclopentane ring. Esterification of the resultant lac­

tone with £ “biphenylcarbonyl chloride yielded compound 20.

Deiodination of _20 with tributyltin hydride followed by cleavage of the ether linkage produced the alcohol (21).

Collins oxidation (51) of _21_ afforded an aldehyde which was

immediately used in a modified Wittig reaction with the

sodium salt of dimethyl-2-oxoheptylphosphonate. Reduction of the resultant ketone (22) was examined extensively (52).

A new reducing agent (2J0 was prepared from thexyl

borane, racemic limonene and t-butyl lithium. When ketone ,co2h l.KI-L n O'"”" 2. v S - ' ^ • OH Cl 19 Py rid in *

r - - . 1. Bu3SnH, C6H6,5S' ^ **.^ ^ ^ C H 2OH 20 2 . BBr-

1.CrOjfPyridin*^ 21 ? ° 2. (M *O L PC h {!

c (m *)2c h {m #)2

22 HPA/THF/ Et20/C 5H12 ,

FIG. 10 The Corey bicycloheptene route to natural prostaglandins - Part II. 25

22 was treated with this new reagent (73) in the presence of hexamethylphosphoramide at -120° C in a solvent mixture of tetrahydrofuran, ether and pentane, the desired 15a alcohol (24) was found to predominate over the 153 isomer by a ratio of 82:18 (53). Only small amounts of double bond j1 reduction product were detected.

Apparently at -120° C, compound 21 is predominantly frozen into a conformation in which the biphenyl group is lined up with the enone side chain. It is believed that electronic interaction of the two groups is most favorable when the enone is in the s-cis conformation. With this particular arrangement the hydride attack of a bulky reduc­ ing reagent must come from the opposite side of the molecule, yielding the desired 15a isomer.

After chromatographic purification of compound 24, the p-biphenyl protecting group was hydrolyzed. The resul­ tant diol was converted to the bistetrahydropyranyl deriva­ tive. Subsequent reduction with diisobutylaluminum hydride yielded the hemiacetal (2J>) . The hemiacetal group is syn­ thetically equivalent to an alcohol, aldehyde system.

After addition of the carboxyl side chain by means of a second Wittig reaction, the Cg hydroxyl can be selectively oxidized for entry into the E series of prosta­ glandins. Thus compound ^6 can be readily transformed into

PGF2a or PGE2 . It is important to note that this Wittig reaction produced exclusively the cis- ^ double bond 26 OH l.K.CO^/M«OH

25

THP

OH O THP © No © $ 3 P * " ^ n^ C 0 2 DMSO

OH PGP,

p g e 2 THP 26

2.AcOH, H20 h 2 ,

5 % P d /C

AcO H , H2 0 OH PGF, 1*C

PGE. THP THP 2 7

FIG. 11. The Corey bicycloheptene route to natural prostaglandins-Part III. 27

isomer.

Selective hydrogenation of the cis double bond in

the intermediate 26^ yielded compound 2^7 (54). This inter­ mediate (27) could be converted to either PGF. or PGE. . — la 1 The syntheses of PGF3a and PGE^ were achieved by a slightly modified route (55).

Two additional approaches were developed by Corey and co-workers for the synthesis of intermediates similar

to 2_1 ( 56 , 57). Each method started with cyclopentadiene.

Miyano and associates at G. D. Searle developed a

short, nonstereoselective synthetic route to dd-PGE^ and d_l-PGFja (58-60). This approach utilized two successive aldol addition steps to form a eyelopentenone ring (see

Figures 12 and 13).

Styrylglyoxal (^0), one of the starting materials, was prepared by oxidation of benzalacetone (_29) ■ Mild hydrolysis of dimethyl 3-oxoundecan-1,11-dioate (2ji) pro­ vided an aqueous solution of its dipotassium salt, which was condensed directly with 30 in aqueous citrate buffer to

yield compound 3_1. Treatment of ^1 with dilute aqueous alkali produced the cyclodehydration product (32). Oxida­

tive cleavage of the styryl double bond afforded an oily mixture of aldehydes (_^3) .

Reduction of 3_3 with zinc in cold aqueous acid yielded the saturated aldehyde (34) , which was then treated with n-hexanoylmethylene triphenylphosphorane. The product 28

m *o2c 2 8 0^ |s«0 2/ K O H / H2 0 ID io x o

f°2K} 0 ^ Y

L I *

Ph 4.5

CO,H

HO

31 .CO«H HO

O

32 Dioxant / u O HO CHO 33

FIG. 12. The Miyano synthesis of dJL-PGI;. and d 1 - TGI7. Part I. a 29

CHO HO

34

HO

+

dj -11 -*p i-15-d •hydro-PGE,

N aB H 35

3 5 N q (C N )B H 3 \ dl-PGE, + dl-15-.pi-PGE,

FIG. 13. The Miyano synthesis of dl-PGE. and d_l-PGF, - Part II. was a mixture of cU-15-dehydro-PGE^ (3_5) and dl- 11-epi-15

dehydro-PGEj in approximately equal quantities. These

isomers were separable by either partition or silica gel

chromatogrnphy.

Sodium borohydride reduction of compound 3j> pro­ duced all four possible stereoisomers, including racemic

PGF^a . Selective reduction of the 15-ketone group of 3_5

with sodium cyanoborohydride yielded raceinic PGE^ along

with racemic 15-epi-PGE^.

A series of related syntheses developed by E. J.

Corey and his co-workers (61-64) all utilized either an

aldol cyclization or a vinylogous aldol cyclization to

form the cyclopentane ring of the prostaglandins. This work includes the first synthesis of the resolved enan-

tiomers of PGE^, along with routes to dl_PGFla, dl-PGF^^,

dl-PGA1 , and dl-PGBj.

Two different stereoselective routes to the pros­

taglandins employ aromatic precursors. In the first syn­

thesis, a research group at Merck Sharp 5 Hohme (65)

proceeded from a methoxyindanol to d_l-PGE^. In the secon

R. B. Woodward’s group (66) devised an elegant pathway to

dl-PGF2 a from phloroglucinol.

A more recent synthesis by the chemists at Merck

(67) has produced the resolved enantiomers of PGE^. Thei

approach utilizes an initial Diels-Alder cyclization with

maleic anhydride. 31

Sih and associates (68-70) have developed an inter­

esting synthesis of PGE^ methyl ester which features a microbiological carbonyl reduction and the use of organo-

lithium reagents for conjugate addition.

A synthesis of til - PGE^-methoxime was developed at

the Ciba Pharmaceutical Company (71). Unfortunately, final attempts to convert this material to PGH^ were relatively unsuccessful.

Prostaglandins of the A series are readily obtained

by acetic acid dehydration of the corresponding E compounds

(72). In addition, a direct synthesis of the A prostaglan­ dins has recently been reported (73).

Prostaglandins of the B series are available from

sodium hydroxide treatment of the corresponding members of

the A or E series (74). Also, numerous direct syntheses of

d_l-PGBj have been developed (75-79). Some of these routes

involve as few as seven steps.

The A series of prostaglandins has been converted

to both the E and F series (80). In addition, a procedure

for the conversion of PGA2 to PCC£ has been published (81).

The Purpose of the Research

The objective of this research was to develop syn­

thetic routes for conversion of the mold metabolite

terrein (36) to one or more series of prostaglandins (e.g.

the E series, see Figure 14). The synthesis of PGB^ from 32

>^s.COuH

> 10

OH p g e 2

RO

CHO RO 37

FIG. 14. Coversion of terrein to one or more series of prostaglandins. 33 terrein (36) would require four basic transformations:

A) introduction of a side chain a to the carbonyl group,

B) reduction of the double bond within the ring, C) elabor­ ation of the propenyl side chain of terrein either through oxidation of the terminal methyl group or through oxidative cleavage of the propenyl double bond, and D) reductive elimination of the hydroxyl group a to the ketone. However, it was felt that the fourth transformation might not be necessary. That is, a new series of prostaglandins with an extra hydroxyl group at C10 might prove to have useful bio­ logical properties.

Consideration was given to the synthesis of compound

37, or a similar derivative. It was believed that this in­ termediate would provide rapid entry to several series of prostaglandins.

There are a number of reasons why terrein appears to be a desirable starting material for prostaglandin synthesis.

Since it is an optically active compound, the resolution of enantiomeric products would be avoided. It already has oxygen functions for the prostaglandin ring at and with the correct oxidation level and the correct stereo­ chemistry for access to the E series. Also, the propenyl side chain contains the potential C13 ^ double bond with the required trans configuration. Finally, ample quanti­ ties of terrein are available from fermentation. 34

Previous studies on terrein

In 1935 two strains of Aspergillus terreus grown at

24° C on Czapek-Dox solution with glucose as the sole source of carbon were found to produce a new substance (82). This compound (CgH^gOj, mp 127° C) was named terrein.

A publication concerning the molecular constitution of terrein appeared in 1937 (83). However, the actual

structure and absolute stereochemistry of terrein were not determined until 1955 (84). Barton was able to assign the stereochemistry on the basis of a degradative conversion of

terrein to a derivative of (+)-tartaric acid. Just prior to

this structure determination, new sources of terrein were

found (85).

In 1965 Birch studied the biosynthesis of terrein

(86). He demonstrated that it had a polyketide origin. Its unusual feature 6f two linked "carboxyl" carbons was thought

to result from contraction of a six-membered ring precursor.

The isolation of three chlorinated metabolites from

fermentation of Sporormia affinis was reported in 1969 (87).

These compounds were structurally similar to terrein.

The total synthesis of racemic terrein from cis-1 ,4 - bisbenzyloxy-2,3-epoxycyclopentane was recently reported

(88). EXPERIMENTAL

Methodology

1. The infrared spectra were taken on a Perkin-Elmer

Model 257 infrared spectrophotometer.

2. Ultraviolet spectra were determined on a Cary

Model 15 recording spectrophotometer.

3. Proton magnetic resonance spectra were taken in deuterochloroform or carbon tetrachloride with tetramethyl- silane as an internal standard on a Varian Model A-60 A instrument. These spectra are reported in 5 (ppm) units.

4. The mass spectra were determined on a Du Pont

Model 21-491 mass spectrometer by Mr. Edward Fairchild of

The Ohio State University.

5. Melting points were measured on a Thomas-Hoover apparatus. These values are uncorrected.

6. All reported temperatures are in degrees centi­ grade .

' 7. Microanalyses were determined by Midwest Micro-

Laboratories, Indianapolis, Indiana.

8. Thin layer chromatography was performed using

5 X 20 cm glass plates spread with EM Reagents silica gel G.,

Cat. No. 7731.

35 36

9. Thick layer chromatography was performed using

20 X 20 cm glass plates spread with EM Reagents silica gel

G. , Cat. No. 7731 .

10. Adsorption column chromatography utilized either

EM Reagents silica gel 60 (particle size 0.063-0.200 mm,

70-230 mesh ASTM) or Woelm neutral alumina, grade III.

Purity of Reagents and Solvents

1. All reagents used in this investigation were initially of analytical purity as purchased. Any further purification steps are noted.

2. Analytical grade benzene, diethyl ether, pentane,

1,2-dimethoxyethane (glyme) and tetrahydrofuran were all purified further by refluxing with lithium aluminum hydride, followed by distillation. The distillates were stored over sodium wire.

3. Reagent grade methanol was first refluxed with magnesium turnings and then carefully distilled. The distil­ late was stored over type 3A molecular sieves.

4. Tertiary butyl alcohol of analytical purity was refluxed with sodium for six hours and then distilled. The distillate was stored over type 4A molecular sieves.

1-Iodo-7-octene (38) (89) . A dry 500-ml 3-neck flask equipped with a septum inlet, pressure-equalizing funnel, and a magnetic stirrer was flushed with N2 and then maintained 37 under a static pressure of the gas. The flask was charged with 100 ml of THF and 1,7-octadiene (33.06 g, 0.30 M). Con­ version to the trialkylborane was achieved by dropwise addi­ tion of a 1,00 M solution of borane in THF (103.2 ml, 0.31 M of hydride) over a period of 40 min. This solution was stir­ red for 1 hour at room temperature, then 2 ml of methanol were added to destroy the excess hydride.

Iodine (56.0 g, 0.221 M) was added all at once, fol­

lowed by dropwise addition of a methanolic solution of sodium hydroxide (74 ml of a 3.0 M solution, 0.222 M) over a period of five minutes. The reaction mixture was poured into 300 ml of water containing 10 g of sodium thiosulfate pentahydrate to remove the excess iodine. After shaking, the mixture

separated into two layers which were then separated. The

lower layer was extracted with petroleum ether (30-60°). The petroleum ether extract and the top layer were combined. The resultant solution was dried over magnesium sulfate and the

solvent was removed. Distillation first removed a small quantity of 1 ,7-octadiene, then 1-iodo-7-octene was collected as a pale yellow liquid (18.5 g [26%], bp 67-69° at 0.35 mm

Hg); ir cm"1 (CHClj): 3075, 3000, 2925, 2855, 1630, 1460,

1430, 1170, 995, 910; nmr (CDClj): 6 1.2-2.2 (br m, 10H, aliphatic protons), 6 3.18 (t, J = 7Hz, 2H, -CI^I), 6 4.8-

5.2 (m, 2H, -HC = CH2), 6 5.5-6.2 (br m, 1H, -CH = CH2) . 38

7 -Iodoheptanoic acid (39). I-Iodo-7-octene (38)

(1.81 g, 7.60 mM) was emulsified in 50 ml of dioxane-water,

3:1, by vigorous stirring. An aqueous solution of potassium permanganate and sodium bicarbonate was added in portions until a drop of the resultant mixture gave a purple color on filter paper. The mixture was allowed to stir for 12 hours at room temperature. Then a sufficient quantity of aqueous sodium bisulfite solution was added to dissolve the brown precipitate. The solution was acidified with dilute HC1 and extracted with chloroform. The chloroform extract was ex­ tracted with saturated sodium bicarbonate solution. Acidi­ fication of the aqueous phase with dilute HC1 initiated slow crystallization. The white crystals were filtered, washed with cold water, and dried to yield 0.43 g (221) of the desired compound (^9); mp 49-51°; ir cm ^ (CHCl^): 3300-2500

(broad), 1710, 1460, 1430, 1410, 1285, 1190; nmr (CDClj):

6 1.2-2.2 (br m, 8H, aliphatic protons), 6 2.38 (t, J =

6Hz, 2H, -CH2 -C02H), 6 3.20 (t, J = 6.5Hz, 2H, -CH2I),

6 10.9 (s, 1H, -C02H).

7-Iodo-1-heptanol (40). First Procedure (90). A so­ lution of borane in THF was slowly added to a solution of acid ^9 (200 mg, 0.781 mM) in 4 ml THF hydrogen. This addi­ tion required 1.0 ml of a 1.0 M solution of borane in THF.

Examination by tic (SiGel G, benzenedioxane, 9:1) indicated that all the starting material had reacted. The mixture was 39 evaporated to dryness and distributed between benzene and water. The benzene phase was dried over sodium sulfate and concentrated to a pale yellow oil (145 mg, 771); ir cm"*

(CHC13): 3610, 3440, 3010, 2940, 2860; nmr (CDClj): 5 1.2-

2.1 (br m, 10H, aliphatic protons), 6 2.92 (s, 1H, -OH),

6 3.17 (t, J ■ 7Hz, 2H, -CH2I), 6 3.58 (t, J * 6Hz, 2H,

-CH20H).

Second Procedure (9J_). A solution of l-iodo-7- octene (3J3) (100 mg, 0.42 mM) in 5 ml ethyl acetate, main­ tained at a temperature of -78°, was treated for approxi­ mately 45 sec. with a slow stream of ozone in oxygen until the solution turned blue. The solution of ozonide was then added rapidly to a vigorously stirred ice cold mixture of sodium borohydride (32 mg, 0.845 mM) in 10 ml of absolute ethanol. The mixture was stirred for 2 hr at ice-bath temperature and an additional 4 hr at room temperature.

Then the mixture was poured into 100 ml of water contain­ ing 3 ml of concentrated HC1. After the ethanol and ethyl acetate were removed by distillation under reduced pres­ sure, the hydrolysis mixture was saturated with sodium chloride. The oil which formed was separated and the aque­ ous fraction was extracted with diethyl ether. The combined oil and ether extracts were washed successively with water, aqueous sodium carbonate, and water and dried over anhydrous sodium sulfate. After concentration, the residual yellow oil (86 mg) from the organic phase was chromatographed over 40 silica gel using hexane. The major eluant was concentrated to an almost colorless oil (45 mg, 44%). By nmr, ir and tic, this product was identical to that prepared using the first procedure.

THP-ether of 7-iodo-1-heptanol (41) (92). 7-Iodo-

1-heptanol (2.00 g, 8.26 mM) was transferred as a neat solution to a round bottomed flask stoppered with a serum cap. A solution of dihydropyran (1.05 g, 12.5 ntM, 1.51 eq.) in 2.0 ml anhydrous ether was transferred to the flask (via syringe). The resultant mixture was stirred at room tem­ perature. Then a solution of £-toluenesulfonic acid (85 mg,

0.494 mM, 0.06 eq.) in 5.5 ml ether was added to the flask.

This mixture was stirred for 1 hr. at room temperature, and then washed with a saturated NaHCO^ solution. The ethereal phase was dried (anh. Na^O^), filtered and concentrated to a yellow oil (2.70 g). The oil was purified by column chromatography using neutral alumina (Woelm) of activity grade III. The column was developed with methylene chloride,

The purified product was a pale yellow oil (2.61 g, 97%); ir cm"1 (CHC13): 3010, 2940, 2860, 1145, 1030; nmr (CDClj):

6 1.42 (m, 10H, medial heptyl protons), 6 1.62 (m, 6H,

^ 2 % * CH., CH 2, 6 3.18 (t, J = 7Hz, 2H, -CH2I, 6 3.3-4.1

6 4.57 (br s, 1H, anomeric proton). 41

This THP-ether (41) was hydrolyzed quantitatively to 7-iodo-1-heptanol (40j by dissolving in a 1:1 acetic acid-water mixture and heating at 80° for five minutes.

Tetrakis [iodo (tri-a~butylphosphine)copper (1)1

(42) (93). 13.15 g (0.069 M) of copper (I) iodide [Ventron,

Cu 2 l2 : stock #26110, lot #090773 was used; (Ventrol, "Cul ultrapure": stock #87816, lot #032773 was found to be grossly impure and did not give successful results)] was dissolved with stirring in a solution of 130 g of potassium iodide in 100 ml of water. Since the resulting solution was yellow, and not colorless, it was stirred for 15 min. with

1 g of decolorizing charcoal at room temperature and filter­ ed by suction through 5 pieces of filter paper. The re­ sulting colorless solution was next shaken vigorously for

5 min. with 12.5 ml (10.15 g, 0.050 M) of freshly distilled

(1.2 mm Hg, 75® C) tri-n-butylphosphine in a 250 ml glass- stoppered Erlenmeyer flask until the greasy mass initially formed changed to white granular crystals ('1 min.) The crystals were collected on a Buchner funnel and washed free of any occluded copper (I) iodide with several 10-ml por­ tions of saturated potassium iodide solution. They were then similarly washed with distilled water and 951 ethanol, and air-dried. The product was purified by recrystalliza­ tion using a hot mixture of 115 ml of absolute ethanol and

75 ml of isopropyl alcohol. The yield of long, white needles (mp. 75-76°C; lit. mp 75®C) was 14.30 g (73i) . 42

Attempted conjugate addition of THP-ether 41 to trans-2-methyl-2-pentenal (94,95). A solution of the THP- ether of 7-iodo-1-heptanol (£1) (0.186 g, 0.570 mM) dis­ solved in 1.5 ml of diethyl ether was transferred to a round bottomed flask. The flask was stoppered with a serum cap and cooled to -78°. Then 1.22 ml of a 0.96 M tertiary- butyllithium solution (1.17 mM) was injected into the flask.

The resultant solution was stirred at -78° for two hours.

A solution of tetrakis [iodo(tri-n-butylphosphine) copper (I)] (0.112 g, 0.285 mM) dissolved in 3 ml of ether was prepared in a second round bottomed flask. This flask was also stoppered with a serum cap and cooled to -78°.

Next, the solution in the first flask was transferred to the second flask (via syringe). The resulting mixture was stirred at -78° for one hour.

Then a solution of trans-2-methyl-2-pentenal (0.028 g, 0.285 mM) dissolved in 2.5 ml of ether was transferred to the reaction mixture. The reaction mixture was stirred at -78° for an additional one-half hour, before being warmed to 0°. The mixture was stirred for one final hour at 0 °.

The reaction mixture was quenched by the addition of aqueous ammonium sulfate, and then extracted with ether.

The ethereal phase was dried (anh. Na 2 SO^') , filtered, con­ centrated to a yellow oil (0 . 2 2 1 g ) , and purified by column chromatography using silica gel. Three compounds were 43

collected from the column: tri-n-butylphosphine, trans- 2 -

methyl-2-pentenal and the unreacted THP-ether of 7-iodo-l-

heptanol (^1) (0.130 g).

Attempted addition of 1-iodo-7-octene ( M O to

benzophenone. The reaction vessel used in this experiment was a modified culture tube (18 X 150 mm), A 14/20 female

joint had been fused to the top of the tube and a side arm

(with a slight upward angle) had been added 5 cm below the

top of the tube (see Figure 15). When this apparatus was

used with a one-half inch stirring bar and a small volume of

solvent, a mixing action capable of pulverizing lithium

metal could be achieved.

The aforementioned reaction vessel, containing a magnetic stirring bar, was purged with argon while being heated with a flame. The vessel was then maintained under

a positive pressure of argon. After it had cooled to room

temperature, 4 ml of ether were added by injection.

An alloy of 99:1 lithium-sodium (0.132 g, 18.8 mM of

Li) was rinsed with diethyl ether to remove a coating of mineral oil and added to the reaction vessel. The vessel was cooled to -78°, and its contents were allowed to stir

for 15 minutes. Then a solution of 1 -iodo-7-octene (38)

(2.00 g, 8.40 mM) dissolved in 4 ml of ether was injected dropwise into the reaction vessel over a period of 15 minutes. The temperature of the vessel was slowly raised 44 to -25°, at which temperature the solution became cloudy-- signifying that a reaction had begun. The temperature of the reaction mixture was maintained between -25° to -15° for 3.5 hours. During this time period, a considerable quantity of a white precipitate was formed. The remaining lithium was very shiny. Upon subsequent warming to 0°, most of the precipitate went into solution and the remain­ ing metal acquired a dull appearance.

Next a solution of benzophenone (1.53 g, 8.40 mM) dissolved in 5 ml of ether was injected dropwise into the reaction vessel. The mixture immediately took on an intense blue color. (This color gradually lightened with time.) The reaction mixture was stirred for an additional hour at 0 °.

The reaction mixture was quenched by being poured into a saturated aqueous ammonium sulfate solution. The resultant mixture was extracted with ether. The yellow ethereal phase was decolorized by washing with a solution of sodium thiosulfate. The organic phase was dried (anh.

Na 2 SO^), filtered and concentrated to an oil (2 . 1 2 g).

Examination by tic (SiGel G, benzene-chloroform, 1:1) revealed the presence of two products--unreacted benzo­ phenone and a new compound (which was not 1 -iodo-7-octene).

The new compound was the dimerization product, i.e.,

1,15-hexadecadiene (£3). Purification of this product by silica gel column chromatography yielded a colorless liquid 45

14/20 Foma I* Joint

/20 Malo Joint

J Sint*rod Glass Frit

m M/20 Mai* Joint

FIG. 15. Laboratory glassware prepared for organolithium reactions. 46

(0.735 g, 791); ir cm ' 1 (CHC13) : 3075, 3000, 2925, 2855,

1630, 1460, 1430, 995, 910; nmr (CDClj): 5 1.2-2.2 (br m,

24H, aliphatic protons), 6 4. 75-5. 15 (m, 4H, -CH = C ^ ) ,

6 5.48-6.17 (br m, 2H, -CH = C H ^ .

7-Bromoheptanoic acid (44) . Ethyl 7-bromohep- tanoate (obtained from K § K Laboratories, Inc.) was trans­ ferred to a 250 ml round bottom flask. The ester was dis­ solved in 75 ml of glyme, then 75 ml of 10% a q . KOH was added to the flask. This mixture was stirred magnetically at room temperature for 18 hours. The mixture was then acidified (brought to pH 1) by the addition of dil. HC1.

Next the glyme was removed Jji vacuo. The resultant mixture was extracted with anhydrous ether. The combined ether ex­ tracts were extracted with saturated NaHCO^ solution. The combined NaHCOj extracts were acidified with dil. HC1 (to pH 1). At this point the solution became cloudy, but no crystals formed. So, the resultant mixture was extracted with chloroform, The combined CHClj extracts were dried

(anhyd. Na^SO^), filtered, and concentrated to a pale green liquid (1.69 g, 771) which solidified to give 7-bromo- heptanoic acid as a white solid upon cooling to 0°. Tic examination (silica gel G, CHClj) indicated that the product consisted of a single component. The structural assignment was verified by the following spectral data: ir cm 1

(CHC13) : 3300-2500 (broad) 1710, 1410, 1285, 1260, 920; nmr 4 7

(CDC13): 6 1.20-2.15 (br m, 8 H, aliphatic protons, 6 2.36 V

(t, J = 6.5Hz, 2H, -CH2 -C02 H), 6 3.39 (t, J = 6.5 Hz, 2H,

-CH2 Br), 6 10.5 (s, 1H, -C02 H).

7-Bromo- 1-heptanol (4_5) . First Procedure. A solu­ tion of borane in tetrahydrofuran (1.0 mM of BHj) was in­ jected into a flask containing ethyl 7-bromoheptanoate

(50 mg, 0.21 mM) in 4.0 ml tetrahydrofuran. A reflux con­ denser with attached drying tube was connected to the flask.

The mixture was then refluxed for 1 2 hrs. After refluxing was stopped, water was added to the flask. The THF was removed jm vacuo. This mixture was extracted with anhydrous ether. The combined extracts were dried (anhyd. Na^O^) , filtered, and concentrated to a colorless oil (48.3 mg).

This product was purified by silica gel G thick plate chromatography (CHClj). The oil was found to consist of at least four components. The major component (33 mg, 80%) was

7-bromo-1-heptanol (£5); ir cm ^ (CHCl^) : 3610, 3455 , 3005,

2940, 2860, 1460, 1250, 1050; nmr (CDClj): 6 1.2-2.1 (br m,

10H, aliphatic protons), 6 2.40 (s, 1H, -OH), 5 3.39 (t, J =

6.25Hz, 2H, -CH2 Br), 6 3.60 (t, J = 6.25Hz, 2H, -CH2 OH).

Second Procedure. 7-Bromoheptanoic acid (£4) (1.32 g, 6.31 mM) was dissolved in 25 ml tetrahydrofuran in a round bottom flask that contained a magnetic stirring bar and was stoppered with a serum cap. Then a solution of borane in THF (8.0 mM of BH^ or 24.0 mM of hydride was injected into the flask slowly, with concomitant venting 48

(using a second syringe needle). The addition of diborane initiated a brisk evolution of hydrogen. Examination by tic (silica gel G, MeOH-CHCl^, 1:49) indicated that the starting material had been converted to product. The mix­ ture was evaporated to dryness and distributed between benzene and water. The benzene phase was dried (anh.

Na 2 S0 ^), filtered, and concentrated to a pale yellow oil

(1.12 g, 91%). By nmr, ir and tic, this product was ident­

ical to that prepared using the first procedure.

THP-ether of 7-bromo-1-heptanol (46). 7-Bromo-l- heptanol (4j0 (1.09 g, 5.59 mM) was transferred as a neat liquid to a round bottom flask that contained a magnetic stirring bar. A solution of dihydropyrane (705 mg, 8,38 mM,

1.5 eq.) in 1.0 ml anhydrous ether was added to the flask.

The resultant mixture was stirred at room temperature. Then a solution of j>-toluenesulfonic acid (58 mg, 0.337 mM, 0.06 eq.) in 4.0 ml anhydrous ether was added to the flask. This mixture was stirred for 1 hour at room temperature.

The reaction mixture was diluted with 10 ml of ether and then washed with saturated sodium bicarbonate solution.

The ethereal phase was dried (anh. Na 2 SO^), filtered, and concentrated to a pale yellow oil (1.600 g). The oil was purified by column chromatography using neutral alumina

(Woelm) of activity grade III. The column was developed with methylene' chloride. The product from the column was 49 further purified by stirring its solution in benzene for

24 hrs. with anh. CaCl^ (to remove the last trace of the starting alcohol). The final product (1.482 g, 95%) was a pale yellow oil.

Verification of the structural assignment was pro- _ i vided by the following spectral data: ir cm (CHClj):

3005, 2940, 2860, 1455, 1440, 1355, 1260, 1130, 1075, 1030,

910, 875; nmr (CDCl^): 5 1.42 (m, 10H, medial heptyl

qi 2 protons), 6 1.63 (m, 6 H, CiU^ ) , 6 3.39 (t, J =

6 4.56

(br s, 1 H, anomeric proton).

Attempted addition of THP-ether £ 6 to carbon dioxide

(96). In this experiment the previously described modified culture tube (see Figure 15) was used along with a new appa­ ratus. A funnel with a sintered glass frit (diameter of

1 cm) was fused to two 14/20 male joints, one on each end

(see Figure 15). When used together, these two pieces of

glassware permitted the filtration of a mixture under a positive pressure of an inert gas.

The modified culture tube (containing a magnetic

stirring bar) was heated in an oven at 1 2 0 °, purged with

argon and then maintained under a positive pressure of

argon. After it had cooled to room temperature, 2.5 ml of

ether were added by injection. An alloy of 99:1 lithium- 50 sodium (0.034 g, 4.85 mM of Li) was rinsed with ether and added to the reaction tube. The tube was cooled to -15°, and its contents were vigorously stirred. Several drops of neat THP-ether 46 were injected into the tube. Since no reaction occurred (even after two additional injections), the temperature of the reaction vessel was allowed to rise slowly. Not until having reached room temperature (28°) did the reaction mixture give signs of a reaction. Then the lithium became shiny and a small amount of a white precipitate was formed.

The rest of the THP-ether (46) (a total of 0.600 g,

2.15 mM) was injected dropwise in portions over a period of

25 minutes. Following this addition, the reaction mixture was vigorously stirred for 5,5 hours at room temperature.

Next, using the previously described filtering apparatus, the reaction mixture was filtered under argon into a 2 -neck round bottom flask. After th„ yellow filtrate was diluted with 2 ml of ether, anhydrous carbon dioxide was allowed to bubble through this solution for 30 minutes. Then aqueous ammonium sulfate was added to the filtrate.

The resultant mixture was extracted with ether. The ethereal phase was washed with saturated NaHCO^ solution.

(The NaHCO^ extract was found to contain virtually no organ­ ic material.) Then the ethereal solution was dried (anhy.

Na 2 S0 ^ ), filtered, and concentrated to a yellow oil (0.3547 g). Purification of this oil by thick layer chromatography 51

(silica gel G, CHClj) yielded the dimerization product

(0.236 g, 55%), i.e. the bistetrahydropyrany1 ether of 1,14-

tetradecadiol (47); ir cm ^ (CHClj): 3005, 2940, 2860, 1450,

1355, 1040; nmr (CDC1,): 6 1.28 (m, 24H, medial tetradecyl

/ ^ \ protons),

), 6 4.57 (br s, 2H, anomeric protons). c»2 1 * <■0*O ' " ™ 2 Attempted addition of 2-chloro-1,1-diethoxyethane

(48) to benzophenone (97). 2-Chloro-1,1-diethoxyethane (48)

(obtained from Aldrich, Lot No. 081437) was washed twice

with aq. sat, NaHCO^ solution. Then the reagent was stirred

for 14 hours over anh. C a C ^ and anh. Na2 C0 ^. The final

purification step was distillation from anh. Na2 C0 j (bp 33°

at 0.175 mm Hg).

The modified culture tube (see Figure 15), contain­

ing a magnetic stirring bar, was connected to a (14/20)

Claisen head adapter. This adapter was attached to both a

10-ml (14/20) pressure-equalizing dropping funnel and a

(14/20) reflux condenser. The connective tubing leading

from the source of anhydrous argon to the side arm of the

reaction tube was a rubber hose. The intact assembly was

purged with argon and then maintained under a positive pres­

sure of this gas. All of the glassware was heated with a

flame to remove water vapor.

After the reaction tube had cooled to room 52 temperature, 3 ml of benzene were added by injection through the side arm. Next the alloy of 99:1 lithium-sodium (0.154 g, 0.022 M) was rinsed with benzene and added to the reac­ tion vessel. Vigorous stirring at room temperature was com­ menced.

1.5 ml of 2-chloro-1,1-diethoxyethane (4ji) (1.53 g,

0.010 M) were transferred to the dropping funnel. Eight drops of this reagent were added to the reaction tube, how­ ever, no reaction ensued. Therefore, the mixture was heated (by a mineral oil bath) to initiate refluxing. At this point, the lithium became shiny and a white precipi­ tate was forming. The remaining 2 -chloro-1 ,1 -diethoxyethane

(48) was added dropwise over a period of 20 minutes. Thirty minutes later further heating was discontinued, and the reaction mixture was allowed to cool to room temperature.

The mixture then consisted of a thick white precipitate plus unreacted lithium suspended in a brown solution. This mix­ ture was diluted with an additional 3 ml of benzene.

The Claisen head was removed and the reaction mix­ ture was filtered (through the filtering apparatus shown in

Figure 15), under argon, into a 2-neck round bottom flask.

The rb flask was then removed from the filtering apparatus and maintained under a positive pressure of argon at room temperature. A solution of benzophenone (1.82 g, 0.010 M) dissolved in 7.5 ml of ether was injected into the flask.

The resultant solution was stirred for one hour. Next, 53

aqueous ammonium sulfate was added to the flask. Following

removal of the organic phase, the aqueous layer was ex­

tracted with ether. The combined organic layers were dried

(anh. Na 2 S0 ^), filtered, and concentrated to a colorless oil

(1.859 g). Examination by tic . (silica gel G, CHCl^) and ir

(CHClj) revealed that the product contained virtually no

components other than unreacted benzophenone.

Attempted addition of 2-bromo-1,1-diethoxyethane

(49) to benzophenone. A solution of 2 -bromo-1 ,1-diethoxy-

ethane (49) (60 ml, 78.6 g) dissolved in 60 ml of reagent

grade dichloromethane was washed in succession with 40 ml

portions of distilled water, sat. NaHCO^ solution and sat.

NaCl solution. Then the solution was dried (anh. Na 2 SO^),

filtered and concentrated to a yellow oil. The oil was

distilled from anh. Na 2 C0 3 (bp 53° at 0.20 mm Hg).

The modified culture tube (see Figure 15) , with a

magnetic stirring bar, was heated in an oven at 1 2 0 °,

purged with argon and then,maintained under a positive

pressure of this gas. After it had cooled to room temp­

erature, ether ( 6 ml) was added by injection through the

side arm. A 99:1 1ithium-sodium alloy (0.118 g, 16.8 mM of Li) was rinsed with ether and added to the reaction

tube. This mixture was vigorously stirred.

Twelve drops of neat compound 4_9 were injected

into the reaction vessel through the side arm. Immediately 54 the metal pieces became shiny and the reaction mixture be­ came cloudy. The temperature of the reaction tube was lowered to -5°. Over the next 35 minutes the remainder of the neat 2-bromo-1,1-diethoxyethane (££) (a total of 1.500 g, 7.61 mM) was injected dropwise, in portions, through the side arm of the reaction tube. The reaction mixture was stirred for an additional 1.5 hours at -5°. At the end of this time, the small remaining portion of lithium began to tarnish qnd there was a relatively large quantity of a white precipitate.

The reaction mixture was then filtered under argon into a 2 -neck rb flask that contained a magnetic stirring bar. The receiving flask was cooled to 0° during the fil­ tration. Twice 3-ml portions of ether were injected into the reaction tube through its side arm to rinse the pre­ cipitate on top of the sintered glass frit of the filtering apparatus (see Figure 15). After filtration, the receiving flask was maintained under a positive pressure of argon at

0 °.

A solution of benzophenone (1.387 g, 7.61 mM) dis­ solved in 5.5 ml of ether was injected dropwise into the receiving flask over a period of 15 minutes. The resultant mixture was stirred for an additional 2 hours at 0°. Then aq. (NH^)^ SO^ was added to the reaction mixture. The organic layer was subsequently removed and the aqueous layer was extracted with ether. The combined ethereal phases were 55

dried (anh. Na 2 S0 ^), filtered and concentrated to a pale yellow oil (0.136 g). Examination by ir (CHClj) and tic

(silica gel G, CHCl^) indicated that the product consisted almost exclusively of unreacted benzophenone.

THP-ether of 2-bromoethanol (51) . A solution of 2- bromoethanol (50) (obtained from K 5 K Laboratories, Inc.,

Lot No. 8694-A) (23 ml, 40.5 g) was prepared by dissolving the neat alcohol in reagent grade dichloromethane (23 ml).

This solution was washed in succession with 15-ml portions of distilled water, sat. NaHCO^ solution and sat. NaCl solution. Then the solution was dried (anh. Na2 S0 ^) , fil­ tered and concentrated to an oil. The oil was purified by distillation from anh. Na 2 CO^ (bp 32-34° at 0.175 mm Hg).

A solution of purified 2-bromoethanol (_50) (26.102 g, 0.209 M) dissolved in 22 ml of ether was transferred to a 500-ml Erlenmeyer flask. Next a solution of dihydropy- rane (26.355 g, 0.313 M, 1.5 e q .) dissolved in 22 ml of ether was added to the flask. The resultant mixture was stirred at 0°. Then a solution of p-toluenesulfonic acid

(2.158 g, 0.0125 M, 0.06 eq.) dissolved in 100 ml of ether was added to the flask. The Erlenmeyer flask was allowed to warm to room temperature, and the reaction mixture was stirred for an additional 2.5 hours.

Then the ethereal reaction mixture was washed with

50-ml portions of sat. NaHCO^ solution and sat. NaCl solu­ tion. Next the organic phase was dried (anh. Na2 S0^) , filtered and concentrated to a pale yellow oil. Purifica­ tion of the oil by distillation (bp 65-67° at 0.30 mm Hg) from anh. Na2 COj yielded a colorless liquid (30.212 g, 69%) ir cm' 1 (CHC13): 2940, 2860, 1455, 1440, 1355, 1275, 1205,

1180, 1140, 1080, 1025, 965; nmr (CDClj) : 6 1.67 (m, 6 H,

^ c h 2

CH„ NC H J , 6 3.32-4.23 (m, 6 H, BrCH,CH,0 THP and p 2 i - 2

), 6 4.66 (br s, 1H, anomeric proton).

Q. 0 " Attempted reaction of THP-ether 5^L with lithium. A magnetic stirring bar was placed in the modified culture tube (see Figure 15), which was heated in an oven at 120° for 1 hour. Next the reaction tube was purged with argon and then maintained under a positive pressure of this gas.

When it had cooled to room temperature, 5 ml of ether were injected into the reaction tube through the side arm.

An alloy of 99:1 lithium-sodium (0.111 g, 15.8 mM of Li) was placed in a beaker containing pentane. The corrosion on the surface of-the metal was removed by scrap­ ing with a spatula. Then the metal wire was cut into 12 pieces, which were transferred to the reaction tube.

Vigorous stirring of the mixture of lithium in ether (at room temperature, 23°) was begun.

Twelve drops of neat THP-ether ^1 were injected in­ to the reaction tube through the serum cap which had been used to stopper the vessel. Although the reaction mixture 57 became slightly cloudy, the Li pieces never acquired the shiny spots which are indicative of reaction. The remainder of the THP-ether (_51_) (a total of 1.500 g, 7.17 mM) was dis­ solved in 2,5 ml of ether. This solution was injected into the reaction tube dropwise, in portions, over a period of

30 minutes. At the end of this addition, however, there were still no signs of reaction. Even brief heating of the reaction mixture to initiate refluxing failed to produce a discernible reaction. After the reaction mixture had stir­ red for 4 days at room temperature under argon, there still had been no reaction between the lithium and the THP-ether of 2 -bromoethanol (JU) .

Attempted conjugate addition of THP-ether 5^1 to trans-2-methyl-2-pentenal. A 3-neck, 100 ml round bottom flask (containing a magnetic stirring bar) and a 25-ml pres­ sure-equalizing dropping funnel were heated in an oven at

120°. Then the flask and the funnel were connected, flushed with argon and maintained under a positive pressure of this gas. After the glassware had cooled to room temperature, magnesium turnings (0.232 g, 9.54 mM, 2.0 equiv.) and tetra­ hydrofuran (5 ml) were added to the flask.

A solution of both THP-ether 5_1 (1.000 g, 4.77 mM,

1.0 equiv.) and 1,2-dibromoethane (0.898 g, 4177 mM, 1.0 equiv.) dissolved in 20 ml of THF was prepared. This solu­ tion was injected into the dropping funnel. Then a portion 58

(4 ml of the solution) was added to the flask, but no reac­ tion ensued. However, after a few pieces of Mg were crushed with a glass rod, the Mg immediately began to react. The remainder of the solution in the dropping funnel was added dropwise over a period of 40 minutes. The reaction mixture was then stirred for an additional 4 hours at room tempera­ ture, during which time practically all of the Mg was con­ sumed and a large quantity of a white precipitate was formed.

Next the rb flask was cooled to -78°. A solution of tetrakis [iodo (tri-n-butylphosphine) copper (I)] (42)

(0.940 g, 2.39 mM, 0.5 equiv.) dissolved in 8 ml of ether was injected into the dropping funnel and subsequently added dropwise to the flask. The resultant mixture was stirred at

-78° for 1.5 hours. Then a solution of trans-2-methy1- 2- pentenal (0.470 g, 4.77 mM, 1.0 equiv.) dissolved in ether

(4 ml) was injected directly into the rb flask. The resultant mixture, which temporarily acquired a pale green color, was stirred at -78° for 1 hour. In succession the reaction mix­ ture was warmed and allowed .to stir at 0 ° for 2 hours and at room temperature (23°) for 1 hour.

The reaction mixture was then diluted with 2 equiva­ lent volumes of ether and poured into a 10% a q . HC1 solution.

The ethereal layer was separated and the aqueous layer was extracted with ether. The combined ethereal phases were dried (anh, ^£50^) , filtered and concentrated to a yellow 59 oil (3.242 g). Purification of the oil by column chromatog­ raphy using silica gel yielded three products: tri-n- butylphosphine, unreacted trans- 2-methyl-2 -pentenal and 5- hy droxy pen tana 1 ( 5j0 (0.335 g , 691).

When compound S2^ was spotted on a silica gel tic slide and treated with 2,4-dinitrophenylhydrazine spray reagent, a yellow derivative was formed. The following spectral data suggest that compound 5_2 exists predominantly as a hemiacetal in solution: ir cm 1 (neat): 3520-3200

(broad), 2940, 2860, 1140, 1080, 1020; nmr (CDC1,): 6 1.57

(m, 6 H , CH 2 lH 2 ) , 6 2.4 (br s, 1H, ), <53.4-

■ 0 ^ 0 - ^ C T 0 H

4.1 (m, 2H, c O ), 6 4.83 (br, s, 1H, anomeric proton). 0 - ° '

An ethereal solution of THP-ether _51 was shaken in a separatory funnel with a 101 aq. HC1 solution. The ethereal phase was then found to contain a new product which was identical by tic (silica gel G, CHClj) and ir (neat) to compound 5 2.

Trans -5-methyl-l,5-octadien-4-ol (53) (98 ,99).

Tetraallyltin (0.311 g, 1.10 mM) dissolved in 3.0 ml of ether was stirred with 2.2 ml of 2.0 M (4,40 mM) phenyl lithium in a centrifuge tube under argon. This mixture was stirred for

1 hour at room temperature. 60

During that stirring period, a 3-neck 50-ml round bottom flask was heated in an oven at 1 2 0 °, purged with argon and then maintained under a positive pressure of argon. Recrystallized cuprous iodide (0.380 g, 2.00 mM) was transferred to the flask, followed by di-n-butyl sul­ fide (0.70 ml, 0.587 g, 4.01 mM). After a few minutes of stirring, all of the Cul went into solution. Then 2 . 0 ml of anhydrous ether were injected into the flask.

After the one hour period of stirring, the reaction mixture in the tube was centrifuged. The supernatant brown solution was transferred via syringe to a pressure-equaliz­ ing dropping funnel. In succession, two 5 ml portions of ether were used to recover any additional allyllithium from the centrifuge tube.

The flask was cooled to -78°. Next the mixture of allyllithium in the funnel was added dropwise over a period of 10 minutes. The resulting mixture in the flask became red, then orange, and finally bright yellow. The reaction mixture in the flask was stirred for a total of 1 hour at

-78°. The dropping funnel was rinsed with 1.0 ml of ether.

(In one variation, the flask was cooled at this point to

-116?)

A solution of trans-2-methyl-2-pentenal (0.196 g,

2.00 mM) dissolved in 5.0 ml of ether was injected into the dropping funnel. This solution was added dropwise to the flask, causing the reaction mixture to become blood-red. 61

After 1 hour at -78° the reaction mixture was added dropwise to a vigorously stirTed 151 aq. HC1 solution (100 ml). Am­ monium hydroxide was added to bring the mixture to pH 8 . The aqueous layer then became blue.

The ether layer was removed and the aqueous layer was extracted with ether. The combined ethereal phases were dried (anh. Na2 SO^), filtered and concentrated to a yellow oil (0.813 g). Purification of this oil by silica gel column chromatography afforded the 1 ,2 -addition product (_53) as a colorless oil (0.188 g, 67%); ir cm * (neat): 3520-3240

(broad), 3080, 2965, 2940, 2880, 1645, 1460, 1380, 1310, 1050,

1020, 1000, 916, 865; nmr (CDClj): 6 0.95 (t, J = 7Hz, 3H,

^ \ / C = ) , 6 1.59 (s, 3H, C = ), 6 1 .65- 2.40 (m, 4H, C\(2 tit CH3 CH, H^ ^£H-CH2-CH=CH2

C = Cv \ 6 2.62 (s, 1H, -OH), 6 3.93 (t, J = CH,■ I " S Me ./ ~ 2 Me

I H 7Hz, 1H, -C-H), <5 4.76-6.10 (m, 4H, and-CH - CH,). ■ _ / ~ -L I Et OH

The uv (MeOH) spectrum of compound ^3 displayed no absorption maximum, only end absorption. This spectrum pro­ vided a sharp contrast to that of the starting material, MeOH trans-2 -methyl-2-pentenal [uv *max : 227 nm (log e 4.28)]. 62

Methyl 5-hexenoate (54). The procedure used to form the diallyl lithium cuprate was exactly the same as that given for the preparation of compound j^3. Addition of a solution of methyl acrylate (0.172 g, 2.00 mM) dissolved in

5 ml of ether to the diallyl lithium cuprate reaction mix­ ture at -78° did not produce a color change. The final reaction mixture was stirred for 1 1/2 hours at -78°. The reaction mixture was worked-up in the same way as in the preceding procedure to yield a yellow oil (0..789 g) . Puri­ fication of the oil by silica gel column chromatography

(9:1 benzene-chloroform as the eluting solvent mixture) yielded the conjugate addition product as a colorless oil

(0.201 g, 79%); ir cm- 1 (neat): 2955, 2925, 2870, 1740, 1640,

1460, 1450, 1435, 1380, 1260, 1220, 1165; nmr (CDC13): 6 1.4-

2.6 (br m, 6 H, CH2 = CH (CH2 ) 3 C02 Me ) , 6 3.66 (s, 3H, -OCHj),

6 4. 76-5.16 (m, 2H, -CH = CH2) , <5 5.4-6.1 (br m, 1H, -CM =

CH2); uv ^ a x ” 1 2 0 4 nm (log e 2-65)-

3-Carbomethoxy-4 (S), 5(R)-dibenzoxy-2-cyclopenten-1- one ethylenethioketal (56) (100). 4 (_5) , 5 (R) - Dibenzoxy-3 - formyl-2 -cyclopenten-l-one ethylenethioketal (^5) (1 0 1 )

(0.150 g, 0.352 mM) was dissolved in a mixture of ether (0.5 ml) and methanol (3 ml). This solution was transferred to a

50-ml round bottom flask that contained a magnetic stirring bar.

In a separate flask a mixture of sodium cyanide 63

(0.091 g, 1.85 mM, 5.26 equiv.), acetic acid (that had been distilled from P 2 O 5 ) (0.111 g, 1.85 mM, 5.26 equiv.), meth­ anol (9 ml) and activated manganese dioxide (obtained from

Winthrop Laboratories) (0.642 g, 7.38 mM, 21.0 equiv.) was prepared. Next sufficient additional acetic acid (5 drops) was added to bring the mixture to pH 6 . This mixture was added to the solution of aldehyde 5ji. The resultant mix­ ture was stirred at room temperature for 18 hours.

The reaction mixture was then filtered through a fun­ nel with a sintered glass frit. The residual MnC^ was thoroughly washed with methanol, and the filtrate was mixed with an aqueous NaCl solution. The resulting mixture was concentrated on a rotary evaporator to remove methanol and was subsequently extracted with ether. The Et2 0 extract was dried (anh. Na 2 S0 ^), filtered and concentrated to a pale yel­

low oil. Purification of the oil by silica gel column chroma­

tography (CHClj as the eluting solvent) afforded the methyl ester (56) as a colorless, viscous liquid (0.115 g, 72%).

Crystallization from methanol produced white needles; mp 123-

125°; ir cm " 1 (CHC13): 2920, 1725, 1450, 1310, 1260, 1100,

1070; nmr (CDCl^): <5 3. 32 (br s, 4H, ethylenethioketal pro­ tons), <5 3.75 (s, 3H, -C02 CH3), 6 5.81 (d of d, J = 1Hz,

1.5Hz, 1H, H-4), 6 6.32 (d, J = 1.5 Hz, 1H, H-5),6 7.06 (d,

J = 1Hz, 1H, H-2), 5 7.18-8.25 (br m, 10H, benzoate protons);

uv 2 2 9 nm (1°S e 4. 41) , 272 nm (log c 3.23). in 3 x Anal: Calculated for ^23^20^2^6: 4.42; 64

S, 14.05; 0, 21.03; Found: C, 60.69; H , 4.50; S, 13.91; 0,

21.14.

5 (R)-Benzoxy-3-carbomethoxy-2 -(2-propenyl)- 3-cyclo-

penten-l-one ethylenethioketal (57). A diallyl lithium

cuprate mixture was prepared by following a previously de­

scribed procedure (see the synthesis of compound 53J on a one-

tenth scale. A solution of tetraallyltin (0.031 g, 0.110 mM)

dissolved in ether (1 ml) was stirred with 0.22 ml of 2.0 M

(0.440 mM) phenyl lithium in a centrifuge tube under argon

'for 1.5 hours, at room temperature. Meanwhile a 3-neck 25-ml

round bottom flask (containing a magnetic stirring bar) was

heated in an oven at 1 2 0 °, purged with argon and then main­

tained under a positive pressure of this gas. After it had

cooled to room temperature recrystallized cuprous iodide

(0.038 g, 0.200 mM) was added to the flask, followed by di-n-

butyl sulfide (0.07 ml, 0.059 g, 0.401 m M ) . As soon as the

Cul went into solution ether (0.5 ml) was injected into the

flask.

At the completion of its stirring period, the reaction

mixture in the tube was centrifuged. The supernatant solu­

tion was transferred via syringe to a 1 0 -ml pressure-

equalizing dropping funnel that was maintained under a

positive pressure of anhydrous nitrogen. A portion of ether

(1.5 ml) was injected into the centrifuge tube. The result­

ant mixture was stirred and then the tube was centrifuged 65

again. This second supernatant solution was also trans­

ferred to the dropping funnel. The entire process was

repeated with an additional portion of ether (1.5 ml) in

order to recover any residual allyllithium from the centri­

fuge tube.

The flask containing the Cul mixture was then cooled

to -78°. Next the mixture of allyllithium in the funnel was

added dropwise over a period of 10 minutes. The resulting

mixture in the flask soon became bright yellow. After the

empty dropping funnel was rinsed with ether ( 1 ml), the

reaction mixture was stirred for 1.5 hours at -78®.

A solution of compound 5j> (0.091 g, 0.200 mM) dis­

solved in a mixture of benzene (0,1 ml) and ether (0.9 ml) was injected into the dropping funnel. The dropwise addition

of this solution caused the resultant mixture to become milky yellow. After stirring for 1 hour at -78° the reaction mixture was poured into a vigorously stirred 15% aq. HC1

solution ( 1 0 0 ml).

The ethereal layer was removed and the aqueous layer was extracted with ether. The combined ethereal phases were

dried (anh. N a 2 SO^), filtered, and concentrated to a yellow oil. Isolation of reasonably pure 57_ proved to be difficult even though the product mixture was not complex. The major contaminant was di-n-butyl sulfide. The crude reaction mixture was initially purified by silica gel column chromatog raphy (using pet ether, C I ^ C ^ and CHCl^ as the eluting 66 solvents). The combined fractions which contained Sl_ (an of 0.55, slightly greater than that of starting material, by tic examination--silica gel G, CHCl^) were purified by two successive stages of preparative thick layer chromatography

(silica gel G, CHGl^) to afford compound 57 as a pale yellow oil (0.038 g, 51%; ir cm " 1 (CHClj): 2940, 2915, 2845, 1720,

1440, 1255, 1210, 1075, 1005; nmr (CDC13) : 6 1.5-2.1 (br m,

3H, -CH2 - CH=CH2 and H-2), 6 3.31 (s, 4H, ethylene thioke- tal protons, 6 3.77 (s, 3H, -CC^CHj), 6 4.83-5.25 (m, 2H,

-CH=CH2), 6 5.55-6.20 (br m, 1 H, -CH=CH2), 6 5.93 (d, J = 3Hz,

1H, H-5), 6 6.83 (d of d, J - 1Hz, 3Hz, 1H, H-4), 6 7.25-8.25

rtI 1 _ * . MeOH. __ (br m, 5H, benzoate protons); uv A : 229 nm (log c 4.16). JT13 X

Terrein diacetate (5Ji) . First procedure (102). Terrein

(36) (0.0308 g, 0.20 mM) was dissolved in a mixture of acetic anhydride (3 ml) and anhydrous sodium acetate (0.020 g, 0.244 mM). The reaction mixture was stirred for 21 hours at room temperature. Then the acetic anhydride was removed by distil­ lation (0.5 mm Hg at 36°). The residue was dissolved in di- chloromethane, filtered and concentrated to yield 58 as a viscous, pale yellow oil (0.0475 g, 99.8%); ir cm 1 (CHClj):

3020, 2920, 1745, 1725, 1645, 1585, 1435, 1370, 1350, 1235,

1180, 1055, 1035, 965; nmr (CDC13 ) : 6 1.95 (d of d, J =

1Hz, 4Hz, 3H, -CH»CH-CH3), 6 2.15 (s, 6 H, -C^C CH3) , 6 5.25

0

(d, J = 2.5 Hz, 1H, 67

CHC1_, ), 6 6.2-6.4 Cm, 3Ht vinyl protons): uv X J • max

274 (log e 4.21).

Second Procedure (103). Terrein (36) (6.50 g, 42.2 mM) was transferred to a 250 ml Erlenmeyer flask that con­ tained acetic anhydride (50 ml) and a magnetic stirring bar.

The resultant slurry was stirred at 0°. Then a solution of anhydrous j>-toluenesulfonic acid dissolved in acetic anhy­ dride (15 ml) was added to the flask. After 5 minutes the flask was warmed to room temperature and the reaction mix­ ture (which was now a pink solution) was stirred for an additional 2 hours.

The acetic anhydride was then removed by distilla­ tion (0.25 mm Hg at 40°), and the residual brown oil was dissolved in ether. The ethereal solution was washed with cold (0°) 0.1 N NaHCOj solution and distilled water. The solution was subsequently dried (anh. Na 2 S0 ^), filtered and concentrated to a pale yellow oil (8.813 g, 8 8 %). Exam­ ination by tic (silica gel G, 49:1 (CHClj-MeOH) revealed that the product was not contaminated with either starting material (36^) or monoacetylated product.

4 (S), 5 (R)-Diacetoxy-3-(trans-1-propenal)-2-cyclo- penten-l-one (59) (104). Terrein diacetate (58) (14.09 g, 68

59.1 mM) was transferred to a 1-1 rb flask that contained a magnetic stirring bar. The starting material was dissolved in 500 ml of xylene and then this solution was brought to reflux with stirring. Next freshly resublimed selenium dioxide (9.19 g, 82.8 mM, 1.40 equiv.) was added in small portions to the reaction mixture, and a Dean-Stark trap (for removal of water) was inserted between the flask and the re­ flux condenser.

After 4.5 hours a second portion of SeC^ (9.19 g) was added to the flask. Four hours later, refluxing was stopped and the hot reaction mixture was filtered. The red filtrate was concentrated to a black oil (14.140 g). Exam­ ination by tic (silica gel G, 1:24 MeOH-CHCl^) revealed that this oil contained at least 6 components. The desired product (_5£) gave an immediate positive test with 2,4- dinitrophenylhydrazine spray reagent. The crude reaction product was charged to a 250 g silica gel column, which was developed rapidly with CHCl^. The fractions which contained aldehyde 5j) were combined to yield a red oil (8.05 g). This product was further purified by the use of silica gel (250 g) column chromatography again (CHCl^ as the eluting solvent) to afford compound 59 as a pale yellow oil (4.41 g, 30%).

Prolonged contact of J>9 with silica gel was found to cause the formation of a new, more polar aldehyde (of undetermined structure) . 69

The desired product (59) was characterized by the

following spectral data: ir cm"* (CHCl^): 3020, 2920, 2820,

2720, 1745-1735 (broad), 1688, 1570, 1370, 1230, 1170, 1110,

1035, 070; nmr (CDCl-j): 6 2.14 (s, 6 H, -02 CCH3), 6 5.21 (d,

Ac J = 3Hz, 1H, ), <5 6.15 (d, J = 3Hz, 1H,

Ac Ac 0 0 « 6-52 (d of d, J « 7Hz, 16Hz, 1H. ^J^CHO) , I c ° V »

0

6 6 6.69 (s, 1H, e K s - ), 5 ?*7.37 (d of d, J = 0.8Hz, 16Hz,

CHO H

1H, CHO ), 6 9.70 (d of d, J = 0.8Hz, 7Hz, 1H, -CHO;

uv A^JC 1 3: 274 nm (log e 4.19), 284 nm (sh, log e 4.11); J11HX ci mass spectrum (isobutane): 254 (12%), 253 (MH+ , 100%),

193 (54%).

4 (S), 5(R)-Diacetoxy-3-(3-hydroxy-trans-1-octenyl)-

2-cyclopenten-l-one (£H^) . A 2-neck 50-ml round bottom flask containing a magnetic stirring bar was heated in an oven at

1 2 0 °, purged with argon and then maintained under a positive pressure of argon. After the flask had cooled to room tem­ perature, a solution of 1-bromopentane (0.889 g, 5.89 mM, 70

1.26 equiv.) dissolved in tetrahydrofuran (3.5 ml) was added by injection. Then a strip of magnesium ribbon (0.156 g,

6.41 mM, 1.38 equiv.) was sanded (to remove its oxide coat­

ing) and cut into small pieces, which were transferred to the flask.

Since formation of the Grignard reagent proceeded exothermically, the temperature of the flask was moderated using a beaker of water at room temperature. After 30 minutes of stirring another portion of THF (3.5 ml) was injected in order to dissolve the white precipitate which had formed. Thirty minutes later ether (3.5 ml) was in­ jected into the flask, which was then cooled to -78°. The cooling process converted the reaction mixture into a white slurry, which still contained a small amount of shiny unreacted Mg.

Next a solution of the starting aldehyde (j>£) (1.175 g, 4.66 mM, 1.0 equiv.) dissolved in a mixture of THF (3.5 ml) and Et^O (3.5 ml) was injected into the flask. The resultant brown slurry began to solidify. Solidification was avoided by the use of vigorous stirring and the injection of more ether (4 ml). The reaction mixture was stirred for an. additional 3.5 hours at -78°. The mixture was then poured into a rapidly stirring 10% aq. HC1 solution. The ethereal layer was removed and the aqueous layer was extracted with ether. The combined ethereal phases were dried (anh. Na 2 SO^), 71 filtered and concentrated to a brown oil (1.297 g) .

The crude reaction product was purified by silica gel (130 g) column chromatography (1:24 MeOH-CHClj as the eluting solvent mixture) to yield a mixture of the two epi- meric alcohols (6j0) . The purified product was a pale yellow oil (0.779 g, 525); ir cm " 1 (CHC13) : 3700-3460

(broad), 2920, 2840, 1745 , 172S (sh) , 1638, 1580, 1370,

1095, 980; nmr (CDClj): 50.90 (t, J = 5Hz, 3H,

\ (CH7 ).CH,, 6 1.10-1.70 (br m, 8 H, 2}4 — 3 * HO

2.14 (s, 3H, -02C CH3), 5 2.15 (s, 3H, -02C CH^), 6 2.57

(br s, 1H, -OH), 6 4.16-4.40 (m, 1H, H OH AcO 0

6 5.22 (d, J = 2.5Hz, 1H ), 6 6.09 (d, J = 2.5Hz

AcO 0

6 6.20-6.35 (m, 3H, vinylic protons);

CHC1 3: 272 nm (log e 4.22). max

Attempted thioketalization of ketone 60 (105). A 10- ml round bottom flask was heated in an oven at 1 2 0 °, stop­ pered with a small serum cap., and allowed to cool to room temperature. A solution of ketone i60 (0.050 g, 1.154 mM)

dissolved in 1 ml of acetic acid (distilled from P 2 ^ 5 ^ was injected into the flask. While this solution was stirring, 72

first 1,2-ethanedithiol (0.03 ml, 0.032 g, 0.342 mM, 2.22

equiv.) and then boron trifluoride etherate ( 2 drops) were

sequentially injected into the flask. The reaction was

mildly exothermic. The reaction mixture was stirred for

1.5 hours at room temperature, during which time the color

of the mixture changed from yellow to brown.

Then ether and distilled water were added to the

reaction mixture. The organic phase was washed with

three times (to remove AcOH), dried (anh. Na 2 SO^) and con­

centrated to a brown oil (0.085 g). The ir (CHCl^) spectrum of the crude product showed no significant absorption at

1725 cm However, examination of this product by tic

(silica gel G, 1:24 MeOH-CHClj) revealed the presence of at

least nine components.

4 (S), 5(R)-Diacetoxy-3-(3-[Z-tetrahydropyranyloxy]- trans-1 - octenyl) - 2- eye 1 open ten-1 - one (61_) - Compound j6j0

(0.050 g, 0.154 mM) was transferred as a neat liquid to a

5-ml round bottom flask, which contained a magnetic stirring bar. This starting material was dissolved in ether (0.25 ml).

Next dihydropyrane (0.02 ml, 0.019 g, 0.232 mM, 1.5 equiv.) was added to the flask. Then a solution of £-toluenesulfonic acid (0.0016 g, 0.0092 mM, 0.06 equiv.) dissolved in ether

(0.25 ml) was transferred to the reaction flask.

The reaction mixture was stirred for 2 hours at room temperature, during which time the color of the mixture changed from pale yellow to brown. Then sat. NaHCOj solution and more ether were added to the flask. After the aqueous phase was removed, the ethereal phase was washed with sat.

NaHCO^ solution, distilled water and sat. NaCl solution.

Next the ethereal phase was dried (anh. Na 2 S0 ^), filtered and concentrated to yield a brown oil (65.7 mg). Purification of the crude product by thick layer chromatography (silica gel G,

1:32 MeOH-CHClj) afforded THP-ether 61 as a pale yellow oil

(44.2 mg, 7 01); ir cm ' 1 (CHClj) : 3000, 2940, 2860 , 1745 ,

1730 (sh), 1640, 1465, 1450, 1440, 1375, 1240, 1125, 1080,

1025, 970; nmr (CDClj): 6 0.89 (t, J = 5Hz, 3H,

THPO

CH3), 6 1.10-1.55 (br m, 8 H, 6 1.63 (m,

OTHP

CH / “J

6H , CH0 CH2 ), 6 2.13 (s, 3H, -02C CH3) , 6 2.15 (s, 3H,

(br s, 1H, anomeric proton), 6 5.20 (d, J = 2.5Hz, 1H 74

6.20 (m, 3H, vinylic protons); uv aE|?^3: 274 nm (log z 4.20), max 326 nm (sh) (log c 3.23).

4 (S), 5 (S) -Diacetoxy-l-(3- [2 - tetrahydropyranyloxy ] -

trans-1 -octenyl)-eye1openten-3-ol (62J . A 2-neck 15-ml round

bottom flask (containing a magnetic stirring bar) was heated

in an oven at 1 2 0 °, purged with argon and then maintained

under a positive pressure of argon. After it had cooled to

room temperaturea solution of ketone 6lI (0.052 g, 0.127 mM)

dissolved in tetrahydrofuran (1 . 1 ml) was injected into the

flask- The flask was then cooled to -78°.

Next 0.14 ml of a 1.107 M solution of lithium tri-

sec-butylborohydride (obtained from Aldrich under the name of

"L-selectride") (0.155 mM, 1.22 equiv.) was injected dropwise

into the flask. This addition caused the color of the re­ action mixture to change from pale yellow to yellowish brown.

After 2 hours of stirring at -78°, a 10% a q . HC1 solution

(0.1 ml) was injected into the flask. The flask was allowed to warm to room temperature and then more HC1 solution (8 drops) was added to adjust the reaction mixture to pH 6 , The

final acidification caused the color of the reaction mixture

to change to pale yellow.

After the acidified reaction mixture had stirred at room temperature for an additional 45 minutes, ether and distilled water were added. The aqueous phase was removed and then the ethereal phase was washed with water and sat. 75

NaCl solution. Next the ethereal phase was dried (anh.

NazS0 4), filtered and concentrated to a pale yellow oil

(69.5 mg). Purification of the crude oil by thick layer chromatography (silica gel G, 1:24 MeOH-CHClj) yielded the reduced product (62) as a pale yellow oil (37 mg, 711); ir cm' 1 (CHClj): 3640-3300 (broad), 2920, 2855, 1735,

1450, 1370, 1125 , 970; nmr (CDClj) : 6 0.89 (t, J = 5Hz,

3H, ^-(CH2 )4 CH3) , 6 1 . 09- 1.52 (br ni, 8 H, ^ (CHZ)4 CH5) ,

THPO THPO

CH, S —2^ 6 1.63 (m, 6 H , CH 7 CH.), 6 2.06 (s, 3H, -07C CH,),

6 2.09 (s, 3H, -02C CHj), 6 2.78-3.05 (br s, 1H, -OH),

<5 3.3-4.2 (m, 4H,

6 4,57 (br s, 1H, anomeric proton), 6 5.15-6.22 (br m, 5H,

vinylic protons and H* ); uv *max 3: 236 nm (log e 3.84);

AcO ci mass spectrum (isobutane): 393 (7%), 309 (141), 277 (24%),

261 (14%), 250 (14%), 249 (100%), 134 (24%), 133 (62%), 115

(21%), 103 (21%), 8 6 (10%), 85 (72%), 84 (14%), 61 (17%). 76

3-(4(S),5(S)-Diacetoxy-1-[3-hydroxy-trans-1-octenyl]- cyclopentenyl) ethyl malonate (63). The diastereomeric mix­ ture of alcohols represented by structure 6_2 (0. 0285 g,

0.0694 mM) was transferred to a 15-ml round bottom flask that contained a magnetic stirring bar. Tetrahydrofuran (0.8 ml) was added to the flask and, after the starting material had dissolved, the flask was cooled to 0°C. Next 0.10 ml of a

1 0 1 (v/v) solution of ethyl malonyl chloride (obtained from

Aldrich, lot No. 112137) in THF (0.0125 g, 0.0831 m M , 1.2 equiv.) was injected into the flask. Then 0.10 ml of an

11.61 (v/v) solution of triethylamine in THF (0.0084 g,

0.0831 mM, 1.2 equiv.) was injected into the flask. Upon the addition of triethylamine the pale yellow reaction mixture became cloudy. The reaction mixture was stirred for 2 hours at 0 ° and was then allowed to warm to room temperature.

After the mixture had stirred for an additional 1.5 hours at room temperature, 0.05 N HC1 solution (1 ml) was added to the flask. Next conc. HC1 reagent (3 drops) was added to adjust the reaction mixture to pH.l. This acidified reaction mix­ ture was stirred for 1 hour.

Ether and distilled water were then added to the reaction mixture. The aqueous layer was removed and the ethereal layer was washed with distilled water, dried (anh.

Na^SO^), filtered and concentrated to a yellow oil (0.036 g ) .

Purification of the crude oil by thick layer chromatography 77

(silica gel G, 1:24 MeOH-CHCl^, 2 developments) afforded the desired product (63) as a pale yellow oil (18.6 mg, 61%); ir cm ' 1 (CHC13): 3600-3420 (broad), 2935, 2870, 1735, 1655,

1460, 1375, 1310. 1125, 980; nmr (CDClj): 6 0.89 (t, J = 6 H z ,

3H, (CH2 )4 CH3), 6 1.06-1.75 (br m, 8 H, (cu2 )4 c h 3 .), HO HO

6 1.28 (t, J * 7Hz, 3H, -C02 CH2 CH3), 6 2.07 (br s, 6 H, -0 2 C-

CHj), 6 3.37 (br s, 2H, -02C CH 2 -C02 CH 2 CH3), 6 3.58 (br s,

1H, -OH), 6 4.19 (q, J = 7Hz, 2H, -C02 CH2 CH3) , 6 4.3-4.5

(m, 1H, v ^ ( C H 2 )4 CH3) , 6 5 .09- 5. 21 (m, 1H, - .- ^ 0 C0 CH2C0 2Et* ■ HO H O w ^ S — Ac

AcO -i 5 5.23-6. 25 (br m, 5H, vinylic protons and o V > uv ^max^' Ac6

237 nm (log e 3.87).

Attempted manganese dioxide oxidation of alochol 6 3 .

The diastereomeric mixture of alcohols represented by struc­ ture 63 (0.050 g, 0.114 mM) was transferred to a 15-ml rb flask which contained a magnetic stirring bar. The starting material was then dissolved in spectral grade chloroform 78

(2.S ml). Next activated Mn02 (obtained from Winthrop Lab­ oratories) (0.200 g, 2.30 mM, 20.2 equiv.) was added to the flask. This reaction mixture was stirred at room tempera­ ture for 48 hours. Then the mixture was filtered through a funnel which had a sintered glass frit. The residual Mn0 2 was thoroughly washed with methanol. The filtrate was dried

(anh. Na2 SO^), filtered and concentrated to a yellow oil

(42.4 mg). Examination of the product by tic (silica gel G,

1:24 MeOH-CHCl3) and nmr (CDClj) revealed that it was unreacted starting material (65).

Attempted biphasic chromic acid oxidation of alcohol

65 (106). Chromium trioxide (0.012 g, 0.120 mM, 1.05 molar equiv., 1.58 oxidative equiv.) was dissolved in distilled water (2 ml) in a 25-ml rb flask. Then a solution of the starting material (alcohol 6_3) (0. 050 g, 0.114 mM) dissolved in ether (2 ml) was added to the flask. The two phases were vigorously stirred together at room temperature for 15 hours.

At the end of this stirring period the aqueous phase had changed in color from yellow to green.

The aqueous phase was then removed, and the ethereal phase was directly spotted on a preparative silica gel G plate. This plate was developed once (1:24 MeOH-CHCl^), revealing one major and three minor bands. Examination of the major band (38,1 mg) by ir (CHCl^) and nmr (CDClj) revealed that it was unteacted starting material (65). 79

4 (S),5(S)-Dibenzoxy-1-formyl-3(R)-hydroxycyclo- pentene (65) (107). 4 (S),5(S)-Dibenzoxy-1-(1,2-dihydroxy- propyl)- 3(R)-hydroxycyclopentene (

1470, 1340, 1255, 1140, 1125, 1100, 900; nmr (CDClj): 6 4.65

(br s, 1H, vinylic proton), 6 7.18-8.18 (br m, 10H, benzoate MeOH protons), 6 9.75 (s, 1H, -CHO); uv \ : 2 29 nm (log e 4.43), max

272 nm (log e 3.40), 279 nm (sh) (log e 3.31); ci mass spec­ trum (isobutane): 353 (MH+ , 2%), 263 (20%), 231 (12%), 123

(100%), 105 (12%). 80

4 (S) , 5(S)-Dibenzoxy-3 (R) - hydroxy-1- (5-oxo - t.rans - 1-

octenyl)-eye 1opentene (6 6 ) (109). A 2-neck 50-ml round bottom flask (containing a magnetic stirring bar) was heated

in an oven at 1 2 0 ®, purged with argon and then maintained under a positive pressure of argon. After it had cooled to room temperature, a 0.03047 M THF solution of the sodium salt of dimethyl (2-oxohepty1)-phosphonate (11.2 ml, 0.341 mM,

1.20 equiv.) was injected into the flask. Next a solution of compound 6jj (0.100 g, 0.284 mM) dissolved in THF (3 ml) was injected into the flask. The resultant mixture was allowed to stir at room temperature for 45 minutes.

The reaction mixture was then diluted with ether and washed with 5% aq. HC1 solution and sat. NaCl solution. The combined aqueous phases were extracted with ether. Next the combined ethereal phases were dried (anh. Na 2 S0 4), filtered and concentrated to a pale yellow oil (0.172 g). Purifica­ tion of the crude product by thick layer chromatography

(silica gel G, 1:24 MeOH-CHClj, 2 developments) afforded compound 6 6 as a colorless oil (0.087 g, 6 8 %); ir cm *

(CHC13): 3 540, 2980, 2910, 1735, 1690 (sh), 1645, 1620, 1470,

1340, 1250, 1140, 1125, 1100, 1010, 895; nmr (CC14) : 6 0.91

(t, J = 5Hz, 3H, (CH2 )4 CH3), 6 1.06-1.70 (br m, 6 H,

0

^ C H 2 (CH2 )3 CH3), 6 2.37 (t, J - 6 Hz, 2H, O \ CH 2 (CH2 )3 CH3), 6 3.66 (br s, 1H, -OH), 6 4.73 (m, 1H 81

BzO OH Bz0 OH

, 6 5.16 (t, J = 3.5Hz, 1H, H* U ), 6 6.27

(d, J = 18Hz,

BzCL OH H H ), 6 7.18 (d, J = 18Hz, 1H, H 0

MeOH. 6 7.25-8.28 (br m, 1GH, benzoate protons); uv \ : 231 nm max

(log e 4.48), 264 nm (log e 4.37); ci mass spectrum (iso­ butane): 449 (MH+ , 2%), 432 (11%), 431 (35%), 327 (20%),

309 (4%), 207 (17%), 205 (55%), 123 (100%).

3 (R)-(4(S),5(S)-Dibenzoxy-1-[3-oxo-trans-1-octenyl)]- cyclopenteny 1) ethyl malonate (6^7). A solution of compound 6 j6

(0.058 g, 0.130 mM) dissolved in tetrahydrofuran (1.5 ml) was transferred to a 15-ml rb flask that contained a magnetic stirring bar. The flask was warmed to 42°. Next 0,19 ml of a 10% (v/v) solution of ethyl malonyl chloride in THF (0.0238 g, 0.158 mM, 1.22 equiv.) was injected into the flask. Then

0.22 ml of a 10% (v/v) solution of triethylamine in THF

(0.016 g, 0.158 mM, 1.22 equiv.) was injected into the flask.

The resultant mixture was stirred for 1 hour at 42°, before being cooled to room temperature.

Ether was then added to the reaction mixture. The 82 mixture was washed with 101 aq. HC1 solution and sat. NaCl solution. The combined aqueous phases were extracted with ether. Next the combined ethereal solutions were dried

(anh. Na 2 S0 4), filtered and concentrated to a yellow oil.

Purification of this oil by thick layer chromatography

(silica gel G, 1:99 MeOH-CHCl^) yielded compound 6_7 as a pale yellow oil (0.0517 g, 71%); ir cm * (CHCl^): 2980, 2910,

1735, 1690 (sh), 1645, 1620, 1475, 1390, 1345, 1290, 1135,

1100, 1045, 1020, 900; nmr (CC14) : 6 0.90 (t, J = 6 Hz, 3H,

(CH2 )4 CH3), 6 1. 05-1. 70 (br m, 9H, CH2 (CH2)3~

CH3 and -C02 CH2 CH3), 6 2.37 (t, J = 6.5Hz, 2H,

\ CH 2 (CH2 )3 CH3), 6 3.34 (s, 2H, -02C CH2 C0 2 CH 2 CH3), 6 4.12

(q, J ** 7Hz, 2H, -C02 CH 2 CH3) , 6 5.58 (t, J = 3Hz, 1H,

Bz O O' Bzn O' „

), 6 5.80 (m, 1H, \ ), 6 6.28 (d, J = 3Hz, S '>

Bz0 O'o' 1H, \ ^ - \ * ), 5 6.33 (d, J = 16. 5Hz, 1H, iT-- H O • & ’ ■

0

(CH2 )4 CH3), <5 6.41 (d, J = 2.5Hz, , 6 7.17 (d,

Hi J = 16.5H2, 1H V ^ T f C C H - ) .CH,) , 6 7.21-8.19 (br m, 10H, H O L 4 A Me* OH benzoate protons); uv * max : 232 nm (log e 4.46), 264 nm

(log e4.36); ci mass spectrum (isobutane): 441 (23%), 432

(27%), 431 (92%), 321 (16%), 320 (12%), 319 (60%), 311 (14%), 83

309 (22%), 308 (15%), 205 (19%), 187 (11%), 161 (100%), 133

(16%), 123 (63%), 115 (24%), 105 (70%).

Attempted cyclization of compound 67 with disodium

salicylate. A 2-neck 15-ml round bottom flask (containing a

magnetic stirring bar) was heated at 1 2 0 ®, purged with argon

and then maintained under a positive pressure of argon.

After it had cooled to room temperature, tetrahydrofuran

(1 ml) and disodium salicylate (0.0011 g, 0.006 mM, 0.10

equiv.) were added to the flask. To this mixture was added

a solution of compound

THF (1 ml). The resultant mixture was stirred vigorously

for 22 hours at room temperature. Then more THF (1 ml) and more disodium salicylate (0.0044 g, 0.024 mM, 0.4 equiv.) were added to the reaction mixture, which was warmed to 40°.

After 24 hours of stirring at 40®, the reaction mixture was diluted with ether and washed with sat. NaCl solution. The ethereal phase was dried (anh. Na2 S0 ^), filtered and concen­ trated to a pale yellow oil (0.0295 g). Examination of the crude product by tic (silica gel G, 1:99 MeOH-CHCl^), ir

(CHClj) and nmr (CCl^) revealed that it consisted almost exclusively of unreacted starting material (67).

Attempted cyclization of compound 67 with sodium methoxide. A 2-neck 15-ml rb flask, which contained a magnetic stirring bar, was heated in an oven at 120°. The 84

flask was then purged with argon and maintained under a posi­

tive pressure of this gas. The flask was cooled to -23°

(CCl^/CC^)- Next a solution of compound ^7 (0.0284 g,

0.050 mM) dissolved in methanol (1.5 ml) was injected into

the flask. Subsequently, a 0.04 M solution of sodium

methoxide in methanol (0.125 ml, 0.0050 mM, 0.100 equiv.) was

injected into the flask. After the resultant mixture was

stirred for 2 hours at -23°, the reaction was quenched by the

addition of a 101 aq. HC1 solution (0.5 ml).

The reaction mixture was allowed to warm to room tem­

perature. Then the mixture was concentrated to a yellow oil.

Purification of this oil by thick layer chromatography (silica

gel G, 1:99 MeOH-CHCl^) yielded two components--unreacted

starting material (67^) (0.0042 g) and the product formed by

hydrolysis of the ethyl malonate group (i.e., ]56) (0.013 g,

67%). The identities of these two products were confirmed by

their respective nmr (CC14) spectra.

1 (R)* 5(R),8 (R)-Benzoxy-4 (R)-carbethoxy-6 - (3-oxo-

trans-1-octenyl)-3-oxo-2-oxabicyclo (3.3.0] oct-6 -ene (6 8 ).

A 2-neck 15-ml round bottom flask (containing a magnetic

stirring bar) was heated at 1 2 0 °,'purged with argon and then maintained under a positive pressure of argon. After it had

been cooled* to room temperature, a solution of compound 67

(0.057 g, 0.102 mM) dissolved in t.-butyl alcohol (3 ml) was

injected into the flask. Next a 0.04 M solution of sodium 85 t^butoxide in _t-butyl alcohol (0.26 ml, 0.0104 mM, 0.10 equiv.) was injected into the flask. The resultant mixture was stirred for 2.5 hours at room temperature. Then the reaction mixture was warmed to 40° and stirred for an addi­ tional 3.5 hours. At the end of that stirring period more of the sodium _t-butoxide solution (0.52 ml, 0. 0208 mM,

0 . 2 0 equiv.) was added to the reaction mixture, which was then stirred at 40° for an additional 24 hours.

Finally, one last portion of the sodium t-butoxide solution (0.52 ml, 0.0208 mM, 0.20 equiv.) was added to the flask and the reaction mixture was refluxed for 16 hours.

Next the rb flask was allowed to cool to room temperature and the reaction was quenched by the addition of 10% aq. HC1 solution (1 ml). The quenched reaction mixture was concen­ trated to a residue, which was purified by thick layer chromatography (silica gel G, 1:99 MeOII-CHCl^, 2 develop­ ments). Seven minor bands and one major band were seen on the preparative plate. Extraction of the major band afforded compound 6 8 as a pale yellow oil (0.012 g, 27%); ir cm ' 1 (CHC13): 2980, 2910, 1790, 1735, 1690 (sh), 1640,

1620, 1475, 1390, 1345, 1290, 1190, 1135 , 1100, 1060, 1020,.

900; nmr (CC14) : 6 0.92 (t, J = 6 Hz, 3H, V'(CH 2 )4 CHj) ,

6 1. 09- 1 . 72 (br m, 9H, V ~ C H 2 (CH^ 3 CH 3 and -C(>2 CH 2 CH3) ,

6 2.50 (t, J = 6 Hz, 2H, V “CH 2 (CH2 )jCHj), 6 3.37 (d, 86

S .CO-jEt

3.5Hz, 1H, V ), 5 4.17 (br d, J = 7Hz, 1H, /%

0 CO,Et

H€ / ), 6 4.32 (q, J = 7Hz, 2H, -C02 CH2 CH3) , 6 5.2 2

2 . CO.Et Bz0

2 V (br d, J = 7Hz , !H, h ' 11 )• 4 5 *8 9 Cm* 1H» .Y?5» > n ' J

H 6 6.20 (d, J = 18Hz, 1H, (CH2 )4 CH3) , 6 6.31 (br s,

rf ^ ? 0 H

1H, 6 7 * 2 6 (d> J = 18Hz» 1H* ^ cS^ tr^ ( CH2)4CH3) ’

- 0 MeOH 6 7.18-8.17 (br m, 5H, benzoate protons); uv *max : 231 nm

(log e 4.24), 26Z nm (log e 4.37); ci mass spectrum (iso­ butane): 442 (161) , 441 (MH+ , 52%), 399 (27%), 351 (14%),

343 (16%), 335 (12%), 321 (11%), 320 (21%), 319 (95%), 314

(20%), 313 (93%), 301 (10%), 299 (20%), 279 (12%), 225 (31%),

223 (100%), 205 (12%), 161 (27%), 123 (24%).

1 DISCUSSION

The initial experimental work involved the syn­

thesis of compounds to be used in conjugate addition reac­

tions with appropriate derivatives of terrein, in order to

introduce a side chain a to the carbonyl group of terrein

(36). The first compound synthesized for this purpose was

the THP-ether of 7-iodo-1-heptanol (41). It was believed

that following a conjugate addition reaction with 41, hydrolysis of the tetrahydropyrany1 ether and subsequent oxidation of the resultant alcohol to a carboxylic acid would provide access to one or more of the PG^ series. The

synthesis of THP-ether 41^ is delineated in Figure 16.

1. B2H6 KMn04 I co 2h NaHCO 1,7 - OctadUn* 1.03

2.NaBH

40 41

FIG. 16. Synthetic routes to the THP-ether of 7-iodo-l- heptanol (41_). 88

1,7-Octadiene was converted to 1-iodo-7-octene (38)

by a hydroboration procedure (89) in 261 yield. Two routes

were found for the preparation of a new compound, 7-iodo-1-

heptanol* (40), from 1 -iodo-7-octene (38). In the first, a

conventional alkaline potassium permanganate oxidation of

the olefin (_38) produced in 224 yield 7-iodoheptanoic acid

(_39), which was then reduced by diborane (90) to 7-iodo-l-

heptanol (£0) in good yield (77%). In the second, l-iodo-7-

octene (38) was converted to its ozonide, which was reduced without isolation by an ethanolic mixture of sodium borohy-

dride (91) to produce the alcohol (£0) in a 444 yield.

7-Iodo-1-heptanol (40) was converted to its tetrahydropyranyl

ether (_41) using a catalytic amount of £-toluenesulfonic acid

in diethyl ether (97% yield) (92).

Crystalline tetrakis (iodo (tri-n-butylphosphine) copper (I) ] (4_2) was prepared in 73% yield from freshly distilled tri-n-butylphosphine and purified cuprous iodide

(see Figure 17). This reagent is soluble in organic sol­ vents and serves as a carrier of cuprous iodide for the in situ generation of organocopper complexes (93). Reagent 4_2 was used in the attempted conjugate addition of the THP- ether of 7-iodo-1 -heptanol (4_1) to trans- 2-methyl -2-pentenal.

The previous reaction failed because the metal- halogen exchange between _t-butyllithium and the primary alkylhalide (41) was unsuccessful at -78°, -23° and 0°. 89

With more vigorous conditions an equilibrium might have been attained, however vigorous conditions might have led to dimerization.

4 (b -C4«9)3P + 4 Cu> ------> [ { (q -C4H9 )3p } c u |]4

42

1. fM.)3CLi, -78°

2. 42 , 78 No Reaction 3.

l.ti-Na (99:1)

38

FIG. 17. Attempted formation of organolithium compounds from primary alkyliodides.

Attempts were made to react another primary alkyl- iodide (38) directly with lithium. When lithium metal con­ taining only 0 .0 0 1 % sodium was used, no reaction occurred with 1 -iodo-7-octene (38) in ether even at room temperature. 90

However, when compound _38 was reacted with a 99:1 lithium- sodium alloy (1 1 0 ) dimerization, ra*ther than addition to benzophenone, occurred. Thus 1,15-hexadecadiene (4_3) was formed in 79% yield.

One literature reference (111) indicated that the reaction of a primary alkyliodide with lithium, once initiated, should proceed predominately or exclusively to the dialkyl coupled product. The same source indicated that the reaction of a primary alkylbromide with lithium is less likely to lead to dimerization. For this reason, synthesis of the bromine analogues of the previously prepared compounds was undertaken (see Figure 18) .

44

(M eO C H^,

45 Ethyl 7-Bromo- h«ptonoat« P -T * 0 H

4 6

FIG. 18. Synthetic routes to the THP-ether of 7-bromo-l-

heptanol (£6 ). Ethyl 7-bromoheptanoate, which was available in the labora­ tory, was converted directly to 7-bromo-1-heptanol (45) in

80% yield by refluxing with diborane in tetrahydrofuran.

Ethyl 7-broraoheptanoate was also hydrolyzed to 7-bromohept- anoic acid (44) in 77% yield using a mixture of aqueous potassium hydroxide in glyme. This carboxylic acid (44) was readily reduced to the alcohol (£5) by treatment with dibor­ ane (91% yield); The THP-ether (46) was formed from 7-bromo-

1-heptanol (45) by the use of dihydropyrane and a catalytic quantity of p-toluenesulfonic acid (95% yield). Compound £6 was purified by neutral alumina (Woelm, activity grade III) column chromatography.

As shown in Figure 19, the tetrahydropyranyl ether of

7-bromo-1-heptanol (£6) underwent dimerization when treated with a 99:1 1ithium-sodium alloy in ether to give the bis- tetrahydropyranyl ether of 1 ,14-tetradecadiol (£7) (55% yield). Since dimerization had resulted from the reactions of an alkyliodide and an alkylbromide with lithium, an attempt was made to form the lithium salt of 2-chloro-1,1- diethoxyethane (4_8). If the conjugate addition of acetal

48 to a derivative of terrein were successful, hydrolysis would then produce an aldehyde. Such an aldehyde should provide access to PG 2 compounds via a Wittig reaction.

A benzene solution of 2-chloro-1,1-diethoxyethane

(48) reacted with the 99:1 1ithium-sodium alloy only when heated to the reflux point (see Figure 19). A subsequent 92

l.U-Na(99:l) ,

’- Y ° N et20, R.T.

2-C O j' E»2O' 0 46

l.li-N a (99: l) , 0 ^6H6 ' ^ _____ No Product 2.

* ( O ' * ® - Et2 0 ' R. T.

1. Li-Na (99:1)

Et20 ' -5 No Product S ' 2. »( On® E»2 0 ' 0°

FIG. 19. Reactions between a 99:1 1ithium-sodium alloy and three primary alkylhalides. 93 condensation with benzophenone was unsuccessful, yielding only unreacted benzophenone. No unreacted acetal (4ji) was recovered. A possible explanation for these experimental results is that compound 48 reacted with lithium to form a carbanion which immediately eliminated ethoxide. The resultant vinyl ethoxide could then have undergone hydro- lysis to acetaldehyde during the reaction work-up (see

Figure 20). The acetaldehyde would then have been removed by aqueous extraction and subsequent concentration of the organic phase.

It was hoped that 2-bromo-1,1-diethoxyethane (49) would react with the 99:1 1ithium-sodium alloy under conditions sufficiently mild to avoid an internal elimi­ nation of ethoxide. However, although this acetal (49) reacted with lithium at -5°, the subsequent attempted condensation with benzophenone was again unsuccessful (see

Figure 19). Therefore, another approach was tried.

The tetrahydropyrany1 ether of 2-bromoethanol (51) was prepared in the belief that it might form a lithium salt without undergoing internal elimination. The success­ ful conjugate addition of this molecule (51) to an ap­ propriate terrein derivative would permit entry to the series. That is, hydrolysis of the THP ether and oxidation of the resultant alcohol to an aldehyde would provide an opportunity for elaboration of this side chain by means of a Wittig reaction. 94

< LI-No (99:1) U 0

/ ® \ 0 + ( ) V OH S = / + LiOEt H

FIG. 20. Proposed reaction between a 99:1 1ithium-sodium alloy and 2-bromo or chloro-1,1-diethoxyethane. 95

An ethereal solution of 2-bromoethanol (M)) was treated with dihydropyrane and a catalytic amount of p-toluenesulfonic acid. Distillation of the crude product from sodium carbonate afforded THP-ether 51_ in 69% yield

(see Figure 21). Surprisingly, this compound (51) did not react with a 99:1 lithium-sodium alloy in ether at room temperature. Thus a copper catalyzed Grignard reaction was attempted.

1,2-Dibromoethane was used with THP-ether 5J_ in or­ der to facilitate reaction with magnesium. The product from this reaction was treated sequentially with tetrakis [iodo

(tri-n-butylphosphine) copper (I)] (£2) and trans-2-methyl-

Z-pentenal at -78° to produce 5-hydroxypentanal (£2) (69% yield) and unreacted trans-2-methyl-2-pentenal. Apparently the 5-hydroxypentanal (52) arose from internal elimination by the Grignard reagent (see Figure 21). The same product

(52) was formed by acid hydrolysis of the starting material

(!!>• All of the preceding work had failed to produce a reagent for the conjugate addition of a side chain at the unsubstituted carbon atom a to the carbonyl group of terrein.

Therefore, a new approach was undertaken. Diallyl lithium cuprate was reacted with model compounds to study which derivatives of terrein would be suitable for a conjugate addition reaction. The conjugate addition of an allyl group 96

OH P-TsOH 8r 50 0 * < T j

Li-Na (99:l)f u „ 51 ^ No Reaction EfjO, R.T.

l.Mg,THF 2.0.5 tq. f{ (n-CH,)3 p}C ull. 3 + B r-^8' E„0, -78° 4 9 3 r U

O® jHO^j + ^ + BrM,® 4------

© V >0 h 3o ® —2______51

52

FIG. 21. Attempted formation of an organolithium reagent from the THP-ether of 2-bromoethanol (51). 97 to an appropriate derivative of terrein was expected to pro­ vide access to a variety of PG^ compounds, since the terminal olefinic double bond of an allyl group should be convertible to an aldehyde.

HO LiCu H

Trane -2 -iw th y l- 53

2 -p «n t« n a l

LiCu 0

OM* OMe Methyl Acrylate 54

FIG. 22. Reactions of diallyl lithium cuprate with a,|J- unsaturated carbonyl compounds.

The reaction of an ethereal mixture of diallyl lithium cuprate (98,99) with trans-2-methyl-2-pentenal at

-78° yielded the 1,2 addition product (see Figure 22). This product, trans- 5-methyl- l,5-octadien-4-ol (S_3) , was obtained in a yield of 671. However, the reaction of diallyl lithium cuprate with methyl acrylate afforded the conjugate addition product, methyl 5whexenoate (^4), in 79% yield. It thus 98

appeared that conversion of terrein to a derivative which

was an oc, B-unsaturated ester might permit conjugate addition

of diallyl lithium cuprate.

4 (S),5(R)-Dibenzoxy-3-formyl-2-eyelopenten-l-one

ethylenethioketal (55) was prepared by the joint efforts of

Drs. K. Shirahata and G. W. Clark (101). The synthesis of

this compound (55) from terrein (.36) is depicted in Figure

23. The a , 3-unsaturated aldehyde (5j>) was converted directly

HO BzO,

HO' Bz 36

Bz BzO BzO

OH

Bz CHO Bz 55 HO HO

FIG. 2.3. The synthesis of 4 (S) , 5 (R)-dibenzoxy-3-formyl-2- cyclopenten-1-one ethylenethioketal (55). to an a, B-unsaturated ester, 3-carbomethoxy-4(S),5(R)- dibenzoxy-2-cyclopenten-l-one ethylenethioketal (56), in 72% yield (see Figure 24) (100). A possible mechanism for this 99 reaction involves formation of a cyanohydrin which is then t oxidized to an acyl cyanide. The intermediate acyl cyanide would presumably be labile to nucleophilic attack by metha­ nol to produce the ester (^6)■

The reaction of an ethereal mixture of diallyl lithium cuprate with compound 5j> led to loss of the allylic

N a C N , A cOH, 55 ------M n O j' M*OH

56 O 57 0

FIG. 24. Conversion of compound 5_5 to 5 (R)-benzoxy-3- carbomethoxy-2 -(2-propenyl)-3-cyclopenten-1 - one ethylenethioketal (5^7) . benzoate by a process that likely follows an SN2' pathway.

Thus 5 (R)-benzoxy-3-carbomethoxy-2-(2-propenyl)-3-cyclo- penten-l-one ethylenethioketal (57) was produced in 51% yield. It might be possible to quench this reaction before elimination of the benzoate group either with a proton source or by further reaction with a ‘'trapping" compound (such as

CISiRj). However, no attempts were actually made to modify the last reaction. Instead derivatives of terrein (36) were prepared in order to study the feasibility of an internal

Michael addition reaction. 100

Terrein diacetate (5jp was produced from terrein

(36) using acetic anhydride and either acid or base catal­ ysis (see Figure 25) (102,103). When acetylation was per­

formed with anhydrous sodium acetate a 99.8% yield of 58^ was obtained. Acetylation with anhydrous j>-toluenesul-

fonic acid gave the diacetate (58) in 88% yield. The side chain methyl group of terrein diacetate (_58) was then oxidized to an aldehyde. A xylene solution of terrein diacetate (58) was refluxed in the presence of freshly resublimed selenium dioxide to produce 4 (S),5(R)-diacetoxy-

3- (trans-1-propenal)- 2-eyelopenten-1-one (59) in 301 yield

(104).

Treatment of compound _59 with one equivalent of n- pentyl magnesium bromide at -78° gave selective reaction with the aldehyde. The resultant 4 (S),5(R)-diacetoxy-3-

(3-hydroxy-trans-1-octenyl)-2-cyclopenten-l-one (60) was obtained in 52% yield. An attempted thioketalization of compound (H) yielded a complicated mixture of products.

4 (S),5(R)-Diacetoxy-3 -(3-[2-tetrahydropyranyloxy]- trans-1-octenyl)- 2-cyclopenten-1-one (61) was prepared from the alcohol (^0) using dihydropyrane and a catalytic quan- . tity of £-toluenesulfonic acid (70% yield). This THP-ether

(61) was reduced with lithium tri-sec-butylborohydride to give a 71% yield of 4 (S),5 (S)-diacetoxy-1 - (3-[2-tetrahydro- pyranyloxy]-trans-1-octenyl)-cyclopenten-3-ol (62). 101

A c£ 0

NaOAc

AcO, HO s«o

AcO CHO

p-TsOH

-78

AcO AcO,

AcO AcO*

HO

Ac OH

BH Nino Uncharac torizod Compounds Ac, Ac

62 63 HO

Mn O2 ' CrO CHCI,

Thr«« Minor Unch< No Reaction Products

FIG. 25. The synthesis of 3-(4(S),5(S)-diacetoxy-1-[3- hydroxy-trans-1-octenyl]- eye1opentenyl) ethyl malonate (63).

Ester if ication of compound 6^2 with ethyl malonyl chloride and

subsequent hydrolysis of the tetrahydropyranyl ether pro­ duced 3- (4(S),5 (S)-diacetoxy-1- [3-hydroxy-trans-1-octenyl]- cyclopenteny1) ethyl malonate (63^ in 61% yield.

Two attempts to oxidize compound 63^ to a ketone were unsuccessful. Treatment of a chloroform solution of ()3 with activated manganese dioxide yielded only unreacted starting material. A biphasic chromic acid oxidation of compound 6_3 produced three minor products, which were not characterized.

In addition, there was a 76% recovery of unreacted starting material from the last reaction. It was felt that a stronger 103 oxidizing reagent would successfully oxidize the allylic alcohol (63). However, no further oxidative attempts were made for two reasons. First, the stereochemistry of the ring carbon attached to the ethyl malonate group was not known, and a subsequent Michael addition reaction (of the methylene carbon of the malonate side chain) would proceed to give the required (a) prostaglandin stereochemistry only if the malonate group had the a configuration. Secondly, a shorter route was developed to an intermediate which had the desired stereochemistry.

4 (S),5(S)-Dibenzoxy-1-(1,2-dihydroxypropy1)-3(R)- hydroxycyclopentene ((^4) was synthesized by Dr. G. W. Clark

(108) as shown in Figure 26. Note that the preparation of the starting compound used in this synthetic step was indi­ cated in Figure 23.

qH

BzO BzO OH OH

BzO* BzO HO

FIG. 26. The synthesis of 4 (S),5(S)-dibenzoxy-1-(1,2- dihydroxypropyl)-3(R)-hydroxycyclopentene (64.).

Compound 6£ was oxidized to an aldehyde (6j>) using sodium meta periodate (see Figure 27) (107). The resultant

4 (S),5(S)-dibenzoxy-1-formyl-3 (R)-hydroxycyclopentene (65) 104

OH No

BzO No 10, 64

BzO CHO 65

NEt3 , OH ,CICOCH2C02Et Bz

N a O M *' MeOH

BzO* Bz 67 66

° 0 2 Na

No Reaction 68

FIG. 27. Conversion of compound 64 to 1 (R),5(R),8(R)- benzoxy-4 (R)- carbethoxy^l)-(3-oxo-trans-1-octeny1)- 3-oxo-2-oxabicyclo [3 . 3. O]oct- 6-ene (IjIQ . 105 was obtained in 91% yield. A modified Wittig reaction (109) between compound &S and the sodium salt of dimethyl (2-oxo- heptyl)-phosphonate produced 4 (S),5(S)-dibenzoxy-3 (R)- hydroxy-1-(5-oxo-trans-1-octeny1)- eye1opentene (66) in 68% yield. Esterification of compound 6>6 with ethyl malonyl chloride was catalyzed by triethylamine to give 3(R)-(4(S)- dibenzoxy-1-[3-oxo-trans-1-octenyl]- eye1opentenyl) ethyl malonate (67) in 71% yield.

A variety of bases were utilized in attempts to in­ duce compound 67 to undergo an internal Michael addition.

Disodium salicylate proved to be too weak a base to effect cyclization. Sodium methoxide, on the other hand, was suf­ ficiently nucleophilic to hydrolyze the ethyl malonate side chain. The use of sodium methoxide at - 23° produced compound

66 in 67% yield. Treatment of a refluxing solution of com­ pound 6^7 in t-butyl alcohol with sodium t_-butoxide yielded the cyclized product which had undergone elimination of the allylic benzoate group. 1(R) ,5(R),8(R)-Benzoxy-4 (R)- carbethoxy-6-(3-oxo-trans-1-octenyl)-3-oxo-2-oxabicyclo

[ 3. 3. 0 ] oct-6-ene ((>£) was thus obtained in 27% yield.

The nmr spectrum of compound displayed a doublet centered at 6 3.37 for the malonate methine proton. The band in the infrared spectrum at 1790 cm '*' is indicative of a five-membered lactone. The uv data for compounds 6IJ and

67 are also instructive; whereas the e values for the 106 a,B,y,5-unsaturated ketone absorption maxima (at 262 nm and

264 nm respectively) are practically identical, the e value

for the benzoate absorption maximum in compound 6ji (at 231 nm) is just over one-half of that for the benzoate absorp­

tion maximum in compound 67^ (at 232 nm). The isobutane chemical ionization mass spectrum of compound £»8 displayed prominent peaks for the protonated molecule ion (m/e 441) and the MH+-benzoic acid ion (m/e 319).

In summary, two procedures were developed for the alkylation of terrein derivatives at the pre-Cg prostaglandin position. One procedure employed a reaction with diallyl lithium cuprate, and the other involved lactonization. In each case an allylic ester function was eliminated, thus losing the pre-C^| hydroxyl group of the E and F series of prostaglandins. Both of the alkylated terrein derivatives

(68 and 57) should be convertible to prostaglandin inter­ mediates with fully elaborated upper side chains.

In addition, this research featured the development of two separate routes for complete elaboration of the prostaglandin lower side chain. One route utilized the reaction of an aldehyde (j>9) with a Grignard reagent. The other route involved a Wittig condensation with a different aldehyde (65).

The next phase of this research program will be to complete the synthesis of prostaglandins. BIBLIOGRAPHY

1. Kurzrok, R. , and Lieb, C. C . , Proc. Soc. Exp. Biol. Med. , _28, 268 (1930) .

2. Goldblatt, M. W. , J. Soc. Chem. Ind., 52, 1056 (1933).

3. Euler, U.S. von, J. Phys., 84, 21P (1934).

4. Bergstrom, S., Ryhage, R., Samuelsson, B., and Sjovall, J., J. Biol. Chem., 238, 3555 (1963).

5. Bergstrom, S., Science, 157 , 382 (1967).

6. Abrahamsson, S., Acta Crystallog., 16, 409 (1963).

7. Cuthbert, M. F. (Ed.), "The Prostaglandins," William Heinemann Medical Books Ltd., London (1973).

8. Bergstrom, S., Carlson, L. A., and Weeks, J. R., Pharm. Rev.,20,1 (1968).

9. Carlson, L. A., Prog. Biochem. Pharm., 3, 94 (1967).

10. Coceani, F., Dreifuss, J, J . , Puglisi, L., and Wolfe, L. S., in "Prostaglandins, Peptides and Amines," Mantegazza, P. and Horton’, E. W. , eds., p. 73, Academic Press, New York (1969).

11. Euler, U. S. von, and Eliasson, R . , "Prostaglandins," Medicinal Chemistry Monographs, Vol. 8, Academic Press, New York (1967).

12. Fried, J., Lin, C . , Mehra, M . , Kao, W . , and Dalven, P., Ann. N.Y. Acad. Sci., 180, 38 (1971).

13. Kaley, G . , and Weiner, R., Ann. N.Y. Acad. Sci., 180, 338 (1971).

14. Kloeze, J., Biochem. Biophys. Acta, 187 , 285 (1969).

15. Nakano, J . , Ishii, T . , and Gin, A. C., Clin. Res., 16, 30 (1968).

107 108

16. Ramwell, P. W. , and Shaw, J. E., Rec. Prog. Hor. Res., 26, 139 (1970).

.17. Schofield, J. G . , Nature (London), 228, 179 (1970).

18. Steinberg, D., Vaughan, M . , Nestel, P. J., and Bergstrom, S., J. Clin. Invest. , 43, 15S3 (1964).

19. Butcher, R. W . , and Sutherland, E. W . , Ann. N.Y. Acad. Sci. , 139, 849 (1967).

20. Hedqvist, P., Acta Physiol. Scand., 75, 511 (1969).

21. Vane, J. R. , Nat. New Biol. , 2_31 , 232 (1971) .

22. Jones, R. L. , Brit:. J. Pharm. , 45 , 144 (1972).

23. Nugteren, D. H., Van Dorp, D. A., Bergstrom, S., Hamberg, M . , and Samuelsson, B. , Nature (London), 212, 38 (1966).

24. Bergstrom, S., Carlson, L. A., and Weeks, J. R., Pharm. Rev., 20, 1 (1968) .

25. Weinheimer, A. J., and Spraggins, R. L., Tetrahedron Lett. , 59, 5185 (1969),

26. Bundy, G. L., Lincoln, F. H . , Nelson, N. A., Pike, J. E. and Schneider, W. P., Ann. N.Y. Acad. Sci., 180, 76 (1971).

27. Zaludek, G . , Amer. Drug., 171, No. 3, 33 (1975).

28. Skarnes, R. C . , and Howard, L. A., (Worcester Foundation For Experimental Biology), U.S. Patent 3,843,467, October 22, 1974 , via C. A, 82^: 29683d (1975).

29. Bergstrom, S., Danielsson, H., Klenberg, D., and Samuelsson, B., J. Biol. Chem., 239, PC4006 (1964).

30. Bergstrom, S., Danielsson, H., and Samuelsson, B., Biochim. Biophys. Acta, 9 0 , 207 (1964).

31. Van Dorp, D. A., Beerthuis, R. K., Nugteren, D. H., and Vonkeman, H . , Nature (London), 203, 839 (1964).

32. Van Dorp, D. A., Beerthuis, R. K., Nugteren, D. H., and Vonkeman, H., Biochim. Biophys. Acta, 90, 204 (1964). 109

33. Anggard, E., and Samuelsson, B. , J. Biol. Chem.., 240, 3518 (1965).

34. Lapidus, M. , Grant, N. H . , and Alburn, H. p., J. Lip. Res., 9, 371 (1968).

35. Daniels, E. G., Krueger, W. C., Kupiecki, F. P., Pike, J. E., and Schneider, W. P., J. Amer. Chem. Soc., 90, 5894 (1968).

36. Granstrom, E., Lands, W. E. M . , and Samuelsson, B., J. Biol. Chem., 243, 4104 (1968).

37. Sih, C. J., Ambrus, G., Foss, P., and Lai, C. J., J. Amer. Chem. Soc., 91, 3685 (1969).

38. Israelsson, U., Hamberg, M . , and Samuelsson, B., Eur. 2* Biochem., 11, 390 (1969).

39. Axen, U., Pike, J. E . , and Schneider, W. P., in "The Total Synthesis of Natural Products," Vol. I, J. ApSimon, ed., pp. 84-86, Wiley*Interscience, New York (1973).

40. Just, G., and Simonovitch, C., Tetrahedron Lett., 2093 (1967).

41. Holden, K. G. , Hwang, B. , Williams, K. R., Weinstock, J., Harmon, M., and Weisbach, J. A., Tetrahedron Lett. , 1569 (1968) .

42. Schneider, W. P., Axen, V., Lincoln, F. H . , Pike, J. E., and Thompson, J. L., J. Amer. Chem. Soc., 90, 5895 (1968).

43. Schneider, W. P., Axen, V., Lincoln, F, H., Pike, J. E., and Thompson, J. L., J. Amer. Chem. Soc., 91, 5372 (1969).

44. Just, G., Simonovitch, C., Lincoln, F. H . , Schneider, W. P., Axen, V., Spero, G. B., and Pike, J. E., J. Amer. Chem. Soc., 91, 5364 (1969).

45. Axen, V., Lincoln, F. H., and Thompson, J. L., Chem. Comm., 303 (1969).

46. Schneider, W. P., Chem. Comm., 304 (1969).

47. Axen, V., Thompson, J. L., and Pike, J. E., Chem. Comm. 602 (1970). 110

48. Corey, E. J., Weinshenker, N. M. , Schaaf, T. K., and Huber, W . , J. Amer. Chem. Soc., S[l, 5675 (1969).

49. Corey, E. J., Koelliker, V. , and Neuffer, J., J. Amer. Chem. Soc., 93, 1489 (1971).

50. Corey, E. J., Ravindranathan, T., and Terashima, S., J. Amer. Chem. Soc., 93, 4326 (1971).

51. Corey, E. J., Schaaf, T. K . , Huber, W. , Koelliker, U., and Weinshenker, N. M . , J. Amer. Chem. Soc., 92, 397 (1970).

52. Corey, E. J., Albonico, S. M. , Koelliker, V., Schaaf, T. K . , and Varma, R. K., J. Amer. Chem. Soc., 93, 1491 (1971).

53. Corey, E. J., Becker, K. B . , and Varma, R. K., J. Amer. Chem. Soc., 94, 8616 (1972).

54. Corey, E. J . , Noyori, R . , and Schaaf, T. K., J. Amer. Chem. Soc., 92, 2586 (1970).

55. Corey, E. J., Shirahama, H. , Yamamoto, H . , Terashima, S., Venkateswarlu, A., and Schaaf, T. K., J. Amer. Chem. Soc., 93, 1490 (1971).

56. Corey, E. J. , Arnold, Z., and Hur.ron, J., Tetrahedron Lett., 307 (1970).

57 . Corey, E. J., and Noyori, R. , Tetrahedron Lett., 311 (1970).

58. Miyano. M . , and Dorn, C. R., Tetrahedron Lett., 1615 (1969).

59. Miyano, M., Mueller, R. A., and Dorn, C. R . , Intra- Science Chem. Rept., 6, 43 (1972),

60. Miyano, M . , and Stealey, M. A. , Chem. Comm., 180 (1973)

61 . Corey, E. J. , Andersen, N. H. , Carlson, R. M. , Paust, J . , Vedejs, E. , Vlattas, I., and Winter, R. E. K., J Amer. Chem. Soc. , 90, 3245 (1968).

62. Corey, E. J. , Vlattas, I., Andersen, N. H . , and Harding K., J. Amer. Chem. Soc. , 90, 3247 (1968).

63. Corey, E. J., Vlattas, I., and Harding, K., J. Amer. Chem. Soc. , 91, 535 (1969). Ill

64. Corey, E. J., Ann. N.Y. Acad. Sci. , 180, 24 (1971).

65. Taub, D . , Hoffsommer, R. D . , Kuo, C. H . , Slates, H. L. , Zelawski, Z. S., and Wendler, N. L., Chem. Comm., 1258 (1970).

6 6 . Woodward, R. B., Gosteli, J . , Ernest, I., Friary, R. J. , Nestler, G . , Raman, H . , Sitrin, R., Suter, Ch., and Whitesell, J. K., J. Amer. Chem. Soc., 95, 6853 (1973).

67. Slates, H. L. , Zelawski, Z. S., Taub, D., and Wendler, N. L . , Tetrahedron, 30 , 819 .(1974).

68. Sih, C. J., Heather, J. B., Peruzzotti, G. P., Price, P Sood, R., and Lee, L. H., J. Amer. Chem. Soc., 95, 1676 (1973).

69. Sih, C. J., Price, P., Sood, R., Salomon, R. G., Peruzzotti, G., and Cassey, M . , J. Amer. Chem. Soc., 94, 3643 (1972).

70. Sih, C. J., Salomon, R. G . , Price, P., Peruzzotti, G. , and Sood, R . , Chem. Comm., 4, 240 (1972).

71. Finch, J., and Fitt, J. J . , Tetrahedron Lett., 4639 (1969) .

72. Pike, J. E., Lincoln, F. H., and Schneider, W. P., J. Org. Chem., 34, 3552 (1969).

73. Corey, E. J., and Grieco, P. A., Tetrahedron Lett., 107 (1972).

74. Andersen, N. H . , J. Lipid Res., 10, 320 (1969).

75. Hardegger, E., Schenk, H. P., and Broger, E., Helv. Chim. Acta, 50, 2501 (1967).

76. Collins, P., Jung, C. J . , and Pappo, R . , Israel J. Chem 6, 839 (1968).

77. Yura, Y . , and Ide, J., Chem. Pharm. Bull. (Tokyo), 17, 408 (1969).

78. Katsube, J . , and Matsui, M . , Agr. Biol. Chem. (Tokyo), 33, 1078 (1969).

79. Morin, R. B.,Spry, D. 0., Hauser, K. L., and Mueller, R. A., Tetrahedron Lett., 6023 (1968). 112

80. Bundy, G., Lincoln, F., Nelson, N., Pike, J., and Schneider, W . , Ann. N.Y. Acad. Sci., 180, 76 (1971).

81. Corey, E. J., and Cyr, C. R . , Tetrahedron Lett., 1761 (1974).

82. Raistrick, H . , and Smith, G., Biochem. J., 29, 606 (1935).

83. Clutterbuck, P. W. , Raistrick, H . , and Reuter, F., Biochem. J. , J31 , 987 (1937).

84. Barton, D. H. R . , and Miller, E., J. Chem. Soc., 1028 (1955).

83. Grove, J. F., J. Chem. Soc. , 4693 (1954).

8 6 . Birch, A. J . , Cassera, A . , .and Jones, A. R . , Chem. Comm.,167 (1965) .

87. McGahren, W. J . , van den Hende, J. H., and Mitscher, L. A., J. Amer. Chem. Soc., 5?^, 157 (1969).

88 . Auerbach, J., and Weinreb, S. M . , Chem. Comm., 298 (1974).

89. Brown, H. C., Rathke, M. W. , and Rogic, M. M . , J. Amer. Chem. Soc., 90, 5038 (1968).

90. Harrison, I. T . , Lewis, B . , Nelson, P., Rooks, W . , Roszkowski, A., Tomolonis, A. and Fried, J. H . , J. Med. Chem., 13, 203 (1970).

91. Benton, F. L., and Kiess, A. A., J. Org. Chem,, 25, 470 (1960).

92. Corey, E. J., Achiwa, K., and Katzenellenbogen, J. A., J. Amer. Chem. Soc., 9^, 4318 (1969).

93. Kaufman, G. B , , and Teter, L. A., Inorg. Syn., 7, 9 (1963).

94. Grieco, P. A., and Reap, J. J . , J. Org. Chem., 38, 3413 (1973).

95. Heather, J. B . , Sood, R. , Price, P., Peruzzotti, G. P., Lee, S. S., Lee, L. F. H . , and Sih, C. J., Tetra­ hedron Lett., 2313 (1973). 113

96. Eaton, P. E. , Cooper, G. F. , Johnson, R. C . , and Mueller, R. H . , J. Org. Chem., 37, 1947 (1972).

97. Gilman, H . , Zoellner, E. A., and Selby, W. M . , J. Amer. Chem. Soc. , 54 , 1957 (1932).

98. Whitesides, G. M . , Fischer, W. F., Filippo, J. S., Bashe, R. W . , and House, H. 0., J. Amer. Chem. Soc., 91, 4871 (1969).

99. House, H. 0., and Fischer, W. F., J. Org. Chem., 33, 949 (1968).

100. Corey, E. J., Gilman, N. W . , and Ganem, B. E., J. Amer. Chem. Soc. , 90, 5616 (1968).

101. Mitscher, L. A., Shirahata, K . , and Clark, G. W . , "unpublished data."

102. Mitscher, L. A., Clark, G. W . , and Showalter, H. D. H . , "unpublished data."

103. Mitscher, L. A., and Hammesfahr, P. D., "unpublished data."

104. Biichi, G. , Foulkes, D. M. , Kurono, M. , Mitchell, G. F., and Schneider, R. S., J. Amer. Chem. Soc., 89, 6745 (1967).

105. Christy, M. E. , Boland, C. C. , Williams, J. G, , and Engelhardt, E. L., J. Med. Chem., 13, 191 (1970).

106. Wideman, L. G . , J. O r g . Chem., 33, 4541 (1968).

107. Ireland, R. E., and Newbould, J., J. Org. Chem., 28, 23 (1963).

108. Mitscher, L. A., and Clark, G. W. , "unpublished data."

109. Corey, E. J., "Proc. Robert A. Welch Foundation Conf. Chem. Res. XII. Organic Synthesis," p. 55 (1969). .

110. Beel, J. A., Koch, W. G . , Tomasi, G. E., Hermansen, D. E . , and Fleetwood, P., J. Org. Chem., 24, 2036 (1959).

111. Coates, G. E. , "Organo-Metallic Compounds," Second Edition, pp. 6-8, John Wiley and Sons, New York (1960),