INFORMATION to USERS the Most Advanced Technology Has Been

INFORMATION TO USERS

The most advanced technology has been used to photo graph and reproduce this manuscript from the microfilm master. UMI films the original text directly from the copy submitted. Thus, some dissertation copies are in typewriter face, while others may be from a computer printer.

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

Oversize materials (e.g., maps, drawings, charts) are re produced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each oversize page is available as one exposure on a standard 35 mm slide or as a 17" x 23" black and white photographic print for an additional charge.

Photographs included in the original manuscript have been reproduced xerographically in this copy. 35 mm slides or 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

Accessing the UMIWorld’s Information since 1938

300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA

Order Number 8820267

Studies toward the total synthesis of trixikingolide

Cheney, Daniel Ley, Ph.D.

The Ohio State University, 1988

UMI 300 N. Zeeb Rd. Ann Arbor, MI 48106

PLEASE NOTE:

In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark V .

1. Glossy photographs or pages_____

2. Colored illustrations, paper or print______

3. Photographs with dark background____

4. Illustrations are poor copy______

5. Pages with black marks, not original copy i S

6. Print shows through as there is text on both sides of page______

7. Indistinct, broken or small print on several pages f S

8. Print exceeds margin requirements_____

9. Tightly bound copy with print lost in spine______

10. Computer printout pages with indistinct print______

11. Page(s)______lacking when material received, and not available from school or author.

12. Page(s)______seem to be missing in numbering only as text follows.

13. Two pages numbered . Text follows.

14. Curling and wrinkled pages

15. Dissertation contains pages with print at a slant, filmed as received ______

16. Other______

UMI

STUDIES TOWARD THE TOTAL SYNTHESIS OF TRIXIKINGOLIDE

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in the Graduate School of the Ohio State University

Daniel L. Cheney, B.S,

*****

The Ohio State University 1988

Dissertation Committee Approved by Dr. Gideon Fraenkel Dr. Leo A. Paquette Dr. John S. Swenton Adviser ” Department of Chemistry To Paivi and ray parents. ACKNOWLEDGEMENTS

I would like to thank Dr. Paquette without whose encouragement, guidance, and inspiration, this doctoral work would not have been possible. I would also like to express my indebtedness to those who played important roles in my development as a scien tist. Among them, Doc, for imparting to me the thrill of facing a synthetic challenge and the endurance to see it through; Ho Shen, for teaching me the importance of knowing the literature, and for serving as a role model for experi mental observation and technique; and of course, Jurgen for teaching me the importance of critical thought. I would also like to acknowledge Denis St. Laurent, Wang, Kevin Moriarty, Jurgen Dressel, Jeff Romine, Dwight MacDonald, Ho Shen Lin and Callie and Tana Pistorius, whose friendship has made these years very precious. I extend my appreciation to Kay Kampsen for her perseverance in completing the task of typing this thesis. I thank my wife, Paivi, for her encouragement, help in the final compilation of this work, and for serving as a source of balance during my graduate years. Finally, I thank my parents for their continual sup port, and for instilling in me (for better or worse) the spirit of adventure. iii VITA

June 30, 1957 ...... Born - Detroit, Michigan January-February, 1979 Scholarship for Language Study, Goethe Institut, Grafing, West Germany May-August, 1982,1983 Summer Student Reserach, Warner Lamberg/Parke- David, Ann Arbor, Michigan 1983 ...... B.S., Summa Cum Laude, Eastern Michigan Univer sity, Ypsilanti, Michigan

PUBLICATIONS "A New Dihydrobenz[a]anthraquinone Antitumor Antibiotic (PD) 116740)" Wilton, J.H.; Cheney, D.L.; Hokanson, G.C.; French, J.C. J. Orq. Chem. 1985, 50, 3936.

FIELDS OF STUDY Major Interest: Organic Chemistry TABLE OF CONTENTS Page ACKNOWLEDGEMENTS ...... iii VITA ...... iv LIST OF SCHEMES ...... viii LIST OF TABLES ...... X LIST OF FIGURES ..... xi CHAPTER I. THE TRIXIKINGOLIDES ...... 1 1.1 Isolation and Structure ...... 1 1.2 Biosynthesis ...... 1 1.3 Retrosynthetic Analysis ...... 4 II. SYNTHESIS OF TRICYCLIC KETONE 17 ...... 6 2.1 Synthesis and Alkylation of Bicyclic Dione 18 ...... 6 2.2 Synthesis of Tricyclic Ketone 17 .... 11 2.3 Summary ...... 21 III. SYNTHESIS AND REACTIONS OF TRICYCLIC UNSATURATED ESTER 16 AND ITS METHYL ENONE ANALOGUE 68 ...... 22 3.1 Synthesis of Tricyclic Unsaturated Ester 16 ...... 22 3.2 Reactivity of Unsaturated Ester 16 and Model System 59 toward 1,4 Addition . 24 3.3 Tricyclic a,|3-Unsaturated Methyl Ketone 68. Synthesis and Reactions ...... 28 3.4 Summary ...... 30

v Table of Contents (continued)

Page IV SECOND ROUTE TO ALDEHYDO-ESTER 15 ...... 31 4.1 Strategy ...... 31 4.2 Nazarov Cyclization Route to 72a .... 32 4.3 Silicon-Mediated Nazarov Cyclization .. 35 4.4 Direct Route to Methyl Ketone 72a ... 42 4.5 Cleavage of the D-Ring ...... 46 4.6 Summary ...... 48 V APPROACHES TO C(4)-C(5) BOND FORMATION .... 50 5.1 The Prins Cyclization ...... 50 5.2 Aldol Condensation ...... 52 5.3 Intramolecular Alkylation ...... 56 5.4 Cleavage of the A Ring ...... 59 5.5 Summary ...... 77 VI ALTERNATIVE APPROACHES AND CONCLUSIONS .... 79 6.1 Ramberg-Backlund Route ...... 79 6.2 Effecting Conformational Change at C ( 2 )...... 81 6.3 Conclusion ...... 84 LIST OF SCHEMES

Scheme Page I Retrosynthetic analysis ...... 5 II Duthaler's synthesis of 18...... 7 III Improved synthesis of 18 ...... 8 IV Attempted phosphorylation of 32 ...... 11 V Double elimination of 37 ...... 14 VI Hydrolysis of 34 and 43 ...... 15 VII Prins cyclization of 42 ...... 18 VIII Synthesis of enone 51 ...... 20 IX Synthesis of tricyclic ketone 17...... 21 X Synthesis of unsaturated ester 16...... 23 XI Mechanisms of carbonylation of 54 and 55 .... 23 XII Functionalization of ester 16 ...... 24 XIII Application of Yamamoto's conditions to 59 .. 27 XIV Application of Oppolzer's conditions to 57 .. 27 XV Synthesis of 15 via ketone 68 ...... 28 XVI Synthesis of methyl ketone 68 ...... 29 XVII Alternate plan to synthesize 15 ...... 31 XVIII Synthesis of divinyl ketones 75a,b ...... 33 XIX Mechanism of the silicon-mediated Nazarov cyclization...... 36 XX Synthetic plan for ketone 72a ...... 36

vii List of Schemes (continued)

Scheme Page XXI Nazarov cyclization of 7 8 a ...... 38 XXII Dimerization of 79 ...... 39 XXIII Acid-catalyzed isomerization of 79afb ...... 40 XXIV Initial synthesis of 72a ...... 41 XXV Trost cyclopentaannulation of 17 ...... 43 XXVI Ozonolysis of enol ether 91...... 46 XXVII Alpha-hydroxylation of 7 2 a ...... 47 XXVIII Summary of synthesis of 15 ...... 49 XXIX Prins cyclization of 15 ...... 50 XXX Methylation-lactonization of 15 ...... 51 XXXI Hydroboration of 99 and 15 ...... 53 XXXII Synthesis of keto-aldehyde 97 ...... 54 XXXIII First attempted synthesis of 105 ...... 57 XXXIV Intramolecular alkylation of 105 ...... 58 XXXV Expansion of the A ring...... 60 XXXVI Synthesis of enol ether 116 ...... 61 XXXVII Hydroboration of 117 ...... 63 XXXVIII Radical cyclization strategy from 119 ...... 64 XXXIX Synthesis of dialdehyde 121 ...... 65 XL Efforts to synthesize 126...... 67 XLI Synthesis of 133 and 134 ...... 68 XLII Mechanism of silylation with TMS-EtOAc ..... 69

viii List of Schemes (continued)

Scheme Page XLIII Synthesis of halo esters 140 and 142...... 71 XLIV Intramolecular alkylation of 140 and 142.... 72 XLV Ramberg-Backlund route ...... 79 XLVI Initial synthesis of 155 ...... 82 XLVII Synthesis of tert-butyldimethy1silvl ether 151 ...... 83 XLVIII Preliminary attempts to alkylate 151 ..... 84 IL Proposed strategy for C(4)-C(5) closure .... 86

ix LIST OF TABLES

Table page 1 The trixikingolides ...... 2 2 Alkylation of dione 18 ...... 9 3 Reduction of 32 ...... 12 4 Prins cyclization of 42 ...... 19 5 Nazarov cyclization of 75 ...... 34 6 Unsuccessful selective ketalization of 101 .... 55 7 Aldol condensation of 98 ...... 56 8 Conditions for attempted alkylation of 105 .... 59 9 Kinetic silyl enol ether formation from 113 ... 62 10 Conditions used for deprotection of 117 ...... 62 11 Kinetic silyl enol ether formation of 133 and 134 ...... 69 12 Conditions for attempted intramolecular alkylation of 140 and 142 ...... 73 13 Thermodynamic silyl enol ether formation from from 101 ...... 80

x LIST OF FIGURES

Figure Page 1. Biosynthesis of a-cedrene ...... 3 2. Proposed biosynthesis of the isocedrene ring skeleton ...... 4 3. C versus 0 alkylation of 18 ...... 10 4. Possible mechanism for chlorination of 37 ...... 14 5. 13C NMR of aldehyde 42 ...... 16 6. Mechanism for Prins cyclization ...... 17 7. Conjugate additions onto unsaturated esters ..... 26 8. Configurations of linear triguinanes...... 32 9. X-ray structure of 105 ...... 75 10. Conformations of 105 ...... 76 11. Conformations of 142 ...... 76 12. Intermediates for C-4,C-5 closure ...... 78 CHAPTER I

THE TRIXIKINGOLIDES

1.1 Isolation and Structure The trixikingolides were first isolated by Bohlraann in 1979 from two species of plant (T. wrightii and T. inula) from the South American genus Trixis (Tribe: Mutisieae) . Twelve in number, they share in common a novel pentacyclic ring system constituted of an isocedrane carbon framework bracketed both by an ene lactol bridge (from C-5 to C-7) and either a lactone, lactol or ether linkage (from C-ll to C-4) (Table 1). Structural elucidation was accomplished on the basis of spectroscopic analysis and chemical degradation. To date, these molecules have not been studied for biological activity.

1.2 Biosynthesis. Although a biosynthetic pathway has not yet been elucidated for structures 1-12, their origin is probably similar to that of a-cedrene and related sesquiterpenes.

1 Table 1 : The trixigingolides.

R" -R’

•CH '■■•• CH3 .....CH.

''OR R’O MebuO R’O

R R’ R R’ R" R R’ 1 H Mebu 7 Sen OH H 11 H Mebu 2 H iVal 8 Sen H OH 12 OSen Mebu 3 OSen Mebu 9 iVal OH H 4 OiVal Mebu 10 iVal H OH 5 OMebu iVal 6 OAc Mebu

Mebu =

OAc = 3 The latter arise from cationic rearrangement of farnesol pyrophosphate (Figure 1)3.

PPO

CH3 CH. CHa

G H cedrene nucleus isocedrene nucleus

Figure 1 : Biosynthesis of « - cedrene.

The pathway for trixikingolides, however, must be distinct in allowing for migration of the C-3 methyl group to 0 5 . This feature is unique among trixikingolides and a few other related sesquiterpenes isolated at the same time from the tribe Mutisieae3. Bohlmann has proposed the following rearrangement to account for this phenomenon (Figure 2). 2x1,2 shift

- H+

CH;

isocedrene nucleus

Rgure 2 : Proposed biosynthesis of the isocedrene ring skeleton.

1.3 Retrosynthetic Analysis. Unsubstituted lactol 13 was chosen as the synthetic target since it was regarded as the parent system of this interesting class of compounds. Our retrosynthetic analy sis is shown in Scheme I. Alkylation of the known bicyclic diketone 18 with a protected propanal unit, followed by cyclization was expected to deliver the tricyclic ketone 17. Homologation to unsaturated ester 16, followed by conjugate addition of an acetaldehyde equivalent from the convex face of the 5

Scheme I

OCH. OCH.

-•CH. \ CH. >

HO* H 13 14 15

OCH,

H H 17 16 molecule with subsequent methyl trapping would provide 15 having cis-oriented aldehyde and ester groups. The second Prins cyclization to form the C4-C5 bond was regarded as a key step in the synthesis, since Dreiding models suggested the presence of some strain in the resulting product. Lactonization, followed by oxidative cleavage of the C15- C14 double bond was anticipated to deliver the dialdehyde. Acid-catalyzed isomerization and epimerization at the C-7 position would then provide the target molecule. CHAPTER II

SYNTHESIS OF TRICYCLIC KETONE 17

2.1 Synthesis and Alkylation of Bicyclic Dione 18. Bicyclic dione 18 was chosen as starting material since it was shown by Duthaler and Maienfisch to undergo alkylation with methyl iodide and propargyl bromide at C-l in high yield (eg 1)^.

NaH, THF; (1 ) RX.rt (83 - 87%)

R = methyl, 19 R = propargyl, 20

Stetter and coworkers first synthesized 18 in 1961 from 2-cyclopentenone in eight steps and 17% overall yield5. Since then, more efficient syntheses of 18 have appeared in the literature, the most recent of which is shown in Scheme 11^'®.

6 Scheme II

1-) OoN OEt

t - BuOK,THF OEt 2.) MeOH, 5 days 21 23 O H 2 , Pd/C EtOH, HOAc

NaOCH, Ether, 0° C OEt

24 O

(40 - 45% overall yield)

This synthesis still suffers from an overall modest yield, takes about two weeks to perform, and requires as its homoenolate source (3-nitro propionate 22, the synthesis of which is low yielding (30%) and does not lend itself to bulk preparation^. Therefore, a more expedient synthesis was developed, which utilized the zinc homoenolate 25 reported by Nakamura (Scheme III)8 . This annulation procedure is the most efficient to date. The conversion is completed in -1-2 days and proceeds in substantially higher yield than the methods mentioned above. 8

Scheme III

OTMS OEt ) 25 2

T M S C I, ether CuBr D M S , HMPA OEt 21 (7 0 % ) 26

NaOCH3 lether, (56% ) or, 1.) H+ 2.) NaOCH3 , ether, (88 % )

When 18 was subjected to the original alkylation conditions used by Duthaler and Maienfisch with 27 as the electrophile, alkylation product was not observed (eg 2).

no reaction

Therefore, an examination of the effects of solvent, tem perature, enolate counter ion, and leaving group was under taken (Table 2). 9 TABLE 2: Alkylation of 18. Entrv Base Solvent Temp(°C) Time(h ) X Yield(%) 1 NaH THF R.T. 40 Br 0 2 NaH THF/DME R.T. 20 Br 0 3 NaH THF/DMF R.T. 20 Br 0 4 NaH THF/DMF 70 24 Br 8 5 NaH DMF R.T. 24 Br 0 6 NaH DMF 80 20 Br 41 7 k 2c o 3 DMF R.T. 20 Br 0 8 k 2c o 3 DMF 80 20 Br 35 9 NaH DMF 50 6 Br-»Ia 55-60 10 NaH DME/HMPA 50 6-8 I 75 a Sodium iodide was added to the reaction mixture.

It is evident from Table 2 that solvent polarity and temperature are the most crucial reaction parameters. In the absence of a polar medium and elevated temperatures, the sodium enolate was largely insoluble. Also, electrophile 27 was shown to be relatively unreactive. When 4- bromobutene was reacted with 18 under similar conditions, a much higher yield of product was obtained in only one-sixth of the reaction time (eq 3; compare with Table 2, entry 9). This pointed to the acetal moiety in 27 (presumably its steric bulkiness) as the source of deactivation. Conse quently, the more reactive iodo analogue 28 was used in its place (entry 10). 10

O-Alkylation (a common side reaction of 1,3 diones) was never observed, possibly reflecting the ring strain which would be present in such a product. Of relevance is Stetter's observation that dione 18 exists in its enol form to the extent of only 1.4% (Figure 3)^. o ORa

H H H 30 31 32

O - alkylation not observed

H H 33 18 (1.4%) (98.6 %) Figure 3: C vs. O - alkylation of 18 . 2.2 Synthesis of Tricyclic Ketone 17. Experiments were now initiated to transform 32 into the diene acetal 34 (eg 4).

/r\ (4)

H H 32 34

This was first attempted in a direct manner using the Ireland procedure®, but unfortunately 32 resisted bis enol phosphate formation, even when resubjected to reaction conditions (Scheme IV). Scheme IV

(RO), P

1.) LD A .TM E D A

2.) (RO)2POCI

same conditions 12 A stepwise procedure was next applied. Reduction of 32 with DIBAL led to a mixture of all three possible diols.

DIBAL, -78°C

Table 3 : Reduction of 32 with Dibal.

HO Ra OH HO Ra OH HO Ra OH 05 db mH H H solvent 37a 37b 37c

CH^C^/hexane 4.5 3.5 2.0 (1:1)

CHgCV hexane/ 7.0 1.0 2.0 THF(9:9:1)

Assignments are based upon symmetry characteristics as determined by 13C NMR and the respective order of elution from silica gel; endo alcohols are assumed to be less polar than their exo epimers. 13 The differences in the stereochemical direction of reduction as a function of solvent are of interest. In dichloromethane, DIBAL first coordinates to a carbonyl oxygen. Once the complex is formed, there appears to be little steric discrimination toward hydride delivery from one face of the molecule or the other. When tetrahydrofuran is added to the medium, the solvent complexes with the reagent, and hydride delivery is now bimolecular in nature. Evidence of this is found in the preponderance of the one stereoisomer (37a) resulting from hydride attack above the convex face of the molecule. An attempt to dehydrate directly the diols with phos phorus oxychloride led to the exclusive formation of an epimeric mixture of mono chloro olefins 38a and 38b (eg 6).

O O POCU 37a-c (6) (35%) H 38a ,b

Dehydrochlorination of 38a,b with strong base did not effi ciently deliver 34; nor were other direct dehydration methods (activated alumina, thionyl chloride, Burgess' salt) effective on 37a-c. Consequently, 37a-c were sub jected to classical mesylation conditions with the inten tion to achieve double elimination. Quite unexpectedly, 14 extensive chlorination took place (Scheme V). One possible explanation for this side reaction is neighboring group

Scheme V

HO Ra OH Et3N, MsCI + CH2CI2 , 0°C H 40 3 7 a -c Ms = S02CH3 DBN, 130°C

H H 34 38a,b (2.2 : 1 ) (45-50% overall yield)

MS2O , pyridine DBN, 130°C 37a-c 39 34 o ° c , CH2CI2 (70%) (85%) assistance from one of the acetal oxygens to form a six- membered intramolecular complex (Figure 4). MsO •7 •o ci- III 40 -OMs 41 Figure 4 : Possible mechanism for chlorination of 37. 15 This explanation accounts concisely for nucleophilic displacement by chloride anion of a neopentyl, secondary mesylate at only 0 °C. A similar mechanism could also account for the chlorination observed earlier in the reaction of the diol mixture 37 with phosphorus oxychloride (eq 6). The chlorinative side reaction was circumvented by using mesyl anhydride in pyridine instead (Scheme V)10. Assessment of DBU, DBN, and potassium tert-butoxide showed DBN to give the highest yield. Diene acetal 34 could not be hydrolyzed directly to 42 due to its stability 11 . Consequently, transacetalization to the dimethoxy acetal 43 was performed prior to complete hydrolysis (Scheme VI).

Schem e VI

H0Ac /H 20 + 34

H H 34 42 M eO H , reflux PTSA,2 h (6 7 % ) (1 5 % ) OCH3 , i-OCHi

H0Ac /H 20 42 34

H (7 8 % ) (6 .4 % ) 43 16 The Cs symmetry of 42 is revealed in its 8 line 13C spectrum shown in Figure 5 (300 Hz, C6D6):

Figure5 : 13C NMR spectrum of aldehyde 42.

An efficient route had now been worked out for the synthesis of 42, an important precursor for the next key step in the synthesis, the Prins cyclization. The Prins reaction is described in Figure 612 17

concerted: •MX,

A B C D

stepwise:

■MX, \ \O

E F G H

Figure 6: Mechanism for Prins cyclization.

The reaction may be of a concerted or stepwise nature. Of importance here is the intramolecular proton abstraction by oxygen through a six atom transition state to form a transposed double bond. When 42 was treated with various Lewis acids, the normal course of reaction was not followed (Scheme VII). Instead, a mixture of chloro alcohols 47a,b was produced. Oxidation to produce a single chloro ketone (Scheme VIII) showed these to be epimeric at c -1 1 (stereochemical assign ment was based on the assumption that the less polar 47a has an endo orientation). When the reaction was done in benzene in place of CH2C12, C-9 was phenylated! This 18 unexpected behavior can perhaps best be rationalized in terms of Scheme VII. The framework of carbocation 44 is too rigid and restrictive to permit the oxygen from abstracting a proton from C-8. The result is a long lived

Scheme VII

10 9 SnCI4. CH2CI2, 0° deprotonation — H OSnCr H 42 44

benzene Cl'

••a

H 47a 47b 48 carbocation which readily traps ambient nucleophiles, e.g., chloride ion when the reaction is done in methylene chloride, and phenyl when the reaction is done in benzene. Table 4 summarizes the various conditions employed toward optimization of this step. Table 4a . Prins Cyclization of 42. Lewis acid (eg) Temp ( °C)______43a 47b____ 48

1 SnCl„ (1.5) 0 34% - -

2 SnClu (1.2) -20 50% - -

3 SnCl* (1.05) -45->-30 42% - -

4 SnClu (1.2) -78-»-10 57% - -

5 TiCl„ (1.2) -78 38% - -

6 SnCl,, (1. 2 )b -30-»-25 58% - -

7 SnCl4 (1. 2)c -22 43% - -

8 Me2AlCl (3) -78->15 57% - - 9 SnCl„ (1.2)d -78-*10 25% 23% 4.4

a All reactions done in methylene chloride. Yields deter mined by mass of isolated products. b Inverse addition of substrate to acid. c High dilution conditions. d Lewis acid was freshly distilled from phosphorus pentoxide.

The original synthetic plan toward 17 would have required regioselective hydrogenation of the C8-C9 double bond. However, the acquisition of 47 lent itself to simple reduction of its carbon-chlorine bond. Before the synthesis was carried forward, five-membered ring formation (as opposed to six) during the Prins reaction was verified on the basis of the following chemi cal transformations (Scheme VIII). 20

Scheme VIII

Oxidation and dehydrochlorination of 47a,b provided a molecule which had all the spectroscopic features of an unsaturated ketone. Its structure can only be assigned as 51, since 50 would be impossibly strained. Furthermore, the facile elimination of chloride in 49 indicates a likely a orientation of the chlorine in the major Prins cyclization product 47a,b. Birch reduction of 51 to obtain 17 directly failed (Scheme IX). However, sodium-ammonia reduction of the carbon-chlorine bond in 47a,b, followed by oxidation, provided in high yield the tricyclic target. 21

Scheme IX

Birch reduction .. 51 ----——TT H 17

PCC (94%) 3 - A sieves .OH Na/NH, 47a,b THF (96%)

H 52

2.3 Summary Tricyclic ketone 17 was obtained from 2-cyclopentenone in eleven steps and 5% overall yield. The sequence makes use of a novel, efficient synthesis of 18. In addition, the Prins reaction of 42 followed an unexpected course to provide chloro alcohols 47a,b, which in turn lent them selves to smooth transformation into 17. CHAPTER III

SYNTHESIS AND REACTIONS WITH TRICYCLIC UNSATURATED ESTER 16 AND ITS METHYL ENONE ANALOGUE

3.1 Synthesis of Tricyclic Unsaturated Ester 16. Tricyclic unsaturated ester 16 was initially syn thesized using vinyl iodide 54 as intermediate (Scheme X)13. Reaction with nickel tetracarbonyl in hot methanol provided the desired product in good yield 14 . Evenso, this procedure suffered from two disadvantages: instability of 54 and the potent toxicity of nickel tetracarbonyl. The alternative procedure using enol triflate 55a was found to work equally well, while avoiding the prolems associated with the above method (Scheme X)^3. (Small amounts of the regioisomer of 55b were detected later in thesynthesis; for discussion see section 4.2). Both methods follow similar mechanistic courses (Scheme XI): each involves vinylation of zero-valent metal 14 15 to form the metal (II) species 56 ' . Carbon monoxide insertion into the sp2 carbon-metal bondfollowed by methanolysis gives the final product.

22 23 Scheme X

1.) H2NNH2 , EtOH Et3N , 70°C 17 l2 , ether (94%) tetramethyl guanidine, (84%)

(89%) LDA; Tf2NPh DME (54%) Ni (CO)4 M eO H , 50°C

OTf OCH Pd(0), C O , MeOH

Et3N , LiCI, DMF (60%) H 16

X-lor OTf

S - solvent or ligand

M - Ni or Pd MeOH

L -ig a n d 24 3.2 Reactivity of Unsaturated Ester 16 and the Model System 59 Toward 1.4 Addition. Cyclopentenyl ester 59 was chosen as a model template for 1,4 addtion. o

o c h 3

The initial strategy called for cuprate addition of an acetaldehyde equivalent from the 3 face of the molecule, followed by trapping of the resulting enolate anion with methyl iodide from the a face. The ethyl vinyl ether cuprate 60 developed by Wollenberg seemed ideally suited to 16 this end (Scheme XII)

Scheme XII o

^ ^ CULi H 60 16 - 2.) Mel

3.) H30 *

However, attempts to react 60 with ester 59 failed; indeed, 1,4 addition was not observed even with cyclohex- enone, one of the substrates reported by Wollenberg to have undergone smooth conjugate addition. An earlier member of the Paquette group similarly found this cuprate to be inert toward different unsaturated carbonyl systems under various conditions 17 25 Attention was therefore next directed to lithium divinylcuprate, since a selective hydroboration-oxidation sequence would provide the desired aldehyde appendage. Following an examination of the literature, some important aspects of copper mediated conjugate additions onto unsaturated esters came to light. First, such systems are generally quite unreactive (exceptions are unsaturated lactones 18 ). House, who postulated that cuprate additions proceed via an electron transfer mechanism, found a strong correlation between the reduction potential of the acceptor system and its reactivity toward dialkyl cuprates19 Empirically, any system with a reduction potential of less than -2.4 (volts vs. SCE) was viewed to be unreactive. House developed additivity rules which allow one easily to estimate the reduction potential of a conjugated carbonyl system bearing any substitution pattern and therfore to deduce its level of reactivity. Thus, the reduction potential for esters 59 and 16 is 2.5 V 19 , suggesting that these molecules should be fundamentally inert toward conventional cuprate addition. Yamamoto later found that such systems were, however, receptive if Lewis acids were involved as promoters (Figure 7) 20 26

estimated reduction potential (volts)

/ ~ k ° r ^culi 2 .5 \ y— \ no reaction > ' OCH, 61 Bu BuCuBFg 61 (30%) OCH; 62

CH. h3c ch3 BuCuBF, 2.5 o — (96%) 011 / = ° h3co 63 64

Figure 7: Conjugate additions onto unsaturated esters

Unfortunately, when such conditions were applied to unsaturated ester 59, a complex mixture resulted (Scheme XIII). 27

Scheme XIII

CuBFa 59 59 3eq 65 (46% ) (13% ) ( 1 1 %)

59 59 65 2eq 66 (13% ) (23 %) (1 1 %)

Bis-addition products such as 66 are commonly observed with organocopper-Lewis acid reagents 20 '21 Oppolzer had shown that ester 67 underwent poor 1,4 addition using an organocopper complex 22 . Upon addition of tri-n-butylphosphine to the reaction mixture, however, product formation occurred in 96% yield. Similar treatment of ester 59 yielded only starting material (Scheme XIV). Scheme XIV

in- R" H3CCuBF3 1 ,4 adduct (96%) n Bu3P

CuBF3 59 no reaction n Bu3P

CuBF3 59 no reaction HMPA 28 Addition of HMPA to the above mixture similarly did not enhance reaction efficiency 23 . At this point in time, Corey's 24 , Johnson's 25 , and Nakamura's 26 work with activated organocopper and organocuprate systems had not yet appeared in the literature. These systems, where chlorotrimethylsilane presumably polarizes the substrate and solvent additives, such as DMAP or TMEDA in some way activate the organometallic complex, characteristically add to unsaturated esters in good yields.

3.3 Tricyclic g.B-Unsaturated Methyl Ketone 68. Synthesis and Reactions. To bypass the above difficulties, the decision was made to adjust the reduction potential of 16 in a more favorable direction by replacing the methoxy group with methyl. Vinyl cuprate addition, methyl iodide trapping, and subsequent haloform degradation would then provide an intermediate very near to the desired aldehydo-ester 15.

Scheme XV o

CH. OCHg r-ch-

17 68 15 29 This required an efficient synthetic conversion of tricyclic ketone 17 to methyl enone 68. Examination of the literature provided no satisfactory means of doing so 27 Therefore, a novel procedure was developed, which utilized palladium chemistry developed earlier by Stille and ? 8 others . Enol triflate 55a (Scheme XVI) was reacted with the vinyl tin species 69 in the presence of Pd(0) to form the diene enol ether 70, hydrolysis of which yielded the methyl ketone 68 in good yield.

Scheme XVI

OTf OEt Pd(PPh3)4 (5%) THF , LiCI, 70°C

EtO ,Sn(CH3)3 H T 69

HOAc, H ,0

CH:

H (56 %, 2 steps)

As expected, 68 was quite reactive toward cuprate addition. For example, reaction with the higher order divinyl cuprate 30 29 was complete in about 5 minutes at -78 °C . Facial selec tivity, however, was unexpectedly poor (eg 7).

O O

CuCNLi: CH3 68 (7) - 78°C, ether H H ( 7 5 % ) 71a 71b

(6.5 : 3.5)

Of the cuprates used, the higher-order species gave the highest yield. The stereochemical assignments of 71a,b were made on the assumption that the major product results from attack on the convex face of the molecule. Wallen berg' s cuprate 60 and the cuprate derived from vinylmag- nesium bromide and a copper(I) salt led to "no reaction" and competing 1,2-addition, respectively.

3.4 Summary An efficient synthetic route to 15 was found. The system, however, was discovered to be largely unreactive toward various organocopper and organocuprate reagents. A synthesis was devised and carried out for the analogous methyl ketone 68. 1,4-Addition was facile, but preliminary experiments showed unexpectedly poor facial selectivity. CHAPTER IV

SECOND ROUTE TO ALDEHYDO-ESTER 15

4.1 Strategy A second approach to functionalization of the C ring was envisioned, which used as its key intermediate the tetracyclic ketone 72a. Oxidative bond cleavage as shown (Scheme XVII) would liberate the cis-locked aldehyde tether and ester functionality.

Scheme XVII o

OCH; ‘'O H ;

H H H 15 72a 17

This strategy was particularly well suited to the molecule for two reasons: cis ring fusion in cyclopentanoid annulation greatly predominates over trans (for reasons of ring strain) and linear triquinane frameworks, as

31 32 constituted by the B, C and D rings of 72a, are most easily elaborated in the cis-anti geometry, not the cis-cis or, all-concave form (Figure 8)30'31.

73 74

cis - anti cis - cis

Figure 8: Configuration of linear triquinanes

4.2 Nazarov Cyclization Route to 72a. Cyclopentenoid annulation via Nazarov cyclization of divinyl ketone 75a was thought to be the most expeditious

O O acid (8 )

H H 75a 76

route to 72a since Birch reduction of the resultant enone 76 and subsequent trapping with methyl iodide was expected 30 to deliver the desired cis-anti product 33 The preparation of 75a was accomplished as follows (Scheme

XVIII):

Scheme XVIII

OTf OTf

17 1.) LDA.DME 2.) Tf2NPh

H (9 H 5 5 a 55b

P d(P P h3) , LiCI, THF co ( Sn(CH3)3 O

H H

7 5 a 7 5 b

( 5 2 % ) ( 12%)

The presence of 55b as a minor product was deduced from the generation of 75b. Although this was the first time 55b was observed, it was presumably generated in earlier reations as well. Formation of 55b as a minor product was quite unexpected since it is kinetically dis- favored and presumably more strained relative to 55a32 34 The inseparable mixture was carried forward into the coupling step, whereupon 75a was isolated in modest yield. Cyclization of 75a under a variety of conditions did not succeed in providing 76 in sufficient yield for the route to be viable. Summarized below are the conditions employed, together with the corresponding yields and pro ducts (Table 5).

TABLE 5. Nazarov cyclization of 75a. entry reagent temp (0 C) result yield H aPOu(85%) rt N.R.

i H HaPO^/HCOaH rt 75% (1:1)

h 77

SnCl* 0-»rt complex mixture 4 FeCl3 0-»rt 76 8% 5 CH3S03H/P205 50 decomposition

6 CH3S03H/P,05 0 (5 min) 76 10 - 20'

The Eaton reagent (Table 5, entries 5,6) gave the best results, but further variation of reaction time and 33 temperature did not raise the yield beyond 20% . The formylation product 77 obtained with the phosphoric-formic acid mixture (entry 2) indicates that Michael addition is 35 faster than the conrotatory closure. This result is unusual (since formic acid is a poor nucleophile) and implies that the sluggishness of the cyclization step might have allowed side reactions to occur under other conditions as well, thus possibly accounting for the overall poor results observed.

4.3 Silicon-Mediated Nazarov Cyclization. Silicon-mediated cyclization was next considered since conditions are often milder and the yields higher than for 0/1 O £ the unadorned Nazarov cyclization . Like the Nazarov cyclization, the silicon-mediated reaction is thought to proceed through a (2 + 2) conrotatory closure. Silicon, however, plays the additional role of stabilizing the carbocation beta to itself through hyperconjugation; nucleophilic attack on the activated silicon atom then produces the double bond regioselectively in the position 34 35 shown (Scheme XIX) ' 36

Scheme XIX

Si(CH3)3 Si(CH3)3

H Si(CH3)3

Literature precedent led to the expectation that 79a would be formed stereoselectively 31 . Kinetic methylation and reduction would then deliver the desired methyl ketone 72a (Scheme XX).

Scheme XX

TMS

Lewis acid 2 steps 72a 37 The cyclization precursor 78a was synthesized as shown in eg (9)31.

Pd(PPh3)4, LiCI 55a,b + Si(CH3)3 (9) CO.THF Sn(CH3)3 J l 80 H H (H3C)3Si 78a 78b

(74 - 80 %) (8 - 10%)

Reaction of 78a with boron trifluoride etherate in toluene succeeded in producing stereoselectively the desired enone 79a, but in only 25% yield (Scheme XXI). This result was unexpected since Stille had earlier shown this catalyst to be the reagent of choice in his reitera tive Nazarov cyclization approach to p-capnellene33,a. Ferric chloride was found to give cyclization product in higher yield, but at the expense of stereoselectivity (Scheme XXI). 38

Scheme XXI

BF3OEt2 78a toluene, r t

H H 79a 79b

(25 %) (0 %) FeCU, 0°C 78a 79a 79b CH,CI2V/I2 (1.5 : 1)

(15 - 70%)

Stereochemical assignment of the inseparable isomeric mixture was based on the expectation that isomer 79a would be of lower energy due to lack of steric congestion present in the concave isomer 79b. This assignment was corro borated later with the X-ray analysis of an advanced inter mediate synthesized from 72a (Figure 9). In addition to the lack of stereoselectivity, the cyclization using ferric chloride as catalyst had one other problem associated with it; specifically, rapid dimerization of the product enone. Spectroscopic data suggested that a formal Michael addition had occurred, presumably through the intermediacy of the iron enolate 81 (Scheme XXII). 39

Scheme XXII

/ / — OFeCI, FeCI prolonged reaction time

It is striking that bond formation occurs readily between such bulky molecules at 0 °C. Dimerization of the cor responding lithium enolate (see Scheme XXIV, below), for example, does not occur. This side reaction was often difficult to control since the rate of reaction was vari able (1-8 h). This in turn may have been connected to the varying level of moisture in the reaction mixture, since Denmark has observed that under strictly anhydrous condi tions ferric chloride alone will not induce reaction^®. 40 The above inseparable isomeric mixture of enones was next equilibrated with p-toluenesulfonic acid in toluene in anticipation of forming enone 76, which was presumed to be the most stable of all three possible conjugated enone isomers. Instead, a different, but useful outcome was observed (Scheme XXIII). Scheme XXIII

o 0

H (1.5 : 1) H 79a 79b

(8 1 % ) PTSA, 150°C

O 0

H (1.5 : 1) H 79a 76

Enone 79a was converted to 72a by a kinetic methylation-

0 7 O O reduction sequence ' (Scheme XXIV). The yield for 41 angular methylation could not be optimized; no side-products were isolated.

Scheme XXIV

1.) LHMDS , THF, - 78°C 79a 2.) Mel (25% ) H 83a

L - Selectride (95% ) -78 °C , THF

1.) L i(3 e q ), NH3,THF 76 0.6 eq tert-BuOH 2.) Mel (10 e q ) H (67% ) 72a

Birch reduction and in situ methylation of enone 77 led to some interesting observations. Use of 1 eq of proton source (tert-butyl alcohol) resulted in little methylation 3 9 . This must have arisen from competitive deprotonation of solvent by the intermediate P carbanion, since when the proton source was omitted from the.reaction mixture, substantial methyl incorporation only at C(ll), 42 (not at C(2)) was observed (30-55%). Eventually, it was found that 0.6 eq of tert-butyl alcohol was optimal. As expected on the basis of the precedent, only the desired cis-anti isomer was obtained 30 . Minor amounts of polyalky- lation products were obtained, and it was sought to sup press their formation by using instead triphenylmethanol as proton source, since the conjugate lithium base formed in situ is too weak to regenerate the lithium enolate of 72a 40 . This, however, only had the effect of reducing the overall yield of product.

4.4 Direct Route to Methyl Ketone 72a. In the previous section, 72a was synthesized from tricyclic ketone 17 in 14% overall yield. As a key inter mediate in a linear synthesis, it was essential that this intermediate be prepared in quantity in a much more effi cient manner. The synthetic solution to this problem was found in a cyclopentannulation method developed by Trost in the early 1970's 41 ' 42 . The procedure, which has seen very little use in natural products synthesis since its introduction (one example by Trost 43 ), is shown m its application to tricy clic 17 in Scheme XXV. Epimeric silyl enol ethers 86arb were quite labile, and therefore did not lend themselves’ to separation. 43

Scheme XXV

P h ^ bf4

KOH, DMSO H

LiNEt2 (85% , 2 steps) TMSCI, DME

TMSO

OTMS

410° C, 30 sec.

(97%) H 85 86

(1 :1 .5 ) MeLi, DME -50° C ; Mel Pd(OAc)2 A

Hr,.. Li/NH3,THF CH

0.5 eq. tert-BuOH; Mel H H H 76 72a 72b

A: 52% B : 44% 29% 44 Initially, these were desilylated with methyllithium and treated with methyl iodide to give the corresponding mixture of methyl ketones, the minor of which was dis carded. More efficiently, it was found that 86a,b could be converged to enone 76 by treating the mixture with stoi- chiometric quantities of palladium acetate 44 . The yield of this reaction was made reproducible by addition of 2,6-di- tert-butylpyridine as a proton sponge. Use of catalytic amounts of palladium reagent with methallyl carbonate or benzoquinone led to extensive hydrolysis of the starting material45,46. Birch reduction as described above then gave 72a selectively in 5 steps and 45% overall yield. The possibility that vinyl cyclopropyl alkoxide (eq 10) would undergo an anionically accelerated 1,3 sigma- tropic shift was briefly examined. In addition to making the reaction (normally done at 330-430 °C) more convenient, the product enolate could be alkylated in a one-pot pro cess. H H 87 88

Using a simple Hiickel-based model, Carpenter has predicted such an acceleration to occur 47 . More recently Danheiser has shown accelerations to take place in rearrangements such as that shown below (eg 11) 48

S 0 2Ph n - Butyllithium THF, - HMPA -78 — - -30°C 89 90

In the event, vinyl cyclopropyl alkoxide 87 was generated in the usual manner at -78 °C; glyme and HMPA were then added to promote separation of the lithium alkoxide ion pair and the reaction mixture was warmed slowly to room temperature and eventually to 80 °C. Progress of the reaction was monitored by TLC against ketone standard, but rearrangement was found not to occur, the cyclopropyl alkoxide degenerating over time to form a complex mixture of products. Despite the lack of success in this experiment, the reaction deserves further attention because of its potential usefulness. 46 4.5 Cleavage of the D ring. An efficient synthesis had now been developed to provide the large quantities of 7 2 a needed to carry the synthesis forward. Originally, it was hoped that oxidative cleavage of the D ring could be accomplished by selective ozonolysis of the enol ether double derivative 91 49 Selectivity for the silyl enol ether bond was never observed, even with very slow and controlled addition (either as a gas or cooled saturated solution) of ozone, in the presence of an indicator dye (sudan red) or reactivity modifier (pyridine) (Scheme XXVI).

Scheme XXVI

NaHMDS, THF _ Oo, _ slow addition , . . 72a ►jti \ \ no selectivity TMSCI / Et3N various conditions

(91%) H 91

An alternative, classical approach was next given atten tion50. This involved a-hydroxylation of 7 2 a followed by oxidative cleavage with lead tetraacetate or periodic acid to provide 15. Hydroxylation was initially atempted by NBS bromination of 91 followed by hydroxide displacement of the 50 a-bromine atom in dimethylformamide . This method worked poorly, and in the end direct oxidation of the enolate of 47 72a was found to deliver 92 in good yield. Of the two reagents used (Scheme XXVII), MoOPh proved to be the most efficient 51 ' 52 .

S ch em e XXVII OH

1.) KHMDS

2.) \ N H H Ph S 0 2Ph 72a 92 ( inseparable from aromatic impurities)

1.) 4 eq LHMDS, THF 72a ------(72-76%) 2.) 6 eq MoOPh THF , -40°C

a-Diketone formation using MoOPh was at first a serious side reaction (up to 30% of the product), but was suppressed by inverse addition of the enolate to a cold slurry of MoOPh. Vedejs also regularly observed 5-15% starting material in his hydroxylations 52 . The -20% starting material initially observed in the above reaction was reduced to 3-8% by using a large excess of reagents 53 Oxidative cleavage with lead tetraacetate in methanol/ben zene (1:3) provided in quantative yield the long sought 48 tricyclic aldehydo ester 15 with all stereocenters in place (eg 12).

OCH

Pb(OAc)4 , 0°C 92------MeOH , benzene (1 : 3) H ( quantitative yield)

4.5 Summary An efficient and stereoselective route was developed from 17 to 72a, utilizing the Trost cyclopentannulation sequence. Hydroxylation of 72a followed by oxidative cleavage provided 15. A summary is presented in Scheme XXVIII. 49

Scheme XXVIII

2 steps

H 78

3 steps

(1.5:1 H 79a 79b

2 steps

H 72a

2 steps

15 CHAPTER V

APPROACHES TO C(4)-C(5) BOND FORMATION

5.1 The Prins Cyclization. The Prins cyclization was expected to go through a chair transition state, with the carbonyl oxygen facing the exterior of the molecule for steric reasons (Scheme XXIX).

Scheme XXIX

H 93

Dreiding models suggested that the resultant six-membered ring could be somewhat strained, but that nonetheless C-4 and C-5 could come within adequate proximity to each other for bond formation.

50 51 Treatment of 15 with a variety of Lewis acids invariably returned unreacted starting material. Use of more forcing conditions such as extended reaction times, excess Lewis acid, or gentle heating only produced base line materials (TLC analysis). Since the intramolecular ene reaction has a small negative volume of activation, application of pressure should in theory enhance reaction rate 54 . Reaction of 15 at 98,000 psi m the presence of tin tetrachloride or dimethylaluminum chloride produced polymeric material and the methylation-lactonization pro duct 96, respectively. The latter product apparently arose from a 1,2 methyl shift of the Lewis acid-aldehyde complex 94, followed by lactonization with the neighboring ester group (Scheme XXX).

Schem e XXX

H OCH3 AIMs2CI ch, 2ci2 15 98,000 psi

H H 94 95

H 96 52 5.2 Aldol Condensation. At this point, the source of difficulty in ring clo sure was far from clear. It was not known, for example, whether the molecular framework was such as to not allow the C-4 and C-5 from coming into close proximity under any circumstances, or whether the difficulty resided in the Prins reaction itself. Therefore, the decision was made to form the same bond but through a different reaction pro cess, i.e., an aldol condensation (Eq 13).

OCH; HO OCH;

acid or base 0

H H

✓ 9 8 0 HO 0 OCH; 'CH;

Initially, it was thought that the requisite precursor 97 could be synthesized by a simple hydroboration-oxidation sequence of the C(5)-C(15) double bond in 15. This site of unsaturation, however, was unexpectedly inert. Alkyl boranes (9-BBN, R.T., 93 °C in a sealed tube, and at 98,000 psi; thexylborane) and even the diborane-DMS and THF complexes failed to add to this species. On the premise that the borate ester formed by hydroboration of the alde hyde function was serving to sterically shield the A ring, the dimethoxy acetal 99 was prepared. Exposure to the same reaction conditions showed it to be similarly unreactive (Scheme XXXI).

Scheme XXXI

o c h 3 'CH 3 alkyl boranes, N R BH3 • THF , or H BH3 • DMS

X - QCH3 99 X - O 15

An alternative synthetic route to 97 became evident when it was found that methyl ketone 7 2 a underwent complete and regioselective hydroboration with 9-BBN in refluxing 55 tetrahydrofuran to provide diol 1 0 0 . The success of this reaction is apparently due to the "tying back" of the tethered functionalities at C-ll and C-2. Whereas direct oxidation of the alkylborane provided dione 1 0 1 in 30% 54 yield56, a two-step sequence involving Swern oxidation of 57 diol 100 resulted in a much higher yield (Scheme XXXII)

Scheme XXXII

1.) 9-B B N , THF .reflux

2.) N aO H , EtO H , H20 2 HO .....

( 9 7 % ) H H 72a 100

TFAA,DMSO, ( 7 2 % ) CH2CI2 Et3N

*CH: T M S O T f, CH 2CI2 , - 78°C O TMSO^ ^OTMS

H H 102 ( 8 4 - 9 6 % ) 101

( 7 4 % ) 1.) LHMDS, THF 2.) M oOPh, -40°C

OCH;

'**■ CH 3 p b ( ° A c ) 4 a c e to n e HCI , r t MeOH, benzene 97 (1 : 3 ) (100%)

H ( 9 1 % ) H

103 104 55 Attempted regioselective ketalization of the 015 carbonyl by conventional methods failed (Table 6). Even tually, the method of Noyori was found to work very effec- 58 tively and in high yield

Table 6. Unsuccessful selective ketalization of 101. conditions results 1 2,4,6-trimethylpyridinium tosylate, after 72 h, ethylene glycol (1.4 eg), refluxing little product benzene 2 PTSA, ethylene glycol, rt no selectivity 2-methoxy-2-ethyl-l,3-dioxolane 3 trimethyl orthoformate, PTSA, no selectivity ethylene glycol, CH2C12, rt 4 ethylene glycol (1.3 eq) impractically PTSA, refluxing benzene slow 5 1,3-propanediol, PTSA no selectivity refluxing benzene

MoOPh hydroxylation of 102 was 10-15% more efficient than for 72a (see Section 4.4), implicating the double bond as a site of possible secondary reaction (oxomolybdenum complexes are known to attack double bonds at higher tern- peratures) 59 . Lead tetraacetate cleavage and hydrolysis delivered the aldol precursor in six steps and in 42% overall yield from 72a. 56 Keto aldehyde 97 was subjected to acidic and basic conditions, but cyclization was not observed (Table 7).

Table 7. Aldol condensation of 97. entry conditions results 1 tert-BuOK/THF no reaction 0 °C 2 tert-BuOK/t-BuOH rapid destruction of R.T. material 3 KOH/MeOH/R.T. destruction of material 4 HCl/HOAc/R.T. no reaction 1.5 days 5 benzene/PTSA only baseline material R.T.-»ref lux after several hours

Since Dreiding models indicated strain to be present in the resulting six-membered ring, it was reasoned that this may induce the retro-aldol process to occur readily. In other words, bond formation might have occurred, but due to the higher energy product, the equilibrium resides far to the starting material side within observable limits (TLC, NMR).

5.3 Intramolecular Alkylations. To probe the above possibility, a precursor was needed where C(4)-C(5) bond formation would proceed without pos sible reversal. Therefore, an intramolecular alkylation 57 was considered, using 105 as its precursor. The original plan to synthesize 105 directly from 15 was abandoned, when attempts to hydroborate the C(5)—C(15) double bond were not successful (Scheme XXXIII).

Schem e XXXIII Br

OCH.

H HO 105

OCH. NaBH4 , MeOH 15 (100%)

106

CBr4 , Ph3P (70%) lutidine

OCH; hydroboration N.R. r ch.

H 107 58 A workable route to 105 is outlined in Scheme XXXIV.

Scheme XXXIV

104 NaBH4 , 0°C MeOH (97%)

(85%) LiBr, acetone reflux, 5 h

OCH. OCH. ''•c h . ‘'C H . base o Oi

H H 110 105

Surprisingly, no lactonization occurred during the aldehyde reduction step. Direct bromination of alcohol 108 by standard methods (triphenylphosphine-carbon tetrabromide; N-bromosuccinimide-triphenylphosphine)^'^ worked poorly. Stepwise mesylation-bromide displacement, however, was found to work well, especially since the last step removed the ketal protecting group in situ. Extensive chlorination occurred in the mesylation step with methylene chloride as solvent and it was later found that this could be minimized 59 by performing the reaction in ether, in which the triethyl- ammonium hydrochloride salt is rather insoluble. Nonethe less, keto bromide 105 prepared according to Scheme XXXIV was always contaminated with about 8% of the corresponding chloro compound from which it could not be separated. Reaction of 105 with various bases led disappointingly to no cyclized product (Table 8).

Table 8 Conditions for the attempted alkylation of 105. entry conditions temp. (°C) result 1 tert-BuOK/THF -78-»rt no reaction 2 KHMDS/THF -78-»rt no reaction 3 NaHMDS/THF -78-»rt no reaction 4 LiHMDS/THF -78->rt no reaction HMDS =: hexamethyldisilazide

5.4 Cleavage of the A Ring. It was now evident that the molecular framework was such as to prevent bond formation between C-4 and C-5. Therefore, approaches were undertaken to open or expand the A ring with the intent of giving the molecule the addi tional flexibility it might need to achieve the required transition state conformation.

Expansion of the A Ring. The first approach to expand the A ring is shown in Scheme XXXV. 60

Scheme XXXV

OCH. r-CH. o o f TBDMS H H 111 112 1

H H 72a 113

Although the "reach" of C-5 to C-4 and the overall flexibility of the A ring in 111 were only marginally increased, it was hoped that this would nevertheless suf fice for bond formation to occur. The synthesis of 113 is outlined in Scheme XXXVI. 61

Scheme XXXVI

Hi...

BF3 OEt2 1 CH2 CI2 72a 9 -B B N .T H F ethanedithiol reflux (7 9 -8 5 % ) (7 2 % ) H H 114 115

PCC (63% ) 3 - A sieves

R3Si = TBDMS X =OTf O RaSi H B 116 113

Dithiolane 114 was prepared readily from 72a. However, use of conventional lithium amide bases failed to deprotonate 113 regioselectively at the less hindered C-14 (Table 9). Consequently, the mixture of silyl enol ether isomers 116 was carried forward with the intention of optimizing this step at a later point in time. It was anticipated that selective removal of the dithiolane group in the presence of the hindered silyl enol ether could be accomplished using non-acidic conditions such as methyl iodide in refluxing aqueous acetonitrile with buffer. In actuality, 116 resisted reaction under these conditions as well as when treated with methyl iodide in refluxing 62 aqueous acetone, both under prolonged reaction periods (Table 10, entries 1,2). On the other hand, use of elec- trophilic or Lewis acidic reagents either rapidly decom posed the substrate (entries 3-5) or displayed no selec tivity (entries 6,8).

Table 9. Kinetic silyl enol ether formation from 113a. entry______base______ratio A:Bb 1 lithium diisopropylamide 1:1.52 2 lithium isopropylcyclohexylamide 1:1.56 3 lithium bis(trimethylsilyl)amide 1:1.2 4 lithium tetramethylpiperidinamide 1:1 a All reactions done at -78 °C in THF. " Ratio determined by capillary G.C.

TABLE 10. Conditions Used for Deprotection of 116. Entry______Conditions______Results 1 aq. acetone, Mel, NaHCOg, N.R. reflux

2 Mel, aq CH3CN, Na2C03, reflux N.R. 3Tl(OCOCF3)3, THF material destroyed 4 ( N H j 2Ce(N03)6, aq CH3CN material destroyed 5I2/MeOH material destroyed

6 Et30+BFlt_; CuSO* rapid deprotection of both groups 7 (PhSeO)20, 60 °C, THF N.R. 6 days 8 NBS, lutidine, aq acetone, 0 °C deprotection of both groups 63 The dithiane 117 was next prepared since deprotection of the six-membered thioketal with methyl iodide is known to be rapid in some cases 6 2 . However, all attempts to hydroborate 117 led only to destruction of material (Scheme XXXVII).

Scheme XXXVII

various 1 ,3 - propanedithiol alkylboranes 72a ------destruction ( CH2CI )2 , Zn(OTf)2 of material

(56%) H 117

The difficulties associated with efforts to expand the A ring prompted us next to consider its cleavage as an alternative means of providing a suitable cyclization candidate.

Cleavage of the C(5)-C(15) bond. Cleavage could most easily be accomplished at C (5)~C(15) or C(15)-C(14).

OCH

'' c h . 64 The first of these options was pursued initially since fewer steps were involved from the known intermediate 15 (Scheme XXXVIII) to the target molecule 119. The radical center resulting from samarium diiodide cleavage of the

Scheme XXXVIII

OCH. OCH. OCH ''CH. hcH. *

O H

119 carbon-iodine bond was then expected to bond with the more reactive C-5 aldehyde center 6 3 . The attempted synthesis of 119 is shown in Scheme XXXIX. Thus, alcohol ester 106 was treated with dihydropyran and pyridinium tosylate acid to provide protected alcohol 120. The latter was oxidatively cleaved under Lemieux- Johnson conditions to produce two products in a 1:1 ratio 6 4 . These were separated on flash silica gel to pro vide what appeared by NMR to be the expected dialdehyde 121 and, as the more polar component, hydrate 122. Unfor tunately, the dialdehyde, upon standing, converted largely to 122. 65

Scheme XXXIX THPO

OCH; dihydropyran 106 P P T S , CH2CI2

(8 3 % ) H 120

Nal04, 0s04 (3 0 % ) H20 , dioxane

THPO THPO

HO '•CH.

HO H o 122 121

To study the possibility of dehydrating 122, the model hydrate 123 (eg 14) was prepared. Heating, distillation under vacuum, or stirring with 4A sieves did not alter its composition. A survey of 1,5 dialdehydes in the literature

HO Nal04, 0s04 (14) H20 , dioxane HO 123 66 revealed that most exist in their hydrated forms.65 , except in those cases where, because of strain or conjugation (124 and 125, respectively), the open form is energetically more favored66.

125

Cleavage of the C(15)-C(14) bond. Two routes were next examined which were anticipated to provide the cyclization intermediates 126 and 127, both with the desired cleavage at the C(14)-C(15) position.

OCH.

c h .

O' X OCH; OCH.

126 127

Synthesis of 126 proved unexpectedly difficult (Scheme XL). Ozonolysis of the bis silyl enol ether derivatives 128 and 129 did not lead to any observable product (for further details on the regioselective generation of 128 and 129, see below). Generation of bisacyloin 130 was 67 similarly inefficient: treatment of 128 and 129 with m-chlorobenzoic acid (with or without buffer; in hexane or methylene chloride) led to a complex product mixture of low mass yield and extensive hydrolysis, respectively. Even tually, small amounts of 130 were obtained using the Vedejs procedure, but the yield could not be optimized beyond 15%. Vedejs suggests that the poor yields commonly realized with relatively unhindered ketones (as in the A ring) stem from cross-condensation of the intermediate metallo-acyloin and another carbonyl functionality 52

Schem e XL

OTBDMS "** C H 3 i.)(p h M e2Si)2NLi 3 MCPBA or 0 3 no observable -7 8 °C ,T H F tr d m ro product 2.) TBDMSOTf TBDMSO

(1 7 % )

OTMS TMS - EtOAc MCPBA or 0 3 no observable - product TBAF, - 78°C THF TMSO (95% )

-< C H 3 (PhMe2Si)2NLi

MoOPh , - 4 0 °C , THF O ( 15 %) 68 At the same time, the synthesis of the alternate cyclization intermediate 127 was proceeding with fewer difficulties, and as a result that route now became the focus of endeavor. The initial synthetic stage is shown in Scheme XLI. Chloroketal 131 was chosen as starting material, since its carbon-chlorine bond was expected to be stable to all interim transformations and, when necessary, could undergo halide exchange to provide the more useful bromide or iodide. Kinetic deprotonation of 132 with lithium hexamethyldisilazide led to a 2:1 regioisomeric mixture of silyl enol ethers, the major product resulting from deprotonation of the more hindered C-5 position (Table 11). The assignment was later verified, when it was shown that oxidative cleavage of the major component led to a product showing a simple AB quartet absorption pattern in the NMR for the C-5 methylene protons. Scheme XU ci

107 (9 5 -1 0 0% )

H 131

(98%) aq HCI, Cl acetone Cl

H R3Si = TBDMS 133 132 R3Si = TMS 134 69 Table 11. Kinetic silyl enol ether formation of 133 and 134a . entry base ratio A;B 1 LHMDS 2:1

2 n-Bu^NF 1:2 TMS-EtOAc

PhMe,Si),NLi 1:20 a All reactions performed at -78 °C in THF.

This was quite unexpected since it seemed contrary to precedent and theory regarding enolate formation. Very hindered bases were next used. The tetra-n-butylammonium fluoride/(trimethylsilyl)ethyl acetate combination interestingly reversed the above ratio 67 . The mechanism by which this reagent silylates is shown in Scheme XLII.

Scheme XUI O KJO" TBA + Bu4NF + OEt TMSF TMSr * OEt A B C D

O -T B A + O o'- O

A, O Et H3Cx '^ 'O E t * • 1 1 * WAA/W C F

O -T B A + OTMS o O'TBA* f^ OEt rs + f^OE‘ VWVAA, TMS B H C 70 . The intriguing feature of this kinetic base is that contrary to alkali metal amides it derives its bulkiness from the cationic portion of the ion pair. Unfortunately, the trimethylsilyl enol ethers were inseparable by TLC, and proved to be labile toward hydrolysis. Excellent regio- selectivity was eventually observed with lithium diphenyl- tetramethyldisilazide (LDPTMS) 6 8 . The reason for these results is unclear. Chelation of the relatively small kinetic base, e.g., LHMDS, with one of the hetero atoms appended to the C ring cannot be discounted on the basis that 113 (where such chelation cannot occur) showed a somewhat different ratio of deprotonation with the same base (see Table 9). Trapping the enolate of 132 with tert-butyldimethylsilyl triflate (TBDMSQTf) afforded the much more stable enol derivative 133. Before proceeding forward, the direct hydroxylation of the enolate of 132 with MoOPh was attemp ted, but resulted largely in the formation of baseline 69 material on TLC. Ozonolysis or m-chloroperbenzoic acid treatment of the silyl enol ether derivative 133 to provide cleavage product or acyloin also worked poorly. Even tually, it was found that reaction of 133 with osomium tetroxide provided 135 in satisfactory yield 70 . The fully functionalized intramolecular alkylation precursors 140 and 142 were then synthesized as shown in Scheme XLIII. 71 Halogen exchange of the chloride in 136 was very slow and consequently resulted in low overall yield. Therefore bromo ketone 105 was carried through the same sequence, giving in the halide exchange reaction an excellent yield of the desired iodo compound 141.

Scheme XLIII

X X x

OCH;

0 s 0 4 , t - BuOH TBDMSO o MeOH acetone, HaO H benzene HO o

X = CI 133 ( 83% ) X = CI 1 3 5 (6 5 % ) X = CI 136 (100% )

X = Br 137 (88% )

Amberlyst -15 HC(OCH3)3 N a l, acetone pyridine, reflux (9 6 % ) (8 7 % )

(/ ^ och3 0 '■ch3 Amberlyst-15 0

h c (o c h 3)3 HjC HjC h 3c o X" T ' (81% ) H jCO 0CH3 A h H jC O OCR 142 141 140

Products 140 and 142 were treated under a variety of alkylative conditions but cyclized material again was not observed. Very forcing conditions were applied but only 72 resulted in elimination to form the olefinic ester 144 (Scheme XLIV).

Scheme XLIV

OCH OCH.

*' CH.

OCH.

X = Br 140 143 X= I 142 + O

144

The inability of 140 and 142 to undergo ring closure was difficult to understand, especially since the inter mediate 145 in Danishefsky's quadrone synthesis cyclizes in good yield (70%)71. TABLE 12. Conditions for attempted akylation of 140 and 142a.

Entry X______Base______Solvent_____ HMPA (temp °C)______Temp Range (0 °C)_____144 145

1 I LHMDS DME 1% (-30) -78 -* RT Xb X

2 I LHMDS THF 1% (-30) -78 -* RT X X

3 I LHMDS DME 10% (-30) -78 -* RT X X

4 I NaHMDS DME - -78 -* RT X -

5 I NaHMDS DME 5% (-30) -78 -» RT X X

6 I NHMDS DME 10% -20 X X

7 I KHMDS DME - -78 -> 0 X X

8 Br NaHMDS DME 10% (-30) -78 -» 0 X X

9 Br KHMDS DME - -78 -» 0 X X a Reactions were assayed by TLC and FT-80 NMR. b 'x' indicates the appearance of the material. 74

...•CH3 LDA.HMPA

CHa THF, -78°C — rt (76% ) O OCH3

145 146

At about this point in the synthesis, X-ray analysis of bromoketone 105 was obtained (Figure 9). The relevant feature to note is the downward puckering of C-2, placing the p-bromoethyl substituent and C-ll methyl group in equatorial positions. One possible reason for this is the relief of 1,3 diaxial interactions between the C-ll methyl and C-9 and C-l protons. Conformations of 105 are shown in Figure 10. Since 105a represents the ring conformation needed for intramolecular alkylation to take place, this may account for the failure of 105 to undergo ring closure (see section 5.3). 75

CH,

'•CH;

Figure 9: X-ray structure of 105. conformation conformation adopted needed for T.S. in the solid state

E = C 0 2C H 3 Figure 10: Conformations of 105

The conformational bias in 105 would be expected to be preserved in the remaining B and C rings of 140 and 142 as well, thus providing an explanation for these systems' inability to react (Figure 11).

X X E

142a 142b

conformation favored needed for T.S. conformation

E = C 0 2C H 3 x = o c h 3

Figure 11: Conformations of 142 77 5.5 Summary. Intermediates 15, 97, 105, 140, and 142 were prepared from 72a and were shown uniformly not to undergo the desired C(4)-C(5) bond formation (Figure 12). X-ray analysis of 105 indicated an equatorial pre ference for the C-2 substituent, thereby offering a pos sible explanation for this unreactivity. 78

( number of steps from

72a .yield) O

H OCH;

CH; Lewis acid (2 steps, 75%)

H 15

OCH; acid or (6 steps, 42%) CH; o base

OCH; base ‘''C H ; (9 steps, 32%) O

105

OCH;

CH;

(13 steps, 19%) 140 H3C O (1 4 steps, 14% ) 142

X = l 140 X = Br 142

Figure 12: Intermediates for C-4.C-5 closure. CHAPTER VI

ALTERNATIVE APPROACHES AND CONCLUSION

6.1 Ramberq-Backlund Route. Large-ring formation followed by ring contraction was next considered in effecting the C(4)-C(5) closure. Aside from providing a bridge between two otherwise distant terminii, large-ring annulation might require less upward puckering of C(2), thus minimizing to some degree the 1,3- diaxial interactions discussed in Section 5.4. The synthe tic strategy chosen was the Ramberg-Backlund rearrangement as outlined in Scheme XLV 73

Scheme XLV

OTBDMS OCH,

TBDMSO

101 128a 147

M>0

OCHj MsO * cHg R a m b e rg - B acklund

Hj CO* HjCO' 148 149 79 80 Regioselective silyl enol ether formation was attempted under thermodynamic control (Table 13).

OSIR' — CH. b ase; R3SiX (16)

H H

101

Table 13. Thermodynamic silyl enol ether formation from 101. b,c entry conditions ratio A :B 1 BMDAa/TMSCl 1:1.3d 2 LDA(1 eq), t-BuOH(0.05 eq) 1. 6:1 HMPA, 50 °C, 1 h; TBDMSOTf 3 KHMDS, HMPA; TBDMSOTf 1 :1.1

4 KH, THF, HMPA, TBDMSCl 1 .6:ld -78 °C ■+ RT 5 KH, THF, HMPA, RT; poor silylation -78 °C, TBDMSOTf

6 KH, DME, HMPA 2.4:1 TBDMSCl, -78 °C -»RT. a BMDA = bromomagnesium diisopropylamide k Ratios determined by capillary G.C. c All silyl enol ethers were inseparable mixtures d Reaction proceeded to 60-70% completion. 81 Eventually, it was found that a slight modification of the Turner procedure (replacement of THF with DME) provided the 74 desired enol derivative in an acceptable yield (entry 6) Attention was next turned to double oxidative cleavage of the C(12)-C(4) and C(15)-C(14) bonds. Efforts to form the intermediate bis-acyloin by treatment of 128a with MCPBA or osmium ttroxide led only to destruction of starting material. Direct acyloin formation using the Davis reagent similarly did not meet with success (eg 17).

1.) KH , HMPA, DM E, rt 101 " O (17)

H 150

6.2 Effecting Conformational Change at C(2). A second approach using 151 as its key intermediate was also being pursued, and in light of the above difficul ties, it now received more attention.

OTBDMS

142b 151

e = c o 2c h 3 x = o c h 3 82 The strategy rested on the hypothesis that replacement of the relatively less sterically demanding C(11)-ester group in 142 with a methylene tert-butyldimethylsilyl ether as in 151 might force the molecule to place the ether group in an equatorial position, and thus to adopt the conformation needed for bond formation. The initial phase for the synthesis of 151 is shown in Scheme XLVI.

Scheme XLVI

OTMS OCH. ‘‘ CH. cOTMS DIBAL, -78°C 105 r° Lr ° o > ^ TM SO TF, - 78°C I— o (86%) H H (8 9 % ) 152 153

H,0+

Br Br OTBDMS OH

'"CH. '‘CH. imidazole O side product o DMF, TBDMSCl

H H 155 154

Protection of the alcohol by the method shown resulted in an inseparable mixture (1:1) of 155 and an unknown side- product. The latter was presumed to be the chlorinated substrate, since NMR properties of the compounds were 83

Scheme XLVII

Br Br OTBDMS OTBDMS

'CH. EtgN, TBDMSOTf PTSA. acetone 153 O -78°C H H (9 5 % ) 156 (9 0 % ) 155

(PhMe2Si)2NLi (100%) TBDMSOTf

OTBDMS OTBDMS TBDMSO CH:

'CH.

TBDMSO O M eO H , PhH NMO

( 70% ) HO ( 71% ) 157 159 158

Nal acetone

TBDMSO CH: TBDMSO CH.

HC(OCH3)3 OCH. Amberlyst-15 OCH. O (7 7 % , 2 steps) O 160 151 84 similar, and the side-product exhibited a shorter retention time on capillary G.C. This problem was circumvented by use of tert-butyldimethylsilyl trifluoromethanesulfonate instead as shown in Scheme XLVII. Preliminary attempts to alkylate internally did not meet with success (Scheme XLVIII).

Scheme XLVIII

TBDMSO CH. TBDMSO CH.

LHMDS, HMPA OCH; 151 + OCH. T H F , -78°C rt OCR OCH. O O 161 151

NaHMDS 161 H li

Lithium hexamethyldisilazide was too unreactive and returned mostly starting material. The sodium analogue quickly eliminated iodide ion (observed from vinyl absorp tion in FT-80 NMR of the crude reaction product).

6.3 Conclusion. Intermediates 15, 97, 105, 150, 142, and 151 were synthesized and found not to undergo the required C (4)—C(5) bond formation. Although the total synthesis of 85 trixikingolide was not realized within the period of this doctoral thesis, several interesting synthetic developments did occur. Most notable among these are:

1) An efficient two-step preparation of bicyclic dione 18, an important starting material for polyguinane synthesis. 2) Utilization of an unusual cyclization-chlorination pathway followed by the Prins reaction of 42. 3) A novel regioselective synthesis of an unsa turated methyl ketone from a cyclic ketone precursor. 4) A stereoselective application of the Trost cyclopentaannulation procedure to a natural product synthesis. The key step remains the C(4)-C(5) ring closure. Work is now underway to replace the iodide of 151 with bromide, thereby permitting the use of more forcing reaction condi tions (e.g., heat and more polar medium) while avoiding the elimination side reaction. The large ring formation-ring contraction sequence discussed earlier remains a promising avenue. The Ramberg- Backlund route was not fully explored. Ozonolytic cleavage 86 of enol ether 128a might possibly provide the desired intermediate 147. An alternative approach which would utilize metallo- cycle 165 is outlined in Scheme IL. The final reductive coupling to yield 166 is quite feasible, since there

Scheme IL

E CH. E CH.

S nR ;

161 162

E CH;

- Pd(0) Pd(0) S n R : CH.

166 165 164

y = o c h 3 E - C 0 2CH3 X - 1 or Br are many examples in the literature of transition metal- mediated annulations to give four-, five-, and six-membered rings75 EXPERIMENTAL

Aldehyde Ester 15. O A dry solution of the acyloin 0 92 (10 mg, 0.043 mmol) in benzene/ methanol (1.5 mL and 0.5 mL) was /jT'CHa prepared and cooled to 5 °C. Lead tetraacetate (20 mg, 0.046 mmol) H was added in one portion. After 5 min 0.5 mL of sodium bicarbonate solution was added and the mixture was filtered through Celite. The filtrate was extracted with benzene. Drying over magnesium sulfate and concentration gave 10.5 mg of a colorless oil (93%; larger scale reactions gave up to 100% yield). IR (neat, cirT^) 3045 (w), 2940 (s), 2850 (m), 2720 (w), 1720 (s), 1460 (m), 1445 (m), 1210 (s), 1130 (s);XH NMR (300 MHz, C6D6) 6 9.35 (m, 1 H), 5.97-5.94 (m, 1 H), 5.46-5.43 (m, 1 H), 3.26 (s, 3 H), 2.50-2.40 (m, 2 H),2.30-2.24 (m,1 H), 2.11-1.92 (m, 3 H), 1.77-1.68 (dd, J = 12.8, 13.7 Hz, 2 H), 1.52-1.43 (t, J = 12.6 Hz, 2 H), 1.13- 0.99 (m, 3 H), 1.01 (s, 3 H); MS m/z_ (M+ ) calcd for C16H2203 262.1569, obsd 262.1535. Anal. Calcd for C16H220 3: C, 73.25, H, 8.45. Found: C, 73.04; H, 8.46.

87 88 Tricyclic Unsaturated Methyl Ester 16. Method A: From 54. To a dry, two-necked flask 0 equipped with a water condenser OCH3 and argon gas inlet was placed sodium methoxide (119 mg, 2.20 mmol), 2 mL of dry methanol, and H nickel tetracarbonyl (0.765 g 4.50 mmol). After 10 min of stirring 54 (94 mg, 0.36 mmol) was added as a solution in methanol and ether. The reac tion mixture was then heated to 48 °C for 2 h, whereupon the excess nickel (0) was destroyed with a methanol/iodine solution. The mixture was poured into ether and washed with 3 x 35 mL of sodium thiosulfite solution. The extract was dried over magnesium sulfate and separated by MPLC (silica gel, 20% ethyl acetate in petroleum ether) to give 41 mg (55%) of a colorless oil; IR (neat, cm"1) 3040 (w), 2930 (s), 1715 (s), 1620 (m), 1440 (m), 1240 (s), 1210 (s), 1040 (s); XH NMR (300 MHz, C6D6) 5 6.66-6.64 (m, 1 H), 5.54-5.28 (m, 2 H), 3.71 (s, 3 H), 3.05 (m, 1 H), 2.67-2.45 (m, 3 H), 2.33 (m, 1 H), 1.97-1.65 (m, 4 H), 1.39-1.34 (m, 1 H); 13C NMR (75 MHz, CDC13) ppm 165.46, 142.17, 137.90, 137.02, 128.75, 66.69, 56.16, 51.58, 51.16, 45.49, 39.65,

33.24, 30.41; MS m/z (M+ ) calcd for C13H1602 204.1140, obsd 204.1146. 89 Method B: Using 55a as Intermediate. A solution of 55a (119 mg, 0.405 mmol), anhydrous triethylamine (82 mg, 0.81 mmol), triphenylphosphine (14 mg, 0.02 mmol), palladium(II) acetate (4.5 mg, 0.02 mmol), and dry methanol (152 mg, 4.76 mmol) in 1 mL of dimethylformamide was prepared under argon and at room temperature. The system was purged with carbon monoxide for 5 min, and then maintained under 1 atm of carbon monoxide gas for 1.5 h. The reaction mixture was poured into water and extracted with ether. The extracts were dried over sodium sulfate and separated as above to yield 60 mg (60%) of 16.

Tricyclic Ketone 17. A slurry of powdered 3A o molecular sieves (30 mg), sodium acetate (19 mg, 0.23 mmol), and H pyridinium chlorochromate was prepared in dry methylene chloride H (5 mL) under an argon atmosphere. A solution of alcohol 52 (125 mg, 0.726 mmol) in methylene chloride (2 mL) was added by a double ended needle. After 1 h, the reaction mixture was diluted with ether (25 mL) and poured over Florisil. The eluate was concentrated to give 116 mg (94%) of a colorless oil which required no further purification; IR (neat, cm-1) 3040 (w), 2940 (m), 90 2865 (m), 2840 (m), 1730 (s), 1450 (w), 1440 (w), 1400 (w), 1360 (w); lH NMR (300 MHz, CDC13) 5 5.68-5.65 (m, 1 H), 5.52-5.49 (m, 1 H), 2.84-2.73 (ddt, J = 17.2, 9.8, 2.3 Hz, 1 H), 2.49-1.64 (series of m, 10 H), 1.50-1.42 (m, 1 H); 13C NMR (75 MHz, CDC13) ppm 222.60, 136.14, 130.17, 65.89, 57.91, 48.19, 41.01, 38.44, 34.93, 32.10, 29.40; MS m/z (M+ ) calcd for Cu Hll(0 162.1044, obsd 162.1036. Anal. Calcd for C u H u O: C, 81.44; H, 8.70. Found: C, 81.13; H, 8.62.

Bicyclic Dione 18. Method A: Hydrolysis/Cyclization of Enol Ether 26. Enol ether 26 (lOOmg, 0.389 mmol) was dissolved in 5 mL of o o tetrahydrofuran and treated with 4 drops of 6N hydrochloric acid. H After being stirred for 1 h, the solution was diluted with ether, washed with 5% sodium bicarbonate solution, and dried to give 72 mg (99%) of a colorless oil which showed identical spectroscopic features to that reported in the literature. This keto ester was next cyclized according to the proce dure of Eaton (88%)6. Overall yield 61%. 91 Method B: Treatment of (26) with Sodium Methoxide. Crude enol ether 26 (180 mg, 0.7 mmol) in 2 mL of ether was added slowly (over 5 min) to a well stirred suspension of sodium methoxide (45 mg, 0.85 mmol) in ether at 10 °C under argon. After 3 h of stirring, the brown suspension was transferred by suction to a stirred buffer solution of 10% potassium dihydrogen phosphate (2 mL). The mixture was extracted with methylene chloride and dried. The product residue was distilled in a Kugelrohr apparatus (0.1 torr, 80 °C) to give 77 mg (56%) of a crystalline material spectroscopi cally identical to 18 prepared by Eaton6.

Silyl Enol Ether Ester 26. To a three-necked, 50-mL flask was placed anhydrous zinc OTMS chloride (1.63 g, 12.1 mmol). The salt was then melted under vacuum OEt (0.1 torr) with a Bunsen burner and allowed to cool to room temperature in vacuo. Argon was flushed back in, 30 mL of ether was added, and the mixture was heated at reflux until the zinc chloride dissolved. The solution was cooled to 92 room temperature and 1-trimethylsiloxy-l-ethoxycyclopropane (5.0 g, 23 mmol) was added via syringe. The mixture was stirred for 1 h at room temperature and heated at reflux for 30 min to insure zinc homoenolate formation. The mixture was cooled to 0 °C, and 2-cyclopentenone (0.82 g, 10 mmol), copper(I) bromide-dimethyl sulfide complex (40 mg, 0.19 mmol), and hexamethylphosphoramide (3.1 g, 18 mmol) were added in that order. The cloudy mixture was stirred at room temperature for 5 h, after which time 4 g of silica gel and 20 mL of hexane were added. The super natant was removed, and the solid was washed with 3 x 20 mL of hexane/ether (1:1). The organics were filtered through Celite and the solvent was removed to give 2.0 g (70%) of nearly pure product as an oil. This material could be used directly in the next reaction or purified further by Kugelrohr distillation (0.2 torr, 73-78 °C). IR (neat, cm-1) 3055 (w), 2955 (m), 2845 (m), 1730 (), 1638 (s), 1445(w), 1370 (m), 1345 (m), 1250 (s), 1180 (s), 1035 (w), 928 (m), 870 (m), 845 (s); XH NMR (300 MHz, C6D6) 5 4.62- 4.60 (m, 1 H), 3.96 (g, J = 7 Hz, 2 H), 2.59-2.53 (m, 1 H), 2.27-1.19 (m, 8 H), 0.97 (t, J = 7.1 Hz, 3 H), 0.14 (s, 9 H); MS m/z (M+ ) calcd for C13H21t03Si 256.1495, obsd 256.1487. 93

Iodo Acetal 28. To a 5L, three-necked flask equipped with a mechanical stir rer, reflux condenser, and calcium chloride drying tube was placed 3L of reagent grade acetone and sodium iodide (500 g, 3.3 mol). Bromo acetal 27 (120 g, 0.62 mol) was added dropwise over 5 min. A white precipitate appeared after 10 min. Following overnight stirring, the mixture was diluted with 4L of ether, filtered, and concentrated to give a dark liquid which was diluted with petroleum ether and washed with sodium thiosulfite solution until the organic layer was light yellow. After concentration, the yellow residue was distilled (0.4 torr, 56 °C; not to exceed 70 °C to avoid decomposition) and stored over copper wire and potassium carbonate. Yield: 121 g (81%); IR (neat, cm ^) 2980 (s), 2870(s), 1385 (s), 1250 (s), 1155 (s), 1140 (s), 1020 (s); XH NMR (300 Hz, C6D6) 6 4.39 (t, J = 5.0 Hz, 1 H), 3.73- 3.68 (m, 2 H), 3.31-3.21 (m, 2H), 2.99-2.94 (t, J = 16.9 Hz, 2H), 2.06-1.97 (m, 2 H), 1.76-1.66 (m, 1 H), 0.66-0.60 (m, 1H); MS m/z. (M+ ) calcd for C6Hn 02I 242.9886, obsd 242.9853. 94 Bicyclic Dione 29. To a solution of 18 (200 mg, 1.5 mmol) and sodium iodide (100 mg, 0.67 mmol) in dry dimethylformamide was added sodium hydride (36 mg, 1.5 mmol). After 1 h at H room temperature, 4-bromo-l-butene (788 mg, 5.84 mmol) was added and the mixture was heated to 50 °C. Afte 24 h, the mixture was poured into 50 mL of 10% potassium dihydrogen phosphate solution and extracted with ether. The combined organic layers were dried and con centrated. The residual oil was purified by MPLC (silica gel, 30% ethyl acetate in petroleum ether) to give 249 mg (90%) of a colorless oil; IR (neat, cm ) 3070 (w), 2940 (s), 1750 (s), 1710 (m), 1635 (m), 1405 (m), 1135 (m), 1090 (m), 915 (m); XH NMR (300 MHz, C6D6) 5 5.64-5.51 (m, 1 H), 4.93-4.84 (m, 2 H), 2.37-0.75 (series of m, 13 H); MS m/js (M+ ) calcd for C12H 160 2 192.1140, obsd 192.1138. Bicyclic Dione Acetal 32. To a dry 12-L, three-necked flask equipped with a 1-L addition funnel, mechanical stirrer, and bubbler, was added dry glyme (2 L), and sodium hydride (60% in

H oil, oil was not removed, 53.3 g, 1.33 mol). Dione 18 (173 g, 1.26 mol) in 800 mL of glyme was added via the addition funnel over 1.5 h, during which time the reaction flask was cooled to about 5 °C with a large ice bath. After 40 min, hexamethylphosphoramide (300 mL) was added, followed by iodo acetal 28 (263 g, 1.80 mmol). The ice bath was removed. The reaction mixture was heated to 40 °C, but product formation remained sluggish. Therefore, an additional 700 mL of glyme and 150 mL of hexamethylphosphoramide were added. After another 16 h, the solution became homogeneous and starting material was consumed. The mixture was cooled to room temperature and added through a large canula (by suction) to a well stirred solution (4 L) of 10% potassium dihydrogen phosphate in a 12-L flask. The aqueous layer was extracted with 3x3 L of ether and the combined extracts were washed with 1000 mL of brine and dried. Concentration and crystallization from ether gave 126 g of material. The mother liquor was con centrated and chromatographed over 1.5 kg of silica gel 96 with 10, 20, and 30% ethyl acetate in petroleum ether as eluent to give an additional 102 g of crystalline product (total yield 72%); mp 95-96 °C; IR (CDC13, cm-1) 2960 (m), 2855 (m), 1760 (s), 1720 (m), 1710 (m), 1405 (m), 1375 (w), 1140 (s), XH NMR (300 MHz, CDC13) 6 4.37 (5, J = 5.1 Hz, 1 H), 3.99 (d, J = 11.8 Hz, 1 H), 3.96 (d, J = 10.8 Hz, 1 H), 3.70(td, J = 2.0, 12 Hz, 2 H), 2.93 (quintet, J = 5.0 Hz, 1 H), 2.43-2.10 (m, 6 H), 2.10-1.96 (m, 1 H), 1.83-1.66 (m, 4 H), 1.51-1.43 (m, 2 H), 1.34-1.28 (m, 1 H): MS m/z (M+ ) calcd for C^H^O,, 254.1518, obsd 254.1455. Anal. Calcd for ClltH20O<*: C, 66.65; H, 7.99. Found: C, 66.78; H, 7.94.

Diene Acetal 34. Dimesylate acetal 39 (160 mg, o-— ^ 1 0.362 mmol) was dissolved in dry

\ benzene and the solvent was removed in vacuo. Diazabicyclono- nene (1 mL) was added and the temperature was raised to 106 °C for 2.5 h, then 140 °C for 15 min. Cold benzene (30 mL) was added to the black mixture, and the resulting solution was washed with 0.5 M hydrochloric acid and 5% sodium bicarbonate solution prior to drying. MPLC purification (silica gel, 2.5% ethyl acetate in petroleum ether) 97 provided 36 mg (69%) of a colorless oil. IR (neat, c m -'1') 3040 (m), 2955 (w), 2920 (s), 2850 (s), 2770 (w), 2715 (w), 2650 (w), 1615 (w), 1605 (w), 1465 (m), 1445 (m), 1425 (m), 1400 (m), 1375 (m), 1395 (w), 1280 (m), 1266 (m), 1235 (m), 1175 (w), 1140 (s), 1110 (m), 1050 (2), 1030 (m), 1000 (s), 960 (w), 940 (m), 925 (m); XH NMR (300 MHz, CDC13) 6 5.51- 5.45 (m, 4 H), 4.92-4.40 (m, 1 H), 4.05-3.97 (m, 2 H), 3.71-3.62 (m, 2 H), 2.64-2.55 (ddt, J = 16.0, 9.0, 1.6 Hz, 2 H), 2.47-2.42 (m, 1 H), 2.05-1.91 (m, 3 H), 1.52-1.49 (m, 4 H), 1.22-1.16 (br d, J = 7.2 Hz, 1 H); 13C NMR (75 MHz, CDC13) ppm 135.77, 128.77, 102.74, 68.03, 66.88, 44.08, 41.13, 32.71, 31.07, 25.87; MS m/z (M+ ) calcd for Cll4H20O2 220.1463, obsd 220.1430. Anal. Calcd. for C11|H20O2; C, 76.36; H, 9.15. Found: C, 76.34; H, 9.17.

Diol Acetal 37. Dione acetal 32 (43 g, 0.17 mol) was dissolved in 650 ml of dry methylene chloride and the solution was cooled to -78 °C. Diisobutylaluminum hydride (1 M in p hexanes, 0.38 mol) was added over 1 h. After an additional 2 h, 600 ml of saturated potas sium sodium tartrate solution was introduced and the mix 98 ture was stirred overnight. The layers were separated and the aqueous layer was extracted with 3 x 500 ml of methylene chloride. The combined extracts were dried and concentrated to give 45 g of crude product which was chro matographed over silica gel (60% ethyl acetate in methylene chloride) to give the three diol isomers 37a-c (38 g, 88%) in the ratio of 4.5:3.5:2.0. If the above DIBAL solution is in tetrahydrofuran , the ratio changes to 7:1:2;

For 37a; IR (neat, cm"1) 3630 (w), 3460 (m), 3010 (m), 2960 (s), 2875 (m), 1240 (w), 1145 (m), 1100 (m), 910

(S); 1H NMR (300 MHz, CDC13) 5 4.48 (t, J = 4.9 Hz, 1 H), 4.90 (br s, 2 H), 4.07-3.99 (m, 4 H), 3.72 (dt, J =12, 2.4 Hz, 2 H), 2.11-1.29 (series of m, 15 H); 13C (75 MHz, CDC13) ppm 102.63, 81.15, 66.71, 56.05, 48.51, 34.62, 33.64, 31.01, 29.32, 25.65; MS m/z (M+-H) calcd for C11(H23Ol< 255.1596, obsd 255.1592. 99

For 37b; IR (neat, cm"1) 3340 (br), 2940 -- "7 (s), 2850 (m), 1150 (s); XH NMR Is0 HO / o h (300 MHZ, C D C lg) 5 4.50 (t, J = 4.7 Hz, 1 H), 4.07 (dd, J = 10, CD 5 Hz, 2 H), 3.81 (t, J = 5.5 Hz, 2 H H), 3.73 (dt, J = 2, 11.7 Hz, 2 H), 2.83 (br s, 2 H), 2.18-1.54 (series of m, 12 H), 1.30

(br d, J = 13.4 Hz, 1 H), 1.10 (m, 2 H); 13C NMR (75 MHz,

CDClg) ppm 102.96, 79.45, 66.82, 59.50, 47.98, 34.13, 31.72, 29.76, 25.66, 24.18; MS m/z (M+-H20) calcd for C^HjjOg 238.1569, obsd 238.1586.

For 37c; IR (neat, cm-1) 3430 (s), 2940 (s), 2855 (s), 1450 (m), 1395 (m), 1370 (m),

HO OH 1235 (m), 1135 (s), 1070 (m); 13C NMR (75 MHz, CDClg) ppm 102.96, 78.63, 78.03, 77.05, 75.63, 75.46, 66.93, H 55.56, 45.67, 33.36, 32.54, 30.58, 29.32, 28.33, 26.64, 25.82; MS m/z (M+-H) calcd for Ci^H2 30^ 255.1596, obsd 255.1599. 100

Chloro Acetals 38a,b. Diol Acetal 37a-c (250 mg, 0.977 mmol) were treated at 0 °C with phosphorus oxychloride (2.95 g, 19.5 mmol). After 24 h at 58 °C, the reaction mixture was quenched by dropwise addition H of water (3 mL) at 0 °C. The aqueous layer was extracted with ether, and the combined extracts were washed with brine, dried over magnesium sulfate, and purified by MPLC (silica gel, 3% ethyl acetate in petroleum ether) to give 87.5 mg (35%) of two colorless oils in equal proportions.

For 38a: XH NMR (CDC13, 300 MHz) 6 5.86-5.81 (m, 1 H), 5.40-5.36 (m, 1 H), 5.10 (dt, J = 20.7, 2.3 Hz, 1 H), 4.47 (q, J = 5.4 Hz, 1 H), 4.08 (dd, J = 4.2, 5.0 Hz, 2 H), 3.79-3.70 (td, J = 2.3, 10 Hz, 2 H), 3.79-1.30 (series of m, 13 H); MS m/z. (M+ ) calcd for C11|H21C102 256.1230, obsd 256.1227.

For 38b. IR (neat, cm”^) 3040 (m), 3070 (s), 2860 (s), 1475 (w), 1440 (m), 1385 (s), 1250 (s), 1145 (s), 960 (m), 990 (m), 880 (m), 860 (m), 740 (m); XH NMR (CDC13, 300 MHz) 101 6 5.75 (m, 1 H), 5.32 (in, 1 H) , 4.50 (d, J = 4.7 Hz, 1 H) , 4.23 (dd, J = 1.7, 2.5 Hz, 1 H), 4.08 (ra, 2 H), 3.75 (td, J

= 11, 1.5 Hz, 2 H), 2.70-1.27 (series of m, 13 H); MS m/z (M+ ) calcd for ClltH21Cl02 256.1236, obsd 256.1232.

Dimesylate Acetal 39. Diol acetal 37a-c (237 g, 0.93

L of dry methylene chloride and the solution was cooled to 0 °C. Pyridine H (462 g, 5.39 mol) in 600 mL of methylene chloride was added to the heavy white suspension and the resultant mixture was allowed to stir at 0 °C for 1 h, and then allowed to warm to 15 °C over an additional hour. Enough 5% sodium bicarbonate solution was added to dissolve the salts. The mixture was poured into water and extracted with methylene chloride. The combined extracts were washed with 5% sodium bicarbonate solution, water, and brine, and then dried over sodium sulfate. Concentration and chromatography over silica gel (700 g, 15% ethyl ace tate in methylene chloride) gave 335 g (85%) of a semi- crystalline product; IR (neat, cm ) 3020 (w), 2950 (m), 2760 (m), 1455 (m), 1350 (s), 1170 (s), 1140 (m); XH NMR (300 MHz, C6D6) 5 4.95-4.90 (m, 2 H), 4.67-4.63 (m, 1 H), 102 3.81-3.74 (m, 2 H) , 3.38-3.28 (m, 2 H), 2.48-2.40 (3 sing lets for 3 isomers, 6 H), 2.15-1.14 (series of m, 13 H); MS m/z^ (M+-CH3S03) calcd for C15H2 505S 333.1372, obsd 333.1425.

Diene Aldehyde 42. Method A: From Diene Acetal 34. Diene acetal 34 (1.59 g, 6.82 o mmol) was dissolved in 75 mL of acetic acid and 75 mL of water. H / T \ The mixture was heated to 48 °C Pl for 50 h, after which time it was cooled to room temperature, neutralized with iced 5% sodium carbonate solution, and extracted with methylene chloride prior to drying. Concen tration and purification by MPLC (4% ethyl acetate in petroleum ether) yielded 740 mg (67%) of a colorless oil and 152 mg (10%) ofstarting material; IR (neat, cm ) 3040 (m), 2920 2s), 2840 (m), 2715 (w), 1720 (s), 1680 (m), 1440 (w), 1145 (m); XH NMR (300 MHz, C6D6) 6 9.36-9.35 (m, 1 H), 5.63-5.60 (m, 2 H), 5.56-5.51 (m, 2 H), 3.46-3.28 (ddt, J = 17.0, 8.8, 2.2 Hz, 2 H), 2.51-2.44 (m, 1 H), 2.41-2.35 (td, J = 9.6, 1.4 Hz, 2 H), 2.13-2.10 (br d, J = 17.0 Hz, 2 H), 1.84-1.53 (t, J = 8.2 Hz, 2 H), 13C NMR (75 MHz, CDC13) ppm 103 202.42, 134.97, 129.47, 67.68, 43.87, 40.99, 39.98, 30.32; MS m/£ (M+ ) calcd for Cn Hu O 162.1045, obsd 162.1082.

Method B, Part 1: By Transacetalization. Diene acetal 34 (1.52 g, 6.89 mmol) and p-toluenesulfonic acid (47 mg, 0.36 mmol) were dissolved in 60 mL of dry methanol and the mixture was stirred at the reflux temperature for 4.5 h. Solid sodium carbonate was added and the solvent was removed in vacuo. The resulting residue was washed with 3 x 50 mL of methylene chloride. Concentration of the combined organic phases gave 1.5 g (for yield, see Part II) of an oil, which was carried on into the hydrolysis step without further purification; IR (neat, cm-^) 3020 (m), 2920 (s), 1442 (m), 1382 (m), 1190 (m), 1120 (s), 1058 (s), 760 (m); 1H NMR (300 MHz, C6D6) 6 5.54-5.48 (m, 4 H), 4.27 (t, J = 4.0 Hz, 1 H), 3.14 (s, 6 H), 2.58-2.37 (m, 4 H), 2.00-1.55 (m, 5 H); MS m/js (M+ ) calcd for C13H20O 2 208.1451, obsd 208.1458.

Part II: The crude product from above was treated with 12 mL of acetic acid and 2 mL of water and the mixture was stirred at room temperature for 1.5 h, and at 0 °C over night. The mixture was neutralized with cold 5% sodium 104 bicarbonate solution and extracted with methylene chloride. The combined organic layers were dried over magnesium sulfate and concentrated to give an oil which was purified by MPLC (4% ethyl acetate in petroleum ether). There was isolated 859 mg (77%) of 42 as a colorless oil and 96 mg (6.3%) of starting material.

Phenyl Alcohol 46. A solution of 42 (190 mg, 1.17 mmol) in 5 mL of dry benzene OH at 0 - 5 °C was treated with tin tetrachloride under argon (0.45 Ph mL, 1.73 mmol). After 1.5 h, the H reaction mixture was quenched with saturated ammonium chloride solution and extracted with ether. The extracts were dried and separated by MPLC (18% ethyl acetate in petroleum ether) to yield 14 mg (56%) of a colorless oil. IR (neat, cm-1) 3360 (broad), 3020 (m), 2920 (s), 2840 (s), 1935 (w), 1860 (w), 1685 (m), 1595 (m), 1490 (m), 1445 (s), 1000 (m), 980 (m), 920 (m), 750 (m), 720 (m), 700 (s); XH NMR (300 MHz, C6D6) 6 7.32-7.16 (m, 5 H), 5.71-5.63 (m, 1 H), 5.55-5.50 (m, 1 H), 4.11 (m, 1 H), 2.61-1.25 (series of m, 12 H); MS m/z. (M+ ) calcd for Cn H 20O 240.1514, obsd 240.1518. 105 Chloro-Alcohol 47a,b. Method A. To an anhydrous solution of

OH 42 (1.0 g, 6.17 mmol) in 50 mL of methylene chloride under argon and -••iCl at -78 °C was added tin tetrachlo

H ride (1.93 g, 7.40 mmol). The mixture was allowed to warm to -30 °C over 1 h during which time a burgundy color appeared. After an additional hour, more tin tetrachloride was added and the reaction mixture was stirred at -10 °C for 1 h. The mixture was cooled to -78 °C, quenched with saturated sodium bicarbonate solution, and extracted with methylene chloride. The combined extracts were washed with brine, dried over magnesium sulfate, and passed through a short Florisil column. MPLC purification (silica gel, 25% ethyl acetate in petroleum ether) provided 48 and 47a,b as three colorless oils in quantities of 57 mg (5%), 313 mg (27%), and 290 mg (24%), respectively. The second and third compounds were shown to be stereoisomers by conver sion to 17. 106

For 47a: IR (neat, cm"1) 3600-3200 (br s), 3070 (m), 2950 (s), 2880 (m) OH 2860 (m), 1460 (m), 995 (s), 730 (s); XH NMR (300 MHz, C6D6) 5 5.55 V ^ S -/ (br s, 2 H), 4.38-4.31 (m, 2 H), H 2.67-2.61 (m, 1 H), 2.33-2.31 (m, 1 H), 2.25-2.21 (t, J = 6.5 Hz, 1 H), 2.08-1.59 (series of m, 8 H); 13C NMR (75 MHz, CDC13) ppm 137.34, 128.69, 75.21, 67.12, 63.10, 60.67, 46.54, 43.03, 39.60, 35.98, 35.29; MS m/z (M+ ) calcd for Cn H 18C10 198.0812, obsd 198.0822.

For 47b: IR (neat, cm"1) 3340 (br) 3040 (m), 2920 (s), 2850 (m), 1450 (m), 1260 (m), 1100 (m), 930 (w), 895 (w), 820 (w); XH NMR (300 MHz C6D6) 6 5.48 (s, 2 H), 4.29-4.21 H (m, H), 3.62-3.53 (m, 1 H), 2.57- 2.48 (dd, J = 13.3, 6.3 Hz, 1 H), 2.32-1.48 (series of m, 9 H), 1.40-1.30 (dd, J = 13.0, 6.2 Hz, 1 H); MS m/z (M+ ) calcd for C1:LH15C10 198.0812, obsd 198.0798. 107

For 48: XH NMR (300 MHZ, C6D6) 6 5.65-5.62 (m, 1 H), 4.25-4.24 (d, OH J = 3.6 Hz, 1 H), 3.56-3.47 (m, 1 H), 2.34-2.20 .(m, 1 H), 2.14-1.49

H (series of m, 11 H). (Assignment based on doublet pattern of CCl-H). MS m/z (M+-H) calcd for Cn HuC10 197.0733, obsd 197.0714.

Method B: Using Titanium Tetrachloride. A solution of 42 (30 mg, 0.185 mmol) in 2 mL of dry methylene chloride was cooled under argon to -78 °C, whereupon titanium tetrachloride (0.222 mmol, 0.38 M in methylene chloride) was added. A green color formed and a precipitate fell from solution. After 5 min, reaction was complete by TLC. Sodium bicarbonate solution (3 mL) was added and the mixture was warmed to room temperature, and extracted with methylene chloride. The combined extracts were dried over magnesium sulfate and the crude product was purified by MPLC to give 14 mg (38%) of 47a and 1 mg (2%) of 47b. 108 Method C. Using Dimethylaluminium Chloride. Diene aldehyde 42 (64 mg, 0.40 mmol) was dissolved in 10 mL of dry methylene chloride and cooled to -78 °C, whereupon dimethylaluminum chloride (0.790 mmol, 1 M in hexane) was added. After warming the reaction mixture during 1 h to 0 °C, the conversion appeared incomplete (TLC analysis) and another portion of catalyst (0.40 mmol) was added. After warming to room temperature, the reaction mixture was quenched with 10% sodium hydroxide solution and extracted with ether. The combined extracts were dried over magnesium sulfate and concentrated. Purification of the residue by MPLC (silica gel, 21% ethyl acetate in petroleum ether) provided 48 mg (56%) of chloro alcohol 47a.

Tricyclic Chloroketone 49. To a slurry of pyridinium chlorochrornate (94 mg, 0.44 mmol) and sodium acetate (7.5 mg, 0.093

• ...Cl•It mmol) in 3 mL of dry methylene H chloride under argon, was added 47 (51 mg, 0.26 mmol) in 3 mL of the same solvent. After 40 min the reaction mixture was diluted with ether and passed through a short column of Florisil. Separation by MPLC (silica gel, 12% ethyl acetate in petroleum ether) provided 44 mg (87%) of a 109 colorless oil; IR (neat, cm -1 ) 3040 (w), 2960 (m), 2925 (m), 2850 (m), 1730 (s), 1450 (w) , 1400 (w), 1350 (w), 1160 (m); XH NMR (300 MHz, C6D6) 5 5.69-5.60 (br s, 2 H), 4.58- 4.53 (m, H), 2.89-1.81 (series of m, 10 H); MS m/z (M+ ) calcd for C n H ^ C l O 196.0655, obsd 196.0630.

Tricyclic Enone 51. Chloro ketone 49 (43 mg, 0.22 mmol) was dissolved in 2 mL of dry methylene chloride. 1,8-Diaza- bicyclo[5.4.0]undec-7-ene (51 mg, 0.321 mmol) was added and the H reaction mixture was stirred overnight at room temperature under an argon atmosphere. The yellow-green mixture was then diluted with methylene chloride and passed through a short Florisil column. The concentrated eluate was purified by MPLC (silica gel, 13.5% ethyl acetate in petroleum ether) to give 31 mg (88%) of a colorless oil; IR (neat, cm-^) 3040 (m), 2920 (s), 2845 (m), 1730 (s), 1630 (m), 1445 (m), 1405 (m); NMR (300 MHz, C6D6) 6 6.42-6.40 (m, 1 H), 5.82-5.79 (m, 1 H), 5.71- 5.67 (m, 1 H), 3.13-3.12 (ddd, J = 18.4, 9.1, 3.7 Hz, 1 H), 2.89-2.80 (dd, J = 15.9, 9.0 Hz, 1 H), 2.64-2.43 (m, 4 H), 2.25-2.19 (complex d, J = 18.0 Hz, 1 H), 1.98-1.90 (m, 2 Hz); MS m/£ (M+) calcd for C1XH 120 160.0874, obsd 160.0881. 110

Tricyclic Alcohol 52. Ammonia (9 mL, previously

,OH dried over sodium) was distilled into a flask equipped with a dry ice condenser, argon inlet, and bubbler, and containing sodium H metal (89 mg, 3.85 mmol). Chloro alcohols 47a,b (176 mg, 0.883 mmol) and ethanol (14 mg, 0.3 mmol) in 1.2 mL of tetrahydrofuran were added dropwise to the above mixture at -78 °C. After 30 min, isoprene was added slowly until the blue color of the reaction mixture was dissipated. Saturated ammonium chloride solution (0.5 mL) was added and the ammonia was evaporated off. The residue extracted with ether and dried. The product was purified by MPLC (33% ethyl acetate in petroleum ether, silica gel) to provide 139 mg (96%) of a colorless oil; IR (neat, cm”1) 3600-3100 (br s), 3040 (m), 2930 (m), 2850 (m), 1605 (w), 1440 (m), 1350 (w), 1070 (w), 990 (w); XH NMR (300 MHz, C6D6) 6 5.60-5.57 (m, 1 H), 5.48-5.45 (m, 1 H), 3.97 (br s, 1 H), 2.57-2.48 (ddt, J = 16.7, 7.9, 2.2 Hz, 2 H), 2.16-1.15 (series of m, 11 H); MS m/z (M+) calcd for Cn H 160 164.1121, obsd 164.1193. Ill

Tricyclic hydrazone 53 Tricyclic ketone 17 (48 mg, N 0.23 mmol), hydrazine hydrate (282 g, 5.60 mmol), ethanol (0.5 mL), and triethylamine (327 mg, 3.23 mmol) were mixed together and H heated to gentle reflux for 1.5 h, after which time the reaction mixture was quenched by pouring it into 10 mL of water. The mixture was extracted with ether, and the combined extracts were dried and con centrated to give 50 mg (94%, crude) of a colorless oil. This was used without purification. IR (neat, cm ^) 3360- 3380 (br m), 3200 (br m), 3040 (m), 2930 (m), 2850 (m), 1735 (w), 1650 (m), 1615 (m), 1440 (m), 1420 (w); XH NMR (300 MHz, C6D6) 6 5.51-5.27 (m, 2 H), 4.60-4.30 (br s, 2 H), 2.65-1.11 (series of m, 12 H); MS m/z. (M+) calcd for

Ch H 16N 2 176.1314, obsd 176.1297

Tricyclic Vinyl Iodide 54. A solution of tetramethyl- guanidine (0.758 g, 6.59 mol) in 7.7 mL of dry ether was trans ferred over 15 min to a solution of iodine (236 mg, 0.935 mmol) in H 5.3 mL of dry ether. A dark pre- 112 cipitate formed and the mixture turned light red. After 10 min of stirring at room temperature, hydrazone 53 (77.3 mg, 0.439 mmol) was added slowly over 25 min. After 30 min, the mixture was diluted with ether and water. The organic layer was removed and washed with 2 N hydrochloric acid, and saturated sodium thiosulfite solution and 5% sodium bicarbonate solutions. The extracts were passed over a short neutral alumina column to give 100 mg of a colorless oil (84%) which was used without further purification; IR (neat, cm"1) 3040 (w), 2940 (s), 2860 (m), 1600 (w), 1445 (s), 1260 (m), 880 (m), 730 (m); XH NMR (300 MHz, C6D6) 6 6.03-6.01 (g, J = 2.2 Hz, 1 H), 5.62-5.58 (m, H), 5.55-5.52 (m, 1 H), 2.94-2.91 (m, H) , 2.71-2.61 (m, H), 2.37 (dt, J = 2.6, 16.5 Hz, 2 H), 1.75-1.65 (m, 3 H), 1.60-0.87 (series of m, 3 H); MS m/z (M+ ) calcd for Cn H 13I 272.0062, obsd 272.0053.

Tricyclic Enol Triflate 55a. To a solution of diisopropyl

OTf amine (0.10 g, 0.99 mmol) in 4 mL of dry dimethoxyethane under argon at 0 °C was added butyllithium (0.903 mmol, 1.55 M in hexanes). After being stirred for 10 min, 113 the mixture was cooled to -75 °C and 17 (133 mg, 0.821 mmol) in 2 mL of dimethoxyethane was added by canula over an 8 min period. The mixture remained colorless. After 50 min, N-phenyltrifluoromethanesulfonimide (302 mg, 0.903 mmol) in 2 mL of dimethoxyethane was added. The reaction was warmed to 0 °C and allowed to stir for 1 h. The sol vent was then removed by reduced pressure. The residue was dissolved in ether and passed over a short Florisil column. Separation by MPLC (silica gel, petroleum ether) provided 214 mg (89%) of an inseparable mixture of major and minor (9:1) regioisomers as a colorless oil . IR (neat, cm"''") 3045 (w), 2935 (m), 2845 (m), 1650 (m), 1419 (s), 1340 (w) , 1245 (s), 1210 (s), 1145 (s), 1123 (m), 1110 (m), 1070 (m), 1042 (m), 1000 (m), 940 (m), 875 (m), 800 (m), 735 (w); XH NMR (300 MHz, C6D6) 6 5.63-5.58 (m, 1 H), 5.57-5.53 (m, 2 H), 3.00-2.98 (m, 1 H), 2.73-2.64 (m, 1 H), 2.6 (dt, J = 17, 2.3 Hz, 1 H) 2.45-2.41 (m, 1 H), 2.39-2.32 (dt, J = 18.7, 3.0 Hz, 1 H), 2.04-1.97 (m, 1 H), 1.78-1.75 (m, 3 H), 1.53-1.48 (m, 1 H); 13C NMR (75 MHz, CDC13) ppm 148.98, 136.01, 130.21, 119.21 (g), 116.35, 54.47, 51.82, 40.44, 40.17, 32.42, 27.37, 65.51; MS m/z (M+ ) calcd for

C i2H13F 303S 294.0538, obsd 294.0529. 114 Methyl Vinyl Ketone 68. Diene enol ether 70 (120 mg) O was dissolved in acetic acid:water ch3 (1:1) and stirred for 2 h. The mixture was neutralized with sodium bicarbonate solution and H extracted with ether. Purifica tion by MPLC (20% ethyl acetate in petroleum ether) gave 47 _ 1 mg (56% for 2 steps) of a colorless oil; IR (neat, cm ) 3065 (m), 3000 (w), 2945 (s), 2940 (s), 2900 (m), 2860 (m), 2840 (m), 2830 (m), 1735 (m), 1660 (s), 1610 (m), 1460 (w), 1440 (m), 1425 (m), 1370 (s), 1340 (w), 1320 (w), 1285 (w),

1235 (S), 1090 (w); XH NMR (300 MHz, C6D6) 6 6.58 (dd, J = 9.3, 3.2 Hz, 1 H), 5.51 (m, 2 H), 3.05 (m, 1 H), 2.65-2.56 (m, 2 H), 2.34-2.26 (m, 1 H), 2.27 (s, 3 H), 2.02-1.95 (m, 2 H), 1.91-1.85 (m, 1 H), 1.77-1.73 (m, 1 H), 1.66-1.54 (m, 1 H), 1.38-1.30 (m, 1 H); 13C NMR (75 MHz, CDC13) ppm 196.56, 147.30, 142.49, 136.99, 128.53, 66.45, 55.63, 51.56, 45.80, 39.49, 33.49, 30.58, 26.87; MS m/jz calcd for C13H 160 188.1201, obsd 188.1203. Anal. Calcd for C 13H 160: C, 82.93; H, 8.56. Found: C, 82.72; H, 8.51. 115 (Ethoxyvinyl)trimethyltin 69 A solution of freshly dis tilled ethyl vinyl ether (94 mmol)

EtO Sn(CH3)3 in 100 mL of dry tetrahydrofuran T was cooled to -78 °C and tert- butyllithium (96 mmol, 1.5 M in hexanes) was added dropwise over a 25 min period. After 45 min of stirring at -78 °C, the reaction mixture was allowed to warm to -15 °C over 1 h. A bright yellow suspension formed. While at -15 °C, the mixture was transferred via a 16 gauge canula to a solution of trimethyltin chloride (17.8 g, 67.9 mmol) in 50 mL of tetrahydrofuran at -78 °C over 15 min. The mixture was warmed to 15 °C and quenched with saturated ammonium chloride solution. The product was extracted into ether and dried over potassium carbonate. Distillation (15 torr, 68-70 °C) provided 11.2 g (71%) of a colorless liquid; 1H NMR (300 MHz, C6D6) 5 4.67 (d, J = 1.7 Hz, 1 H), 4.08 (d, J = 1.8 Hz, 1 H), 3.71 (q, J = 7.0 Hz, 2 H), 1.27 (t, J = 7.0 Hz, 3 H), 0.19 (s with satellites, 9 H). MS m/z (M+-CH3) calcd for C6H13OSn 218.9983, obsd 218.9981. 116 Diene Enol Ether 70 Anhydrous lithium chloride (54 mg, 1.3 mmol) and tetrakis- OEt (triphenylphosphine)palladium (26 mg, 0.022 mmol) were transferred under argon to a 25 mL flask H equipped with a water condenser. A solution of (ethoxyvinyl)trimethyltin (109 mg, 0.463 mmol) and enol triflate 55a (132 mg, 0.449 mmol) in 8 mL of dry tetrahydrofuran (degassed at -78 °C under vacuum) was added. The slurry was heated to 70 °C and stirred for 12 h. The mixture was cooled to room temperature, and was diluted with pentane and poured into 10% ammonium hydroxide. The aqueous layer was removed and the organic extract was washed with 2 x 10 ml of ammonium hydroxide solution and passed over a short Florisil column to yield 117 mg of colorless oil which was used without further purification in the next step (for yield see experimental for 68); IR (neat, cm-1) 3040 (m), 2970 (m), 2920 (s), 1640 (m), 1570 (s), 1440 (m), 1370 (m), 1345 (m), 1260 (s), 1125 (m), 975 (m), 795 (s), 735 (m); XH NMR (300 MHz, C6D6) 5 6.12 (br s, 1 H), 5.53-5.50 (m, 1 H), 5.44-5.41 (m, 1 H), 4.23 (s, 1 H), 4.08 (s, 1 H), 3.51 (q, J = 7.0 Hz, 2 H), 117 3.06 (m, 1 H), 2.55-1.25 (series of m, 9 H), 1.11 (t, J = 7.0 Hz, 3 H); MS m/z (M+ ) calcd for C15H20O 218.1620, obsd 218.1586.

Vinyl Methyl Ketone 71a,b. A slurry of copper cyanide (31 mg, 0.35 mmol) was prepared ch3 under argon in 0.5 mL of dry ether and cooled to -78 °C. Vinyllithium (0.75 mmol) in 1 mL ether was H added. The mixture was warmed to 0 °C and then cooled again to -78 °C, whereupon 68 (40 mg, 0.21 mmol) in 0.75 mL ether was added dropwise over 2 min. The previously green mixture became deep yellow. After 1 h at -78 °C, the reaction mixture was quenched with saturated ammonium chloride/ammonium hydroxide solution (1:1) and extracted with ether. The extracts were poured over a short Florisil column and the product was purified by MPLC (silica gel, 6% ethyl acetate in petroleum ether) to yield 32 mg (75%) of a colorless oil, containing all four pos sible isomers as an inseparable mixture; IR (neat, cm ^) 3400 (w), 3080 (w), 3040 (w), 2430 (s), 2850 (s), 1700 (s), 1635 (w), 1440 (m), 1350 (m), 1270 (m), 990 (m), 915 (s); XH NMR (300 MHz, C6D6) 6 6.07-4.90 (series of m, 3 H), 3.50-1.12 (series of m, 14 H), four singlets at 2.12, 2.11, 118 2.10, 2.09 in a ratio of 2:2:4:1 corresponding to the 4 isomers (3 H); MS m/z (M+) calcd for C15H20O 216.1512, obsd 216.1490.

Tetracyclic Methyl Ketones 72a and 72b. Method A: By Methylation of the Enolate Generated from 86. A solution of 86 (1.43 g, 5.20 mol) was cooled to -78 °C and H CH. methyllithium (5.72 mmol, 1.5 M in ether,

H H low halide content) was 72a 72b added dropwise over 5 min. After 20 min, the reaction mixture was warmed to 0 °C. After an additional 20 min, the mixture was recooled to -78 °C, and methyl iodide (25 mmol, 3.5 g, dried by passage through basic alumina) was added in one portion. After being stirred for 10 min, the mixture was warmed to room temperature, quenched with saturated ammonium chloride solution, and extracted with petroleum ether. The combined extracts were poured over Florisil and purified by MPLC (silica gel, 2% ethyl acetate in petroleum ether) to give 496 mg (44%) of 72a as a white crystalline material of mp 40-42 °C and 326 mg (29%) of its stereoisomer 72b as a colorless oil. 119 For 72a; IR (neat, cm"1) 3055 (w), 2930 (s), 2860 (m), 1730 (s), 1460 (m), 1445 (m), 1405 (w), 1368 (w), 1072 (m), 1028 (m), 749 (w), 732 (w); *H NMR (300 MHz, C6D6) 6 5.40- 5.37 (br s, 2 H), 2.85-2.61 (m, 1 H), 2.54-2.45 (m, 1 H), 2.08-1.20 (m, 12 H), 1.13-1.05 (dd J = 11.3, 13.6 Hz, 1 H), 0.71 (s, 3 H); 13C NMR (75 MHz, C6D6) ppm 221.26, 138.61, 128.61, 68.70, 59.30, 54.78, 50.23, 47.37, 43.26, 39.71, 35.40, 32.85, 29.22, 21.69, 16.55; MS m/z (M+ ) calcd for C15H20O 216.1514, obsd 216.1507. Anal. Calcd for C 15H20O: C, 83.28; H, 9.32. Found: C, 83.05; H, 9.35.

For 72b; IR (neat, cm"1) 3025 (w), 2930 (s), 2845 (m), 1725 (m), 1445 (m), 1400 (w), 1135 (w), 1100 (w), 748 (w), 728 (w); XH NMR (300 MHz, C6D6) 6 5.60-5.57 (m, 1 H), 5.49- 5.46 (m, 1 H), 2.58-2.48 (m, 1 H), 2.23-1.19 (m, 14 H), 1.16 (s, 3 H); 13C NMR (75 MHz, C6D6) ppm 220.61, 138.85, 126.65, 70.64, 62.98, 57.96, 52.36, 50.26, 43.44, 38.80, 38.19, 36.48, 30.56, 24.69, 24.48; MS m/z (M+ ) calcd for C15H20O 216.1514, obsd 216.1504. 120 Method B: Reduction of Enone 79a. To a cold (-78 °C) solution of enone 79a (7 mg, 0.032 mmol) in 2 mL of dry methylene chloride under argon was added L-selectride (0.032 mmol, 1 M in tetrahydrofuran). After 20 min, the cooling bath was removed, the reaction mixture was quenched with saturated ammonium chloride solution and extracted with ether. The combined extracts were dried and finally poured over a short Florisil column. Purification by MPLC (silica gel, 10% ethyl acetate in petroleum ether) gave 6.8 mg (98%) of product as colorless oil.

Method C: Reductive Alkylation of Enone 76. To a three-necked round bottomed flask equipped with a dry ice condenser, argon inlet, and bubbler was placed low sodium lithium wire (20 mg, 2.90 mmol). Ammonia (15 mL, previously dried over sodium) was distilled into the reac tion vessel. After 15 min of stirring at -78 °C, a solu tion of 76 (204 mg, 1.02 mmol) and dry tert-butyl alcohol (38 mg, 0.51 mmol) in 5 mL of tetrahydrofuran was added over 25 min. After 40 min, 0.75 mL of methyl iodide (1.4 g, 12 mmol, previously passed through a short basic alumina column) was added over 1 min. After the first 4-5 drops, the blue color dissipated. A larger volume bubbler was attached, and the mixture was allowed to warm to room 121 temperature slowly. After 3 h, most of the ammonia had evaporated, and saturated ammonium chloride solution was added (2 mL). The reaction mixture was extracted with ether and the combined extracts were dried over magnesium sulfate. Concentration and chromatography as above gave 146 mg (66%) of a crystalline material.

Divinyl Ketone 75a. To a flask under argon atmos phere and containing anhydrous lithium chloride (324 mg, 9.00 mmol) and tetrakis(triphenylphosphine) palladium (144 mg, 0.12 mmol) was added a solution of 55a (610 mg, 2.07 mmol) and trimethylvinyl tin (603 mg, 2.2 mmol) in 21 mL of dry tetrahydrofuran by syringe. The system was purged with carbon monoxide for 15 minutes, before being sealed under a static pressure of carbon monoxide maintained by a Fisher gas bag. The reaction mixture was heated to 55 °C and maintained there for 48 h, after which time the mixture was diluted with pentane, and was washed with water and brine before being passed over a short Florisil column. Separation by MPLC (7.5% ethyl acetate in petroleum ether) provided 207 mg (52%) of a colorless oil identified as the desired product 75a and 48 122 mg (12%) of isomer 75b resulting from the enol triflate isomer 55b; IR (neat, cm-1) 3040 (m), 2925 (s), 2890 (s), 2840 (m), 2860 (m), 1655 (s), 1600 (s), 1405 (s), 1215 (s), 980 (s); *H NMR (300 MHz, C6D6) 5 6.61 (dd, J = 8.4, 17.0 Hz, 1 H), 6.27 (dd, J = 2.2, 17.1 Hz, 1 H), 6.11 (dd, J = 4.2, 2.7 Hz, 1 H), 5.43-5.33 (m, 2 H), 5.28 (dd, J = 2.1, 8.2 Hz, 1 H), 3.24 (m, 1 H), 2.52-0.81 (series of m, 9 H); 13C NMR (75 MHz, C6D6) ppm 187.02, 147.64, 141.70, 137.28, 133.08, 128.83, 126.69, 66.42, 56.47, 51.85, 46.14, 40.01, 33.73, 30.68; MS m/z (M+ ) calcd for C1„H160 200.1201, obsd 200.1196. Anal. Calcd for C1(tH160: C, 83.96; H, 8.05. Found: C, 83.53; H, 7.94.

Divinyl Ketone 75b. XH NMR (300 MHz, C6D6) 6 6.61 O (dd, J = 17.0, 10.3 Hz, 1 H), 6.33 (dd, J = 7.1, 2.2 Hz, 1 H), 5.5- 5.4 (complex d, 2 H), 5.27 (dd, J

= 10.3, 2.6 Hz, 1 H), 3.01 (dd, J H = 8.7, 15.6 Hz, 1 H), 2.86-1.38 (series of m, 10 H); MS m/z, (M+ ) calcd for ClltH160 200.1201, obsd 200.1182. 123 Keto Formate 77. A mixture of phosphoric and formic acids (0.6 mL, 1:1) was H ,L added to 75a (16 mg, 0.08 mmol). After 1.5 h of stirring at room temperature, the mixture was neu- H tralized with sodium bicarbonate solution and extracted with ether. The extract was dried over magnesium sulfate; purification by MPLC (silica gel, 20% ethyl acetate in petroleum ether) yielded 13.5 mg (84%) of a colorless oil; XH NMR (300 MHz, C6D6) 6 7.52 (s, 1 H), 5.85 (br s, 1 H), 5.44-5.34 (m, 2 H), 4.31 (t, J = 6.2 Hz, 1 H), 3.13-3.12 (m, 1 H), 2.45 (dd, J = 16.5, 2.2 Hz, 2 H), 2.55-1.55 (series of m, 10 H); 13C NMR (75 MHz, C6D6) ppm 193.91, 160.30, 146.99, 141.65, 137.24, 128.83, 66.50, 59.32, 56.15, 51.88, 45.93, 39.86, 37.69, 33.73, 30.84; MS m/z. (M+ ) calcd for C15H 180 246.1256, obsd 246.1272.

Trimethylsilyl Divinyl Ketone 78a. To a flask containing anhydrous lithium chloride (360 mg, 8.78 mmol) and tetrakis(triphenylphosphine) palladium under argon was added a solution of (trimethylsilyl)vinyl trimethyl H 124 tin (878 mg, 2.53 mmol) and 55a (706 mg, 2.41 mmol) in 25 mL of dry tetrahydrofuran. The system was purged with carbon monoxide for 15 min and maintained under a static pressure of 1 atm while being heated to 70 °C. After 48 h, the mixture was diluted with pentane and washed with water and brine prior to being passed over a short Florisil column. Separation by MPLC (silica gel, 1.5% ethyl acetate in petroleum ether) provided 522 mg (80%) of the desired product as a colorless oil and 54 mg (8%) of the isomer resulting from the enol triflate isomer 55b; IR (neat, cm-1) 3050 (m), 2960 (s), 2900 (s), 2865 (m), 2845 (m), 1640 (s), 1600 (m), 1420 (m), 1335 (m), 1245 (s), 1225 (s), 990 (s); XH NMR (300 MHz, C6D6) 5 7.43 (d, J = 18.6 Hz, 1 H), 7.06 (d, J = 18.3 Hz, 1 H), 6.24 (dd, J = 4.3, 2.5 Hz, 1 H), 5.46-5.37 (m, 2 H), 3.37-3.34 (m, 1 H), 2.55-2.45 (ddt, J = 18.8, 16.8, 2.1 Hz, 1 H), 2.39-2.29 (m, 1 H), 2.23-1.55 (series of m, 6 H), 1.32-1.23 (m, 1 H), 0.14 (s, 9 H); 13C NMR (75 MHz, C6D6) ppm 186.08, 147.77, 145.01, 141.46, 139.01, 137.36, 128.85, 66.42, 56.70, 51.90, 46.2, 40.07, 33.81, 30.74, -1.72; MS m/z (M+ ) calcd for C17H24OSi 272.1597, obsd 272.1576. Anal. Calcd for C 17H 21

H (m), 995 (ra), 840 (s), 748 (m), 725 (m); XH NMR (300 MHz, C6D6) 6 7.39 (d, J = 18.6 Hz, 1 H), 7.03 (d, J = 18.6 Hz, 1 H), 5.46 (s, 2 H), 3.18-1.42 (series of m, 11 H), 0.02 (s, 9 H); 13C NMR (75 MHz, C6D6) ppm 187.05, 167.40, 144.63, 141.47, 134.17, 132.92, 130.10, 76.83, 43.42, 41.94, 37.47, 36.97, 35.50, 27.06, -1.72; MS m/z (M+ ) calcd for C17H21

Tetracyclxc Enones 79a and 79b. Experiment A: Generation of 79a and 79b. A solution of 78a (32 mg, 0.118 mmol) in dry methy lene chloride (2 mL) was prepared under argon and cooled to -17 °C. Ferric chloride was added (21 mg,

79a 79b 0.13 mmol) and the reaction mixture was warmed to 0 °C 126 and stirred there for 14 h, whereupon it was poured into an equal volume of water and extracted with ether. The com bined extracts were washed with water, sodium bicarbonate solution, and brine prior to drying over magnesium sulfate. Separation by MPLC (silica gel, 17% ethyl acetate in petro leum ether) gave 17 mg (72%) of a colorless, UV active oil, consisting of 79a and 79b as an inseparable mixture in the ratio of 1.6:1 by NMR (300 MHz).

For 79a; IR (neat, cm-'*') 3400 (br w), 3060 (m), 2960 (s), 2880 (m), 1710 (s), 1590 (m), 1455 (m), 1355 (m); XH NMR (300 MHz, C6D6) 6 6.76 (dd, J = 2.6, 6.0 Hz, 1 H), 5.80 (dd, J = 2.2, 5.8 Hz, 1 H), 5.45-5.42 (m, 1 H), 5.18-5.12 (m, 1 H), 3.76-3.70 (m, 1 H), 2.82-0.91 (series of m, 10 H); MS m/z (M+ ) calcd for ClitHlsO 200.1201, obsd 200.1225.

For 79b; IR (neat, cm-'*') 3040 (m), 3050 (s), 3020 (s), 1725 (s), 1680 (s), 1580 (m), 1460 (m), 1445 (m), 1345 (m), 1200 (m), 730 (m). XH NMR (300 MHz, C6D6) 6 6.88-6.82 (dd, J = 5.6, 5.6 Hz, 1 H), 5.78 (dd, J = 5.5, 1.6 Hz, 1 H), 5.36-5.16 (m, 2 H), 2.78-2.70 (m, 1 H), 2.57-2.44 (ddt, J = 17.2, 9.2, 2.2 Hz, 1 H), 2.32 (1, J = 4.7 Hz, 1 H), 2.17 (dd, J = 6.6, 4.4 Hz, 1 H), 2.19-1.43 (m, 6 H), 1.24 (dd, J = 13.1, 7.1 Hz, 1 H), 1.20-1.11 (m, 1 H); MS m/z (M+ ) calcd for ClltH160 200.1201, obsd 200.1201. 127 Experiment B: Generation of 82. A solution of TMS dienone 78a (0.48 g, 1.80 mmol) was treated with ferric chloride (0.300 g, 1.84 mmol) as above in 27 mL of methyl ene chloride to give after 14 h at 0 °C and the same workup and separation as above, 50 mg (10%) of starting material, 36 mg (10%) of 79b, and 39 mg of 82 (67%). The latter product was an inseparable mixture of isomers.

For 82: IR (neat, cm"^) 3040 (w), 2930 (s), 2860 (m), 1730 (s), 1690 (s), 1585 (m), 1445 (m), 1240 (m); 1H NMR (300 MHz, C6D6) 6 6.87-6.75 (2 dd, J = 3, 5.5 Hz; J = 3, 5.3 Hz; 1 H); 5.80-5.76 (m, 1 H), 5.48-5.19 (series of m, 2 H), 3.90-3.88 (m, 1 H), 2.74-1.03 (series of m, 27 H); MS m/z (M+ ) calcd for C28H 320 2 400.2402, obsd 400.2358. 128 Experiment C: Generation of 76. Three portions of the mixture of 76a and 76b 0 (1.5:1) (each 910 mg, 0.045 mmol) were each dissolved in 1 mL of dry toluene and

H placed in sealed tubes. Catalytic sulfonic acid (the first two with p-toluenesulfonic acid and the third with (3- naphthalenesulfonic acid) was added and the mixture was heated to 150 °C for 24 h. All three exhibited approxi mately the same relative amounts of products by TLC analy sis. The mixtures were poured over short Florisil columns and combined. Separation by MPLC (silica gel, 20% ethyl acetate in petroleum ether) yielded 16 mg (59%) of 79 as an oil and 9 mg (22%) of 76 as a crystalline material; mp 68-69 °C; IR (neat, cm-1) 3040 (w), 3930 (s), 2900 (m), 2840 (m), 1680 (s), 1625 (s), 1380 (m), 1225 (w), 740 (w), 725 (w); XH NMR (300 MHz, CDC13) 6 5.60-5.57 (m, 1 H), 5.48-5.45 (m, 1 H), 3.97 (br s, 1 H), 2.57-2.48 (ddt, J = 16.7, 7.9, 2.2 Hz, 1 H), 2.16-1.15 (series of m, 12 H); 13C NMR (75 MHz, C6D6) ppm 201.23, 182.11, 149.45, 136.88, 129.24, 73.49, 51.58, 50.91, 44.61, 40.38, 33.38, 28.20, 25.29; MS m/z (M+ ) calcd for C14H 160 200.1201, obsd 200.1208. 129 Anal. Calcd for Cll4H 160: C, 83.96; H, 8.05. Found: C, 83.83; H, 8.10.

Tetracyclic enone 76. Preparation from 86a,b. Silyl enol ethers 86afb (220 mg, 0.803 mmol) were dissolved in O 4 mL of dry acetonitrile. 2,6-Di- tert-butylpyridine (170 mg, 0.892 mmol) was next added, followed by H palladium acetate (222 mg, 0.993 mmol) in one portion. The solution soon turned brown, then black. After 80 min, the solvent was removed in vacuo and the black residue was washed with methylene chloride. The extract was filtered through Celite and the filtrate was concentrated and chromatographed over preparative grade silica gel in 5, 10 and 20% ethyl acetate in petroleum ether to give 125 mg (78%) of 76. 130 Trimethylsilylvinyl Trimethyltin 80. (A more facile and economical procedure than that reported by

n a Sn(CH3)3 seyferth) To a three-necked 50- mL flask under argon, was placed (H3C)3Si/ 20 mL of dry ether, followed by trans-(trimethylsilyl)-vinyl bromide (2.68 g, 0.0188 mol). The mixture was cooled to - 78 °C and tert-butyllithium (0.0376 mol, 1.5 M in hexane) was added over 15 min. The resulting clear yellow mixture was stirred at -78 °C over 1 h, allowed to warm to -15 °C over 40 min, and then was cooled to -50 °C. The reaction mixture was added dropwise to a well stirred solution of trimethyltin chloride (2.30 g, 0.0157 mol) in 12 mL of ether at -60 °C. The reaction was warmed over 1.25 h to 15 °C and quenched with saturated ammonium chloride solution. The product was extracted with ether, washed with water, and dried over potassium carbonate at 5 °C. Distillation (59-63 °C, 8 torr) gave 3.5 g (85%) of material spectro- 72 scopically identical to that reported in the literature. 131

Methyl Enone 82 To a solution of diisopropyl- araine (19 mg, 0.19 mmol) in 1 ml of tetrahydrofuran was added n- butyllithium (0.18 mol, 1.5 M in hexane) at 0 °C. After 5 min, the H reaction mixture was cooled to -78 °C and enone 79a (29 mg, 0.15 mol) was added in 1 mL of tetrahydrofuran over 20 min. After 1 h, methyl iodide (123 mg, 0.87 mmol) was added. The reaction mixture was allowed to warm and stir at room temperature before being quenched with saturated ammonium chloride solution. The solution was extracted with ether and dried over MgSo4. Purifica tion of the residue by MPLC gave 7.8 mg (25%) of a color less oil; IR (neat, cm ^) 3060 (w), 3045 (s), 2870 (m), 1735 (m), 1700 (m), 1595 (w), 1460 (m), 1380 (w); 'H NMR (300 MHz, C6D6) 5 6.80 (dd, J = 5.5, 4 Hz, 1 H) 5.89 (dd, J = 1.7, 5.7 Hz, 1 H), 5.51-5.25 (m, 2 H), 2.55-1.14 (series of m, 11 H), 1.03 (s, 3 H); MS m/z (M+ ) calcd for C15HiaO 214.1357, obsd 214.1357. 132

Tricyclic Cyclopropyl Oxirane 84. To a solution of 17 (1.5 g 9.3 mmol) in dimethyl sulfoxide o (dry, 20 ml) was added Trost reagent (2.93 g, 9.33 mmol) and powdered potassium hydroxide (1.13 H g, 20.2 mmol). The mixture turned orange-brown. After 8 h, hexane was added and the layers were separated. The hexane layer was washed with 5% sodium bicarbonate solution and dried over sodium sulfate. The resulting binary mixture of 84 and diphenyl sulfide is carried on into the next step. For characterization, the oxirane can be removed selectively by distillation via Kugelrohr at 40 °C and 0.01 torr. For yield see next experimental; IR (neat, cm ^) 3035 (m), 2905-2930 (s), 1435 (m), 1400 (w); XH NMR (300 MHz, C6D6) 6 5.52-5.38 (m, 2 H), 2.60-2.50 (ddt, J = 17, 9.0, 2.2 Hz, 1 H), 2.24-1.37 (m, 11 H), 0.89 (br s, 2 H), 0.66-0.55 (m, 2 H); MS m/z (M+ ) calcd for ClltH180 202.1357, obsd 202.1364. 133

Tricyclic Siloxyvinylcyclopropane 85. The product mixture from the TMSO above reaction was dissolved under argon in 20 mL of dry hexane and diethylamine (1.22 g, 16.6 mmol).

H The solution was cooled to -75 °C and n-butyllithium (10.5 ml, 1.58 M in hexane, 16.6 mmol) was added. After 2-4 min, a white precipitate formed. After an additional 5-6 min, the ice bath was removed. The reaction mixture was warmed to room temperature and stirred for 15-20 min before freshly dis tilled trimethylsilyl chloride (2.74 g, 25.2 mmol) was added. The reaction mixture became clear and dimethoxy ethane (4 mL) was added. After 10 min, the mixture was diluted with 100 mL of dry hexane and the suspension was stored overnight at 5 °C before being filtered over 10 g of Florisil. Concentration and purification by MPLC (silica gel, 15% dichloromethane in petroleum ether) gave 1.47 g (58% for two steps; yields range 48-72%) of a clear oil; IR (neat, cm"''") 3085 (w) , 3040 (w), 3000 (w), 2955 (w), 2930 (m), 2860 (w), 2840 (m), 1443 (m), 1410 (m), 1350 (w), 1260 (m), 1250 (s), 1230 (s), 1228 (s), 1008 (m), 905 (m), 880 (m), 843 (s); XH NMR (300 MHz, C6D6) 6 5.62-5.59 (m, 1 H), 5.50-5.46 (m, 1 H), 5.23 (q, J = 2.1 Hz, 1 H), 3.05-3.01 (m, 1 H), 2.62-2.53 (qt, J = 8.9, 2.3 Hz, 1 H), 2.54-2.47 134 (dt, J = 16.8, 2.0 Hz, 1 H), 2.32-2.25 (dt, J = 17.1, 2.5 Hz, 1 H), 2.26-2.20 (m, 1 H), 1.97-1.70 (m, 4 H), 1.37-1.28 (m, 1 H), 1.02-0.89 (m, 1 H), 0.89-0.76 (m, 1 H), 0.59-0.49 (m, 9 H); 13C NMR (75 MHz, C6D6) ppm 147.88, 138.40, 122.45, 128.00, 67.84, 58.12, 55.36, 51.86, 45.23, 40.24, 34.22, 30.48, 15.54, 11.99, 1.29; MS m/z (M+ ) calcd for C17H26OSi 274.1753, obsd 274.1788. Anal. Calcd for C 17H26OSi: C, 74.39; H, 9.55. Found: C, 74.35; H, 9.55.

Tetracyclic Silyl Enol Ether 86. The siloxy vinyl cyclo pro pane product from above (1.47 g, OTMS 5.36 mmol) was dissolved in dry hexane (5 mL), and placed in a 10- mL addition funnel (alternatively, a syringe pump may be used). Meanwhile, a 1.5 cm x 40 cm pyrex glass column (fitted with a 24/40 male joint at the top, and 24/40 female joint at the bottom) was packed with pyrex glass helices for 20 cm of its height. The column was pretreated by washing it with 5% sodium bicarbonate solution, distilled water, acetone, hexane, and trimethylsilyl chloride. The column was dried under reduced pressure and washed with triethylamine and finally hexane. The column (set in a vertical 135 position) was heated to 430 °C and argon was passed through it at a rate of 15 mL/min. The addition funnel was attached at the top, and a pre-tared 250-mL flask (three necked) , equipped with septa and bubbler was attached at the bottom. The flask was cooled in dry ice and acetone, and the solution of substrate was added dropwise into the column at the rate of 1-2 drops/min. After the addition was complete, the flask was removed and the yellow mixture was concentrated to give 1.43 g (97%) of a yellow oil. This product was not isolated, but carried directly into the next reaction (see experimental for 72a).

Silyl Enol Ether 91. To a cold (-78 °C) solution of methyl ketone 72a (23 mg, 0.105 OSi(CH3)3 mmol) in 2 mL of dry tetrahydrofuran under an argon atmosphere was added sodium hexamethyldisilazide (1 M in tetrahydrofuran, 0.15 mmol). After 1 hr of stirring, trimethylsilyl chloride (34 mg, 0.31 mmol) wasadded. The reaction mixture was allowed to warm slowly toroom temperature after which time sodium sulfate decahydrate (0.13 g) was added. The mixture was diluted with hexane and passed over a short column of silica gel. Concentration gave 27.4 mg (91%) of a 136 colorless oil. IR (neat, cm 3035 (w), 2920 (w), 2840 (m), 1628 (m), 1438 (w), 1365 (w), 1338 (w), 1325 (w), 1250 (w), 1209 (w), 1130 (m), 900 (m), 870 (m), 845 (s); XH NMR (300 MHz, C6D6) 5 5.70-5.67 (m, 1 H), 5.49-5.45 (m, 1 H), 4.36 (m, 1 H), 2.67-2.56 (ddt, J = 2.4, 9.1, 16.9 Hz, 1 H), 2.49 (d, J = 6.5 Hz, 1 H), 2.46-2.43 (m, 1 H), 2.24-2.15 (m, 1 H), 2.11-1.20 (m, 10 H), 1.14 (s, 3 H), 0.17 (s, 9 H) ; MS

Acyloin 92. A solution of diisopropyl amine (58 mg, 0.578 mmol) in 2 mL of dry tetrahydrofuran was cooled to -78 °C under argon and n-butyllithium was added (0.6 mmol, 1.55 M in hexanes). The mixture was warmed to -20 °C, stirred for 10 min, and then cooled again to -78 °C. A solution of ketone 72a (52 mg, 0.239 mmol) in 2 mL of tetrahydrofuran was added to the base over a 5-min period. After 30 min, this was transferred by canula to a stirred suspension of MoOPh (420 mg, 0.970 mol) in 2 mL of tetrahydrofuran at -44 °C over 5 min. After 15 min the suspension had disappeared and a dark orange color materialized. Saturated sodium sulfite solution was added and the mixture was allowed to warm gradually to room 137 temperature. The product was extracted into ether and the extract was washed with 10% hydrochloric acid, sodium bicarbonate solution, and brine. The extracts were dried over magnesium sulfate and purification was achieved by silica gel chromatography (20% ethyl acetate in petroleum ether) to give 39 mg (71%) of a colorless oil and 4 mg (8%) of starting material; IR (neat, cm ■*") 3440 (m), 2935 (s), 2860 (m), 1735 (s), 1442 (m), 1295 (m), 1200 (m), 988 (m), 735 (m); XH NMR (300 MHz, C6D6) 6 (epimeric mixture) 5.47- 5.21 (m, 2 H), 5.01-3.95 (8 line pattern + dd, J = 1.5, 8.5 Hz, 1 H), 2.63-1.05 (series of m, 14 H), 0.84, 0.68 (s, 3 H); MS m/z_ (M+ ) calcd for C15H20O2 232.1463, obsd 232.1453. Anal. Calc for C15H 20S2: C, 77.55; H, 8.68. Found: C, 77.45; H, 8.69.m/z (M+ ) calcd for C18H29OSi 288.1909, obsd 288.1884.

Methyl Lactone 96. A solution of 15 (6 mg, 0.023 H3C mmol) in dry methylene chloride (1 mL) and dimethylaluminum chloride (0.06 mmol, 1 M hexane) was sub jected to 98,000 psi at room temperature for 9 h. The mixture was diluted with methylene chloride and washed with 1 N sodium hydroxide solution. The extract was dried over 138 magnesium sulfate and the product was separated on silica gel (10% ethyl acetate in petroleum ether) to give 3.6 mg (57%) of a colorless oil. By TLC, it appeared that the other methyl isomer was present although in lesser amounts (10-20%); IR (neat, cm-1) 3040 (w), 3030 (s), 2850 (m), 1720 (s), 1443 (m), 1380 (m), 1260 (m), 1200 (m), (1108 (m), 640 (m); NMR (300 MHz, C6D6) 5 5.63-5.60 (m, 1 H), 5.38-5.36 (m, 1 H), 4.11-3.85 (m, 1 H), 1.92-0.84 (series of m, 16 H), 1.02-0.96 (s, 3 H), MS m/z (M+) calcd for C16H2202 246.1620, obsd 246.1594. Minor Isomer: MS m/z^ (M+ ) calcd for C16H2202 246.1620, obsd 246.1595.

Keto Aldehyde 97. Ketal 104 (6 mg, 0.02 mmol) O was dissolved in acetone (2 mL) and 0.3 mL 5% hydrochloric acid solution was added. After 12 h, sodium bicarbonate solution was H introduced, and all volatile materials were removed in vacuo. The residue was extracted with ether and the extract was dried over magnesium sul fate. Chromatography over silica gel (20% ethyl acetate in petroleum ether) gave 5.5 mg (100%) of product as a color less oil; IR (neat, cm-'*') 2960 (s), 2850 (s), 2720 (m), 139 1730-1710 (s), 1460 (m), 1430 (m), 1400 (m) , 1380 (m), 1220 (s); XH NMR (300 MHz, C6D6) 6 9.3 (m, 1 H) , 3.21 (s, 3 H) , 2.32 (d, J = 14.5 Hz, 1 H), 2.28 (d, J = 14.5 Hz, 1 H), 2.09-0.74 (series of m, 13 H), 0.92 (s, 3 H); MS m/z (M+ )

calcd for C16H22° i* 278.1518, obsd 278.1515. Anal. Calcd for C16H2 20l(: C, 69.04; H, 7.97. Found; C, 68.55; H, 8.03.

Tricyclic Acetal Ester 99. Aldehyde 15 (5 mg, 0.019 mmol) and g-toluenesulfonic acid OCH (0.5 mg, 0.003 mmol) were dis f ' " C H solved in 0.5 mL of trimethyl orthoformate and the solution was H stirred for 2 h, at which time sodium bicarbonate solution (3 drops) was added. The mixture was extracted with ether and the extract was dried over magnesium sulfate. Concentration yielded 6 mg (100%) -1 of a colorlessoil which was pure by NMR; IR (neat, cm ) 3040 (w), 2940 (s), 2850 (m), 1722 (s), 1460 (m), 1442 (m), 1211 (m), 1123 (s), 1058 (m); lH NMR (300 MHz, C6D6) 6 6.12-6.10 (m, 1 H), 5.49-5.47 (m, 1 H), 4.41-4.37 (dd, J = 7.4, 4.1 Hz, 1 H), 3.29 (s, 3 H), 3.14 (s, 3 H), 3.14 (s, 3 H), 2.50-2.43 (m, 2 H), 2.15-1.68 (m, 9 H), 1.58-1.49 (dt, 140 >1 = 4, 14 Hz, 2 H), 1.21 (s, 3 H) : MS m/z (M+-OCH3) calcd for C17H250 3 277.1803, obsd 277.1801. Tetracyclic Diol 100. A solution of 72a (38 mg, 0.18 mmol) in dry benzene was OH

CH transferred to a dry flask and the solvent was removed under reduced pressure. A magnetic stir bar, H water condenser, and argon inlet were attached; 1 mL of tetrahydrofuran and 9-BBN (65 mg, 0.53 mmol) were added. The mixture was heated to reflux for 4.5 h. The reaction mixture was cooled to 0 °C and 0.5 mL each of ethanol, 6 N sodium hydroxide, and 30% hydrogen peroxide were added. After overnight stirring at room temperature, the reaction mixture was extracted with methy lene chloride and the extracts were passed through a mag nesium sulfate/Florisil column. Purification by MPLC (silica gel, 70% ethyl acetate in petroleum ether) gave 40 mg (97%) of a colorless oil; IR (neat, cm ) 3370 (br), 2940 (s), 2870 (m), 1560 (m), 1270 (m), 1075 (m), 995 (w), 745 (w); NMR (300 MHz, CDC13) 5 4.31 (m, 1 H), 3.98 + 3.73 (2 t from isomers, J = 8.2 Hz, 5.5 Hz, 1 H), 2.26-1.20 (series of m, 19 H), 0.85, 0.90 (s, 3 H; signals from 2 isomers); MS m/z (M+ ) calcd for C15H2„02 236.1777, obsd 236.1774. 141

Diketone 101. To a solution of dimethyl sulfoxide (53 mg, 0.68 mmol) in 3 mL of dry methylene chloride under argon at -78 °C was added over 10 O min trifluoroacetic anhydride (107 H mg, 0.51 mmol). After 10, min a solution of diol 100 (40 mg, 0.17 mmol) in 3 mL of methylene chloride was added dropwise over 10 min. The reaction mixture was warmed to room temperature over 30 min and stirred at this temperature for 0.5 h, after which time triethylamine (0.2 mL, dry) was added. The mixture was diluted with methylene chloride, washed with water, and dried over magnesium sulfate. Purification by MPLC (silica gel, 25% ethyl acetate in petroleum ether) gave 28 mg (72%) _i of a crystalline product, mp 79-81 °C; IR (KBr, cm ) 2945 (m), 2910 (m), 2880 (w), 2855 (w), 1745 (s), 1730 (s), 1475 (w), 1405 (w), 1180 (w), 1095 (w), 1085 (w), 1040 (w); XH NMR (300 MHz, C6D5) 6 2.41 (t, J = 7.1 Hz, 1 H), 2.20-2.02 (m, 2 H), 1.97 (dd, J = 2.8, 10.0 Hz, 1 H), 1.92-0.64 (series of m, 14 H), 0.618 (s, 3 H); MS m/z^ (M+ ) calcd for C15H210 2 233.1542, obsd 233.1497. Anal. Calcd for C15H2102: C, 77.55; H, 8.68. Found;

C, 77.29; H, 8.68. 142

Ketal Ketone 102. Diketone 101 (54 mg, 0.23 mmol) was dissolved in 0.3 mL of dry methylene chloride. Bis(tri- ""CH methylsilyl) ethylene glycol (49.4 mg, 0.240 mmol) was added and the H solution was cooled under argon to -78 °C. Trimethylsilyl trifluoromethanesulfonate (0.007 mmol, 0.02 M in methylene chloride) was added and the mix ture was stirred for 6 h, after which time dry pyridine (8 mg, 0.01 mmol) was added. The mixture was diluted with methylene chloride and washed with brine. The organic layer was dried over magnesium sulfate. MPLC (silica gel, 10% ethyl acetate in petroleum ether) gave 54 mg (84%) of a semi-crystalline material, mp 71-72 °C (yields on larger scale ranged up to 96%); IR (neat, cm ^) 2940 (s), 2860 (m), 1730 (s), 1450 (s), 1428 (m), 1308 (m), 1105 (m), 1030 (m); XH NMR (300 MHz, C6D6) 5 3.48 (s, 4 H), 3.46 (m, 1 H), 2.16-0.70 (series of m, 16 H), 0.71 (s, 3 H); MS m/z (M+ ) calcd for C17H 2„03 276.1361, obsd 276.1364. 143

Acyloin Ketal 103 To a flask containing 2 mL of OH dry tetrahydrofuran was added n- butyllithium (0.165 mmol, 1.5 M ••"CH; hexane) followed by diisopropyl amine (0.17 mmol). The mixture H was warmed to -20 °C and allowed to stir for 15 min before being cooled to -78 °C, at which time 102 (18 mg, 0.064 mmol) was added in tetrahydrofuran (1.5 mL) over 5 min. After 0.5 h of stirring, the clear enolate solution was added to a stirred suspension of MoOPh (120 mg, 0.271 mmol) in the same solvent (2 mL) at -44 °C over a 3-min period. Shortly thereafter the suspension dissipated and the solution became yellow-green in color. The mixture was warmed over 15 min to -30 °C, quenched with a saturated solution of sodium sulfite, warmed to room temperature, and stirred there for 0.5 h, after which the product was extracted with ether. The extracts were washed with 1.5 M hydrochloric acid solution, sodium bicarbonate solution, and brine before being dried over magnesium sulfate. Concentration and purification on silica gel (50% ethyl acetate in petroleum ether) gave 14 mg (74%) of a crystalline material, mp 85-90 °C (yields ranged 74-88%); IR (neat, cm”1) 3440 (br), 2950 (s), 1730 (s), 1460 (m), 1445 (m), 1425 (m); XH NMR (300 MHz, C6D6) 5 4.05-3.91 (m, 144 2 H), 3.50-3.37 (m, 4 H), 2.73-2.69 (t, J = 7 Hz, 1 H), 2.39-2.35 (t, J = 6.5 Hz, 1 H), 2.16-1.12 (series of m, 13 H), 0.84-0.71 (2 s, 3 H); MS m/z (M+ ) calcd for C17H2„0„ 292.1674, obsd 292.1657. Anal. Calcd for C 17H 2„Ou: C, 69.84; H, 8.27. Found: C, 69.91; H, 8.35.

Aldehyde Ester 104. To a dry solution of 103 (6 o mg, 0.02 mmol) in benzene/methanol OCH; (2:1) at 5 °C was added in one ""CH; portion lead tetraacetate (9.5 mg, 0.021 mmol). After 3.5 min, the H reaction mixture was quenched with sodium bicarbonate solution. The resulting suspension was filtered through Celite and the filtrate was extracted with benzene. The extract was dried over magnesium sulfate and concentrated to give 62 mg (91%, yields were higher on larger scale) of a colorless oil which appeared pure by NMR; IR (neat, cm”1) 2930 (s), 2860 (m), 1720 (s), 1380 (w), 1330 (m), 1205 (m), 1110 (m); 1H NMR (300 MHz, C6D6) 6 3.51 (s, 4 H), 3.24 (s, 3 H), 3.12-3.04 (m, 1 H, 2.95-2.87 (m, 1 H), 2.46-2.41 (t, J = 7.6 Hz, 1 H), 2.31-1.08 (series of m, 14 H), 1.07 (s, 3 H); MS m/z (M+-H) calcd for

CiaH26°5 322.1780, obsd 322.1751. 145

Bromo Ketone 105. Ketal mesylate 109 (190 mg, Br 0.47 mmol, crude) was refluxed in

OCH3 8 mL of dry acetone with lithium 'CH; 3 bromide (170 mg, 2.1 mmol) for 5 O h. The mixture was concentrated H and the residue was extracted with ether and dried over magnesium sulfate. Purification by silica gel chromatography (5, 10, 20% ethyl acetate in petroleum ether) yielded 130 mg (85%)of a crystalline product mp 79-80 °C. The product typically contained 5-8% of the corresponding chloride, as assayed by capillary gas chromatography. (If the sequence is done in the usual manner, i.e., mesylation in methylene chloride, solvent removal and addition of bromination reagents, up to 30% of the chloride is obtained); IR (neat, cm-^) 2940 (s), 2850 (s), 1720 (s), 1400 (m), 1395 (m), 1380 (m); 1H NMR (300 MHz, C6D6) 6 3.19 (S, 3 H), 3.09-3.02 (m, 1 H, 2.90-2.84 (m, 1 H), 2.36-2.34 (m, 1 H), 2.18-0.74 (series of m, 14 H), 1.00 (s, 3 H); MS m/z (M+ ) calcd for C 16H23Br03 342.0830, obsd 342.0817. 146

Alcohol 106. Aldehyde 15 (8.8 rag, 0.032 mmol) was dissolved in dry metha nol (1 mL) and cooled to 0 °C. Sodium borohydride (2.0 mg, 1.1 eq) was added and the mixture was stirred for 5 min. Brine was added and all volatile materials were removed in vacuo. The residue was extracted with methylene chloride and the extract was dried over magnesium sulfate. Removal of the solvent gave 9 mg (100%) of a colorless, viscous oil. Lactonization was not observed during the reaction, workup or storage, but did occur during chromatography on silica gel. IR (neat, cm"1) 3340 (s), 2920 (s), 2860 (m), 1715 (s), 1440 (m), 1210 (m), 1030 (m); XH NMR (300 MHz, C6D6) 6 6.14-6.11 (m, 1 H), 5.50-5.47 (m, 1 H), 3.41-3.25 (m, 2 H), 3.31 (s, 3 H), 2.54-2.43 (m, 2 H), 2.05-1.07 (m, 12 H), 1.17 (s, 3 H); MS m/z (M+) calcd for C16H 2„06 264.1725, obsd 264.1705. 147

Bromo Ester 107. To a solution of carbon Br tetrabromide (28 mg, 0.061 mmol)

o c h 3 in methylene chloride (2 mL) was added triphenylphosphine (25 mg, 0.085 mmol) at 0 °C. After 5 min H a reddish color appeared. Dry pyridine (5 mg, 0.067 mmol) was added as a buffer to inhibit lactonization. After 5 min, alcohol 106 (16 mg, 0.06 mmol) was added in 0.5 mL of methylene chloride. After an additional 5 min, the solvent was removed in vacuo. The wet solid was triturated with 10% ethyl acetate in petroleum ether. Purification by silica gel chromato graphy (5% ethyl acetate in petroleum ether) provided 13.8 mg (70%) of a colorless oil; IR (neat, cm 3045 (w), 2940 (s), 2855 (m), 1725 (s),1460 (m), 1443 (m), 1250 (m), 1240 (m), 1210 (m), 1030 (m),735 (m); XH NMR (300 MHz, C6D6) 6 3.25 (m, 1 H), 5.48-5.45 (m, 1 H), 3.27-3.25 (m, 1 H), 3.24 (s, 3 H), 2.95-2.87 (m, 1 H), 2.53-2.39 (m, 2 H), 2.04-1.37 (series of m, 11 H), 1.10 (s, 3 H); MS m/z. (M+ ) calcd for C16H23Br02 326.0881, obsd 326.0881. 148

Ketal alcohol 108. To a dry methanolic solution

ho of 104 (6 mg, 0.02 mmol in 1 mL of solvent) was added at 0 °C excess sodium borohydride (2.4 mg, 0.064 mmol). After 10 min, brine was H added, the solvent was removed in vacuo, and the residue was extracted with methylene chlori de. The extract was dried over magnesium sulfate and concentrated to yield 6 mg (97%) of a colorless oil; IR (neat, cm"^) 3440 (broad m), 2920 (s), 2865 (s), 1720 (s), 1460 (m), 1440 (m), 1428 (m), 1325 (m), 1210 (s), 1130 (s), 1105 (s), 1040 (m); XH NMR (300 MHZ, C6D6) 5 3.52 (s, 4 H), 3.30 (s, 3 H), 3.45-3.22 (m, 2 H), 2.49-2.44 (t, J = 7.6 Hz, 1 H), 2.42-1.08 (series of m, 15 H), 1.14 (s, 3 H); MS m/z (m+ ) calcd for C18H 280 5 324.2015, obsd 324.1953.

Ketal Mesylate 109. Alcohol 108 (145 mg, 0.448

MsO mmol) was treated with dry triethylamine (181 mg, 1.82 mmol) and OCH. mesyl chloride (117 mg, 0.91 mmol) at -29 °C in dry ether (8 ml). H After 2 h at 0 °C the reaction 149 mixture was diluted with ether, washed with water, and dried over magnesium sulfate. Concentration gave 190 mg (100 %) of a colorless oil; IR (neat, cm-1) 2940 (s), 2870 (m), 1718 (s), 1460 (m), 1448 (m), 1430 (m), 1350 (m), 1330 (s), 1210 (s), 1170 (s), 1125 (s), 948 (s); XH NMR (300 MHz, C6D6) 6 3.90-3.22 (m, 4 H), 3.50 (s, 2 H), 3.26 (s, 3 H), 2.15-2.14 (s, 3 H), 1.05 (s, 3 H), 2.43-0.84 (series of m, 15 H); MS m/z, (M+ ) calcd for C19H30O7S 402.1713, obsd 402.1711.

Thioketal Ketone 113. To a slurry of pyridinium chlorochromate (1.4 mg, 0.052 mmol), sodium acetate (1 mg, 0.01 mmol), and 3 A powdered molecular 0 sieves (25 mg) in dry methylene H chloride (2 mL) was added dropwise a solution of alcohol 115 (9.24 mg, 0.03 mmol) in 2 mL of the same solvent. After 20 min, the dark mixture was diluted with ether and the supernatant was passed through a short Florisil column. MPLC purifiction (20% ethyl acetate in petroleum ether) yielded 5.8 mg (63%) of a colorless oil; IR (neat, cm"1) 2930 (s), 2865 (m), 1730 (s), 1458 (m), 1442 (m), 1400 (m), 1275 (w), 1172 (m); XH NMR (300 MHz, C6D6) 5 3.28-3.13 (m, 4 H), 2.61-1.78 (m, 14 H), 1.70 150 (dd, J = 1.97, 3.3 Hz, 1 H), 1.66-1.32 (m, 2 H), 1.22 (s, 3 H); MS ra/£ (M+) calcd for C17H 21tOS2 308.1258, obsd 308.1252.

Dithiolane Olefin 114, To a solution of ketone 72a (270 mg, 1.25 mmol) in 6 mL of dry methylene chloride under argon was added ethanedithiol (1.06 g, 11.3 mmol) and boron trifluoride H etherate (172 mg, 1.25 mmol). After 8 h, the reaction mixrture was diluted with ether and washed with 3 x 15 mL of 1 N sodium hydroxide and 15 mL of brine. The extract was dried over magnesium sulfate and purified by MPLC (1% ethyl acetate in petroleum ether) to give 270 mg (79%) of a colorless oil; IR (neat, cm ^) 3040 (m), 2915 (s), 2960 (m), 1458 (m), 1445 (m), 1419 (w), 1372 (w), 1275 (w), 960 (w), 935 (w), 735 (m); 1H NMR (300 MHz, C6D6) 5 5.80-5.77 (m,l H), 5.45-5.41 (m, 1 H), 2.29-1.35 (series of m, 19 H), 1.31 (s, 3 H); 13C NMR (75 MHz, C6D6) ppm 127.21, 82.81, 71.82, 60.65, 59.77, 53.63, 51.65, 45.47, 45.35, 39.69, 39.52, 39.07, 38.84, 36.10, 29.82, 29.07, 19.89; MS m/z (M+ ) calcd for C17H 21tS2 292.1319, obsd 292.1354. 151 Anal. Calcd for C 17H21|S2: C, 69.81; H, 8.27. Found: C, 69.90; H, 8.27.

Dithiolane Alcohol 115. A solution of dithiolane 114 (24.7 rag, 0.0845 mmol) in 1 mL of tetrahydrofuran was prepared under argon in a 5-mL, one-necked flask equipped with condenser and gas H inlet. The mixture was heated to 45 °C and 9-BBN (0.318 mmol, 0.5 M in tetrahydrofuran) was added. After 36 h, 0.5 mL of 1 N sodium hydroxide and 0.5 mL of 30% hydrogen peroxide were added at room temperature. After 2.5 h, water was added and the mixture was extracted with methylene chloride. Purification by MPLC (silica gel, 20% ethyl acetate in petroleum ether) provided 19 mg (72%) of a colorless oil and 3.5 mg (15%) of starting material. IR (neat, cm"'*') 3350 (m), 2925 (s), 2855 (m), 1455 (m), 1440 (m), 1415 (w), 1272 (w), 1072 (m), 1055 (m), 730 (m); XH NMR (300 MHz, C6D6) 5 4.31 (quintet, J = 5.5 Hz, 1 H), 4.35-4.28 (m, 4 H), 2.57-1.42 (series of m, 18 H), 1.17 (s, 3 H); MS m/z (M+) calcd for C17H26OS2 310.1425, obsd 310.1421. 152

Silyl Enol Ethers 116a,b. To a cold (0 °C) solu tion of cyclohexylisopropyl- amine (23 mg, 0.165 mmol) in tetrahydrofuran ( 2 mL) under o TBDMS — argon was added n-butyl- H lithium (1.5 M hexane, 0.165 mmol). After being stirred for 40 min, the reaction mixture was cooled to -78 °C and ketone 113 (34 mg, 0.11 mmol) in tetrahydrofuran (1.5 mL) was added over a 5-min period. After 1 h of stirring, tert-butyldimethylsilyl trifluoromethanesulfonate (43 mg, 0.17 mmol) was added. After an additional 15 min at -78 °C, the reaction mixture was quenched with saturated ammonium chloride solution, extracted with ether, and dried over sodium sulfate. Purification by preparative TLC (silica gel, 2% ethyl acetate in petroleum ether) gave 29 mg (68%) of an inseparable mixture of isomers as a colorless oil; IR (neat, cm"1) 2920 (s), 2850 (s), 1640 (m), 1405 (m), 1325 (m), 1250 (s), 1210 (m), 930 (m), 840 (s), 780 (s); 1H NMR (300 MHz, C6D6) 5 5.04, 4.43 (m, 1 H), 2.83-2.68 (m, 4 H), 2.66-1.44 (series of m, 15 H), 1.29, 1.25 (s, 3 H), 0.97 (s, 9 H), 0.28 + 0.17 (s, 6 H); MS m/z (M+ ) calcd for C21H 38OSiS2 362.2090, obsd 362.2007. 153 Dithiane Olefin 117 Method A: To a solution of ketone 72a (17.6 mg, 0.080 mmol) in dry methylene chloride (1 mL) under argon was added 1,3 propanedithiol (53 mg, 0.48 mmol) and boron H trifluoride etherate (22 mg, 0.16 mmol). After 6.5 h at room temperature, the mixture was diluted with ether, washed with 4 x 2.5 mL of 1 N sodium hydroxide and 2 x 2.5 mL of brine, and dried over magnesium sulfate. Concentration and purification on MPLC (2% ethyl acetate in petroleum ether) gave 13.8 mg (56%) of a color less oil. IR (neat, cm 1) 3040 (w), 2930 (s), 1460 (m), 1445 (m), 1420 (m), 1376 (m), 1270 (m), 908 (m); 1H NMR (300 MHz, C6D6) 5 5.78-5.74 (m, 1 H), 5.44-5.41 (m, 1 H), 2.69-2.13 (series of m, 11 H), 1.93-1.61 (m, 7 H), 1.49- 1.38 (m, 3 H), 1.35 (s, 3 H); MS m/z (M+) calcd for C18H 26S2 306.1476, obsd 306.1487.

Method B: To a solution of methyl ketone 72a (90 mg, 0.42 mmol) in 2 mL of dry 1,2-dichloroethane under argon was added zinc triflate (308 mg, 0.84 mmol) and 1,3-propanedithiol (136 mg, 1.26 mmol). The reaction mixture was refluxed for 154 6 h, whereupon the mixture was cooled, diluted with petro leum ether, and washed with water (4 x 13 mL), I N sodium hydroxide solution (12 mL), and brine (12 mL). The com bined extracts were dried over sodium sulfate and separated as above to give 90 mg (70%) of 117 as a colorless oil and 20 mg (21%) of starting material.

Tetrahydropyranyl Ether 120. Alcohol 106 (22 mg, 0.08

THPO mmol) was treated with pyridinium tosylate (2 mg, 0.009 mmol) and hcH dihydropyran (27 mg, 0.33 mmol) in methylene chloride (2 mL) at room H temperature for 6 h. Solid sodium bicarbonate (3 mg) was added and the mixture was filtered through a pad of neutral alumina. Concentration of the filtrate provided 23 mg (83%) of a colorless oil; IR (neat, cm"1) 3040 (w), 2930 (s), 2860 (m), 1720 (s), 1460 (m), 1440 (m), 1210 (m), 1235 (m), 1220 (m), 1030 (m); 1H NMR (300 MHz, C6D6) 6 6.14 (m, 1 H), 5.48 (m, 1 H), 4.57 (m, 1 H), 3.79 (m, 2 H), 3.51-3.30 (m, 5 H), 2.55-2.52 (m, 2 H),

2.05-1.05 (series of m, 20 H); MS m./z (M+ ) calcd for

C2iH3 2 ® k 348.2300, obsd 348.2323. 155

Preparation of 121. To a solution of tetrahydro- THPO pyranyl ether 120 (6.0 mg, 0.017 mmol) in dioxane (0.5 mL) and water (0.3 mL) was added sodium dihydrogen phosphate (4.7 mg), disodium hydrogen phosphate (4.8 mg), osmium tetroxide 0.0005 mmol, 0.026 M in tert-butyl alcohol) and sodium periodate (7.3 mg, 0.034 mmol). After 2 h of stirring, the mixture was diluted with methylene chloride and washed with sodium thiosulfite solution and brine before being dried over magnesium sulfate. Concen tration and purification by flash chromatography (1 g silica gel,20% ethyl acetate in petroleum ether) provided 2.2 mg (30%) of a colorless oil which converted rapidly (over 10 hours) to a more polar compound. IR (neat, cm ) 3400 (br), 2940 (s), 2870 (s), 2720 (w), 1720 (s), 1460 (m), 1450 (m), 1380 (m), 1220 (m); *H NMR (300 MHz, C6D6) 6 9.63 (s, 1 H), 9.25 (m, 1 H), 4.56 (m, 1 H), 3.82 (m, 2 H), 3.43 (m, 2 H), 4.329 + 4.326 (s for two isomers, 3 H), 2.89-0.83 (series of m, 19 H), 1.10 + 1.09 (s for two isomers, 3 H); MS m/z^ (M+-C5H90) calcd for C17H 230U 279.1598, obsd 279.1607. 156

Tricyclic Hydrate 123. A solution of ketone 17 (50 mg, 0.31 mmol) in 11 mL of dioxane o and 5 mL of water was treated with disodium hydrogen phosphate (90

HO mg, 0.64 mmol), soidum dihydrogen H phosphate (84 mg, 0.71 mmol), sodium periodate (130 mg, 0.62 mmol) and osmium tetroxide (0.0078 mmol, 0.2 M in tert-butyl alcohol) at room tempera ture for 3 h. Water was then added and the mixture was extracted with methylene chloride. The extract was dried over magnesium sulfate, concentrated and purified by silica gel chromatogaphy (ethyl acetate) to give 20 mg (31%) of a viscous, colorless oil; IR (neat, cm ) 3450 (br), 2930 (s), 2880 (m), 1730 (s), 1450 (m); 1H NMR (300 MHz, C6D6) 5 4.59-3.80 (series of m, 2 H), 2.43-1.01 (series of m, 12

H), MS rn/z (M+-OH) calcd for Cn H 1503 195.1021, obsd 195.1044. 157

Bis(tert-butyldimethylsilyl)Enol Ether 128a,b. Method A: Under kinetic control Lithium bis(di me thylphenyl si lyl )amide (0.208 mmol) was pre- OTBDMS pared as described above •"CH'3 in 0.75 mL of dry tetra- TBDMSO hydrofuran under argon.

H At -78 °C, a solution of diketone 101 (12 mg 0.052 mmol) in 1 mL of the same solvent was added over 10 min. After 20min, the enolate was quenched with tert- butyldimethylsilyl trifluoromethanesulfonate (55 mg, 0.22 mmol). Ether was added and the mixture was poured over a short Florisil column. The product was purified on silica gel to give 4 mg (17%) of 1281,b as an inseparable mixture in the ratio of 1:9. The low yield was presumably due to hydrolysis during chromatography; IR (neat, cm 1) 3065 (w), 2960 (s), 2920 (s), 2900 (m), 2865 (s), 1650 (m), 1478 (m), 1460 (m), 1375 (w), 1368 (w), 1338 (m), 1260 (s), 850 (s), 788 (m); XH NMR (300 MHz, C6D6) 6 4.63 (m, 1 H), 4.34 (m, 1 H), 2.64-2.42 (series of m, 6 H), 2.11-0.84 (series of m, 7 H), 1.13 (s, 3 H), 0.981 (s, 9 H), 0.975 (s, 9 H), 0.169, 0.159, 0.137 (singlets, 12 H); MS m/z (M+ ) calcd for C27H4802S12 460.3193, obsd 460.3208. 158 Method B: Under thermodynamic control. To a cold solution (-78 °C) of diketone 101 (56 mg, 0.24 mmol) and tert-butyldimethylsilyl chloride (100 mg, 0.62 mmol) in dimethoxyethane (4 mL) and hexamethylphosphoramide (0.5 mL) was added potassium hydride (oil free, 85 mg, 2.1 mmol). The mixture was slowly warmed to room temperature and after 2 h, an additional portion of tert- butyldimethylsilyl chloride (0.67 mmol) was introduced. After 12 h of stirring, the yellow mixture was diluted with ether, poured over a Celite column, and topped with a small amount of Florisil. The filtrate was diluted with petro leum ether, washed with water and brine, and dried over sodium sulfate. The extract was passed down a short neu tral alumina column and concentrated to give 100 mg (93%) of a colorless oil. Analysis by capillary GC showed a mixture of regioisomers 128a and 128b to be present in a 2.4:1 ratio. 159

Bis(Trimethylsilyl)Enol Ethers 129a,b. Method A: To a solution of bis(dimethylphenylsilyl)amine (48 mg, 0.17

OTMS mmol) in 1 mL of dry

'3 tetrahydrofuran under

TMSO argon at 0 °C was added n-butyllithium (0.194 H mmol, 1.5 M hexane). After 15 min, the mix ture was cooled to -78 °C and a dry solution of diketone 101 (9.8 mg, 0.042 mmol) in 1 mL of tetrahydrofuran was added dropwise over a 16 min period. The mixture was stir red for 1 h. A solution of dry triethylamine (36 mg, 0.36 mmol) was added followed by freshly distilled trimethylsilyl chloride (36 mg, 0.34 mmol). After warming to room temperature, volatile materials were removed in vacuo. The residue was triturated with hexane and the resulting sus pension was filtered through a sodium sulfate column. Concentration gave a mixture of bis(dimethylphenylsilyl)- amine and bis enol ethers 129a,b as a colorless oil (63 mg, 95%); IR (neat, cm-1) 3060 (w), 2960 (s), 1640 (sw), 1460 (w), 1440 (w), 1330 (s), 1290 (s), 1255 (s), 1240 (s), 1210 (s), 1130 (w), 1120 (w), 930 (m), 910 (m), 900 (m), 840 160 (s), 760 (m); XH NMR (300 MHz, C6D6) 6 4.64-4.62 (m, 1 H), 4.39-4.31 (m, 1 H), 2.83-1.34 (series of ra, 13 H), 1.13 (s, 3 H), 0.16 (s, 18 H); MS m/z. (M+ ) calcd for C21H3 602Si2 376.2322, obsd 376.2281.

Method B: To a dry solution of diketone 101 (10 mg, 0.043 mmol) and (trimethylsilyl)ethyl acetate (29 mg, 0.18 mmol) in tetrahydrofuran (3 mL) at -78 °C under argon was added a solution of nearly anhydrous tetrabutylammonium fluoride (0.0025 mmol, 1 M in tetrahydrofuran) in 1 mL of tetrahy drofuran over a 23-min period. After being warmed slowly to room temperature, and 1 h of stirring at this tempera ture, the reaction mixture was worked up. All volatile materials were removed in vacuo. The residue was dissolved in hexane and flashed over a plug of silica gel. Concen tration gave 14 mg (82%) of pure product which by relative integration of the enol vinyl proton absorptions appeared to be an 8:1 mixture of enol ethers. 161

Bis-Acyloin 130. To a solution of bis(di- OH methylphenylsilyl)amine (104 mg, 0.365 mmol) in tetrahydrofuran (5 mL) at -78 °C and under argon was added n-butyllithium (0.375 mmol, 0 1.5 M in hexane). The mixture was HO warmed to 0 °C for 15 min and then cooled down to -78 °C. A solution of diketone 101 (16 mg, 0.069 mmol) in 1 mL of tetrahydrofuran was next added over 26 min. After being stirred for 30 min, the homo geneous mixture was added to a stirred suspension of MoOPh (287 mg, 0.663 mmol) in 1 mL of tetrahydrofuran at -40 °C. After 20 min, the reaction mixture was quenched with saturated sodium sulfite solution and extracted with ether. The extract was washed with 10% hydrochloric acid solution, sodium bicarbonate solution, and brine prior to drying over magnesium sulfate. Chromatography over silica gel (ethyl acetate) gave 3 mg (16%) of a colorless oil; IR (neat, cm 1) 3420 (br), 2940 (s), 2860 (m), 1730 (s), 1448 (m), 1400 (m), 1370 (m); 2H NMR (300 MHz, C6D6) 6 4.43-3.74 (series of m, 2 H), 3.00-0.83 (series of m, 15 H), 1.57, 1.27, 1.26, 1.09 (singlets from all four posible isomers, 3 H); MS m/z^ (M+ ) calcd for C-LgHaoO,, 264.1379, obsd 264.1381. 162

Chloro Ketal 131.

A dry solution of 108 (9.0 mg, 0.027 mmol) and triphenylphosphine (42 mg, 0.16 mmol) in carbon tetrachloride (1 mL) was refluxed for 14 h. The mixture H was concentrated and separated over silica gel (10% ethyl acetate in petroleum ether) to give 9-9.5 mg of a colorless oil (95-100%); IR (neat, cm ) 2940 (s), 2860 (m), 1720 (s), 1460 (m), 1440 (m), 1430 (m), 1380 (w), 1330 (m), 1205 (s), 1110 (s); *H NMR (300 MHz, C6D6) 5 3.50 (s, 4 H), 3.23 (s, 3 H), 3.26-3.17 (m, 1 H), 3.11-3.03 (m, 1 H), 2.97-2.41 (t, J = 7.9 Hz, 1 H), 2.34- 2.14 (m, 3 H), 1.94-0.85 (series of m, 11 H), 1.07 (s, 3 H); MS m/z^ (M+ ) calcd for C18H27C10„ 342.1598, obsd 342.1559.

Chloro Ketone 132. Ketal 131 (9.0 mg, 0.026 mmol) was dissolved in 1.5 mL of acetone. A drop of 10% hydro OCH. chloric acid solution was added and the mixture was stirred for 10

H h at room temperature. Sodium bicarbonate solution was added 163 and the mixture was concentrated. The residue was extracted with ether, the latter extract was dried over magnesium sulfate, and the product was chromatographed over silica gel (10-20% ethyl acetate in petroleum ether) to yield 8.0 mg (98%) of a crystalline material, mp 74-76 °C; IR (neat, cm"1) 2940 (s), 2860 (m), 1730-1715 (s), 1400 (m), 1380 (w), 1310 (w), 1200 (s), 1130 (m), 1120 (m); XH NMR (300 MHz, C6D6) 5 3.21 (s, 3 H), 3.28-3.16 (m, 1 H), 3.08-3.00 (m, 1 H), 2.42-2.27 (dd, J = 19.5, 6.4 Hz, 2 H), 1.60 (s, 3 H), 2.17-0.72 (series of m, 13 H); MS m/z (M+ ) calcd for C16H23C103 298.1336, obsd 298.1316.

Chloro Enol Ether 133. To a -78 °C solution of n- ci butyllithium (0.7 mmol, 1.5 M hexane) was added bis(dimethyl- OCH; ■CH; phenylsilyl)amine (0.7 mmol, 200 o mg). The clear reaction mixture TBDMS' H was warmed to -10 °C and stirred there for 15 min before being cooled down to -78 °C. Meanwhile, a dry benzene solution containing chloroketone 132 (100 mg, 0.336 mmol) was transferred to a dry flask and the benzene was removed in vacuo. Dry tetrahydrofuran (6 mL) was added and the solution was added dropwise to the kinetic base over a 22-min period. After 0.5 h at -78 °C, 164 tert-butyldimethylsilyl trifluoromethanesulfonate (neat, 185 mg, 0.701 mmol) was added in one portion. After 5 min 0.5 mL of triethylamine were added. The mixture was diluted with ether and washed with water. The organic layer was dried over sodium sulfate and the product was separated by preparative TLC silica gel (16 g, 5% ethyl acetate in petroleum ether) to give 116 mg (83%, yields -1 range 83-93%) of a colorless oil; IR (neat, cm ) 3060 (w), 2955 (s), 2860 (s), 1725 (s), 1640 (s), 1460 (m), 1440 (m), 1390 (w), 1380 (w), 1330 (m), 1250 (m), 1220 (m), 840 (s), 780 (m); XH NMR (300 MHz, C6D6) 5 4.59 (m, 1 H), 3.25 (s, 3 H), 3.24 (m, 1 H), 3.10 (m, 1 H), 2.81 (dt, J = 10.2, 2.0 Hz, 1 H), 2.58 (d, J = 17 Hz, 1 H), 1.91-1.19 (series of m, 11 H), 1.07 (s, 3 H), 0.98 (s, 9 H), 0.161 (s, 6 H); MS m/z (M+ ) calcd for C22H 37C103Si 412.2201, obsd 412.2154. 165

Chloro Enol Ethers 134a,b.

A solution of chloroketone 132 (54 mg Cl 0.18 mmol) and ethyl OCH3 (trimethylsilyl)acetate (73 mg, 0.45 mmol) in 3 TMSO mL of dry tetrahydro- H furan was prepared under an argon atmosphere, and cooled to -78 °C. A solution tetra-n-butylammonium fluoride (0.013 mmol, 0.1 M in tetra hydrofuran) in 2 mL of tetrahydrofuran was added over a 25 min period. After stirring for 1 h at 0 °C, the reaction mixture was allowed to warm gradually to room temperature. Triethylamine (3 drops) was added, the mixture was diluted with pentane, washed with water and dried over sodium sulfate. Concentration gave 65 mg (99%) of an inseparable mixture of 134a,b as a colorless oil. The ratio of 134a:134b was in this experiment 1:10 (NMR); in other experiments the ratio dropped to as low as 1:2. IR (neat, cm"1) 3020 (w), 2920 (s), 2830 (m), 1705 (s), 1625 (s), 1425 (m), 1315 (m), 1235 (s), 1205 (s), 910 (m), 835 (s); XH NMR (300 MHz, C6D6) 6 5.35, 4.62 (2 m, 1:10, 1 H), 3.23 (s, 3 H), 3.22-2.99 (m, 2 H), 2.85 (d, J = 16.9 Hz, 1 H), 2.62 (d, J = 16.6 Hz, 1 H), 2.52-1.19 (series of m, 11 H), 166 1.09, 1.01 (2 s, 3 H), 0.21, 0.19 (2 s, 9 H); MS m/z (M+ ) calcd for C19H 310 3SiCl 370.1731, obsd 370.1681.

Chloro Acyloin 135 To a solution of silyl enol Cl ether 133 (69 mg, 0.17 mmol) and

OCH. N-methylmorpholine N-oxide (60 mg 0.51 mmol) in 3.5 mL of acetone o and 0.5 mL water was added osmium H HO tetroxide (0.0034 mmol, 0.026 M in tert-butyl alcohol). After 5 h of stirring, 300 mg of sodium hydrogen sulfite and 300 mg of magnesium silicate were added. After an additional 0.5 h, the suspension was filtered through Celite. The filtrate was adjusted to pH 7 with dilute hydrochloric acid solution and the mixture was concentrated in vacuo. The residue was further acidified to -pH 3 and extracted with ethyl acetate. The extracts were dried over magnesium sulfate and the crude product was chromatographed over silica gel (50% ethyl acetate in petroleum ether) to give 45 mg (65%) of a colorless oil; IR (neat, cm"1) 3430 (m), 3080 (w), 3030 (w), 2940 (s), 2860 (m), 1740 (s), 1720 (s), 1480 (m), 1460 (m), 1450 (m), 1430 (m), 1380 (m), 1315 (m), 1210 (m), 1120 (s); XH NMR (300 MHz, C6D6) 6 3.55 (d, J = 5.5 Hz, 1 H), 3.18 (s, 3 H), 3.12-3.06 (m, 1 H), 2.97-2.91 (m, 1 H), 2.42 (d, J = 26 Hz, 167 1 H), 2.40 (d, J = 26 Hz, 1 H), 2.08 (t, J = 7.5 Hz, 1 H), 1.86 (g, J = 5 Hz, 1 H), 1.76-0.89 (series of m, 10 H), 0.97 (s, 3 H); MS m/z (M+ ) calcd for C16H23C10„ 314.1285, obsd 314.1314.

Chloro Aldehyde Diester 136. Chloro acyloin 135 (7.5 mg, Cl 0.25 mmol) was dissolved in 1 mL

och3 0f benzene and 0.5 mL of dry methanol, and cooled to 0 °C. Lead tetraacetate (12 mg, 0.026 mmol) was added in one portion. After 5 min, the reaction mixture was quenched with sodium bicarbonate solution and poured over Celite. The organic layer was washed with brine and dried over magnesium sul fate. Concentration gave 8.5 mg (100%) of a colorless oil which appeared as clean product by NMR; IR (neat, cm ^) 2965 (s), 2880 (m), 2740 (w), 1720 (s), 1465 (m), 1450 (m), 1435 (m), 1360 (m), 1270 (m), 1260 (m), 1225 (m), 1200 (m),

1180 (m), 1115 (m), 1050 (m), 800 (w), 750 (w), 730 (w); 1U NMR (300 MHz, C6D6) 6 3.45-2.99 (m, 2 H), 3.28 (s, 3 H), 3.20 (s, 3 H), 2.91(d, J = 16.5 Hz, 1 H), 2.67 (d, J = 16.5 Hz, 1 H), 2.41-0.82 (series of m, 11 H), 0.94 (s, 3 H), 1.11 (t, J = 7.0 Hz, 1 H); MS m/z (M+-H) calcd for C17H2i,ClOs 343.1282, obsd 343.1255. 168

Bromo Enol Ether 137 Procedure identical to that

Br used for chloro enol ether 133 Bromo ketone 105 (133 mg, 0.388 o c h 3 mmol) in 4 mL of tetrahydrofuran o TBDMS' was added to a solution at of H lithium tetramethyldiphenyldisil- azide in 8 mL of the same solvent at -78 °C over 0.5 h. After 30 min, the reaction mixture was quenched with tert- butyldimethylsilyl trifluoromethanesulfonate (205 mg, 0.776 mmol). The workup was similar to that for 134. There was obtained 156 mg (88%) of a colorless oil; IR (neat, cm ^) 3060 (w), 2950 (s), 2930 (s), 2900 (m), 2860 (s), 1725 (s), 1640 (s), 1460 (m), 1388 (w), 1380 (w), 1330 (w), 1250 (s), 1225 (s), 840 (s), 785 (w); 1H NMR (300 MHz, C6D6) 6 4.62 (m, 1 H, 3.23 (s, 3 H), 3.24 (m, 1 H), 2.89 (m, 2 H), 2.58 (d, J = 17 Hz, 1 H), 1.91-1.19 (series of m, 11 H), 1.08 (s, 3 H), 0.98 (s, 9 H), 0.173 (s, 3 H), 0.167 (s, 3 H), MS m/z. (M+ ) calcd for C22H37Br03Si 456.1695, obsd 456.1680. 169

Bromoacyloin 138. Procedure identical to that Br for chloroacyloin 135. For

OCH. example, 32 mg (0.070 mmol) of CH. bromo enol ether 137 yielded 16 mg 0 (66%) of a colorless oil; IR HO (neat, cm"^) 3460 (br), 2940 (s), 2860 (m), 1740 (s), 1460 (m), 1445 (m), 1430 (m), 1380 (w), 1310 (m), 1230 (m), 1110 (m); XH NMR (300 MHz, C6D6) 6 3.57 (d, J = 5.3 Hz, 1 H), 3.29-2.72 (series of m, 4 H), 3.17 (s, 3 H), 2.46 (d, J = 18.5 Hz, 1 H), 2.37 (d, J = 18.5 Hz, 1 H), 2.09 (t, J = 7.5 Hz, 1 H), 1.91-0.94 (series of m, 12 H); MS m/z. (M+ ) calcd for C16H23BrOl4 358.0779, obsd 358.0763.

Bromo Diester 139. Procedure identical to that Br of the chloro analogue 136 (0.46 mmol). A similar ratio of reagents and the same workup, provided 16.8 mg (94%) of a crystalline solid, mp 82-84 °C; IR 1 ( , 2980 (m), 2940 (m), 2875 (m), 1720 (s), 1710 (s), 1435 (m), 1380 (m), 1235 (m), 1210 (m), 1200 (m), 1110 (m); XH NMR (300 MHz, C6D6) 6 9.79 (m, 1 H), 3.29 170 (s, 3 H), 3.20 (s, 3 H), 3.22-2.86 (m, 2 H), 2.88 (d, J = 17 Hz, 1 H), 2.64 (d, J = 17 Hz, 1 H), 2.37 (t, J = 9 Hz, 1 H), 2.10-0.89 (series of m, 10 H), 0.94 (s, 3 H); MS m/js (M+-C02CH3) calcd for C15H 22Br03 329.0757, obsd 329.0750.

Bromodimethoxy Acetal 140. Diester 139 (16.8 rag, 0.043 Br mmol) was treated with 2 mL of trimethyl orthoformate and CH3 Amberlyst-15 (20 mg) for 2 h. The

h 3c o mixture was filtered, and all volatile materials were removed in vacuo to provide 18 mg (96%) of a colorless oil. IR (neat, cm"1) 2950 (s), 2870 (s), 2820 (m), 1720 (s), 1460 (m), 1445 (m), 1430 (m), 1380 (w), 1360 (w), 1225 (s), 1190(s), 1120 (s), 1065 (s), 855 (w), 900 (w), 795 (w); XH NMR (300 MHz, C6D6) 6 4.57 (d, J = 7 Hz, 1 H), 3.39 (s, 3 H), 3.28 (s, 3 H), 3.22 (s, 3 H), 3.16 (s, 3 H), 3.36-2.91 (m, 2 H), 2.82 (d, J = 15 Hz, 1 H), 2.73 (d, J = 15 Hz, 1 H), 2.08 (dd, J = 5.0, 12.0 Hz, 1 H), 1.90-0.96 (series of m, 10 H), 1.04 (s, 3 H); MS m/z (M+-C02CH3) calcd for C16H28BrO 403.1119, obsd 403.1145. 171

Iodo Diester 141. A solution of bromo diester I 139 (46 mg, 0.14 mmol) in acetone

OCH. (2 mL) containing 1 drop of pyri dine was treated with sodium iodide (91 mg, 0.61 mmol) at the reflux temperature for 12 h. The reactionmixture wasconcentrated in vacuo and the residue was washed thoroughly with ether. The extract was washed with brine and dried over magnesium sulfate. Concentration gave 46 mg (87%) of a colorless oil which was used without purification; IR (neat, cm"’*') 2980 (m), 2940 (m), 1720 (s), 1710 (s), 1460 (m), 1445 (m), 1435 (m), 1220 (m), 1200 (m), 1175 (m); XH NMR (300 MHz, C6D6) 6 3.28 (s, 3 H), 3.25 (s, 3 H), 3.35-2.78 (m, 2 H), 2.90 (d, J = 17.0 Hz, 1 H), 2.67 (d, J = 17.0 Hz, 1 H), 2.38 (t, J = 9.5 Hz, 1 H), 2.14-1.14 (series of m, 8 H), 0.95 (s, 3 H); MS m/z (M+-C02CH3) calcd for C15H22I03 309.1702, obsd 309.1706. 172

Iododimethoxy Acetal 142. Diester 141 (46 mg, 0.13 mmol) was treated with 1 mL of trimethyl orthoformate and 60 mg of Amberlyst-15. After 2 h, fil tration and concentration gave 56 mg of crude product which was chromatographed over silica gel (20% ethyl acetate in petroleum ether) to yield 48 mg (89%) of a colorless oil; IR (neat, cm"1) 4.59 (d, J = 7.5 Hz, 1 H), 3.39 (s, 3 H), 3.27 (s, 3 H), 3.23 (s, 3 H), 3.18 (s, 3 H), 3.16-2.64 (m, 2 H), 2.83 (d, J = 15.0 Hz, 1 H), 2.73 (d, J = 15.0 Hz, 1 H), 2.11 (dd, J = 5.5, 12.0 Hz), 1.90-0.84 (series of m, 10 H), 1.03 (s, 3 H); MS m/z (M+-CH1(0) calcd for C18H2 705 450.0903, obsd 450.0874.

Bromo Ketal 152. To a cold solution (-78 °C) of 105 (195 mg, 0.570 mol) and Br bis(trimethylsilyl)ethylene glycol OCH. (197 mg, 0.684 mmol) in methylene CH; chloride (2 mL) was added tri I—or ° H methylsilyl trifluoromethanesulfonate (27 mg, 0.028 mmol). After 4 h of stirring, dry pyridine was added. The mixture was diluted with methylene 173 chloride, washed with water and dried over magnesium sul fate. Purification by silica gel chromatography (10% ethyl acetate in petroleum ether) yielded 195 mg (89%) of a colorless oil; IR (neat, cm"^) 2940 (s), 2860 (s), 1720 (s), 1460 (m), 1440 (m), 1425 (m), 1325 (m); XH NMR (300 MHz, C6D6) 6 3.51 (s, 4 H), 3.23 (s, 3 H), 3.23-3.04 (m, 1 H), 2.95-2.87 (m, 1 H), 2.44 (t, J = 7.9 Hz, 1 H), 2.29 (d, J = 14.0 Hz, 1 H), 2.21-1.14 (series of m, 13 H), 1.07 (s, 3 H); MS m/z. (M+ ) calcd for C 18H27BrO„ 388.1072, obsd 388.1007.

Alcohol 153. A cold (-78 °C) solution of Br OH 152 (30 mg, 0.078 mmol) in methyl ene chloride (2 mL) was treated '*CH: with DIBAL (0.025 mmol, 1 M in hexanes) for 20 min. Saturated H potassium sodium tartrate solution (2 mL) was added and after 1 h of stirring at room tempera ture, the product was extracted with methylene chloride and dried over magnesium sulfate. Concentration gave 24 mg (86%) of a colorless oil, which was carried into the next reaction without further purification; IR (neat, cm-^) 3440 (br), 2940 (s), 2860 (m), 1335 (m), 1105 (s), 1035 (s); XH NMR (300 MHz, C6D6) 6 3.51 (s, 4 H), 3.09-2.86 (series of 174 m, 4 H), 2.99 )d, J = 3.4 Hz, 1 H), 2.12-1.13 (series of m, 15 H) , 0.76 (s, 3 H); MS m/z. (M+-C2HltO) calcd for C5H23Br02 314.0881, obsd 314.0864.

Reto-alcohol 154. A solution of ketal 153 (15 Br mg, 0.042 mmol) in acetone (1 mL) OH was treated with 10% hydrochloric

VCH; acid at room temperature for 1 h. O Solid sodium bicarbonate was added H and the mixture was concentrated in vacuo. The residue was dissolved in methylene chloride, filtered, and dried over magnesium sulfate. Concentration gave 11 mg (86%) of 154 as a colorless oil which was used without further purification; IR (neat, cm ^) 3450 (br), 2930 (s), 2830 (s), 1730 (s), 1440 (m), 1395 (m), 1170 (m), 1040 (m), 810 (w); XH NMR (300 MHz, C6D6) 5 3.10-3.04 (m, 1 H), 2.92-2.80 (m, 3 H), 2.22-0.77 (series of m, 16 H), 0.69

(s, 3 H); MS m/z (M+ ) calcd for C15H23Br02 314.0881, obsd 314.0866. 175

Bromo Ketal 156.

Br To a cold (-78 °C) solution OTBDMS of 153 (72 mg, 0.20 mmol) in methylene chloride (4 mL) was added dry triethylamine (101 mg, 1 mmol) and tert-butyldimethylsilyl H trifluoromethanesulfonate (211 mg, 0.8 mol). After 20 min of stirring, brine was added, and the mixture was extracted with methylene chloride. The combined extracts were dried over magnesium sulfate and concentrated to give a cloudy residue. Purification by chromatography on neutral alumina in ether yielded 88 mg (95%) of a colorless oil; IR (neat, cm-1) 2945 (s), 2920 (s), 2840 (s), 1415 (m), 1410 (m),

1335 (w), 1250 (m), 1095 (s), 850 (s), 835 (s), 775 (m); lE NMR (300 MHz, C6D6) 6 3.52 (s, 4 H), 3.30 (d, J = 9.6 Hz, 1 H), 3.21 (d, J = 9.6 Hz, 1 H), 3.17-3.09 (m, 1 H), 2.99- 2.91 (m, 1 H), 2.20-1.11 (series of m, 15 H), 0.96 (s, 9 H), 0.87 (s, 3 H), 0.03 (s, 6 H); MS m/z (M+-CltHg) calcd for C19H32Br03Si 415.1304, obsd 415.1372. 176

Ketone 155. Ketal 156 (101 mg, 0.217mmol) was treated with p-toluenesulfonic OTBDMS acid (3 mg, 0.017 mmol) in acetone (10 mL) at room temperature for 24 h. Sodium bicarbonate solution was added and volatile materials were removed in vacuo. The residue was patitioned between methylene chloride and water, and the organic phase was dried over magnesium sulfate. Purification of the residue by silica gel chromatography (10% ethyl acetate in petro leum ether) provided 82 mg (90%) of a colorless oil; IR (neat, cm"1) 2950 (s), 2910 (s), 2840 (s), 1735 (s), 1465 (m), 1460 (m), 1255 (s), 1090 (s), 835 (m); XH NMR (300 MHz, C6D6) 6 3.29-3.02 (series of m, 4 H), 2.25 (d, J = 15.9 Hz, 1 H), 2.19-2.10 (m, 2 H), 2.08 (d, J = 15.5 Hz, 1 H), 1.92-0.88 (series of m, 11 H), 0.94 (s, 9 H), 0.80 (s, 3 H) , 0.01 (s, 6 H); MS m/z. (M+-CltH9) calcd for C17H28Br02Si 371.1041, obsd 371.1105. 177

Silyl Enol Ether 157. Lithium bis(dimethylphenyl- Br silyl)amide (0.145 mmol) was OTBDMS prepared as above in 1 mL of dry tetrahydrofuran under an argon

TBDMSO atmosphere. The mixture was

H cooled to -78 °C and a solution of ketone 155 (25 mg, 0.058 mmol) in 1 mL of tetrahydrofuran was added over a period of 15 min. After 30 min of stir ring, tert-butyldimethylsilyl trifluoromethanesulfonate (46 mg, 0.18 mmol) was added. After 5 min, the progress of reaction was arrested by adding triethylamine, diluting with pentane, and washing with water. The organic layer was dried over sodium sulfate and concentrated. Purifica tion of the products by silica gel chromatography (2% ethyl acetate in petroleum ether) yielded 30 mg (100%) of two regioisomers (3.2:1, cap. GC), the major of which was the desired bromide; IR (neat, cm ^) 3060 (w), 3040 (w), 2950 (s), 2920 (s), 2850 (s), 1640 (m), 1465 (m), 1455 (m), 1250 (s), 1090 (m), 835 (s), 775 (m); lH NMR (300 MHz, C6D6) 6 4.71 + 4.62 (1:3.2) (m, 1 H), 3.31-1.18 (series of m, 17 H), 0.98 (s, 9 H), 0.96 (s, 9 H), 0.84 (s, 3 H), 0.17 (s, 6 H), 0.02 (s, 6 H); MS m/z (M+ ) calcd for C27H5Br02Si2 542.2568, obsd 542.2608. 178

Acyloin 158. A solution of enol ether 157 Br (30 mg, 0.058 mmol) in acetone OTBDMS (1.2 mL) and water (0.16 mL) was treated with osmium tetroxide O (0.0029 mmol, 0.026 M in tert-

HO H butyl alcohol) and 4-methylmorpholine N-oxide (20 mg, 0.17 mmol) at room temperature for 2 h. Magnesium silicate (60 mg) and sodium hydrogen sul fite (15 mg) were added and the mixture was allowed to stir for 1 h before being filtered thorugh Celite. One drop of 10% hydrochloric acid was added to the filtrate, and the latter was concentrated. The residue was dissolved in ethyl acetate and washed successively with 10% hydrochloric acid, 5% sodium bicarbonate solution, and brine. After being dried over magnesium sulfate, the product was purified by silica gel chromatography (50% ethyl acetate in petroleum ether) to provide 18 mg (71%) of a colorless oil; 1H NMR (300 MHz, C6D6) 5 3.55 (dd, J = 1.0, 6.0 Hz, 1 H), 3.11 (d, J = 19.3 Hz, 1 H), 3.01 (d, J = 19.4 Hz, 1 H), 3.04-2.95 (m, 1 H), 2.83-2.75 (m, 1 H), 2.32 (dd, J = 1.0, 17.9 Hz, 1 H), 2.10 (d, J = 17.8 Hz, 1 H), 1.91-0.99 (series of m, 11 H), 0.92 (s, 9 H), 0.74 (s, 3 H), -0.016 (s, 3 H), -0.02 (s, 3 H), MS m/z (M+-C„H9) calcd for C 17H28Br03Si 387.0991, obsd 387.0923. 179

Aldehyde Ester 159. To a solution of 158 (16 mg, 0.037 mmol) in benzene (1 mL) and methanol ygPMiilor\ a i i (0.5 mL) at 0 °C was introduced lead tetraacetate (20 mg, 0.045 mmol). Br—' O HaCO^J After 3 min of stirring, 5% sodium T H O bicarbonate solution was added. The mixture was filtered and the filtrate was extracted with ethyl acetate. The combined extracts were dried over mag nesium sulfate and concentrated to give 15 mg (70%) of a colorless oil which was used without further purification; IR (neat, cm-1) 2950 (s), 2920 (s), 2850 (m), 2715 (w), 1730 (s), 1710 (s), 1405 (m), 1250 (m), 1195 (m), 1085 (m), 835 (m), 875 (w); XH NMR (300 MHz, C6D6) 6 9.81 (d, J = 2.0 Hz, 1 H), 3.28 (s, 3 H), 3.17 (d, J = 9.8 Hz, 1 H), 3.1 (d, J = 9.7 Hz, 1 H), 3.14-3.00 (m, 1 H), 2.95-2.87 (m, 1 H), 2.54 (d, J = 16.3 Hz, 1 H), 2.44 (d, J = 16.3 Hz, 1 H), 2.30-1.29 (series of m, 11 H), 0.93 (s, 9 H), 0.73 (s, 3 H), 0.00 (s, 6 H); MS m/z (M+-C„H9) calcd for Cx8H3oBrO^Si 417.1097, obsd 417.1095. 180

Iodo Acetal 151

Bromide 159 (11 mg, 01023 mmol) was dissolved in 1 mL of jpniiCA ai t acetone and treated with sodium

I iodide (35 mg, 0.23 mmol) at room OCH3 temperature for 12 h. The mixture T H o was diluted with pentane and filtered. The filtrate was concentrated and the residual oil was redissolved in trimethyl orthoformate (1 mL). Amber- lyst-15 ion exchange resin (20 mg) was introduced and the mixture was stirred for 4 h. Filtration and concentration gave a cloudy oil which was purified by silica gel chroma tography (5% ethyl acetate in petroleum ether) to provide 10 mg (77%, overall yield) of a colorless oil; IR 2950 (s), 2920 (s), 2840 (s), 1730 (s), 1405 (m), 1250 (m), 1090 (s), 830 (m), 770 (m); XH NMR (300 MHz, C6D6) 5 4.58 (d, J = 7.3 Hz, 1 H), 3.38 (s, 3 H), 3.31 (d, J = 9.6 Hz, 1 H),3.22 (s, 3 H), 3.19 (d, J = 9.7 Hz, 1 H), 3.180(s, 3 H), 2.94- 2.89 (m, 1 H), 2.73-2.64 (m, 1 H, 2.50 (s,2 H), 2.40 (t, J = 9.2 Hz, 1 H), 2.07 (dd, J = 6.1, 10.5 Hz, 1 H), 1.97-1.21 (series of m, 9 H), 0.97 (s, 9 H), 0.86 (s, 3 H), 0.04 (s, 6 H); MS (M+-CltH9) calcd for C20H31IO5Si 511.1376, obsd 511.1346. REFERENCES

1. Bohlmann, F.; Zedro, C. Chem. Ber. 1979, 112. 435.

2 . (a) Ohta, Y.; Hirose, Y. Chem. Lett. 1972, 263; Gutsche, J.R.; Maycock, E.T. Tetrahedron 1968, 24, 859. (b) Ibid. 1968, 24/ 43* 3. Bohlmann, F.; Zedro C. Chem. Ber. 1979, 112, 427. 4. Duthaler, R.O.; Maienfisch, P. Helv. Chim. Acta 1984, 67, 856. 5. Stetter, H.; Krtiger-Hansen, I.; Riza, M. Chem. Ber. 1961, 94/ 2102. 6 . Eaton, P.E.; Mueller, R.H.; Carlson, G.R.; Cullison, D.A.; Cooper, G.F.; Chou, T.; Krebs, E. J . Am. Chem. Soc. 1977, 99, 2751. 7. Gelbard, G.; Colonna, S. Synthesis 1977, 1, 113.

8. (a) Nakamura, E.; Kuwajima, J. J. Am. Chem. Soc. 1977, 99, 7360. (b) Ibid. 1983, 105, 651. (c) Nakamura, E.; Kuwajima, I. J. Am. Chem. Soc. 1984, 106. 3368. 9. Ireland, R.E.; Mueller, R.H., Willard, A.K. J . Am. Chem. Soc. 1972, 9.4, 5098.

10. (a) Field, L.; Settlage, P.H. J . Am. Chem. Soc. 1953, 76. 1222. (b) Paul, R.C.; Sharma, S.K. Chem. and Ind. 1971, 702.

11. (a) Stowell, J.C.; Keith, D.R. Synthesis 1979, 132. (b) Stowell, J.C. J. Org. Chem. 1976, 41., 560. 12. Snider, B.B.; Karras, M.; Price, R.T.; Rodini, D.J. J . Org. Chem. 1982, 41_, 4538.

13. Barton, D.H.R.; Bushiandes, B.; Fourrey, J.L. Tetra hedron Lett. 1983, 24., 1605.

181 182 14. (a) Corey, E.J.; Hegedus, L.S. J. Am. Chem. Soc. 1969, 91, 1233. (b) Corey, E.J.; Semmelhack, M.F.; Hegedus, L.S. J. Am. Chem. Soc. 1968, 90, 2416, 2417. 15. Cacchi, S.; Morera, E.; Ortar, G. Tetrahedron Lett. 1985, 26, 1109. 16. Wollenberg, R.H.; Albizadi, K.F.; Peries, R. J. Am. Chem. Soc. 1977, £9, 7368. 17. Belmont, D., Doctoral Dissertation, The Ohio state University, 1985. 18. For examples see: Boeckman, R,K.; Heckendorn, D.K.; Chinn, R.L. Tetrahedron Lett. 1987, 2J3, 3551. (b) Rehnberg, N.; Frejd, T.; Magnusson, G. Tetrahedron Lett. 1987, 28, 3589. 19. House, H. Acc. Chem. Res. 1976, 7, 59. 20. Yamamoto, Y.; Yamamoto, S.; Yatagai, H.; Ishihara, Y.; Maruyama, K. J . Org. Chem. 1981, 47, 3263. 21. Kindt-Larsen, T.; Bitsch, V.; Anderson, I.G.; Jart, A.; Munch-Peterson, J. Acta. Scand. 1960, 14, 151. 22. Oppholzer, W.; Loher, H. Helv. Chim. Acta 1981, 64, 2802. 23. Magnus, P.; Quagliato, D. J. Org. Chem. 1985, 50, 1621. 24. Corey, E.J.; Boaz, N.W. Tetrahedron Lett. 1985, 6015, 6019. 25. Johnson, C.R.; Marren, T.J;; Tetrahedron Lett. 1987, 28, 27. 26. Horiguchi, Y.; Matsuawa, S.; Nakamura, E.; Kuwajima, I. Tetrahedron Lett. 1986, 27, 4025, 4029. 27. (a) Taber, D.F.; Gunn, B.F. J. Org. Chem. 1979, 44, 450. (b) Corey, E.J.; Boger, D.L. Tetrahedron Lett. 1978, 5. (c) Wroble, R.R.; Watt, D.S. J. Org. Chem. 1976, 41_, 2939. (d) Pelletier, S.W.; Prabhaker, S. J. Am. Chem. Soc. 1968, 90, 5318. (e) Newman, M.S. J. Am. Chem. Soc. 1953, 75, 4740. 183

28. (a) Scott, W.J.; Crisp, G.T.; Stille, J.K. J. Am. Chem. Soc. 1984, 106, 4603. (b) Labadie, J.W.; Tueting, D.; Stille, J.K. J. Org. Chem. 1983, 4j3, 4634. (c) Labadie, J.W.; Stille, J.K. J. Am. Chem. Soc. , 1968, 90^, 5518. 29. Lipshutz, B.H. Tetrahedron Lett. 1983, 24., 127. 30. Iyoda, M.; Kushida, T.; Kitami, S.; Oda, M. J . Chem. Soc. Chem. Commun. 1986, 1049. 31. See for example ref. 30; also see (a) Crisp, G.T.; Scott, W.J.; Stille, J.K. J. Am. Chem. Soc. 1984, 106, 7500. (b) Stevens, K.E.; Paquette, L.A. Tetrahedron Lett. 1981, 4393. (c) Little, R.D.; Carroll, G.L. Tetrahedron Lett. 1981, 4393. (d) Curran, D.P.; Rakiewicz, D.M. J. Am. Chem. Soc. 1985, 107, 1448. (e) Curran, D.P.; Rakiewicz, D.M. Tetrahedron 1985, 41, 3943. 32. Agosta, W.C.; Wolff, S.J. J. Org. Chem. 1975, 40, 1699. 33. Eaton, P.E.; Carlson, G.R.; Lee, J.T. J. Org. Chem. 1973, 23, 4071. 34. Denmark, S.E.; Jones, T.K. J. Am. Chem. Soc. 1982, 104, 2642. 35. Denmark, S.E.; Jones, T.K. Helv. Chim. Acta 1983, 66, 2378. 36. Denmark, S.E., Jones, T.K. Helv. Chim. Acta. 1983, 66, 2397. 37. Lee, R.A.; McAndrews, C.; Patel, K.M.; Reusch, W. Tetrahedron Lett. 1973, 12.t 965. 38. Taber, D.F.; Petty, E.H.; Raman, K. J. Am. Chem. Soc. 1985, 107, 196. 39. Stork, G.; Rosen, P.; Goldman, N.; Coombs, R.V.; Tsuji, J. J. Am. Chem. Soc. 1965, 82/ 275. 40. Hightower, L.E.; Glasgow, L.R.; Stone, K.M.; Albert son, D.A.; Smith, H.A. J. Org. Chem. 1970, 35, 1881. 184 41. (a) Trost, B.M.; Bogdanowicz, M. J. Am. Chem. Soc. 1973, 9j>, 289. (b) Trost, B.M.; Scudder, P.H. J. Orq. Chem. 1981, 46., 506. (c) Salaiin, J.R.; Conia, J.M. Tetrahedron Lett. 1972, 4J5, 2849. (d) Denis, J.M.; Conia, J.M. Ibid. 1972, 45, 4593. 42. Salaiin, J.R.; Conia, J.M. Tetrahedron Lett. 1972, 45, 2849. (b) Denis, J.M.; Conia, J.M. Ibid. 1972, 45, 4593. 43. Trost, B.M.; Nishimura, Y.; Yamamoto, K.; J. Orq. Chem. 1978, 43., 1012. 44. Ito, Y.; Hirao, T.; Saegusa, T. J. Orq. Chem. 1987, 43, 1011. 45. Shimizu, I.; Minami, I.; Tsuji, J. Tetrahedron Lett. 1983, 24' 1797. 46. Thiessen, R.J. J . Orq. Chem. 1971, 3j5, 752. 47. Carpenter, B. Tetrahedron 1978, 34.' 1877. 48. Danheiser, R.L.; Bronson, J.J.; Okano, K. J . Am.Chem. Soc. 1985, 107, 4579 and references therein. 49. Heathcock, C.; Clark, R.D. Tetrahedron Lett. 1974, 24, 2027. 50. Ihara, M.; Toyota, M.; Fukumoto, K. J . Chem. Soc. Perkin Trans. I 1986, 2151. 51. (a) Davis, F.A.; Stringer, O.D. J. Orq. Chem. 1982, 47, 1774. (b) Davis, F.A.; Vishwakarma, L.C.; Billmers, J.M.; Finn, J. Ibid. 1984, 49, 3243. 52. Vedejs, E.; Engler, D.A.; Telschow, J.E. J. Orq. Chem. 1978, 43, 188. 53. Niwa, H.; Wakamatsu, K.; Hida, T.; Niiyamu, K.; Kigoshi, H.; Yamada, M.; Nagase, H.; Suzuki, M.; Yamada, K. J. Am. Chem. Soc. 1984, 195, 4547. 54. Matsumoto, K.; Sera, A. Synthesis 1985, 1000. 55. Scouten, C.G.; Brown, H.C. J. Org. Chem. 1973, 38, 4092. 185 56. (a) Ramana Rao, V.V.; Devaprabhakara, D.; Chandrase- kran, S. J. Organomet. Chem. 1978, 162, C9. (b) Rao, C.G.; Kulkarni, S.U. Ibid. 1979, 172, C20. (c) Brown, H.C.; Kulkarni, S.U.; Rao, C.G. Synthesis 1980, 151. 57. Omura, K.O.; Sharma, A.K.; Swern, D. J. Org. Chem. 1976, 41, 957. 58. Tsunoda, T.; Suzuki, M.; Noyori, R. Tetrahedron Lett. 1980, 21, 1357. 59. (a) Mimoun, H.; Serce de Roch, L.; Sajus L. Tetra- hedron 1970, 26, 37. (b) Tolstikov, G.A.; Dzhemilev, U.M.; Yut'ev, V.P. Zh. Orq. Khim 1972, 8, 2204. (c) Sharpless, K.B.; Townsens, J.M.; Williams, D.R. J. Am. Chem. Soc. 1972, 94, 295. (d) Arakawa, H.; Morooka, Y.; Ozuki, A. Bull. Chem. Soc. Jpn. 1974, 47, 2958. (e) Sharpless, B.; Michaelson, R.C. J. Am. Chem. Soc. 1973, 95, 136. 60. Hooz, J.; Gilani, S.S.H. Can. J. Chem. 1968, 46^, 86. 61. Bose, A.K.; Bansi, L. Tetrahedron Lett. 1973, 40, 3937. 62. St. Laurent, unpublished results, The Ohio State University. 63. (a) Molander, G.A.; Etter, J.B. J. Orq. Chem. 1986, 51, 1778. (b) Molander, G.A.; Eter, J.B. Tetrahedron Lett. 1984, 25, 3281. 64. Pappo, R.; Allen, D.S.; Lemieux, R.U.; Johnson, W.S. J . Org. Chem. 1956, 21, 478. 65. Evans, D.A.; Illig, C.R.; Saddler, J.C. J. Am. Chem. Soc. 1986, 108, 2478. See also 65a. 66. (a) Ogino, T.; Mochisuki, K. Chem. Lett. 1979, 443. (b) Kon, K.; Isoe, S. Tetrahedron Lett. 1980, 21, 3399. 67. Nakamura, E.; Murofushi, T.; Shimizu, M.; Kuwajima, I.; J. Am. Chem. Soc. 1976, 9£, 2346. 68. Masamune, S.; Ellingboe, J.W.; Choy, W. J . Am. Chem. Soc. 1982, 104, 5526. 186 69. Rubottom, G.M.; Pelegrina, D.R. Tetrahedron Lett. 1974, 4319. (b) Hassner, A.; Ruess, R.H.; Pinnick, H.W. J. Org. Chem. 1975, 40, 3427. 70. (a) Kenny, M.J.; Mander, L.N.; Sehti, S.P. Tetrahedron Lett. 1986, 22, 3927. (b) McCormick, J.P.; Tomasik, W.; Johnson, M.W. Tetrahedron Lett. 1981, 22, 607. 71. Danishefsky, S.; Vaughn, K.; Gadwood, R.; Tsuzukima, K. J. Am. Chem. Soc. 1980, 102, 4262. 72. Seyferth, D.; Vaucrhn. L.G. J. Organomet. Chem. 1963 1, 138. 73. (a) Paquette, L.A. Org. React. 1977, 25, 1-71. (b) Becker, K.B.; Labhart. M.P. Helv. Chim. Acta 1983, 1090. 74. (a) Orban, J. ? Turner, J.V. Tetrahedron Lett. 1983. 24, 2697. (b) Ibid. 1984, 25, 5099. 75. (a) Mori, M.; Oda, I.;Ban, Y. Tetrahedron Lett. 1982, 23, 5315. (b) Mori, M.; Kubo, Y.; Ban, Y. Ibid. 1985, 26, 1519. (c) Mori, M.; Kanda, N.; Bank, Y. J. Chem. Soc. Chem. Comm. 1986, 1375. (d) Piers, E.; Friesen, R.W. J . Org. Chem. 1986, 51_, 3405.