INFORMATION TO USERS
This reproduction was made from a copy of a document sent to us for microfilming. While the most advanced technology has been used to photograph and reproduce this document, the quality of the reproduction is heavily dependent upon the quality o f the material submitted.
The following explanation of techniques is provided to help clarify markings or notations which may appear on this reproduction.
1.The sign or “ target” for pages apparently lacking from the document photographed is “Missing Page(s)”. I f it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure complete continuity.
2. When an image on the film is obliterated with a round black mark, it is an indication of either blurred copy because of movement during exposure, duplicate copy, or copyrighted materials that should not have been filmed. For blurred pages, a good image o f the page can be found in the adjacent frame. If copyrighted materials were deleted, a target note will appear listing the pages in the adjacent frame.
3. When a map, drawing or chart, etc., is part of the material being photographed, a definite method of “sectioning” the material has been followed. It is customary to begin filming at the upper left hand comer of a large sheet and to continue from left to right in equal sections with small overlaps. If necessary, sectioning is continued again—beginning below the first row and continuing on until complete.
4. For illustrations that cannot be satisfactorily reproduced by xerographic means, photographic prints can be purchased at additional cost and inserted into your xerographic copy. These prints are available upon request from the Dissertations Customer Services Department.
5. Some pages in any document may have indistinct print. In all cases the best available copy has been filmed.
University Micrcxilms International 300 N. Zeeb Road Ann Arbor, Ml 48106 8519043
Wiedeman, Paul Edward
A RELAY APPROACH TOWARD THE SYNTHESIS OF ( + )-PLEUROMUTILIN
The Ohio State University Ph.D. 1985
University Microfilms International 300 N. Zeeb Road, Ann Arbor, Ml 48106 A RELAY APPROACH TOWARD THE
SYNTHESIS OF (+)-PLEOROMUTILIN
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree Doctor of Philosophy in the Graduate
School of The Ohio State University
By
Paul Edward Wiedeman, B.S.
* * * * *
The Ohio State University
1985
Reading Committee: Approved by
Dr. Leo A. Paquette ^ Dr. Heinz G. Floss niUr G, Dr. Anthony W. Czarmk ______Adviseiiser Department of Chemistry To Mom, Dad, and Chris
ii ACKNOWLEDGMENTS
I would like to thank Professor Leo A. Paquette for his guidance throughout my graduate studies. His continuous enthusiasm, encouragement, and support during the course of the research presented in this dissertation are greatly appreciated. The partial collaboration with Dr. Philip C.
Bulman-Page was both stimulating and rewarding. The assistance given by colleagues/friends in the Paquette group was invaluable.
I would like to thank the faculty of the Department of
Chemistry of Butler University: Drs. Bob Pribush, Bub
Carlson, Paul Quinney, Joe Kirsch, Bill Scott, and Shannon
Lieb for initiating and fostering my interest in chemistry.
The initial research experiences and desire to pursue a graduate education were both gained from Dr. Jim McCarthy and my other friends at Merrell Dow Pharmaceuticals.
I would like to thank Mrs. Marlene Pease for her patience and expertise in typing the manuscript.
The friendship and support of Tina Kravetz, Rick
Roberts, Gary Drtina, and Phil Page throughout my graduate career has been of inestimable worth. Columbus will always
iii be remembered as a special place due especially to the Neal family, particularly Elaine.
Above all, none of this would have been possible without the love, support, and encouragement of my family. VITA
December 8, 1957...... Born - Kent, Ohio
1980...... B.S., Cum Laude, Butler University, Indianapolis, Indiana
1980-198 1 ...... University Fellow, Department of Chemistry, The Ohio State University, Columbus, Ohio
1981-198 4 ...... National Science Foundation Fellow, Department of Chemistry, The Ohio State University, Columbus, Ohio
HONORS AND AFFILIATIONS
Finalist, Graduate Student Research Competition, Department of Chemistry, 1985. Graduate Student Alumni Research Award, 1984. Honor Society of Phi Kappa Phi. American Institute of Chemists Award, 1980. Cizlak Fellowship, 1978-1980. Outstanding Freshman Chemistry Award, 1977. Member, American Chemical Society.
PUBLICATIONS AND PRESENTATIONS
"A Relay Approach to (+)-Pleuromutilin. I. JDe Novo Synthesis of a Levorotatory Tricyclic Lactone Subunit." Leo A. Paquette and Paul E. Wiedeman. Tetrahedron Lett.. in press.
"A Relay Approach to (+)-Pleuromutilin. III. Direct Degradation of the Natural Product to the Key Diketone Intermediate and its Chemospecific Functionalization." Leo A. Paquette, Paul E. Wiedeman, and Philip C. Bulman-Page. Tetrahedron Lfi±±., in press.
v "Stereospecific Syntheses of the Four Diastereomeric 2- Amino-5-Phenoxycyclopentanols." James R. McCarthy, Paul E. Wiedeman, Albert J. Schuster, Jeffrey Whitten, Robert J. Barbuch, and John C. Huffman. Submitted for publication in J. Orq. Chem.
"An Approach to the Chiral Synthesis of Pleuromutilin. Part I. Total Synthesis of a Key Degradation Product." Paul Wiedeman, Philip C. Page, and Leo A. Paquette. Presented at the 15th Central Regional Meeting of the American Chemical Society, Oxford, Ohio, May, 1983.
FIELD OF STUDY
Major Field: Organic Chemistry
vi TABLE OF CONTENTS
Pass DEDICATION ...... ii
ACKNOWLEDGMENTS ...... iii
VITA ...... v
LIST OF SCHEMES ...... ix
LIST OF TABLES ...... xi
LIST OF FIGURES ...... xii
CHAPTER
I. PLEUROMUTILIN - A STRUCTURALLY UNUSUAL ANTIOBIOTIC
a. Introduction ...... 1
b. Pleuromutilin - Isolation and Biological Activity ...... 3
c. Pleuromutilin - Structural Elucidation ...... 7
d. Biogenesis of Pleuromutilin ...... 13
e. Synthetic Studies on Pleuromutilin ...... 15
f. Strategy - A Relay Approach Toward Pleuromutilin ...... 18
II. SYNTHESIS OF THE FIRST RELAY INTERMEDIATE
a. Construction of the Indanone Nucleus ...... 21
b. Functional Group Differentiation and Resolution of the Indanone Nucleus ...... 31
c. Synthesis of the Lactone Portion of the First Relay Intermediate ...... 39
vii Page
d. Degradative Sequence to the First Relay Intermediate ...... 61
III. PREPARATION OF THE SECOND RELAY INTERMEDIATE
a. Elaboration to the Second Relay Intermediate...... 65
b. Degradative Route to the Second Relay Intermediate...... 78 IV. STUDIES DIRECTED TOWARD COMPLETION OF THE PLEUROMUTILIN FRAMEWORK
a. Introduction to Cyclooctane Formation ...... 83
b. Preliminary Investigations ...... 84
c. The Mukaiyama Approach ...... 87
d. The Photochemistry Approach ...... 93
e. The Free Radical Approach ...... 101
f. The Carbonium Ion and Related Intermediates Approach ...... 105
g. The Anion Approach ...... 112
h. The Car bene Approach ...... 115
i. The Dieckmann Condensation Approach ...... 116
j. S u m m a r y ...... 118
EXPERIMENTAL ...... 120
REFERENCES ...... 254
viii LIST OF SCHEMES
Scheme £ag£
I Degradative Studies of Pleuromutilin ...... 10
II Transannular 1,5-Hydride Shift ...... 12
III 1,5-Hydride Shift Established Relationship of C(3) and C(ll) ...... 12
IV Biogenesis of Pleuromutilin ...... 14
V An Approach to Pleuromutilin via an Anionic Oxy-Cope Rearrangement ...... 16
VI Gibbons' Synthesis of (±)-Pleuromutilin ...... 17
VII A Relay Approach Toward Pleuromutilin ...... 20
VIII Synthesis of 3-Keto Ester 17 ...... 22
IX Attempted Cyclopentenone Annulations ...... 27
X Construction of the Indanone Nucleus ...... 29
XI Functionality Differentiation and Resolution of the Indanone Nucleus ...... 36
XII Formation of the Pendant Lactone ...... 41
XIII "Unnatural" Lactone Diastereomers ...... 46
XIV 3-Elimination Competition with Selenium-Induced Elimination ...... 48
XV Dual Alkylation of a Model Compound ...... 49
XVI An Alternative Ethylidene Lactone Formation Attempt ...... 50
XVII Unexpected Problems with an Ethylidenation Protocol ...... 51
ix Eags
XVIII Completion of the Relay Synthesis ...... 56
XIX Absolute Configurations and Optical Rotations of the Lactones and Lactone Precursors ...... 59
XX Degradative Route to (-)-50 ...... 63
XXI Attempted Lactol Functionalization ...... 68
XXII y-Lactol Functionalization ...... 69
XXIII Dithiane Alkylation Approach ...... 70
XXiv Introduction of the Last Two Framework Carbon Atoms ...... 75
XXV Degradative Sequence to an Advanced Intermediate ...... 80
XXVI Attempted Michael Closure ...... 85
XXVII Reactivity of 68 with Electrophiles ...... 87
XXVIII Differentiation of the Ketone Pair of 68 ...... 90
XXIX Attempted Mukaiyama and Related Ractions ..... 92
XXX The Photochemistry of 76 ...... 95
XXXI Preparation of 86 ...... 99
XXXII Possible Copper(I) Promoted Photocycloaddition ...... 101
XXXIII Synthesis of Free Radical Precursors 88 and 89 ...... 102
XXXIV Sulfonium Ion C hemistry ...... 107
XXXV Carbonium Ion Chemistry...... 110
XXXVI Anion Closure Endeavors ...... 113
x LIST OF TABLES iEaJble £as£ 1. Mukaiyama and Related Reaction Conditions .... 92
2. Photochemical Reaction Conditions ...... 96
xi LIST OF FIGURES
Figure Page
1. Selected Basidiomycotina Metabolites ...... 3
2. Other Pleuromutilins...... 5
3. Stereoselectivity in the Alkylation of 17 .... 24
4. 90 MHz 1H NMR Spectrum of 1 8 ...... 25
5. 90 MHz 2H NMR Spectrum of 2 0 ...... 30
6. 300 MHz 1H NMR Spectrum of 23 ...... 33
7. 300 MHz *H NMR [Eu(hfc)3] Shift Study of 23 .. 38
8. 300 MHz 1H NMR Spectrum of 3 2 ...... 43
9. 300 MHz 1H NMR Spectrum of 33 ...... 44
10. Assignment of the JS-Configuration In 48 and 4 9 ...... 53
11. 300 MHz 1E NMR Spectrum of 4 8 ...... 54
12. 300 MHz 1H NMR Spectrum of 4 9 ...... 55
13. 300 MHz 2H NMR Spectrum of 51 ...... 57
14. 300 MHz *H NMR Spectrum of 50 ...... 58
15. 300 MHz *H NMR Spectrum of <-)-50 via the Degradative Route ...... 64
16. 300 MHz 1H NMR Spectrum of 6 5 ...... 72
17. 300 MHz *H NMR Spectrum of 6 6 ...... 74
18. 300 MHz 1H NMR Spectrum of 6 7 a ...... 76
19. 300 MHz *H NMR Spectrum of 6 8 ...... 77
xii £ag£
20. 300 MHz 1H NMR Spectrum of 71 82
21. 300 MHz 1H NMR Spectrum of 76 89
22. 300 MHz 1H NMR Spectrum of 82 97
23. 300 MHz 1H NMR Spectrum of 89 ...... 103
24. 300 MHz 1H NMR Spectrum of 90 108
25. 300 MHz XH NMR Spectrum of 93 Ill
xiii CHAPTER I
PLEOROMDTILIN - A STRUCTURALLY UNUSUAL ANTIBIOTIC
la. Introduction
To the organic chemist, the term "natural product"
refers to an organic molecule that may be isolated from a source or sources found in either the plant or animal
kingdoms. These natural products have constituted highly suitable targets to the organic chemist in his/ her attempts to further synthetic methodology and strategy. As knowledge has expanded and expertise been refined, the challenges approached have increased in both range and complexity. Besides serving as fasci nating targets for chemical study, natural products frequently exhibit biological activity that is benefi cial to man. Synthesis of the parent compound and select derivatives, particularly if the substance is produced only in minute quantities from its natural source, enables further study and eventually enhances understanding of our environment and its workings about us.
1 One general class of natural products are the terpenes.* These are molecules constructed from the five-carbon subunit, isoprene, leading to cyclic and acyclic structures of great diversity. One abundant source of terpenoid metabolites are fungi and of par ticular interest for this dissertation, the subdivision 2 basidiomycotina. The basidiomycetes are especially fruitful in producing sesquiterpenoids. A few selected examples include coriolin, a hirsutane showing anti biotic and antitumor activity; marasmic acid, an anti bacterial marasmane; sterpuric acid, a compound isolated from the fungus responsible for "silver leaf disease" and representing the sterpuranes; and lacta- rorufin A, an antibiotic lactarane. Amazingly this diverse collection of structural classes are all believed to be formed from farnesyl pyrophosphate via a humulene intermediate (Figure 1). Less prevalent are diterpene metabolites. In fact only one class, the cyathanes, has numerous representatives. They are derived from geranylgeranyl pyrophosphate through a cyclization-rearrangement process. Another diterpene produced by several basidiomycetes is pleuromutilin. corlol1n sterpurlc acid
marasmlc acid TO OH HO OH OH 2 lactarorufin A cy.athln Ag
Figure 1. Selected Basidiomycotina Metabolites
lb. Pleuromutilin - Isolation and Biological Activity 4
Pleuromutilin (1) was isolated in the early 1950's by Kavanagh and coworkers from several species of basi diomycetes: Pleurotus mu.ti.lULS, £. Passeckerianus. and 3 Drosophila substrata. The colorless, crystalline com pound attracted initial attention due to its signifi cant in vitro antibiotic activity against gram positive bacteria and low animal toxicity.
Since then, several other pleuromutilins have been found. The majority involve a modification of the gly colic ester subunit. Thus 14-acetylmutilin (2a) has been obtained in minor amounts while isolating pleuro mutilin. Several derivatives are known in which the hydroxyl moiety of the glycolic ester subunit is esterified by the fatty acids: oleic acid (2b), 9 4 linoleic acid (2c), or A -eicosenoic acid (2d). The same hydroxyl group has also been observed to be involved in a glycosidic linkage with a-D-xylose (2e) when isolating pleuromutilin from a strain of Elipto- pilus pseudo-pinsitus and has been named A40104A.^ Two members of this class of compounds have an additional hydroxyl group in the tricyclic diterpene ring system.
A40104B(3) was isolated as a minor fermentation product in the production of pleuromutilin. The investigators have assigned the additional hydroxyl group to the framework as indicated by mass spectral data but have 5 not specified its location or stereochemistry. Also discovered as a by-product of the fermentation reaction is tetrahydroxy-19-mutilen (4)® (Figure 2).
• ) R-C0C h3 b) R-COCH2OCO(CH2 )7CH-CH(CH2 )7CH3 c) R-COCH2OCO(CH2 )7CH-CHCH2CH-CH(CH2 )4CH3 d> R«COCH2 OCO(CH2 ) 7 CH-CH(CH2 )gCH3
• ) R»C0CH20-<^-0H OR HO OH f ) R-COCH2 SCH2CH2 N E t2
•* OH OH
Figure 2. Other Pleurcautilins
A great deal of effort has been exerted toward understanding the mechanism of action and improving the
7 potency of pleuromutilin. Through the development of semisynthetic derivatives in a systematic investiga tion, the structure-activity relationship has been established. The cyclopentanone ring is essential. The C(ll) hydroxyl group must not be esterified or oxidized. An acyl residue must esterify the C(14) hydroxyl group to produce antibacterial activity. The vinyl group is not necessary. The acyl group deter mines the potency of the derivative. The pleuromutilin derivative currently in veterinary use is tiamulin
(2f). During the course of this investigation/ it was discovered that pleuromutilin is also active against the growth of mycoplasms/ the causative agents of several infectious diseases found in both man and animals.
Pleuromutilin and its derivatives act as protein synthesis inhibitors. The antibiotic binds to ribo somes and impairs peptide chain elongation. Resistance develops to the drug in bacteria in which the ribosomes have been altered and can no longer bind the compound.
An effort continues toward improving the efficacy of pleuromutilin derivatives and to determine the Q unusual chemical reactivity of these compounds. Some recent advances include the understanding that the metabolism of pleuromutilin is initiated by oxidation 8 c of the cyclopentanone ring. Contraction to a cyclo- butanone may alleviate this problem. Alsof inversion of the configuration of the C(6) methyl group reduces the affinity between substrate and enzyme and appar- 8i ently reduces the rate of metabolism of the drug.
None of these improvements would be possible without the elaborate structural elucidation studies performed earlier.
Ic. Pleuromutilin - Structural Elucidation
Shortly after Kavanagh's isolation of pleuro mutilin, Anchel performed a preliminary chemical study g of the compound. The originally proposed molecular formula, C 22H34°5' was confirmed. An easily reduced olefin was noted. The oxygen functionality was estab lished as consisting of a hindered ketonic carbonyl, two hydroxyl groups, and either an ester or lactone.
Subsequently, two research groups headed by Pro fessors Arigoni and Birch independently elucidated the structure of pleuromutilin through elaborate degrada tive studies which will be briefly described.*0 Alka line hydrolysis of pleuromutilin (1) produced glycolic acid and a dihydroxyketone, C20H32°3' which was iater named mutilin (5). It was assumed that no structural changes had occurred during this mild hydrolysis owing to the similarity of the ORD curves of the starting
material and product.
Spectroscopic data for mutilin added to the pic
ture. The infrared spectrum indicated the presence of
a cyclopentanone and a vinyl group besides the two
hydroxyl moieties. The 1H NMR data showed that the vinyl group was attached to a fully substituted center,
and the molecule contained four methyl groups, two
secondary, and two tertiary.
Further chemical studies followed based on this
information. No carbon unsaturation besides the vinyl
group was observed with exhaustive hydrogenation. This was confirmed by the stability of the dihydro compound
to ozonolysis. Therefore, a tricyclic molecule was
established. Bromination of dihydropleuromutilin
diacetate showed the presence of three hydrogen atoms on the a carbonyl sites.
The C 2 0 molecular formula, three rings, and the presence of a vinyl group suggested that pleuromutilin
could be a diterpene derived from a geranylgeranyl pre cursor. This was confirmed by the in vivo incorpora tion of I2-14C] mevalonic lactone, a known terpene precursor, in a tracer experiment. Labeled [l-C*^] acetic acid was similarly employed and appeared at eight sites, one being the vinyl methylene as shown by ozonolysis, on the pleuromutilin framework. Pleuro mutilin was thus determined to be a diterpene, and the remainder of the elucidation studies was predicated on this fact.
Treatment with selenium at 300°C of mutilin and other derivatives all produced 6,7-dimethyl-l-indanone.
Thus, a cyclohexane was determined to be fused to the cyclopentanone, and the position of two of the four methyl groups was established.
The location of the oxygen functionality was investigated through the following set of reactions
(Scheme I). Both hydroxyl moieties were oxidized in mutilin (5) to yield trione 8 indicating that both are secondary. Oxidation of pleuromutilin (1) followed by hydrolysis furnished hydroxy diketone 6. A different hydroxy diketone (7) was formed by sequential acyla- tion, oxidation, and hydrolysis of mutilin. Both were oxidized to the previously obtained trione. The infrared spectra indicated the ketone carbonyl groups to be located on ring(s) larger than five carbons.
Exposure of trione 8 to potassium hydroxide in refluxing ethanol followed by hydrogenation afforded diketo cyclopentenone 10 and established one hydroxyl 10
Scheme I. Degradative Studies of Pleuromutilin
OH
OH
1. H2Cr04 1• Ac20 2. "OH 2. H2Cr04 B. "OH
A
OH
1. KOH, EtOH, A 1. KOH, EtOH, A 2. H2 2. Ho 3. H2Cr04
\
9 10 11 as being 6 to the carbonyl. Similar treatment of 6 coupled with oxidation gave y-lactone 9 which placed the hydroxyls in a 1,4-relationship.
Further stereochemical information was obtained from experiments in which 1,5-hydride shifts were involved. Dihydromutilin monoacetate (11) yielded the anhydro compound 12 when treated with phosphorous oxychloride in pyridine (Scheme II). The vinyl group of mutilin was converted to the vicinal diol 13 with osmium tetroxide. Alkaline treatment produced glyco- laldehyde and the keto diol 14 (Scheme III). The stringent steric requirements for these hydride shifts thus supplied insight into the stereochemistry of the involved sites.
The second site of connection for the eight- membered ring with the indanone nucleus could be either at C(5) or C(6). However, the deshielding observed for the tertiary methyl group could only be accounted for by a C(5)/C(14) bond.
The configuration of the various ring fusions and absolute stereochemistry were established by both research groups with ORD studies on several compounds.
Many other experiments were performed to complete a detailed chemical structural elucidation of 12
Scheme II. Transannular 1,5-Hydride Shift
Scheme III. 1,5-Hydride Shift Established Relationship of C(3) and C(ll)
HCCHoOH 13 pleuromutilin.11 Subsequently, x-ray crystallographic analysis of the monobromoacetate of mutilin verified 12 the relative stereochemistry of pleuromutilin.
Id. Biogenesis of Pleuromutilin
As was indicated in the previous section, determi nation of the biosynthetic pathway along which pleuro mutilin is derived came hand in hand with the struc tural elucidation.10'13 With the realization that pleuromutilin is a diterpene, a scheme was postulated to account for its formation (Scheme IV). A proton initiated cyclization of all-trans geranylgeranyl pyrophosphate produces a labdenyl cation. The indane nucleus is formed by means of the following sequential events: hydride shift, methyl shift, hydride shift, ring contraction, and deprotonation. Closure between the allylic pyrophosphate and the isopropenyl group via a solvolytic process is now possible. The cation is quenched by a 1,5-hydride shift and the addition of water. Secondary oxidation and introduction of the glycolic acid unit complete the biosynthesis of pleuro mutilin. An elaborate set of labeling experiments 14 2 3 involving C, H, and H were used in conjunction with 14
Scheme IV. Biogenesis o£ Pleuromutilin
OPP ‘OPP
PPO
OPP
01
H the degradative studies to confirm this hypothetical sequence.
Ie. Synthetic Studies on Pleuromutilin
Pleuromutilin presents an interesting synthetic target. The structurally unusual diterpene is com prised of a rather rigid tricyclic carbon network arrayed in pseudo-propellane fashion. The five-f six- and eight-membered rings all share two carbon atoms.
The six-, and eight-membered rings also have an addi tional atom in common. Prominent in the structure and presenting a significant synthetic challenge are the eight contiguous chiral centers, seven of which are located on the eight-membered ring. Three of the centers are quaternary. The unusual chemical reacti vity adds to the challenge. to date a prelimi nary approach and one total synthesis of racemic pleuromutilin have been published.
Kahn wished to utilize an anionic oxy-Cope reac tion to construct the eight-membered ring as shown
(Scheme V ) . 1 4 This model study proceeded from cyclo- hexenone in four steps and 20% overall yield. No subsequent work has been published on this approach. 16
Scheme V. An Approach to Pleuromutilin via an Anionic Oxy-Cope Rearrangement
- -
Ri
0 .
>
A much different strategy was employed by Gibbons
in the sole total synthesis of racemic pleuromutilin. * 5
This approach was initiated with a sequential Michael reaction to produce in one step the indane nucleus con taining four of the eight stereocenters found in the parent molecule (Scheme VI). Subsequent transformations led to a key tetracyclic intermediate which upon Grob fragmentation gave the tricyclic ring system of the natural product. Functional group manipulation and construction of the vinyl-methyl quaternary center from a ketone complete this elegant approach to pleuromutilin. Scheme VI. Gibbons' Synthesis of (±)-Pleuromutilin
PhCHO
1. CHoCONHBr
HO OH 14 steps
OBx
OH OH
1. diester IfIcatlon — - » 2. selective .OH OH hydrolysis 18
If. Strategy - A Relay Approach Toward Pleuromutilin
It was decided to address the synthetic challenges of pleuromutilin in a linear manner(Scheme VII). The initial target became construction of the indanone nucleus complete with the methyl groups in the proper configuration. Subsequently, a pendant y -lactone was to be formed stereoselectively with the introduction of two more asymmetric centers and all but two of the requisite skeletal carbon atoms. These last two carbon atoms were to be added next/ thereby supplying an intermediate suitable for closure of the eight-membered ring and with at least two of the remaining four stereogenic centers being installed in the process.
This would leave reduction of one carbonyl and selec tive ester formation with one of the hydroxyl moieties to complete the synthesis of pleuromutilin. A resolu tion was to be performed at an early stage to introduce optical activity into the synthesis.
After a review of the degradative studies (Scheme
I)f it became apparent that a few of the proposed synthetic intermediates were amazingly similar to some of the compounds obtained via degradation. It was felt that the degradation scheme could be modified to supply a second source of these key intermediates which will
be referred to as relay compounds.
This approach offers several advantages. First?
it permits a comparison of synthetically prepared material with that degraded from a source of known absolute configuration. This insures that the proper » enantiomer is being dealt with and that the asymmetric centers have been properly established. Second, since the degradative schemes are shorter, a secondary source of advanced intermediates becomes available to assist in completion of the synthesis. Finally, as the degra dative routes to the relay intermediates are shorter, the relay intermediates should become available at an earlier time than would be practical through the synthetic route. This allows exploration of advanced stages of the synthetic sequence while earlier phases of the synthetic route are being developed. In short, a relay synthesis is one that involves merger of a practical synthesis of an advanced intermediate with a
short degradation of the natural product . 1 6 Scheme VII. A Relay Approach Toward Pleuromutilin
C09Et ' CO-Et
J
OR J
\ -
OR CHAPTER II
SYNTHESIS OF THE FIRST RELAY INTERMEDIATE
Ila. Construction of the Indanone Nucleus
Pleuromutilin may be considered as a substituted
indanone with an elaborate five carbon bridge connect
ing C(5) and C(9) as designated by Arigoni's numbering
system. Accordingly, an efficient synthesis of the
indanone nucleus became the initial goal.
These efforts commenced from the cyclohexanone
carboxylate 17, a compound previously described in the 17 literature. This substance is obtained from a three-
step process involving the relatively inexpensive
starting materials ethyl acetoacetate and ethyl croto- nate (Scheme VIII). A Michael condensation between the
starting materials followed by cyclization furnished
diketo ester 15 in good yield (74%). This substance, which exists primarily as its vinylogous acid tautomer, was subsequently transformed to the corresponding vinylogous acid chloride upon treatment with phos phorous trichloride. Use of 1.2 molar equivalents of
21 22 phosphorous trichloride instead of the 0.33 molar equivalent specified in the literature nearly doubled the reaction yield to a more acceptable 54% level.
Vinylogous acid chloride 16 was reduced to 3 -keto ester
17 on a Paar hydrogenator in the presence of a palla dium catalyst. Excellent yields were obtained only when a proton sponge such as triethylamine was placed in the reaction mixture to neutralize the hydrogen chloride produced during the reduction. In the absence of a suitable baser a significant quantity of polymeric material/ which was assumed to arise from an acid catalyzed process/ was isolated.
Scheme VIII. Synthesis of 8 -Keto Ester 17
E t O H , A 15
EtOH, NEt3 16 17 23
A more recent procedure has been published which 18 would reduce the sequence by one step. The essential difference is the use of crotonaldehyde to replace ethyl crotonate. This eliminates the vinylogous acid to acid chloride transformation. However, the isola tion of the cyclized intermediate is complex and the reported yields are only slightly greater than that for the established procedure. Therefore, this route was not investigated.
At this point, the second methyl group found on the cyclohexane ring of the natural product was intro duced stereoselectively. The selectivity of the alkylation was based on the precedent established by 19 20 Piers and Evans for intermediates in the syntheses of aristolone and bakkenolide-A, respectively. In both cases, alkylation of 2,3-dimethyl-6-u-butylthiomethyl- enecyclohexanone afforded an approximately 4:1 mixture of cisto trans-methyl products. In our case, the methyl group was added, and a reversal of the cis/trans ratio of the methyl substituents was anticipated.
The transition state characteristics of the anionic species are important here. The cyclohexanone ring should be somewhat flattened due to the additional stabilization provided by the carboethoxy moiety. It 24
was anticipated that the 3-methyl group would occupy a
pseudo-axial site, and that alkylation would occur to
provide the desired trans-dimethyl compound (Figure 3).
P-CH
Figure 3. Stereoselectivity in the Alkylation of 17
In practice, 3-keto ester 17 was treated with so
dium hydride in dimethylformamide. Subsequent exposure
to methyl iodide resulted in quantitative conversion to
alkylated products. GC analysis indicated that the
major component comprised 80.5% of the product mixture
and it was assumed to be the desired iranE compound 18
(Figure 4). The cis-isomer accounted for 12%, and the
remaining 7.5% was apparently polyalkylated material.
Separation was achieved on a large scale by spinning
band distillation. The distillation fractions
containing unacceptable levels of undesired isomers were pooled and later recycled through the purification
procedure. T T T
Figure 4. 90 MHz XH NMR Spectrum of 18 26
With the cyclohexanone carboxylate in hand, efforts were directed next toward annulation of a cyclopentenone ring. The protocols investigated involved nucleophilic addition to, the ketone with a three carbon unit which would then be cyclized to yield the sought-after material.
The procedure initially tried involved Trost's 21 diphenylsulfonium cyclopropylide chemistry. This reagent was expected to produce an oxaspiro group which upon treatment with lithium diethylamide and workup would yield a protected vinyl cyclopropanol. Subse- 22 quent pyrolysis and oxidation would furnish the desired cyclopentenone (Scheme IX). Unfortunately, despite numerous attempts involving a variety of conditions, the sulfonium ylide failed to add to the carbonyl of 18.
Our next endeavors focused on Nazarov cycliza- tions. The first nucleophile employed was 1,1- 23 dichloroallyllithium. Frustratmgly, only starting material was recovered from these reactions (Scheme
IX). 27
Scheme IX. Attempted Cyclopentenone Annulations
OSIMe
The brevity of a Nazarov cyclization maintained interest in this transformation which soon proved fruitful. It was possible that the ketone carbonyl of
18 was more hindered than originally believed. The procedure which proved to be most effective was that 24 introduced by Raphael. The Grignard of the tetra- 25 hydropyranyl ether protected propargyl alcohol added selectively to the ketone to produce a diastereomeric mixture of alcohols 19 in 71% yield. These were not separated and need not be purified for the next step.
This reaction seemed sensitive to the solvent used.
Yields were much higher when ether was the major solventr but some tetrahydrofuran was required to maintain solubility of the magnesium salts. Ensuing 28 treatment with 60% sulfuric acid in ethanol supplied cyclopentenone 20 in 59% yield (Scheme X and Figure 5).
Shorter reaction times resulted only in deblocking of
the protected alcohol to give 2 1 , suggesting this may be the first step of the many involved in this process.
When 21 was resubmitted to the reaction conditions, it
afforded the desired 2 0 .
Construction of the indanone nucleus was thereby completed. The five- and six-membered rings had been installed containing twelve of the twenty skeletal carbon atoms of the natural product in just three steps from the known starting material with two of the requi site asymmetric centers established. The carboethoxy moiety provided access for addition of the remaining framework subunits, and the cyclopentenone represented the latent functionality to be used for closure of the eight-membered ring at C(9). 29
Scheme X. Construction of the Indanone Nucleus
CO.Et NaH, DMF BrMgCsCCH20THP CO^Et + Mel Et20-THF 17 18
OH
20
OH Figure 5- 90 MHz NMR Spectrum of 20
U) o 31 lib. Functional Group Differentiation and
Resolution of the Indanone Nucleus
Next, efforts were directed towards differentia tion of the two carbonyl groups contained within inda
none 2 0 , in preparation for elaboration of the carbon skeleton. The enone of 20 proved recalcitrant towards protection as any oxygen-containing acetal. Only starting material was recovered from reactions with ethylene glycol and 1,3-propanediol with a variety of protic catalysts. No reaction was observed in attempts to form the dimethyl acetal from trimethyl orthofor- 26 mate, likewise with an assortment of catalysts.
Similar results were obtained under neutral conditions
employing 1 ,2 -bis(trimethylsiloxy)ethane and tri- 27 methylsilyl tnfluoromethanesulfonate.
However, experiments involving the more nucleo- philic 1,2-ethanedithiol efficiently provided 22 (98%)
(Scheme XI). Apparently, the increased nucleophilicity of sulfur over oxygen was paramount to the success of this transformation. It was hoped that a dimethyl thioacetal could be formed to provide a more distinc tive NMR handle than the ketal produced with ethanedithiol. However, the utilization of 32 methanethiol under a variety of reaction conditions, or of thiomethyltrimethylsilane, did not supply the desired material indicating the importance of entropy factors for an efficient reaction in this case. It should be mentioned that the aforementioned ketaliza- tion was affected with protic catalysis in refluxing benzene with concomitant removal of water. Lewis acids were completely ineffective as catalysts for promoting this reaction.
With the carbonyls thus differentiated, reduction with diisobutylaluminum hydride gave 23 (93%) (Scheme
XI and Figure 6 ). Other reducing agents were investi gated, but none matched the efficiency of diisobutyl aluminum hydride. Obtaining this alcohol was very important as it became the compound from which optical activity would be introduced into the synthetic sequence.
A resolution was to be performed at a relatively early stage in the synthetic sequence to separate the two enantiomers in the racemic mixture produced so far.
This is important from a biological viewpoint since enantiomers frequently exhibit different activity in living systems. HO
ft aIT
I . ■ I ,____ l . t____ . . I t i >- -- i. ■- I i J-----t-n- - » .1 .______t__ - i - 7.0 6.H 5.0 5. 2 4. b 4.0 J . 4 2 7 0 2.2 1 . b 1.0 . 2
Figure 6 - 300 MHz H NMR Spectrum of 23 OJ U) 34
Initially, this resolution was to be achieved via
the corresponding carboxylic acid of 2 0 either as an ester with an optically active alcohol or as an ammo nium salt with an appropriate amine. The ester proved 29 resistant to hydrolysis or nucleophilic cleavage.
Potassium hydroxide in ethanol-water mixtures resulted only in some decomposition of the starting ester.
Lithium iodide in dimethylformamide, sodium chloride in dimethylsulfoxide-water, sodium phenylthiolate in ether, trimethylsilyl iodide in acetonitrile, and potassium phenylselenide in tetrahydrofuran all failed to cleave ester 20. Some of these methods may have worked better with the corresponding methyl ester.
However, this was not pursued due to the results about to be described in resolving alcohol 23. Carboxylic acid 24 was attained by refluxing 20 in a mixture of 30 acetic acid-hydrochloric acid-water but only in 50% yield (Scheme XI). This was determined to be too great of a loss to accept at this point and the route was abandoned.
Instead, resolution of alcohol 23 through a suitable carbamate was anticipated to resolve the dilemma. The first isocyanate employed was that derived from (+)-endo-bornylamine hydrochloride and 35
phosgene, a reagent previously used successfully in 31 this research group. With an excess of this
isocyanate, reaction proceeds slowly to completion over
approximately one week. The diastereomeric carbamates
attained were not readily separable by chromatography.
This requirement was successfully met in the
carbamates from (£)-(+)-a-methylbenzyl isocyanate and 32 alcohol 23. This isocyanate may be isolated but was more conveniently reacted directly with 23 to form 31 33 carbamates 25, and 26 ' which proved to be separable by HPLC on silica gel using recycling and band shaving techniques (Scheme XI).
Typically, strong acids or strong bases are used to hydrolyze carbamates. Acidic conditions were immediately ruled out because of the sensitivity of the thioketal, and basic conditions were disappointing in yielding little hydrolyzed material. Pirkle's tri- chlorosilane-induced cleavage delivered the 34 enantiomeric alcohols 23.
Neither enantiomer was obtained optically pure by this resolution. The extent of purity was determined by a NMR study employing tris13-(heptafluoropropyl- hydroxymethylene)-d-camphorato]europium(III) as the chiral shift reagent. The shift reagent was added in 36
Scheme XI. Functionality Differentiation and Resolution of the Indanone Nucleus
CCLH HCl-H0Ac-H,0 24
20 22
HO 23
OCH |L.h PhCH3 , DMAP, A
H_C—
HS1C13, NEt3 HS1C13, NEt3 PhCH3, A PhChU* A
OH HO (+>-23 37
small aliquots to effect gradual shift of the methyl
singlet downfield. At 15 molar percent [EufhfcJj], the
singlets were sufficiently resolved to be integrated
. with a planimeter (Figure 7). The enantiomeric mixture
enriched with the levorotatory isomer was determined to
have 56% ee. The dextrorotatory enantiomer was
enriched to an extent of 78% ee. Lacking sufficient
material to complete the synthesis, the dextrorotatory
enriched material was diluted with racemic 23 to pro
vide material of 46% ee. For convenience, the rota
tions reported for the optically active compounds
leading to the first relay intermediate are extrapo
lated to optical purity, on the assumption that no
optical activity is lost. The experimental and
extrapolated values are reported within Scheme XIX. 25 The IalD established in this manner for the
enantiomers of 23 is 38.6°. OH
(+)-23 HO (+)-23
(-)-23 (-)-23
X _L X x X
u> Figure 7. 300 MHz 1H NMR [Eu(hfc)3l Shift Study of 23 00 39 lie. Synthesis of the Lactone Portion of the
First Relay Intermediate
Each enantiomer was subjected to the subsequent reactions. To minimize confusion while discussing the reaction sequence/ only the enantiomer with the abso lute configuration of pleuromutilin will be depicted.
A summary of the optical rotations is given within
Scheme XIX and full details for the synthesis of each are given in the Experimental section of this thesis.
It should be mentioned that the absolute configuration of each enantiomer was not known until proper comparison was eventually made at the first relay.
The subsequent goal became construction of a Y- lactone from neopentyl alcohol 23 in order to take advantage of the high degree of stereoselectivity observed in alkylations of such species. Oxidation of the alcohol to the corresponding aldehyde would allow nucleophilic addition of an organometallic reagent to produce directly the necessary secondary alcohol found in the lactone as well as the requisite carbon atoms.
Collins oxidation became the method of choice to 35 furnish aldehyde 27 (80%). Other chromium oxidizing agents gave consistently lower yields/ and the Corey 40
and Swern oxidation procedures resulted in significant
decomposition, apparently arising from the lability of
the thioketal under such conditions.
Molecular models indicated that there should be
little preference exhibited for either the primed or
doubly primed conformations of 27. Both structures,
however, were clearly seen to be blockaded to attack
from the direction cofacial with the thioketal. In
line with this analysis, a 1:1 mixture of 28 and 29 was
obtained from addition of the functionalized Grignard
reagent (92% combined yield) (Scheme XII). These
diastereomers were readily separated
chromatographically.
Just as with the absolute configuration of the
enantiomers of 23, the configuration of the secondary
hydroxyl moiety of 28 and 29 was not known initially.
Both were carried through the subsequent reaction se
quence, and all were assigned an absolute configuration
upon reaching the relay intermediate. It should be
noted that four compounds were in hand, both enantio
mers of two diastereomeric pairs. Once again, to minimize confusion the transformations soon to be
discussed will be depicted only with the absolute
configuration of pleuromutilin. No difference was 41
Scheme XII. Formation of the Pendant Lactone
27* THF JO
OH
OH
28 29 TsOH aq acetone
OH
30 32
34 observed in reactivity regardless of the enantiomer or diastereomer used in any of the reactions leading toward the relay intermediate.
The critical lactone ring was now elaborated by 36 acid hydrolysis of 28 and subsequent Jones oxida- 37 tion. The intermediate lactol 30 was not purified because of its instability to silica gel. Due to the presence of 1/3-propanediol liberated during hydroly sis/ titration with Jones reagent to a persistent end point proved most efficacious in supplying lactone 32
in up to 58% yield for the two-step sequence (Figures 8 and 9). Unfortunately/ the thioketal protecting group was also removed during hydrolysis of the 1,3-dioxane/
but re-exposure to 1 f2 -ethanedithiol under the pre viously employed reaction conditions once again resulted in suitable differentiation of the enolizable sites (Scheme XII).
All that remained for completion of the synthesis of the first relay intermediate was a double alkylation to form the quaternary center at the a-position of the lactone. Already in 29, the configuration of the hydroxyl group has been inverted from the natural configuration/ a task not readily done via degradation of the natural product. This could potentially prove A A 1.14
PPM
Figure 8 300 MHz 1H NMR Spectrum of 32 I H
i .
Figure 9. 300 MHz *H NMR Spectrum of 33 lU important owing to the significance of this site with 7 regard to biological activity. Another opportunity presented itself for formation of "unnatural" analogs in producing this quaternary center. The ^-position of the lactone was purposely left unsubstituted to take advantage of the great degree of diastereoselectivity observed in the alkylation of five-membered endocyclic enolates on the face opposite of that to a substituent 3 8 in the y~position. The first series examined was used to produce the inverted configuration at the lactone site.
This sequence was developed using Y-butyrolactone as a model compound. Following methylation, «-methyl-
Y-butyrolactone was treated with phenylselenoacetal- 39 dehyde under the conditions developed by Clive and 40 Kowalski to introduce a vinyl group and form 36.
This protocol was chosen over other vinyl cation 41 synthons because of the relatively mild conditions involved and compatibility with functionality in the actual system.
Hence, the lithium enolate of 34 was alkylated with methyl iodide to yield 38 with none of the epimer observed. Treatment with phenylselenoacetaldehyde and subsequent elimination established the desired 46 quaternary center in 40 (Scheme XIII). In the methyl
ation sequencer the presence of hexamethylphosphoramide
(HMPA) improved yields . 3 8 ' 4 2 Excess methyl iodide
apparently alkylated the sulfur atoms of the thioketal
and induced eventual hydrolysis during aqueous extrac
tions in the workup, resulting in recovery of the cor
responding enone and low yield of the sought product.
The diastereomer 39 was derived from 35 under identical conditions.
Scheme XIII. "Unnatural” Lactone Diastereomers
1. LDA
2. Mel-HMPA CH
34 38
1. LDA
2. PhSeCH2CH0 CH 3. CH3 S02C1, NEt3
40
1. LDA 1. LDA
2. Mel-HMPA H 2. PhSeCH2CH0 35 3. CH3S02C1, {■pi, NEt, H 39 47
A simple reversal of the above procedure was
anticipated to produce the relay intermediate. If the
vinyl group should equilibrate after formation to yield
an ethylidene lactone/ deconjugative alkylation was
still expected to occur on the lactone face opposite to
the Y-substituent. The Clive-Kowalski procedure
appeared to produce a mixture of vinyl and ethylidene
lactones which could be separated. Upon alkylation of
the mixture/ 43 was the major product isolated (Scheme
XIV). P-Elimination was obviously competing with
selenium induced elimination which has been observed by
Kowalski for unhindered ketones/ and what had been
accounted for as selenium residues were in fact 41/ the
sometimes major component/ and 42. Because of a lack
of a suitable method for removal of the vinyl sele- 43 nxde and an unwillingness to introduce an unnecessary
step/ this route was abandoned.
Two different methodologies soon lent themselves
to our needs. In a very straightforward manner/ the 44 lithium enolate of model compound 44 was condensed 45 with acetaldehyde. The intermediate aldol product was not isolated but instead was immediately converted
to its mesylate and eliminated with 1 /8 -diazabicyclo-
[5.4.0]undec-7-ene (DBU) to ethylidene lactone 45. The 48 yield was acceptable, but a more elegant procedure was desired.
Scheme XIV. 8 -Elimination Competition vith Selenium-Induced Elimination
0 1. LDA______o 0
°\ / 2. PhSeCH2CHO + 3. CH3 S02C1, NEt3 42 . R=H
41b R=SePh 0 t r 1' LDA ^ / CH3 2. Mel, HMPA
43 *SePh
A very promising possibility was the one-pot 46 conversion pioneered by Tanaka. Treatment of 44
sequentially with 2 . 2 equivalents of diisopropyl amide, bislmethoxy(thiocarbonyl)Idisulfide, and acetaldehyde directly gave 45 identical spectroscopically with that previously prepared but in a somewhat higher yield
(Scheme XV). Other protocols were either investigated and found less desirable than those already examined or determined to be incompatible with existing function- 47 ality. Ethylidene lactone 45 was deconjugatively methylated with lithium isopropylcyclohexylamide, methyl iodide, and HMPA to yield 46 having the desired quaternary center. The choice of base is rather criti cal as less hindered bases may add in Michael fashion to the ethylidene lactone.45b,47c
Scheme XV. Dual Alkylation of a Model Compound
1. LDA s 1. LICA 0 0 0 2. CHeOCS-+2 3. CHjCHO
1. LDA \ I I \ 1 I 2. CH3CH0 °? '+ t 1 3 . ch3 so2c i 45 46 44 4 . DBU
With the realization that the directing effect of the lactone ring was not necessary for introduction of the ethylidene group, an alternative in which this functionality would be incorporated in construction of the lactone warranted consideration. The crux of this approach involved opening of a terminal epoxide with the dianion of crotonic acid and subsequent cyclization 48 of the intermediate Y-hydroxy acid. Therefore, aldehyde 27 needed to be transformed to an epoxide with an additional methylene group (Scheme XVI). Repeated efforts did not furnish this material, despite use of a 49 variety of reagents and conditions. The steric hindrance encountered with attempted ester hydrolysis
at this center had returned. Accordingly, the
previously established methodology was reinstated.
Scheme XVI. An Alternative Ethylidene Lactone Formation Attempt
Lactone 34 was submitted to the elegant sequence developed by Tanaka for synthesis of a-alkylidene-Y- butyrolactones which had been so successful in the model system. However, upon addition of bis[methoxy=
(thiocarbonyl)] disulfide the reaction mixture became dark and turbid immediately, a result previously not observed. The only product recovered, albeit in low yield and not consistently from run to run, was the
allylic thiol 47. It appears that a transient
episulfide was not collapsing to the desired product
but was experiencing ring opening (Scheme XVII). The
thioketal, the functionality not represented in the model system, was not compatible with these reaction
conditions.
Scheme XVII. Unexpected Problems With An Ethylidenation Protocol
1. LDA
34
•5H
47
Ethylidene lactone 48 was formed by condensation of the lithium enolate of 34 with acetaldehyde. Subse 45 quent ^-elimination via the mesylate afforded the long sought after ethylidene lactone (Figures 11/12 and
Scheme XVIII). The £-configuration has been assigned based on comparison of the *H NMR data of 48, 49, and
45 with the £- and E-a-ethylidene-yvalerolactones 50 reported in the literature. The vinyl methyl of the
E-isomer is held in the deshielding region of the carbonyl and has a shift of 2.09 ppm. The E-isomer1s vinyl methyl is not under this influence and appears at
1.82 ppm. The chemical shifts of the vinyl methyl groups of 45, 48, and 49 are seen at 1.84, 1.83, and
1.85 respectively, all correlating well with the data for the E-isomer (Figure 10).
For completion of the relay synthesis, ethylidene lactone 48 was deconjugatively methylated by exposure to lithium isopropylcyclohexylamide in the presence of
HMPA and subsequent addition of methyl iodide (Scheme
XVIII). This sequence was implemented to take advan tage of the facial stereoselectivity observed in alkylation of butyrolactones on the face opposite to the y-substituent. Only a single levorotatory stereoisomer 50 (Figures 13 and 14), identical in all respects with the degradation product described in the adjoining section, was isolated in up to 51% yield.
Comparison of the degradation product enabled 53
CH CH 1.83 ppm Figure 10. Assignment of the £-Configuration In 48 and 49 assignment of the absolute configuration of the r- position of the lactone which could not be done pre viously. From this assignment/ the configuration at the a-position follows. The structures and optical rotations of the enantiomers of the diastereomeric pair of compounds carried through this sequence are presented in Scheme XIX. c/1 Figure 11. 300 MHz NMR Spectrum of 48 J t 1 Figure 12. 300 MHz 1H NMR Spectrum of 49 56 Scheme XVIII. Completion of the Relay Synthesis LDA * NEt 48 LI y i-o HMPA-Mel J______,_____ j . L Figure 13. 300 MHz H NMR Spectrum of 51 U1 -j Figure 14. 300 MHz NMR Spectrum of 50 Ul 00 59 Scheme XIX. Absolute Configurations and Optical Rotations of the Lactones and Lactone Precursors 25 26 [+28.8°] [+13.1°] (+)-23 f—23 +38.3°[+17,4°] 38.9 [-21.8 ] (+)-27 C—)—27 +111.3°[+51.2°] -111.7°[-62.6°] M (—)—28 29* (+)-28 -0.5°[-0.2°] +8.5°[+3.9°] +16.2°[+9.1°] +4.6°[+2.6°] OH 30*' 31*' 31* 30* Note: The bracketed values were obtained experimentally, and those outside the brackets are extrapolated to optical purity. 60 Scheme XIX. (Continued) 30*' 31*' 31* 30* (-)-32 (+)-33 (-)-33 (+)-32 -14.6°[-6.6°] +38.3 °[+17.6°] -37.5°[-21.0°] +20.5°[+11.5°] H V - > (+)-34 (+)-35 (-)-35 (-)-34 +5.7 °[+2.6 °] +10. 3 ° [ +4.8 ° ] -14.5°[-8.1°] -1 2 .0 °[-6 .7 °] (-)-48 (+)-49 (-)-49 (+)-48 +36.3°[+16.7 °] -41.6°[-23.3°] +27.5°[+15.4 °] (-)-50 (+)-51 (-)-51 (+)-50 -10 .0 °[-4 .6 °] +6.3 ° [ +2.9 ° ] -14.0°[-7.8°] +8.9 ° [+ 5 .0 °] relay lid. Degradative Sequence to the 51 first Relay Intermediate As was mentioned in the introduction, a relay synthesis offers several advantages. If none of the asymmetric centers are perturbed in degradation of the natural product to a relay intermediate, verification of absolute configuration is possible. If the degra dative sequence, which should be shorter in length, is completed sooner than the synthetic route, investiga tion of advanced stages of the synthetic path becomes possible at an earlier time. Finally, degradation to relay intermediate provides a secondary source of the material. All of these were taken advantage of in pursuit of the first relay intermediate. 11a In a review of Arigoni's degradative work, it was obvious that lactone 9 was very similar to lactone 50, a proposed compound of the synthetic sequence (Scheme I). A modification of this degradation sequence would provide lactone 50 at a much earlier period than would be practical for the ite novo synthesis of levorotatory tricyclic lactone 50. In actuality, only four laboratory manipulations are required to achieve this end result (Scheme XX ) . 1 6 Oxidation of pleuromutilin (1) with pyridinium chlorochromate efficiently (8 6 %) gave pleuromutilone (52) without effecting the glycolic ester function ality. Tiamulin (2f) was similarly treated to provide tiamulone (53) (97%). When either 52 or 53 were heated at the reflux temperature with 1 0 % potassium hydroxide in ethanol a three-step process occurred. The glycolic ester was saponified, the cyclooctanone ring was rup tured via a retrograde Michael reaction, and intra molecular cyclization of the hydroxy diketone so gen erated yielded hemiacetal 54 (83%). The hemiacetal was converted to tricyclic lactone 55 (60%) by oxidative cleavage of the ethyl group. This process, encountered earlier with other tertiary lactols,**a'b probably pro ceeds by C— > 0 migration of the ethyl group to the chromate ester oxygen with ejection of HCrO^”. Water or the equivalent adds to the oxonium-stabilized intermediate which eventually loses the ethyl carbons as acetaldehyde to produce 55, l a ] ^ * -16.9° (CHCl-j). Dithioketalization under the conditions previously developed provided levorotatory 50 (78%), identical in all respects with the same material, (-)-50, obtained via the novo synthetic route (Figure 15). 63 This degradative study established the absolute configurations of the enantiomers of 50 and 51. As a result of its brevity, appreciable quantities of this tricyclic lactone were available before the de novo synthesis was completed and permitted examination of the synthetic sequence presented in the next chapter. Scheme XX. Degradative Route to (-)-50 \ PCC 10* KOH CH 2 C 1 ^ EtOH, A OR OR 52 1 r«coch2oh 53 Zf R=C0CH2SCH2 CH 2 NEt2 PCC hsch2 ch2sh CH 2 C12 TsOH, CgHg» A H H I 54 55 Figure 15. 300 MHz *B NMR Spectrum of (-)-50 via the Degradative Route CHAPTER III PREPARATION OF THE SECOND RELAY INTERMEDIATE 51 Ilia. Elaboration to the Second-Relay. Intermediate With completion of the synthesis of the levorota- tory tricyclic lactone 50, many of the synthetic challenges found in pleuromutilin have been addressed. Eighteen of the twenty skeletal carbon atoms have been established in their proper absolute configuration. This accomplishment includes two of the three rings and two of the three quaternary centers found in the natural product. Half of the stereogenic centers have been incorporated, and all of the oxygen atoms of the mutilin framework are intact. The primary task remaining to confront is introduction of the last two carbon atoms and cyclization to form the eight-membered ring. Functional group manipulation and possibly an epimerization would complete the synthesis of pleuromutilin. Introduction of the remaining skeletal carbon atoms was immediately approached. Thus, addition 65 66 of ethyllithium gave tertiary lactol 56 (90%) (Scheme XXI). Unfortunately, all attempts to trap the hydroxy ketone tautomer of this hemiketal failed. No success was observed with a variety of silylation and esterifi- cation conditions, and destruction of starting material occurred with exposure to ethanedithiol under Dean- Stark conditions. These results may arise from the high degree of steric congestion in the oxygenated region of the molecule. To circumvent this situation, recourse was made to an examination of the reactivity of the corresponding secondary Y-lactol. Accordingly, lactone 50 was reduced with diisobutylaluminum hydride to give 57 (98%). Interestingly, when methanol was used in the workup to quench excess diisobutylaluminum hydride, the alcohol exchanged with water to produce acetal 58 to an appreciable degree. Turning to an entirely aqueous workup prevented the conversion. Conditions for the functionalization of secondary Y-lactols which had proven unfruitful with 56 were reconsidered. These compounds are somewhat ambiguous in their reactivity. There is precedent for both 52 capture of the hydroxy aldehyde tautomer as well as 53 conversion of the hemiacetal to the acetal. Exposur to either acid chlorides or silyl chlorides may have produced the protected lactols, but the products were too unstable to be fully characterized. When silyl triflates were employed, the same methyl acetal 58 reappeared with a methanolic workup (55%). The remarkable facility of this reaction prompted exploitation of these experimental results (Scheme XXI). Consequently, attempts were made at dioxolane formation with ethylene glycol and a protic catalyst. The lactol was again not captured in its open form but transformed to acetal 59. The question then arose as to whether the greater nucleophilicity of sulfur could again be of assistance* Ethanedithiol was substituted for ethylene glycol and afforded 60 (91%). Pyridinium p-toluenesulfonate was the catalyst employed in this reaction, and the same result was obtained with p-toluenesulfonic acid 68 Scheme XXI. Attempted Lactol Functionalization OH 56 OR EtL 1 Et,0 0 1. i-Bu9AlH OH 50 57 1. (t-Bu)Me2SlOTf 1. 1-Bu 2 A1H 2. MeOH lutldlne 2. MeOH OCH, OH 58 59 54 and boron trifluoride. A review of the literature suggested that the appropriate Lewis acid may be more efficacious in producing the desired dithiolane. This in fact was borne out. Dithiolane 61 was produced from 57 upon treatment with ethanedithiol in 55 the presence of aluminum chloride (58%). Nixed acetal 60 was also converted to 61 with the same catalyst and without additional ethanedithiol (Scheme XXII). Titanium tetrachloride was frequently the best catalyst for this transformation with other Y-lactols but in this case yielded a 2:1 mixture (69% overall) of deblocked enone 62 and 61. Scheme XXII. y-Lactol Functionalization ■SH 60 A1C1 OH U "I AT Cl 57 OH 61 HSCH,CH,SH T1C1 OHOH 62 70 In order to introduce the remaining framework carbon atoms as efficiently as possible, dithiane 63 (49%) was prepared from 57 with 1 ,3-propanediol in the presence of titanium tetrachloride. It was anticipated that the hydroxyl group of 57 could be protected, and that the dithiane anion could then be alkylated (Scheme XXIII). However, attempted protection of the alcohol resulted in decomposition of the starting material. Scheme XXIII. Dithiane Alkylation Approach OH HSCHoCHoCHo SH T1C1 57 63 OH NaH PhCH2Br 1. base 2. EtX Concurrently, efforts were directed toward protection of the hydroxyl moiety in 61. Initial results were not especially promising. N-±-butyl- 56 dimethylsilyl, N-methyltrifluoroacetamide, a reagent reported to be efficient in the silylation of hindered alcohols, and Jt-butydimethylsilyl chloride with 4-dimethylaminopyridine gave no reaction at all. Use 57 of i-butyldimethylsilyl triflate may have produced the desired protected alcohol but the crude product decomposed upon attempted purification and was never C Q fully characterized. Reaction with benzyl triflate rapidly produced a single compound 64 (81%). However, spectral data confirmed that reversion to a benzylated mixed acetal had occurred. Eventually, success was attained by deprotonation of 61 with methyllithium and alkylation of the alkoxide at low temperature with only a slight excess of methyl iodide yielding 65 (90%) (Figure 16). With the hydroxyl moiety blocked, the carbonyl groups could be unmasked without fear of intramolecular cyclization. All attempted hydrolyses with heavy metal thio- 59 philes resulted in destruction of the starting material. Oxidative hydrolysis employing iodine or N-bromosuccinimide^ 0 also destroyed the starting material. Deprotection was successfully achieved via 5-methylation and subsequent hydrolysis with excess OCH -f J. I PPM Figure 16- 300 MHz NMR Spectrum of 65 -j to 73 methyl iodide in aqueous acetone ® 1 (Figure 17 and Scheme XXIV). The stage was once again set for the introduction of the remaining two skeletal carbon atoms. Despite the neopentyl nature of the aldehyde carbonyl group in 6 6 / chemoselective addition of one equivalent of ethylmagnesium bromide in ether at -100°C was observed to give largely the desired secondary alcohol 67 (Figure 18) (47% after chromatography) as a separable diastereomeric mixture. Apparently# this selectivity is possible because of the equally congested surround ings of the cyclopentenone moiety. Use of ethyllithium was also investigated# but those reactions yielded a myriad of products. Oxidation with pyridinium chloro- 23 chromate uneventfully provided keto enone 6 8 # (alp -65.5° (CHCl^)# in quantitative yield (Scheme XXIV and Figure 19). The level of optical purity of ( - ) - 6 8 and its precursors is expected to be quite high since none of the stereocenters were disturbed in the transforma tions from optically pure (-)-50. Thus# synthesis of an advanced intermediate containing the requisite framework and functionality for completion of the synthesis of (+)-pleuromutilin has been achieved. OCH PPM Figure 17. 300 MHz NMR Spectrum of 66 75 Scheme XXIV. Introduction of the Last Two Framework Carbon Atoms OH 64 1. MeL 1 2. Mel Mel, NaHCO aq acetone OCH 65 6 6 0CH3 EtMgBr Et20, -100°C PCC CelIte OCH (-)-68 67 OH OCH AJ rtr PPM Figure 18. 300 MHz 1H NMR Spectrum of 67a Figure 19. 300 MHz NMR Spectrum of 68 78 Illb. Degradative Route to the Second Relay Intermediate Once again, a pivotal synthetic intermediate was noted to be remarkably similar to a compound obtained degradatively by Arigoni. It is quite obvious that trione 10 (Scheme I) resembles 68. Intramolecular cyclization was prevented in Arigoni's sequence by oxidation of the C(14) hydroxyl moiety of mutilone (6 ) prior to cyclooctanone cleavage. A second relay intermediate could be attained in our route to pleuromutilin simply by converting the C(14) hydroxyl group of mutilone to the corresponding methyl ether before the eight-membered ring is opened. This modification was successfully achieved. The levoratory diketone 68 was obtained from either (+)-pleuromutilin (1) or tiamulin (2f) in only four steps (72% overall yield). The first transformation required in this sequence is conversion of the C(ll) secondary hydroxyl group to the homologous ketone. Oxidation of either antibiotic with pyridinium chiorochrornate effected this transfor mation (97%) as previously reported for the first relay degradative sequence. Subsequent saponification was realized when methanol was utilized as solvent in this 79 11a mild (5% KOH) alkaline hydrolysis. Retro-Michael fragmentation was circumvented probably because of the lower reaction temperature, and (-)-mutilone (6 ) was obtained in 87% yield (Scheme XXV). Blocking the hindered secondary alcohol as its methyl ether could now be addressed. Reaction condi tions had to be selected to prevent alkylation at any of the several enolizable sites. Page initially considered the use of methyl iodide in the presence of 6 2 silver oxide, but no methylation was observed. Alternatively, he examined the possibility of methyl- thiomethyl ether formation followed by reduction to the methyl ether. There are numerous procedures for making 6 3 methylthiomethyl ethers, and the protocol most effi cacious in this instance was that developed by Pojer61a'6:*c which furnished methylthiomethyl ether 69 nearly quantitatively. Unfortunately, desulfurization was not so productive. Treatment of 69 with either nickel boride or Raney nickel6 1 3 ' ® 4 resulted in either recovery of starting material or desulfurization with concomitant reduction of the vinyl group. Ultimately, a most satisfactory solution was discovered. Extended treatment of 6 with two to three equivalents of 2 ,6 -di-±-butylpyridine and methyl 80 Scheme XXV. Degradative Sequence to an Advanced Intermediate Ra-NI SCH 69 DMSO-HOAc-AcoO 5* KOH MeOH A 52* 53 MeOTf, CH-Cl KOH EtOH,£ OCH 68 och3 OCH 3 70 OCH trifluoromethanesulfonate® 5 produced predominantly 70 and some of 71 which are readily separated chromato- graphically. Enol ether 70 may then be hydrolyzed to 71 upon exposure to perchloric acid in aqueous tetra- hydrofuran or p-toluenesulfonic acid in aqueous acetone. However, for greater convenience# water may be added to the original reaction mixture upon consump tion of mutilone. Triflic acid is liberated and 70 is hydrolyzed quantitatively to 71,taJp25 -38.5 (CHC13>, over 1 h. It is the only product isolated upon workup (Figure 20). The 2,6-di-±-butylpyridine is expensive but is recovered during the chromatographic purifica tion and may be recycled. With the C(14) hydroxyl group suitably blocked, the previously utilized bond scission conditions were employed. Thus, heating 71 to the reflux temperature in ethanolic potassium hydroxide gave ( - ) - 6 8 (85%) (Scheme XXV), identical in all respects with the material previously obtained via the synthetic route. Keto enone 6 8 may be reduced with sodium borohydride in methanol to yield the diastereomeric mixture of hydroxy enones 67 obtained earlier. Completion of this relay linkup conveniently provides an alternate source of an advanced intermediate paramount for culmination of our pleuromutilin synthesis. JL L JL. Figure 20. 300 MHz NMR Spectrum of 71 CHAPTER IV STUDIES DIRECTED TOWARD COMPLETION OF THE PLEOROMUTILIN FRAMEWORK IVa. Introduction to Cyclooctane Formation Having secured both synthetic and degradative routes to the advanced diketo bicyclic intermediate (-)-6 8 , attention was directed toward completion of the synthesis of (+)-pleuromutilin. Already intact are the twenty carbon atoms found in the diterpenoid skeleton and all the oxygen bearing sites. Two of the three rings and four of the eight stereogenic centers/ including two of the three quaternary positions found in the natural product/ have been established. The primary hurdle to overcome in completion of this strategy is formation of the final cyclooctane C-C bond. The latent stereochemistry in ( - ) - 6 8 is expected to guide enantiospecific installation of two and possibly three of the last four chiral centers with any cyclooctane closure reaction. This leaves adjustment of the oxidation level of one carbonyl group/ 83 84 deblocking of the methyl ether, and subsequent esteri- fication to accomplish before obtaining pleuromutilin. The efforts directed towards completion of this endeavor are described in the remainder of this chap ter. The reactions utilized to this end are grouped according to reaction type and/or reactive intermediate. 51 IVb. Preliminary Investigations The most direct means to the desired target would be a Michael reaction between the cyclopentenone and either the enol or enolate of the ketone found in 6 8 . In the degradative work, the retro-Michael fragmenta tion occurred with rupture of the cyclooctane ring, but this required the reflux temperature of ethanol to hap pen. Saponification at the lower temperature/ carried out in refluxing methanol, did not cause ring opening. Accordingly, a range of bases and conditions were uti lized to examine this Michael reaction. Unfortunately, numerous attempts with hydroxide, alkoxide, fluoride, and amine bases produced only recovered starting material. Some Lewis acids were also employed, but the same results were observed (Scheme XXVI). Scheme XXVI. Attempted Michael Closure bases or Lewis acids OCH 68 The question then arose as to whether or not the position adjacent to the ketone could be functionalized to examine other reactive intermediates and untested reactions. The desired results were not obtained but were informative. Exposure of ketone 6 8 to an elec- trophilic source such as phenylselenenyl chloride or pyridinium hydrobromide perbromide^ produced what appears to be a mixture of 72 and/or 73. This indi cated the lability of the methyl ether which will eventually have to be removed and provides a possible reversible intramolecular protection of both the vinyl and C(14) hydroxyl moieties. Formation of the enolate of 6 8 with lithium diisopropylamide followed by exposure to phenyl selenenyl chloride also did not result in functionali- zation of the a position of the ketone but rather the site adjacent to the enone carbonyl (Scheme XXVII). 87 Scheme XXVII. Reactivity of 68 with Electrophiles 68 °CH3 B r o c h3 1. LD A 2. P h S e C l 0 a n d / o r - ~ B r PhSe 7 2 73 IVc. The Mukaiyama Approach The Mukaiyama reaction refers to a C-C bond form ing reaction between a silyl enol ether and an acetal or an a,8 -unsaturated carbonyl (or masked carbonyl) to 67 yield aldol and Michael products. The cyclopentenone of 6 8 can serve as a suitable acceptor, but the ketone could not be converted directly to the corresponding silyl enol ether as determined in preliminary studies. A method was required to differentiate the pair of ketone functionalities as a prelude to achieving cyclization. 6 8 The protocol developed by Kuwajima splendidly met this requirement. Accordingly, keto enone (-)-6 8 , was transformed to a mixture of the bis- and mono-silyl enol ethers 75 and 76, respectively. The mixture was then treated with tributylin fluoride in the presence of bis(tri-fi-tolylphosphine)palladium(II) chloride ( 8 mol %) as catalyst. This resulted in kinetic desily- lation of the less sterically congested silyl enol ether, in this case, the silyl dienol ether subunit. Following chromatographic purification, only (-)-76, IalD^ -34.9° (CCl^) (Figure 21), was isolated in 84% overall yield as a single stereoisomer uncontaminated by either 75 or keto enone 6 8 (Scheme XXVIII). Because of the reaction conditions utilized, kinetic deproto nation in the presence of hexamethylphosphoramide, and the fully substituted nature of the a' carbon, the (£)- 69 configuration has been assigned to 76. Somewhat greater chemoselectivity was observed in the initial silylation reaction when i-butyldimethylsilyl tri- 57 flate was employed for the direct conversion of 6 8 to 77 (50%), I“1D 2 5 - 21.5° (CC14). The bis-silyl enol ether was also produced, but the silyl dienol ether was not effected by exposure to tributylin fluoride and the palladium catalyst. OSi(CH3) , Figure 21. 300 MHz NMR Spectrum of 76 00 vo 90 Scheme XXVIII. Differentiation of the Ketone Pair of 68 2. T M S C 1 75 OCH 68 OCH 1. L D A, H M P A 2. TBDMSOTf P h H , A OSI OSKCH,). 7 7 OCH, " H 1 0 78 OCH 76 OCH With the acquisition of these substrates? the Mukaiyama reaction was examined. Although enones have been used as Michael acceptors,^® and examples exist of the construction of medium-sized rings with this chemi stry, ^ a , 0UJ. exampie involving a tetrasubstituted enone and formation of an eight-membered ring was going to be an austere test of this reaction's capabilities. Examination of a molecular model of 76 revealed that only 1,4-addition on the 3 face of the molecule could be expected as a result of conformational restraints. Despite the variety of conditions used, none produced the desired cyclization, and all resulted in eventual reversion to 6 8 (Scheme XXIX). A similar closure 71 reaction catalyzed by trimethylsilyl triflate was also tried with the same results. Could it be that under the reaction conditions employed a retrograde Michael fragmentation was occurring after cyclooctane formation? This was investigated with mutilone methyl ether 71, a compound that should be formed by the Mukaiyama reaction. Exposure of 71 to titanium tetra chloride in methylene chloride over an extended period of time from -78° to 25°C yielded only recovered starting material. The experimental conditions used in these attempted cyclizations are compiled in Table 1. A related protocol developed by Corriu was evaluated. Cesium fluoride in the presence of either tetramethyl or tetraethyl orthosilicate was found to be 77 an efficient catalyst to promote Michael reactions. It is believed that the orthosilicate generates the base for the reaction at the cesium fluoride surface, and traps the enolate as the silyl enol ether which reacts immediately In situ. No intramolecular examples were cited, but one example with a highly substituted enone was given. Numerous trials with 6 8 gave no reaction (Scheme XXIX). 92 Scheme XXIX. Attempted Mukaiyama and Related Reactions Lewis acids OCH OCH 76 CsF 71 S1(0R) Lewis acids OCH 68 Table 1. Mukaiyama and Related Reaction Conditions Lewis Acid or Triflate Solvent Temperature Reference TiCl 4 ch2 ci2 0°C 70,72 o i 00 TiCl 4 ch 2 ci2 n 70,72 TiCl4 /Ti(QiPr ) 4 ch2 ci2 -40°C 70,73 ZnCl2 Et2 0/CH2 C1 2 20°C 74 BF 3 *OEt2 ch2 ci2 20°C 75 ZnBr2 ch 2 ci2 20°C 76 TMSOTf ch 2 ci2 -78°C 71 93 Acetals derived from a,3 -unsaturated aldehydes also give Michael products in the Mukaiyama reaction 70 under appropriate conditions. To this end, attempts were made to ketalize the enone of 6 8 . This substrate proved recalcitrant under standard Dean-Stark condi tions with either p-toluenesulfonic acid or 2,4,6- 7 8 collidinium p-toluenesulfonate, a catalyst observed in this research group to promote ketalization of an a ,B-unsaturated ketone in the presence of a saturated ketone. At this point, it seemed that a Mukaiyama reaction was not going to deliver the desired cyclization. This can be attributed to any of several factors. Perhaps the tetrasubstituted enone is not a good Michael accep tor for steric reasons. The acyclic silyl enol ether may never adopt a conformation suitable for cycliza tion. The entropy factor involved with formation of the eight-membered ring may be too great for the reaction to be favored. ivd. The— PJbLOtochemistry Approach To surmount the possible difficulties discussed in the previous section, an alternative was sought. One that appeared especially attractive was the possibility of coaxing 76 into a 12+21 intramolecular photocyclo- 79 addition involving the enone and silyl enol ether, the more electronically reactive double bond. The issue of forming an eight-membered ring is thus circum vented and reduced to creating a fused [6,4] carbo- cyclic substructure 79 (Scheme XXX). Should 79 be obtained, exposure to fluoride ion would be expected to 80 induce fragmentation and deliver 80 having the pleuromutilin framework. The conformational restraints revealed by an examination of molecular models imply that the only regioisomer possible in this reaction is 79. The reaction conditions including the lamp, 81—86 filter, solvent, and duration were all varied. A compilation of these is presented in Table 2. For the most part, either no reaction was observed or decompo sition/polymerization ensued. However, the conditions 86a used by Cargill proved most efficacious. Irradia tion of dilute methylene chloride solutions of 76 with a 450 W Hanovia lamp through pyrex for 4-6 h resulted in formation of a single photoisomer (53%). Initially, it was not clear what product had been formed. Spec tral data indicated the loss of both enone and methyl 95 Scheme XXX. The Photochemistry of 76 hv 0CH3 OCH OCH3 76 h v 79 c h 2ci pyrex OSKCH,) 81 82 ether functionality. Extensive NMR work established the reaction product to be 82, [a]D^ +67.0° (CHCl^), (Figure 22 and Scheme XXX). Intramolecular photocyclo- addition was not occurring in the originally desired sense. Instead, hydrogen atom abstraction from the methoxyl group by the a-cyclopentenone carbon is 87 strongly favored. Collapse of intermediate biradical 81 yielded 82. This result spawned two ideas. First, the methoxyl protecting group was interfering with the 96 Table 2. Photochemical Reaction Conditions Lamp ■Filter ^eluent Time Result 450 W Hg pyrex 5 h decomposition C 6 H 6 450 w Hg pyrex Et20 3 h no reaction 450 w Hg pyrex MeOH 1 h decomposition 450 w Hg pyrex hexane 30 h polymerization 450 w Hg pyrex (ch3 )2 co/c6 h 6 14 h decomposition 450 w Hg corex 16 h decomposition C 6 H 6 450 w Hg corex Et20 6 h no reaction 450 w Hg vycor (CH3 )2CO 4 h decomposition 450 w Hg uranium hexane 24 h no reaction glass 450 W Hg uranium cyclohexane 72 h no reaction glass 2537 A — MeOH 1 h no reaction 2537 A — MeOH 5 h decomposition 2537 A — hexane 2.5 h no reaction 2537 A — hexane 4.5 h no reaction 2537 A — hexane 1 0 h decomposition 3500 A — hexane 3 h no reaction 3500 A — hexane 6 h no reaction 3500 A — hexane 1 0 h decomposition 450 W Hg pyrex ch 2 ci2 15 h 82 450 W Hg pyrex CH2 C 1 2 2 h 82 450 W Hg pyrex ch 2 ci2 6 h 82 T Figure 22. 300 MHz NMR Spectrum of 82 98 desired transformation. Would a change in the protect ing group produce the cycloaddition? Second, a C-C bond had been formed. Can the substituents involved with this reaction be modified to give an intermediate which would be useful in our synthetic scheme? The first question will be addressed next while the second remains to be exploited. To disrupt the conformation favorable for hydrogen atom transfer, a protecting group without hydrogens on the atom attached to the C(14) oxygen was needed. The corresponding t-butyldimethysilyl ether seemed appro priate. Accordingly, mutilone (6 ) was converted upon treatment with i-butyldimethylsilyl triflate in the 57 presence of 2,6-lutidine predominantly to 83 and in lesser part to 84. Subsequent acid-promoted hydrolysis converted 83 to 84. The cyclooctane ring in 84 was ruptured by the previously employed conditions to furnish 85, +42.7° (CHClj), in 71% overall yield from 6 (Scheme XXXI). Silyl enol ether 8 6 was formed in a 1:1 mixture with the bis-silyl enol ether 87 (58% overall). Unfortunately, 87 could not be converted to 8 6 with the Kuwajima protocol. Photolysis under the previously successful conditions as well as many others given on Table II did not lead to any characterizable photoproducts. Generally* decomposition or no reaction was obtained. This silyl enol ether* 8 6 * may be more moisture sensitive since hydrolyzed material was fre quently recovered from the reaction mixture. This substrate needs to be investigated further to determine the scope of its reactivity. Scheme XXXI. Preparation of 86 (±-Bu)Me,S10Tf hcio4 -thf-h2o 10* KOH EtOH* A 1. LDA 2. THSC1-HHPA SI (CH.) OSKCH& (CHg^SiO 100 In this same vein, some other blocking groups have been considered. Bulky protecting groups such as £- 8 8 89 butyl or trityl may have the additional advantage of forcing the acyclic chain into a conformation above the bicyclic enone. Unfortunately, neither group could be installed on the hindered hydroxyl moiety of mutilone (6 ). Another area that merits investigation are the 90 experimental conditions developed by Salomon. If silyl enol ether 76, which appears to be a fairly stable molecule, could be reduced stereoselectively to provide the 3-allylic alcohol, copper(I) catalysis may be beneficial in promoting the photocycloaddition (Scheme XXXII). One last area that should be investigated involves 87a solid state photochemistry. Scheffer and Trotter have observed different photoreactivity of the same enone depending on whether it is in a crystalline phase t or in solution based on "steric compression control." Although 76 is an oil, manipulation of either the silyl or methoxyl groups may yield a crystalline compound. Alternatively, 76 could be mounted on a solid support prior to irradiation. The whole photochemistry approach and its many options warrant further examination. 101 Scheme XXXII. Possible Copper(I) Promoted Photocycloaddition OSKCH,). CuOTf OH OH 76 0CH3 0CH3 OCH IVe. The free Radical Approach Free radical carbon-carbon bond formation has played an increasingly active role in the synthesis of 91 complex organic compounds. The relative insensiti vity to steric hindrance permitting bond formation with the creation of quaternary centers is especially attractive. It was felt that the enone moiety of the bicyclic indenone nucleus could serve as an acceptor in 92 a conjugate free radical addition. The free radical precursor could be either an a-halide or selenide of the acyclic ketone. Such derivatives should be readily available from silyl enol ether 76. In fact, exposure 102 93 of 76 to phenylselenenyl chloride and N-bromo- succinimide 9 4 gave rise to 8 8 (69%) and 89 (85%) (Figure 23), respectively, as mixtures of diastereomers (Scheme XXXIII). The acquisition of these compounds makes available precursors necessary to examine a new reactive intermediate in the pursuit of the pleuro- mutilin framework. a-Bromoketone 89 proved to be a more easily purified and stable derivative, and subsequent experiments employed this material. Scheme XXXIII. Synthesis of Free Radical Precursors 8 8 and 89 PhSe PhSeCl# PhH och3 88 76 Br ' * NBS, THF, 0°C o c h 3 69 ] - t i l... __ L ___L J ___.___.___.___L 103 Figure 23. 300 MHz H NMR Spectrum of 89 104 There is little precedent for C-C bond formation 95 involving a free radical adjacent to a carbonyl. Usually, the oxygen bearing center is maintained in a reduced state through the free-radical reaction and then subsequently oxidized. Upon treatment of 89 with 96 tri-n-butyltin hydride /azobisisobutyronitrile in refluxing benzene, a mixture of products was obtained. The enone remained untouched, and at least some reduc tion of the bromide seems to have occurred, although not completely as evidenced by mass spectrum data. In an attempt to prevent this reduction, which appears to be happening more rapidly than the desired cyclization, a tin radical precursor having no tin- hydrogen bonds was considered to be potentially useful. Hexaphenylditin and hexamethylditin have been used in similar situations in a photochemically initiated 97 reaction. So far either no reaction or decomposition has been observed in attempts at cyclization with these reagents. Because of the directness of this approach and many variables yet to be investigated experimentally, the free radical C-C bond formation approach should not be discounted at this point as a viable method for obtaining the pleuromutilin framework. Reduction of the acyclic ketone to the corresponding alcohol if both reduction of the bromide or epoxide formation can be QQ prevented/ may be beneficial. Also, a vinyl radical may be useful and the precursor readily prepared from 99 76 as shown below. SUCH,). Br OSKCH,) fl-Bi^SnH '“*% Br 2 » AIBN, P h H , £ OCH3 OCH3 IVf. The Carboniua Ion and Related Intermediates Approach Because of the lack of success experienced with the Mukaiyama-type cyclization attempts, other methods were sought to effect the desired transformation. Recall that ethylene glycol would not ketalize 6 8 under 106 a variety of experimental conditions. Selective thio- acetal formation of enones in the presence of ketones has been observed . 1 ® 0 For example, exposure of 6 8 to ethanedithiol resulted in formation of a nearly 1 : 1 mixture of mono- and bis-dithioketals (61% overall), 90 and 91, respectively (Scheme XXXIV and Figure 24). Bis-dithioketal 91 cannot be selectively hydrolyzed to 90, but it can be hydrolyzed with some degree of success with the previously employed conditions6 1 back to 6 8 for recycling through the process. Dithioketal 90 was anticipated to lead to a sub strate suitable for a thionium ion-induced cyclization. Trost has developed this reaction, and successfully used it with silyl enol ethers and vinyl silanes to produce rings containing up to seven members in very good yields. The catalyst is dimethyl(methylthio)= sulfonium fluoroborate. 1 0 1 Accordingly, 90 required conversion to its silyl enol ether, a compound which has so far proven elusive. The use of silyl chlorides resulted in decomposition of 90, while silyl triflates have failed to react. Once this functionalization is properly achieved, investigation of this protocol should proceed. It will be interesting to see if 1,4- addition is possible with this reaction, an extension 107 not yet considered in the published literature. Again, conformational restraints are expected to suppress 1 ,2 - addition. Scheme XXXIV. Sulfonium Ion Chemistry M e l , N a H C O OCH H S C H o C H o S H 91 1 PhH, TsOH,A OCH 68 OCH 90 OSiR.'3 SH OCH OCH OCH I 1 108 Figure 24. 300 MHz NMR Spectrum of 90 109 Carbonium ion-induced cyclizations offer a different alternative. Sutherland has found a ,3-epoxy ketones to be effective initiators of olefin cycliza tion processes upon treatment with either protic or 102 Lewis acids. The positive charge exists predomi nantly on the e-position to avoid the instability associated with the site adjacent to the ketone carbonyl. Repeated treatment of enone 6 8 with alkaline 103 hydrogen peroxide afforded a 15:85 mixture of diastereomeric epoxides 92 (75%) (Scheme XXXV). Conversion to either the trimethylsilyl enol ether 93 (Figure 25) or i-butyldimethylsilyl enol ether 94 was achieved with methodology developed for the Mukaiyama reaction precursors, usually with significant amounts of recovered starting material. Anomolously, the cyclopentenone was transformed to its silyl enol ether 95 in one instance. Only preliminary investigation of the cyclization reaction has been accomplished. A myriad of products was obtained from the addition of tin tetrachloride to 93 in methylene chloride at -78°C, none in sufficient quantity to be characterized because of the small reaction scale. Should subsequent efforts achieve 110 Scheme XXXV. Carbonium Ion Chemistry N a O H OCH o c h 3 68 92 OR Lew 1 s acid 0 HO OCH o c h 3 3 94 R=S1Me2 (i-Bu) (CH,),SIO OCH OCH cyclizationf a mixture of diastereomeric a-hydroxy ketones might be obtained. After removal of the unnecessary hydroxyl groupf epimerization should occur to give exclusively the stereochemistry found at the ring juncture in the natural product.8,1°k OSKCH,). OCH. T* r ■i ■T r— / c 0 ®PM 111 Figure 25. 300 MHz 1H NMR Spectrum of 93 112 IVg. The Anion Approach This chapter section will discuss a number of potential cyclization procedures involving an anion as the reactive intermediate. In a continuing examination of the Michael reaction, silyl enol ether 76 was reacted with benzyltrimethylammonium fluoride or tetra- n-butylammonium fluoride at various reaction tempera tures in tetrahydrofuran. Uniformly, keto enone 6 8 , the hydrolysis product, was the only material isolated from these reactions (Scheme XXXVI). A modified Reformatsky reaction was also consi- 104 dered. The Yamamoto conditions of diethylalummum chloride and activated zinc have been successful in construction of other medium-sized ring com pounds. 72c'T h e a-bromoketone 89 was already in hand from our free radical reaction endeavors. Unfortunately, the sole substance obtained upon workup was the protonated enolate 6 8 (Scheme XXXVI). Before leaving this area, two related but somewhat different methodologies demand consideration. Obvi ously, all the enolates generated to date have not become involved in a Michael reaction as was antici pated. Perhaps a more stabilized, softer enolate 113 Scheme XXXVI. Anion Closure Endeavors THF OCH 0CH3 76 68 OCH3 89 will provide a more usefully reactive species. Such a compound could be derived from the selective addition of methyl lithiotrimethylsilylacetate1 ® 6 followed by desilylation or other 2-carbon subunit. The anion of the 3 -keto ester may very well undergo 1 ,^a d d i t i o n . * ® 7 The ester moiety may then be used to introduce the remaining methyl group. Alternatively, it may be subsequently decarboxylated or possibly directly reduced to the methyl group required for the pleuromutilin skeleton. 114 A 3 o c h 3 OCH och 3 66 1 0CH3 o c h 3 If the enone above can be reduced stereoselec- tively and converted to its acetate, cyclization catalyzed by palladium(0 ) becomes a viable alterna- 108 tive This methodology has been used to form many medium to large cyclic structures including eight- membered ring compounds in high yields. The disadvan tage to such an approach would involve differentiation of the double bonds to reinstate the displaced oxygen. Obviously, much more effort should be directed toward this anion area before discarding its possible utility. 115 IVh. The Carbene Approach A reactive intermediate not investigated at all to date involves a carbene intermediate. Possible pre cursors for this reaction are the diazo ketone derived from the acid chloride of 6 6 , or the g-keto ester just discussed in the previous section after treatment with tosyl azide. If the carbene forms the desired cyclo propane through reaction with the enone double bond rather than a cyclobutanone by reaction with the vinyl group, reductive cleavage of the cyclopropane should deliver the pleuromutilin tricyclic system. The remaining methyl group would have to be introduced to complete the skeleton. o o c h 3 66 0 OCH o c h 3 3 116 CH,0, OCH 0CH3 66 CH,0, CH.O-C OCH3 °ch3 IVi. The Dieckmann Condensation Approach This last approach to pleuromutilin would involve closure of the eight-membered ring via a Dieckmann condensation reaction. This advantage to this route is that it avoids the creation of a quaternary center in forming the cyclooctane. There are two routes to the Dieckmann precursor. The first involves selective reduction of the enone carbonyl moiety in an ester 117 derived from 6 6 . Subsequent Claisen rearrangement would deliver the material suitable for ring closure. Alternatively, 1,4-addition of a two carbon sub unit to the same ester enone would directly provide a compound for condensation. Neither ethyl lithiotri- methylsilylacetate, copper-doped Grignard reagents, nor cuprates have successfully added in a conjugate manner to the enone of model compounds such as 6 8 in preliminary studies. Much hope lies in this approach as it comes closest to mimicking the biogenesis of the natural product. The Claisen approach should provide the precursor in excellent yield if stereoselective reduction to the allylic alcohol can be achieved. It leaves again the introduction of an oxygen at the C(3) position from an olefin with the vinyl group also present. The 1,4-addition route seems it will not provide the precursor in exceptional yield, but it does leave the cyclopentanone intact. iCH ~ \ - OCHj och3 OCH '3 CH,0 CH OCH 0CH3 0 CH,0 OCH I o c h 3 och'3 o c h 3 IV j. Summary Our synthesis of (+)-pleuromutilin rests at an advanced stage. Currently, all of the skeletal carbon and oxygen atoms are in place. Four of the eight asymmetric centers have been established inclusive of two of the three quaternary carbons. Two of three 119 rings found in the natural product are intact/ and an acyclic chain is poised to complete the third and introduce at least two and possibly three of the remaining stereogenic centers by ring closure. This latent stereochemistry results from conformational restraints imposed by the bicyclic nucleus initially established. The chiral centers of the acyclic group were produced by taking advantage of the facial selectivity observed during alkylation of a butyrolac- tone. Chirality was initially introduced via resolu tion/ and absolute stereochemistry established by comparison with a degradation product. All of this has been accomplished in an efficient stereoselective manner. Undoubtedly/ the efficiency of the synthesis was greatly assisted by the relay concept. This enabled several portions of this linear sequence to be worked on simultaneously and frequently earlier than would be practical from a purely .de nam 2 synthesis. The primary task remaining is obviously formation of the eight-membered ring. Much effort has been exerted on this endeavor and has begun to limit the scope on the types of reactive intermediates which may affect the desired closure. Many protocols have yet to be examined/ and some suggestions with regard to these are presented in the preceding sections. EXPERIMENTAL General Procedures Infrared spectra were recorded on a Perkin-Elmer Model 467 instrument. Proton magnetic resonance spectra were recorded with Varian T-60, Varian EM-390, Bruker WP- 200 and Bruker WM-300 spectrometers. Carbon spectra were recorded with Bruker WP-80 and Bruker WM-300 spectro meters. Mass spectra were determined on AEI-MS9 and Kratos MS 25 spectrometers at an ionization potential of 70 eV. Elemental analyses were performed by the Scandanavian Microanalytical Laboratory, Herlev, Denmark. Melting points were determined on a Thomas-Hoover melting point apparatus and are uncorrected. All optical rota tions were recorded with a Perkin-Elmer Model 241 polarimeter and concentrations are expressed in g/100 mL. All solvents were pre-dried via standard methods. Unless otherwise indicated, all reactions involving non-aqueous solutions were performed under an inert atmosphere. 120 (±)-Ethyl 4-Hydroxy-6-methyl-2-oxo-3-cyclohexene-l- 17 carboxylate (15). In a 2 L three-necked Morton flask equipped with a mechanical COgEt stirrer/ addition funnel/ and HO condenser topped with a drying tube (CaSO^) was prepared sodium ethoxide by slow addition of small pieces of sodium (36 g, 1.56 mol) to ethanol (500 mL). The solution was cooled in an ice bath and ethyl acetoacetate (218 g, 1 . 6 8 mol) was introduced from the addition funnel. Subsequently/ ethyl crotonate (161 g, 1.41 mol) was also added through the addition funnel. The reaction mixture was permitted to warm to room temperature over 0.5 h and gently refluxed for 4 h. The salt of the product began to precipitate shortly after reflux was initiated. After cooling/ the precipitate was collected by suction filtration and rinsed with small portions of ethanol. The salt was dissolved in water (600 mL) and the aqueous solution was extracted with ether (3x300 mL) before being acidified to pH 1 with 62] hydrochloric acid. Extraction with ether (4x300 mL) followed. The combined organic phases were washed with water until neutral, dried (Na2 S 0 4), and freed of solvent. The resultant white solid was dried at 40°C under vacuum. Yield (206 g, 74%): mp 89.5-91.0°C (lit1 7 mp 89°C); IR (CHCI3 ) cm" 1 3110, 3020, 2960, 2660, 1735, 1600, 1450, 1402, 1367, 1346, 1302, 1225, 1172, 1078, 1022, 839? 2H NMR (60 MHz, CDCl-j) 6 5.43 (s, 1H) , 4.18 (q, J. = 7Hz, 2H), 3.47 (d, J = 3Hz, 1H) , 3.13-2.00 (series of m, 4H), 1.30 (t, J = 6 Hz, 3H), 1.07 (d, i = 6 Hz, 3H) ? mass spectrum calcd (M+ ) jd/£ 198.0892, obsd 198.0887. (±)-Ethyl 4-Chloro-6-methyl-2-oxo-3-cyclohexene-l- carboxylate (16).17 A solution of diketo ester 15 (206 g, 1.04 mol), chloroform (400 mL), and phosphorous trichloride (113 mL, 1.29 mol) was stirred overnight at the reflux temperature. After removal of the solvent, the residue was slowly diluted with ice water (600 mL) while being cooled in an ice bath. The mixture was extracted with ether (2x400 mL and 123 3x200 mL). The combined organic phases were extracted with 2 % sodium hydroxide solution until neutral and dried (Na2 S04). The solvent was removed under vacuum and the remaining oil was distilled to furnish 16 (121 g, 54%): bp 100-110°C/0.25 mm (lit1 7 bp 127-130°C/7-8 mm); IR (neat) cm- 1 2965, 1730, 1670, 1610, 1455, 1350, 1150, 1080, 1030, 842, 812; 1H NMR (60 MHz, CDC13 ) 6 6.15 (s, 1H), 4.18 (q, J = 7Hz, 2H), 3.40-2.20 (series of m, 4H), 1.26 (t, J[ = 7Hz, 3H), 1.12 (d, J = 5Hz, 3H) . 17 (±)-Ethyl 2-Methy1-6-oxocyclohexanecarboxylate (17). A solution of 16 (10.0 g, 46.2 mmol) and triethylamine (6.5 mL, 46 mmol) in ethanol (60 mL) was exhaustively hydrogenated in a Paar hydrogenator at room tem perature over 5% palladium on charcoal as the catalyst. The catalyst was removed by filtration through Celite. Several similar batches were processed using a total of 92.61 g (0.427 mol) of the vinylogous acid chloride. The solvent from the combined filtrates was removed in vacuo. The residue was 124 dissolved in water (500 mL) and the mixture was extracted with petroleum ether (3x200 mL). The combined organic phases were dried (NajSO^) and evaporated. Distillation of the residue gave 17 (72.31 g, 92%): bp 69-75°C/10.25 mm (lit1 7 bp 113-114°C/10mm); IR (CHCI3 ) cm" 1 2930, 1740, 1710, 1650, 1608, 1450, 1368, 1150, 1025; 1H NMR (60 MHz, CDCI3 ) 6 4.22 (q, j; = 7Hz, 2H) , 3.05 (d, 1 = 10Hz, 1H) , 2.57-1.50 (series of m, 7H), 1.30 (t, a = 6 Hz, 3H), 1.05 (d, il = 6 Hz, 3H) ; mass spectrum calcd (M+ ) m/£ 184.1099, obsd 184.1104. <±)-Ethyl (IB*,2B*)-1,2-Dimethy1-6- oxocyclohexanecarbo^late (18). To a suspension of sodium hydride (2.60 g, 0.108 mol) in dry dimethylformamide (450 mL) in a 1 L three-necked flask COgEt equipped with a mechanical stirrer and addition funnel was slowly added B-ketoester 17 (20.0 g, 0.108 mol). After hydrogen evolution had ceased, methyl iodide (33.9 g, 0.239 mol) was introduced. The reaction mixture was 125 heated at 60°C with stirring for 72 h, at which time GC analysis (6 ', 5% SE 30, 150°C) showed a mixture of the desired trans isomer, the cis isomer, and polyalkylated material to be present in a ratio of 80.5:12:7.5, respectively. The reaction mixture was worked up in three batches, with each initially partitioned between ether (400 mL) and water (400 mL). The aqueous layers were extracted with ether (2x200 mL). The combined organic phases were then back-washed with water (3x100 mL) prior to drying ( ^ 2 8 0 4 ). After removal of solvent, the products were separated by spinning band distilla tion, bp 97.5-104.0°C/7-10 mm. The fractions containing 90% or more of the desired trans isomer 18 (10.0 g, 47%), a colorless oil, were reserved for further use while the remainder was recycled: IR (neat) cm- 1 2940, 1740, 1715, 1448, 1372, 1245, 1230, 1196, 1020, 944; *H NMR (90 MHz, CDClj) 6 4.11 (q, J = 7Hz, 2H), 3.00-1.45 (series of m, 7H), 1.31 (s, 3H), 1.24 (t, 1 = 6 Hz, 3H), 1.13 (d, 1 = 6 Hz, 3H), 13C NMR (20 MHz, CDC13) ppm 208.45, 171.37, 60.93, 43.81, 39.87, 30.34, 25.48, 21.11, 18.87, 17.05, 14.20; mass spectrum calcd (M+) m/£ 198.1256, obsd 198.1260. Anal. Calcd for C1 1 H 1 8 ° 3 : C, 66.64; H, 9.15. Found: C, 66.59; H, 9.15. 126 Tetcabydro-2-(2-propynyloxy)-2B-pyran . * ® 9 A 500 mL three-necked flask equipped with a mechanical stirrer, thermometer, and h c = c h 2o O addition funnel was charged with propargyl alcohol (100 g, 1.78 mol) and dihydropyran (150 g, 1.78 mol). The mixture was cooled (C02 (s)/CCl4 ) as concentrated hydrochloric acid (5 mL) was added dropwise and very slowly from the addition funnel. Stirring was continued for 4 h with the temperature of the reaction mixture maintained at 0°C to -10°C. Subsequent distilla tion furnished the product (204 g, 82%): bp 36-40°C/0.25 mm (lit*®® bp 47-50°C/3.5-5 mm); IR (neat) cm-* 3290, 2920, 2100, 1440, 1340, 1260, 890, 860, 805; *H NMR (60 MHz, CDCI3 ) 6 4 , 7 9 (br Sf *H)f 4 , 2 3 (dt' d = 1 3 and 3Hz' 2H), 3.90-3.13 (m, 2H) , 2.43 (tt, ,1 = 12 and 2Hz, 1H), 2.00-1.00 (series of m, 6 H). 127 (±) -Ethyl (lB*r 6JB*) -2-Hydroxy-l, 6-dimethy 1-2- [3- IItetrahydro-2B-pyran-2-yl]oxy]-l- propynylJcyclohexanecarboxylate (19). In a 250 mL three-necked flask fitted with a mechanical stir- rer, condenser, and septum were combined ether (60 mL), tetra- hydrofuran (20 mL), and ethyl- magnesium bromide (38 mL, 1.0 H in ether). Tq this was slowly added from a syringe at room temperature tetrahydro-2-(2-propynyloxy)-2JH-pyran (5.30 g, 37.8 mmol). After ethane evolution had ceased, B-ketoester 18 (5.00 g, 25.2 mmol) was also added from a syringe. After being stirred for 15 h at room tempera ture, the reaction mixture had become yellow and a gray solid had precipitated. TLC analysis (25% ethyl acetate in petroleum ether) showed no starting material remain ing. Ice was added, and the product was extracted into ether-petroleum ether (1:1). The organic phase was dried (Na2 S04) and evaporated. The crude product may be used as is in the next step or purified with a Waters Prep 500 HPLC (silica gel, 20% ethyl acetate in petroleum ether) to yield 19 (6.09 g, 71%) as a viscous yellow oil: IR (neat) cm- 1 3460, 3300, 2950, 1730, 1460, 1450, 1352, 957, 911, 880, 825; XH NMR (90 MHz, CDC13) 6 4.75 (br s, 1H) , 4.25 (s, 2H) , 4.15 (q, J[ = 6 Hz, 2H) , 3.58 (br S, 2H), 2.50-1.38 (series of m, 12H), 1.35 (s, 3H), 1.27 (t, jI = 6 Hz, 3H), 0.95 (d, jl - 8 Hz, 3H); mass spectrum calcd (M+-C5 HgO) m/£ 254.1518, obsd 254.1513. (±)-Ethyl (IB*,6JB*)-2-Hydroxy-2-(3-hydroxy-l-propynyl)-6- methylcyclohexanecarboxylate (21). Ethanol (2 mL) and 19 (200 mg, 1 . 0 1 mmol) were combined in a 1 0 mL flask. After cooling to 0°C with an ice bath, concentrated OH sulfuric acid (1 mL) was slowly added. The reaction mixture was stirred for 45 min followed by neutralization with satur ated sodium bicarbonate solution. The aqueous mixture was extracted with ether and dried (NajSO^) before evaporation of the solvent. The product was separated and purified by MPLC (silica gel? 25% petroleum ether in ethyl acetate) to produce 21 (128 mg, 50%) as a white solid: mp 62-63°C? IR (CHClg) cm” 1 3400, 2940, 1705, 129 1441, 1372, 1275, 1230, 1068, 1010; 1H NMR (90 MHz CDCI3 ) 6 4.23 (s, 2H), 4.12 (q, J. = 7Hz, 2H), 3.10-2.80 (m, 1H), 2.40-1.40 (series of m, 8 H), 1.45 (s, 3H), 1.25 (t, J = 7Hz, 3H), 0.97 (d, J. = 7Hz, 3H); mass spectrum calcd (M+-H 2 0) m/£ 236.1412, obsd 236.1407. (±)-Ethyl (4B*r 5E*)-4,5,6 ,7-Tetrahydro-4,5-dimethy1-3- ozo-4-indancarboxylate (20). A 25 mL two-necked flask equip ped with an addition funnel and mechanical stirrer was charged with ethanol (2 mL) and 19 (996 mg, 2.94 mmol). This solution was cooled in an ice bath to 0°C, and concentrated sulfuric acid (3 mL) was added slowly dropwise with resultant exotherm and color change to dark brown. After 4.5 h, the reaction mixture was neutralized with saturated sodium bicarbonate solution, extracted with ether, and dried (Na^O^). After solvent removal under vacuum, the crude product was purified by HPLC (silica gel; 25% petroleum ether in ethyl acetate) to yield 20 (406 mg, 59%) as a yellow oil: IR (neat) cm * 130 2930, 1730, 1700, 1645, 1445, 1380, 1300, 1032, 975; *H NMR (90 MHz, CDCI3 ) 6 4.10 (q, J = 7Hz, 2H), 2.67-2.00 (m, 6 H), 1.90-1.50 (m, 3H), 1.45 (s, 3H), 1.20 (t, J = 7Hz, 3H) , 0.95 (d, jl = 6 Hz, 3H) ; 13C NMR (20 MHz, CDC13 ) ppm 207.00, 173.62, 172.95, 141.10, 60.50, 45.45, 39.20, 34.77, 29.74, 27.79, 27.19, 21.36, 16.08, 14.32; mass spectrum calcd (M+) m/£ 236.1412, obsd 236.1427. Anal. Calcd for Ci4H20°3! c , 71.16; H, 8.53. Found: C, 70.90; H, 8.67. (±)- (4.B*, 5£*)-4,5,6,7-Tetrahydro-4,5-dimethyl-3-oxo-4- indancarboxylie Acid (24). A 25 mL round bottom flask equipped with a reflux condenser was charged with concentrated C0 2 H hydrochloric acid (5 mL) acetic acid (5 mL), water (5 mL), and ester 20 (250 mg, 1.06 mmol). The solution was brought to gentle reflux. TLC analy sis (2 0 % ethyl acetate in petroleum ether) showed no starting material after 16 h. The reaction mixture was diluted with ether, extracted with water, and dried 131 (MgSO^). The solvent was removed under reduced pressure with high vacuum as required to eliminate the last traces of acetic acid. Purification was accomplished using a Chromatotron (silica gel; 2 mm plate; ethyl acetate- methanol-petroleum ether, 6:1:13) to furnish the desired carboxylic acid 24 (125 mg, 57%) as a brown solid: mp 144-148°C; IR (CHCI3 ) cm" 1 2970, 2650, 1710, 1648, 1430, 1255, 860; 2H NMR (200 MHz, CDCI3 ) 6 8.93 (br s, 1H), 2.57-2.42 (m, 2H), 2.40-2.36 (m, 4H), 1.79-1.68 (m, 3H), 1.47 (s, 3H), 1.05 (d, 3. = 6.3 Hz, 3H); mass spectrum calcd (M+) m/£ 208.1099, obsd 208.1104. (±>-Ethyl (6 '**,7 'JB*)-4' ,5 • ,6 * ,7 '-Tetrahydro-6 • ,7 •- dimetbylspiroII,3-dithiolane-2,1'-indanl-7'-carboxylate (22). A 500 mL flask equipped with a Dean-Stark apparatus was charged with enone 20 (9.99 g, 42.3 mmol), benzene (200 mL), 1,2- ethanedithiol (50 mL, 0.60 mol) and p-toluenesulfonic acid monohydrate (1 . 0 g), and brought to the reflux temperature. After overnight heating, TLC analysis (2 0 % ethyl acetate in petroleum ether) showed the reaction to be complete. The reaction mixture was diluted with chloroform (250 mL) and extracted with 1U sodium hydroxide solution (4x500 ml) followed by drying (Na2 SO^) and solvent evaporation. The product was purified chromatographically (silica gel; 1 0 % ethyl acetate in petroleum ether) giving 22 (12.92 g, 98%) as a light brown solid: mp 70-73°Cy IR (neat) cm * 2920, 1725, 1445, 1368, 1221, 1181, 1105, 1035, 900, 809, 724; *H NMR (90 MHz, CDCI3 ) 6 4.08 (q, J = 7Hz, 2H), 3.47-2.67 (m, 4H), 2.40 (br s, 2H), 2.33-1.90 (m, 4H), 1.80-1.37 (m, 3H) , 1.50 (s, 3H) , 1.25 (t, jl = 7Hz, 3H) , 0.91 (d, jl = 6 Hz, 3H); 13C NMR (20 MHz, CDC13) ppm 173.95, 143.22, 134.48, 75.88, 59.28, 49.42, 47.04, 39.61 (2 carbons), 37.62, 33.59, 26.70, 26.22, 21.36, 15.87, 13.35; mass spectrum calcd (M+) m / s 312.1218, obsd 312.1225. 133 (±>-( 6 \B*,7 '**)- 4 •,5',6 •,7 '-Tetrahydro-6 ',7 • - dimethylspiro[l?3-dithiolane-2,l'-indan1-7'-methanol (23). A 2 L flask was charged with dry toluene (100 mL) and ester 22 (27.65 g? 88.5 mmol). Diisobu- tylaluminum hydride (265 mL? OH 0.265 mol? 1H in hexane) was slowly added via a syringe. After 3 h of stirring at room temperature? TLC analysis (2 0 % ethyl acetate in petroleum ether) showed the reaction to be complete. The reaction mixture was quenched with methanol-water (1:1). The gelatinous aluminum salts were dissolved in a minimum of 1 0 % hydrochloric acid solution. The layers were separated and the aqueous phase was further extracted with ether. The organic phases were combined? dried (MgSOj)? and concentrated An vacuo. The product was purified with a Waters Prep 500 HPLC (silica gel? 10% ethyl acetate in petroleum ether)? furnishing 23 (22.17 g? 93%) as a white solid: mp 47-50°C; IR (neat) cm" 1 3460? 2935? 1630? 1420? 1365? 1270? 1018? 808? 728; 2H NMR (300 MHz? CDCI3 ) 6 3.85 (ABq? jl = 11.6Hz? Au = 35Hz, 2H), 3.35-3.28 (m? 134 4H), 2.45-2.29 (mf 4H), 2.06-2.01 (m, 3H), 1.70-1.65 (m, 3H), 1.27 MHz, CDCI 3 ) ppm 144.77, 136.42, 77.43, 65.78, 49.37, 42.38, 39.81, 39.42, 39.28, 33.93, 27.53, 26.02, 23.55, 15.39; mass spectrum calcd (M+) m/£ 270.1112, obsd 270.1119. (+)-endo-Bornylamine Isocyanate.31 A 250 mL three-necked flask equipped with a mechanical stirrer, reflux condenser, and gas inlet tube was charged with NCO endo-bornylamine hydrochloride (10.0 g, 52.7 mmol) and freshly distilled toluene (100 mL). Phosgene was bubbled through this refluxing mixture for 4 h. The phosgene was first passed successively through traps containing cottonseed oil and concentrated sulfuric acid respectively. Excess phosgene was decomposed in a 2 0 % sodium hydroxide solu tion trap connected to the condenser. Upon completion of the reaction, nitrogen was passed through the mixture to remove residual phosgene. The isocyanate (9.60 g, 100%) 135 was isolated following solvent removal as an off-white solid: mp 64-67°C; rajp2 5 + 45.6°(£ 9.30, CHCI3 ); IR (CDCI3 ) cm" 1 2960, 2260, 1422, 1389, 1350, 1245, 840; XH NMR (90 MHz, CDCI3 ) 6 3 , 7 3 (br d' J = 10Hz, 1H), 2.50- 0.97 (series of m, 7H), 0.87 (s, 3H), 0.83 (s, 6 H). (fi)-(-)-a-Methylbenzyl Isocyanate. In a 250 mL three-necked flask fitted with a mechanical stir rer, gas inlet tube, and conden ser topped with a drying tube (CaSO^) were combined toluene (150 mL) and £-(-)-a-methylben- zylamine (10.00 g, 82.5 mmol). Dry hydrogen chloride was bubbled through the solution, resulting in precipitation of the hydrochloride salt. The drying tube was removed and replaced by a gas inlet adapter connected to a trap containing 20% sodium hydroxide solution. Phosgene was then bubbled through the mixture after first passing through traps of cottonseed oil and concentrated sulfuric acid. The reaction mixture was refluxed for 4 h with continued exposure to phosgene. The solution was purged with nitrogen for 1 h and the solvent was removed by distillation. The isocyanate (9.08 g, 75%) was obtained after distillation: bp 69.5-70.5°C/2 mm (lit11® bp 55- 56°C/2.5 mm); ( a ] D 2 5 -9.73° neat (lit1 1 0 Ecx 1 D 2 0 -9.2° (neat)); IR (neat) cm" 1 2980, 2260, 1495, 1453, 1378, 1348, 1312, 1280, 1205, 1062, 1028, 980, 755, 695; 1 U NMR (60 MHz, CDCI3 ) « 7.30 (br s, 5H), 4.70 (q, J = 7Hz, 1H), 1.57 (d, j I = 7Hz, 3H) . (B)~(+)-a-Methylbenzylamine Isocyanate. B-(+)“a~Methylbenzyl isocyanate was prepared analogously to its fi-enantiomer from £-(+)-«- methylbenzylamine (10.0 g, 82.5 CH3 mmol). Following distillation, the product was obtained as a colorless liquid (4.92 g, 40%): bp 70-71°C/2 mm (lit1 1 0 bp 55-56°C/2.5 mm);[a] D 2 5 +9.61° (neat) (lit1 1 0 U J p 1 9 +9.6° (neat)). 137 (+) -16'E» 7'JB)-4 • r 5 •, 6 • f 7 •-Tetrahydro-6 •, 7 •- dimethylspiroII,3-dithiolane-2,1'-indanl-7'-yllmethyl [ (aJB)-cx-Methylbenzy 11 carbamate (25) and (+)-I(6,£,7»S)-4',5',6',7,-Tetrahydro-6,,7'- dimethylspiroII,3-dithiolane-2,1'-indan1-7'-yl1methyl [ ( o r ) - ai-Methylbenzylcarbamate (26). In a 250 ml three-necked flask equipped with a gas inlet tube/ and condenser topped with a gas inlet adapter were combined dry toluene (130 mL) and B-(+)-a- methylbenzylamine (2.86 mLf 22.2 mmol). Hydrogen chloride was passed through the gas inlet tube above the surface of the solvent resulting in precipitation of the hydrochloride salt. When no further precipitation was observed/ the tube was lowered beneath the solvent surface to insure complete transformation of the amine to the corresponding salt. The gas inlet adapter was connected to a trap containing 2 0 % aqueous sodium hydroxide/ and phosgene was bubbled through the reaction mixture after first going through traps of cottonseed oil and sulfuric acid, the mixture was brought to the reflux temperature and exposure to phosgene was continued for 40 min after the disappearance of all of the precipitate. Subsequently* the solution was flushed with nitrogen for 1 h to remove residual phosgene. Alcohol 23 (2.00 g* 7.39 mmol) dissolved in a minimum quantity of toluene containing 4-dimethylamino- pyridine (200 mg) was then introduced. The reaction mixture was kept at reflux and monitored by TLC analysis (10% ethyl acetate in petroleum ether). The reaction was complete after 48 h. The solution was cooled* diluted with ether* and extracted with saturated sodium bicar bonate solution and brine before being dried (MgSO^) and freed of solvent. The diastereomers were separated by HPLC (Waters Prep 500; silica gel; 20% ethyl acetate in petroleum ether) employing recycling and band shaving techniques. Overall yield 2.69 g (87%): faster eluting diastereomer 26: 877 mg (28%); slower eluting diastereomer 25: 860 mg (28%); recovered mixture 951 mg (31%). Spectral data for 26: [alD 2 5 +13.1° (£ 1.96, CHCI3 ) ; IR (CDCI3 ) cm- 1 3450, 2962* 2930* 1710* 1500* 1448* 1408, 1371* 1330* 1278* 1221* 1055; *H NMR (200 MHz, CDCI 3 ) 6 7.34-7.24 (m* 5H)* 139 4.86 (br s, 1H), 4.51 (ABq, 2 = 10.8Hz, Au ■ 31Hz, 2H>, 3.35-3.16 (m, 4H), 2.38-1.96 (series of m, 5H), 1.86-1.20 (series of m , 8 H), 1.48 (d, jl - 6 .6 Hz, 3H), 0.88 (d, jl = 6.9Hz, 3H); 13C NMR (20 MHz, CDC13 ) ppm 156.19, 144.06, 142.20, 136.14, 128.60, 127.19, 125.85, 76.73, 67.78, 50.79, 49.51, 40.57, 40.06, 39.23, 36.74, 33.92, 26.00, 24.60, 24.28, 22.75, 15.08; mass spectrum calcd (M+) m/£ 417.1796, obsd 417.1821. Spectral data for 25: (alD 2 5 +28.8° (£ 1.72, CHCI3 ) ; IR (CDC1 3 ) C" 1" 1 3450' 2960, 2935, 1710, 1500, 1448, 1408, 1371, 1330, 1278, 1221, 1055; XH NMR (200 MHz, CDClj) 6 7.31-7.24 (m, 5H), 4.89 (br S, 1H), 4.50 (ABq, J = 10.7Hz, Au - 26Hz, 2H), 3.35-3.15 (m, 4H), 2.35-1.22 (series of m, 10H), 1.47 (d, 2. = 6.5Hz, 3H), 1.36 (s, 3H) , 0.91 (d, 1 = 5.1Hz, 3H) ; 13C NMR (20 MHz, CDCI3 ) ppm 156.09, 142.14, 136.14, 130.83, 128.60, 127.19, 125.85, 76.73, 67.85, 50.86, 49.51, 40.57, 40.12, 39.17, 36.55, 33.93, 25.88, 24.47 (2 carbons), 22.75, 14.96; mass spectrum calcd (M+ ) m/£. 417.1796, obsd 417.1821. 140 (+)-(6*B, 7'B>-4,,5',6',7'-Tetrahydro-6',7'- dimethy lspiroll,3-dithiolane-2,1 ’-indan)-7'-methanol and (-J-te'B^'BJ^'^'re'^'-Tetrahydro-e*^*- dimethylspiroIl,3-dithiolane-2,l,-indan]-7'-methanol (23). Carbamate 26 (200 mg, 0.47 mmol) was dissolved in toluene (10 mL) containing triethylamine (0.14 mL, 1.0 mmol). As the reaction mixture was brought to reflux, trichlorosilane (0.10 mL, 1.0 mmol) dissolved in toluene (2 mL) was added dropwise over several minutes. After 6 h at the reflux temperature, TLC analysis (15% ethyl acetate in petroleum ether) showed the reaction to be complete. The reaction mixture was cooled, diluted with ether (40 mL), washed with saturated ammonium chloride solution (2x20 mL) and dried (MgSO^) before removal of solvent in vacuo. The product was purified by MPLC (silica gel; 15% ethyl acetate in petroleum ether) affording the optically active alcohol (-)-23 ( 8 8 mg, 6 8 %), laJp2 5 -38.9° (n 4.78, CHCI3 ). The diastereomer 25 (200 mg, 0.479 mmol) was treated identically producing the enantiomeric alcohol (+)-23 (80 141 mg, 62%), [a]^"* +38.3° (£ 4.86, CHClj). The spectral data for the enantiomers were identical to those of the racemate 23. Note; Neither enantiomer was obtained optically pure by this procedure. The extent of purity was deter mined by a NMR study employing tris13-(heptafluoro- propylhydroxymethylene)-d-camphorato]europium(III) as the shift reagent. The shift reagent was added in small aliquots up to a total of 80 molar percent. In both cases, those spectra obtained from the samples containing 15 molar percent rEuthfc)^] were best suited for deter mining relative peak areas of each enantiomer with a planimeter. The peak chosen for study was the methyl singlet. Cleavage of the carbamate mixture enriched in 25 produced an enantiomeric mixture enriched with the levorotatory enantiomer. The ratio of the two enantio mers was 77.9 to 22.1 or 56% ee with [a)D^ -21.8° (c 4.78, CHC13). The sample enriched with the dextro rotatory alcohol in a ratio of 89.1 to 10.9 or 78% ee with [a]D^ +29.4° (£ 4.86, CHClg). There was in insufficient quantity of the dextrorotatory material to complete the synthetic sequence. Consequently, the substance was further diluted with racemic alcohol to 25 o provide material of 46% ee, IalD +17.4 (£ 5.07, 142 CHCl-j). The optical rotations reported above and for compounds 27 through 51 are extrapolated to optical purity. (+)-(6,JB)-4' ,5' ,6 ' ,7'-Tetrahydro-6 ' ,7*-dimethylspiro[l,3- dithiolane-2 ,1 '-indanJ-7'-carboxaldehyde and <-)-<6'S,7'jS)-4',5'6',7'-Tetrahydro-6',7'- dimethylspiroII,3-dithiolane-2 ,1 '-indanl-7'- carboxaldehyde (27). In a 500 mL three-necked round bottom flask fitted with a mech anical stirrer, solid addition tube, and Claisen adapter with a septum and inlet for the nitro gen line were combined dry methylene chloride (140 mL) and pyridine (33.0 mL, 4.09 mol). Chromium trioxide (20.43 g, 0.204 mol, dried over Pj O j at 100°C/0.5 mm) was slowly added. The Collins rea gent was permitted to form over 20 min. The addition tube was replaced with an addition funnel, and (-)-alcohol 23 (5.53 g, 20.4 mmol) dissolved in methylene chloride (2 mL) was added dropwise. Stirring was continued for 45 min. The reaction mixture was diluted with ether (1600 mL) and extracted in turn with 5% sodium hydroxide solution (3x1000 mL), 5% hydrochloric acid solution (2x1000 mL), 5% sodium bicarbonate solution (2x1000 mL), and brine (1000 mL). The organic layer was dried (MgS(Dj), and the solvent was removed under vacuum. Chromatographic purification on silica gel (elution with 10% ethyl acetate in petroleum ether) afforded (—)—27 (4.41 g, 80%) as a white solid: mp 103-104°C; ra. ] ^ 3 -111.7° (£ 6.49, CHC13); IR (CDCI3 ) cm” 1 2960, 2750, 1729, 1445, 1335, 1292, 828; 2H NMR (90 MHz, CDC13) 6 9.79 (s, 1H), 3.37-3.00 (m, 4H), 2.57-2.03 (m, 6 H), 1.80- 1.53 (m, 3H), 1.47 (s, 3H) , 0.97 (d, jI = 6 Hz, 3H) ; I3C NMR (20 MHz, CDC13) ppm 206.70, 145.65, 134.97, 77.07, 51.58, 49.52, 40.66 (2 carbons), 39.32, 34.53, 27.79, 26.40, 20.15, 15.78; mass spectrum calcd (M+ ) m/£. 268.0956, obsd 268.0963. Anal. Calcd for ci4H20OS2 : C' 62.64; H, 7.51. Found: C, 62.57; H, 7.55. Comparable oxidation of (+)-alcohol 23 (6.22 g, 23.0 mmol) yielded (+)-aldehyde 27 (3.79 g, 61%), (aJD^ +111.3°. 144 2-(2-Bromoethy1)-1,3-dioxane.111 A I L three-necked flask equipped with a gas dispersion tube and condenser topped with a drying tube (CaSO^) was charged with methylene chloride (300 mL) and acrolein (56 g, 1.0 mol). Hydrogen bromide was bubbled through the solution for 4.5 h at room temperature. Excess hydrogen bromide was removed by bubbling nitrogen through the mixture for 2 0 min. Subsequently/ 1 /3-propanediol (76 g r 1.0 mol) and p-toluenesulfonic acid monohydrate (1 . 0 g) were added and stirring was continued overnight. The reaction mixture was quenched with saturated sodium bicarbonate solution/ washed with water/ and dried (MgSO^). After concentra tion in vacuo, the bromoacetal was distilled: bp 38°C/0.05 mm (litli;Lbp 68-70°C/3 mm). Yield 6 6 g (34%). IR (neat) cm" 1 2950/ 2850/ 1375/ 1240, 1125, 1005, 870; 1H NMR (60 MHz, CDC13) « 4.72 (t, i = 5Hz, 1H), 4.33-3.67 (m, 4H), 3.47 (t, J. = 7Hz, 2H), 2.40-1.80 (m, 3H), 1.57- 1.13 (m, 1H). 145 (-)-(a£,6'B,7'B)-a-(2-m-Dioxan-2-ylethyl)-4' ,5' ,6' ,7'- tetrahydro~6',7'-dimethylspi ro11,3-dithiolane-2,1'- indanl-7'-methanol/ and (+) - (a JJ, 6 '£) - a- (2-m-Dioxan-2-ylethyl )-4',5',6',7'- tetrahydro-6', 7'-dimethylspi ro11,3-dithiolane-2,1 1- indanl-7'-methanol (28); and (a S, 6 '£, 7 '£)- a-(2-m~Dioxan-2-ylethy1)-4',5',6 ',7'- tetrahydro-6 ',7'-dimethylspiroII, 3-dithiolane-2 , 1 1- indanl- 7 '-methanol, and (aB,6,S,7,S)-«-(2-jD-Dioxan-2-ylethyl)-4' ,5' ,6' ,7'- tetr ahydr o-61,7 '-dimethylspi rod, 3-di thiolane-2,1 '- indan1-7'-methanol (29). Into a 250 ml three-necked flask equipped with a reflux condenser and addition funnel were placed a few iodine crystals, magnesium turnings (952 mg, 39.2 mmol), and tetrahydrofuran (18 ml). 2 - (2-Bromoethyl)-1,3-dioxane (6.37 g, 32.6 mmol) dissolved in tetrahydrofuran ( 1 2 ml) was added through the addition funnel. Reflux began spontaneously and heating was continued for an additional hour. During this period, 24 ml of tetrahydrofuran was introduced at a rate which did not interrupt the reflux. After an additional 1.5 h, aldehyde 27 (1.75 g, 6.53 mmol) dissolved in a minimum quantity of tetrahydrofuran was added. Heating was continued overnight. After 10 h, TLC analysis (1 0 % ethyl acetate in petroleum ether) indicated the reaction to be complete. The cooled reaction mixture was quenched with saturated ammonium chloride solution/ followed by 1 0 % hydrochloric acid to dissolve the magne sium salts. The product was extracted into ether/ dried (MgSO^)/ and freed of solvent. Separation of the diastereomers was performed with a Waters Prep 500 HPLC (silica gel; 25% ethyl acetate in petroleum ether). Overall yield 2.30 g (92%). The faster eluting diastereomer 28 was isolated as a light yellow oil (1.17 gr 47%). The slower eluting diastereomer was obtained as a white solid (1.13 g, 45%): mp 116-120°C. Spectra data for 28: IR (CDC13) cm” 1 3420/ 2960/ 1425/ 1250/ 1140/ 995; 1H NMR (90 MHz, CDCI3 ) 6 4.64 (t, J. = 5Hz, 1H) , 4.43-3.53 (series of m, 5H), 3.38 (br s, 4H), 2.73-1.23 (series of m, 16H), 1.40 (s, 3H), 1.13 (d, Z = 7Hz, 3H); 13C NMR (20 MHz, CC14) ppm 143.76, 140.18, 103.29, 77.98, 75.13, 67.30 (2 carbons), 50.91, 46.42, 40.54, 39.02, 34.71, 147 33.98/ 30.77, 29.80, 28.82, 27.85, 26.76, 25.12, 17.72; mass spectrum calcd (M+-C 3 H 7 02 ) m / & 309.1347, obsd 309.1325. Analogous Grignard reactions were performed with both enantiomers of 27. (-)-Aldehyde, U ] D -111.7°, (4.40 g, 16.4 mmol) delivered 2.88 g (45.7%) of optically active 28, [alp2 3 +4.6°(.c 5.6, CHCI 3 ). The enantiomer was obtained from reaction of (+)-aldehyde, (alp +111.3°, (3.79 g, 14.1 mmol) in a yield of 3.09 g (57%), ralp2 5 -0.5° (£ 10.6, CHCI3 ). Spectral data for 29: IR (CDCI3 ) cm" 1 3460, 2960, 1425, 1250, 1140, 1075; 1H NMR (90 MHz, CDCI 3 ) 6 4.59 (t, & = 5Hz, 1H), 4.14-3.65 (series of m, 5H), 3.35-3.25 (m, 4H), 2.46-1.43 (series of m, 16H), 1.38 (s, 3H) , 0.96 (d, jI = 7Hz, 3H) ; 13C NMR (20 MHz, CDCI3 ) ppm 160.98, 152.42, 108.79, 78.62, 73.16, 64.24 (2 carbons), 42.33, 37.42, 30.80, 29.89, 29.28, 24.61, 23.28, 21.58, 14.96, 14.11, 13.02, 10.53, 7.50; mass spectrum calcd (M+-C 3 Hg 0 2) m/£ 308.1268, obsd 308.1319. Optically active Grignard products were isolated from the reactions employing the enantiomeric aldehydes described above. Use of (-)-aldehyde 27 supplied 4.15 g 148 (6 6 %) of (+)-29*, [a] D 2 5 +16.2° <£ 5.65, CHCI3 ). The (+)-aldehyde furnished 3.03 g (56%) of 29*', I“ ) D 2 5 +8.49° (£ 12.0, CHC13). (6 R, 7JJ)-4,5,6 ,7-Tetrahydr0 - 6 ,7-dimethy1-7-[(2£)- tetrahydro-5-hydroxy-2-furyll-l-indanone and (6S,7£)-4,5,6,7-Tetrahydro-6,7-dimethyl-7-[(2£)- tetrahydro-5-hydroxy-2-fury]-l-indanone (31). y-Hydroxyacetal 29* (4.15 g, 1 0 . 8 mmol), [a)D +16.2°, was combined with p-toluenesulfonic acid monohydrate (400 mg), ace tone (100 mL), and water (50 mL) in a 250 mL round bottom flask. The mixture was heated overnight at the reflux tempera ture. TLC analysis (50% ethyl acetate in petroleum ether) showed the reaction to be complete. The mixture was concentrated under vacuum. The residue was diluted with water (200 mL) and washed with methylene chloride (4x100 mL) before being dried (MgS04). The solvent was evaporated leaving 31* (3.18 g) as a crude brown oil which was not purified but used immediately in the next 149 step. IR (neat) cm- 1 3420, 2930, 1690, 1622, 1450, 1343, 1270, 1210, 1170, 1060, 808, 732. The enantiomer 29*' (3.03 g, 7.88 mmol) talp +8.49°, produced the corresponding lactol 31*' (2.50 g) in a like manner. (6R, 7E)-4,5,6,7-Tetrahydro-6,7-dimethy1-7-[(2B>- tetrahydro-5-hydroxy-2-furyl]-l-indanone and (6£, 7£)-4,5,6,7-Tetrahydro-6,7-dimethy1-7-1(2S)- tetrahydro-5-hydroxy-2-furyll-l-indanone (30). Lactol 30* (2.22 g) was prepared analogously to its diastereomer 31 reported on the previous page from (+)-28 (2.88 g, 7.49 mmol), [cclD +4.6°. IR (neat) cm- 1 3430, 2930, 1690, 1620, 1380, 1285, 1030, 980, 730. The enantiomer 30*' (2.34 g) was obtained from (-)-28 (3.09 g, 8.03 mmol), [a)n -0.5°. 150 (+)- (5fi)-Dihydro-5-[(4£, 5£)-4,5,6 ,7-tetrahydro-4,5- dimethyl-3-oxo-4-indanyll-2 (3jfl) -furanone and (-) - {5JJ) -Dihydro-5- [ (4fi, 5£) -4,5,6 ,7-tetrahydr0-4,5- dimethyl-3-oxo-4-indanyll- 2 (3fl)-furanone (33). The crude lactol 31* (3.18 g) was dissolved in acetone (125 mL) in a 250 mL three-necked flask equipped with an addition funnel, mechanical stirrer, and thermometer. The solution was cooled to approximately 4°C in an ice bath. The reaction mixture was titrated with Jones reagent until the solu tion maintained a persistent orange color, then quenched with water, extracted with ether, and dried (MgS04). After removal of the solvent under vacuum, the product was evaluated by IR. If a hydroxyl moiety was observed, the material was resubmitted to the reaction conditions. The desired lactone was ultimately purified with the aid of a Chromatotron (silica gel? 4 mm plate; 50% ethyl acetate in petroleum ether) to afford (—)—33 (1.82 g, 58% from 31*) as a colorless oil: -37.5° (.c 6.63, CHC13). The enantiomer (+)-33 (1.11 g, 45%), [a3+38.3° (£ 6.01, CHClj) was produced analogously from 31*' (2.50 g): IR (CHC13 ) cm- 1 2970, 1770, 1690, 1624, 1460, 1380, 1285, 1168, 972, 905; 2H NMR (300 MHz, CDC13 > <5 4.49 (dd, jl = 8 . 6 and 6.9Hz, 1H), 2.68-1.97 (series of m, 10H), 1.60-1.41 (m, 3H), 1.27 (s, 3H), 1.00 (d, £ = 6.9Hz, 3H); 13C NMR (20 MHz, CDC13) ppm 209.27, 180.40, 177.71, 138.04, 83.30, 41.27, 40.95, 35.45, 29.70, 29.06, 28.55, 27.28, 24.78, 19.42, 15.39; mass spectrum calcd (M+ ) m / & 248.1412, obsd 248.1418. (-)-(5B)-Dihydro-5-I (4£,5£)-4,5,6,7-tetrahydro-4,5- dimethyl-3-oxo-4-indanyll-2 (3JQ) -furanone and (+)-(5£)-Dihydro-5-[(4£,5S)-4,5,6 ,7-tetrahydro-4,5- dimethyl-3-oxo-4-indany11-2(3fi)-furanone (32). In like fashion, the enantio meric lactols derived from (-)-28 (3.09 g, 8.03 mmol), (alp -0.5°, and (+)-28 (2.88 g, 7.49 mmol), Ia]D +4.6°, were trans formed to (-)-32 (0.951 g, 41%), 152 U ] D 2 5 -14.6° (£ 5.03, CHCI3 ), and (+)-32 (1.017 g 46%), [a] D 2 5 +20.5° (fi 5.72, CHC13 > respectively: IR (CHCI3 ) cm" 1 2930, 1770, 1690, 1618, 1450, 1382, 1238, 1170, 990, 907; 1H NMR (300 MHz, CDC13 > 6 4.87 (t, jl = 6.5Hz, 1H) , 2.46-2.28 (m, 8 H), 2.00-1.87 (m, 2H), 1.70-1.62 (m, 3H), 1.22 (s, 3H), 0.97 (d, J = 6.5Hz, 3H); 13C NMR (20 MHz, CDCI3 ) ppm 208.44, 177.33, 176.12, 140.34, 82.53, 40.75, 38.58, 35.26, 29.45, 29.13, 27.66, 26.83, 24.21, 18.52, 16.74; mass spectrum calcd (M+) m/£ 248.1412, obsd 248.1418. (+)-(5£>-Dihydro-5-I (6 '£, 7 '£)-4■,5•,6',7'-tetrahydro- 6 ',7'-dimethylspiroII,3-dithiolane-2 ,1 '-indanl-7’-yl]- 2 (3£)-furanone and (-)-C5£)-Dihydro-5-E(6 '£,7•£>-4•,5•,6 •,7'-tetrahydro- 6 ',7'-dimethylspiroll,3-dithiolane-2 ,1 '-indanl-7'-yll- 2 (3£)-furanone (34). A 100 mL round bottom flask equipped with a Dean-Stark apparatus was charged with enone 32 (762 mg, 3.07 mmol), benzene (45 mL), p-toluenesulfonic acid 153 monohydrate (70 mg), and 1,2-ethanedithiol (3.0 mL, 37 mmol). The reaction mixture was brought to reflux and monitored daily by TLC analysis (20% ethyl acetate in petroleum ether). More ethanedithiol (1.5 mL) was added incrementally until no further progress in the reaction was observed. The reaction mixture was diluted with ether, washed several times with 1 U sodium hydroxide solution, dried (MgSO^), and freed of solvent. The desired product was separated from remaining starting material by MPLC (silica gel; 25% ethyl acetate in petroleum ether). The recovered starting material accounted to 113 mg, while 34 (479 mg, 48% or 57% based on recovered starting material) was obtained as an off- white solid: mp 105-107°C; IR (CgDg) cm ^ 2920, 1775, 1465, 1425, 1270, 1190, 1015, 985, 895, 735; 1H NMR (300 MHz, CDC13) 6 4.72 (dd, Z » 9.3 and 6.4Hz, 1H), 3.38-3.12 (series of m, 4H), 2.51-2.02 (series of m, 10H), 1.69- 1.57 (m, 3H), 1.51 (s, 3H), 1.02 (d, Z - 6.9Hz, 3H); 13C NMR (20 MHz, CgDg) ppm 175.80, 144.79, 137.27, 83.01, 78.04, 49.64, 43.93, 41.69, 40.23, 38.83, 34.04, 29.37, 27.36, 26.33, 26.15, 20.08, 16.74; mass spectrum calcd (M+) m/£ 324.1218, obsd 324.1226. Reaction of optically active (-)-32 (951 mg, 3.83 mmol), (alp -14.6°, produced (+)-34 (551 mg, 44%), [a] D 3 3 154 +5.7° (£ 6.76, CHC^)* Correspondingly, (+)-32 (1.017g, 4.09 mmol), £alD +20.5°, yielded (-)-34 (599 mg, 45%), [a] D 2 5 -12.0° (£ 5.70, CHCI 3 ). (+)-(5 S )-Dihydro-5-[(6 •£,7'£)-4•,5•,6 ',7'-tetrahydro- 6 ',7'-dimethylspiroll,3-dithiolane-2 ,1 '-indanl- 7 '-yll- 2 (3fl)-furanone and (-)-<5B)-Dihydro-5-[6'S,7 'S)-4',5',6',7'-tetrahydro- 6',71-dimethylspiroII,3-dithiolane-2,1'-indanl-7'-yll- 2 (3fl)-furanone (35). This diastereomer was prepared in a manner analogous to 34 from 33 (413 mg, 1.66 mmol). The product 35 was obtained as an off-white solid (464 mg, 8 6 %): mp 99-101°C; IR (CgDg) cm” 1 2960, 1780, 1425, 1250, 1195, 1025, 855, 693; 2H NMR (200 MHz, CgDg) 6 4.52 (dd, 1 = 8.3 and 7.1Hz, 1H), 2.99-2.72 (m, 4H), 2.41-1.37 (series of m, 13H), 1.44 (s, 3H), 0.82 (d, jl = 6 .8 Hz, 3H) ; 13C NMR (20 MHz, CgDg ) ppm 175.86, 145.64, 135.75, 82.59, 78.04, 49.76, 43.57, 41.20, 40.35, 155 39.08, 34.22, 28.82, 27.30, 26.15, 25.12, 20.63, 15.89; mass spectrum calcd (M+) m/£ 324.1218, obsd 324.1213. Anal. Calcd for ci7H24®2S2 : Cf ®2.92; H, 7.45. Found: C, 62.69; H, 7.44. Optically active (-)-33 (1.819 g, 7.32 mmol), Cc*]D -37.5° yielded (-)-35 (558 mg, 24%), [alp2 5 -14.5° (£ 4.77, CHCl^). Analogously, (+)-33 (951 mg, 3.83 mmol), [alD +38.3°, furnished (+)-35 (271 mg, 19%), [alD 2 5 +10.3° (£ 5.04, CHCI3 ). (±)- 5fi*)-Dihydro-3-methyl-5-[ ( 6 '£*,7 '£*)- 4',5',6 ',7'-tetrahydro-6',7'-dimethylspiro[l,3- dithiolane-2fl ,-indanl-7,-yl]-2(3fl)~furanone (37). Lithium diisopropylamide was prepared from diisopropylamine (0.23 mL, 1.62 mmol) and n- ✓CH, butyllithium (1.01 mL, 1.62 mmol, 1.60 U in hexane) in tetrahydrofuran (4 mL) in a 50 mL three-necked flask fitted with an addition funnel and septum. This solution was cooled to -78°C, and lactone 35 (439 mg, 1.35 mmol) dissolved in tetrahydrofuran (5 mL) was added dropwise. After 1 h, HMPA (0.23 mL, 1.32 mmol) and methyl iodide (0.10 mL, 1.62 mmol) were added. Stirring was continued for 8 h at -78°C, followed by warming to 0°C over 2 h. The reaction mixture was quenched with water, extracted with ether, and dried (MgS04). After concentration in vacuo. MPLC (silica gel; 1 0 % ethyl acetate in petroleum ether) was used to separate the recovered starting material (97 mg) from the desired methylated lactone 37 (249 mg, 55% or 70% based on recovered starting material) as a yellow oil: IR (CgDg) cm” 1 2980, 1790, 1435, 1385, 1260, 1205, 1015, 865, 745, 705; 1H NMR (200 MHz, CgDg) 6 4.79 (dd, J = 8.1 and 5.9Hz, 1H), 3.39-3.08 (series of m, 4H), 2.95-1.51 (series of m, 12H) , 1.46 (s, 3H), 1.25 (d, jl = 7.3Hz, 3H), 0.99 (d, JL = 6 .6 Hz, 3H) ; 13C NMR (20 MHz, CgDg) ppm 178.83, 146.36, 135.42, 80.37, 78.07, 49.47, 43.74, 41.77, 40.57, 38.76, 34.61, 34.28, 32.91, 27.61, 26.13, 20.12, 16.40, 15.91; mass spectrum calcd (M+) m / e 338.1374, obsd 338.1331. <±) - < 3£*, 5£ *) -Dihy dr o-3-me thy 1-5- E (6 '£*, 7 '£*) - 4', 5',6 ',7'-tetrahydro-6',7'-dimethylspiroII,3- dithiolane-2rl ,-indan]-7'-yl]-2(3fl)-furanone (38). Lactone 38 was prepared analo gously to lactone 37 from 34 j (448 mg, 1.38 mmol). Recovered CH3 starting material (70 mg) was separated from the desired product (255 mg, 55% or 65% based on recovered starting material) by MPLC (silica gel; 10% ethyl acetate in petroleum ether). The methyl ated lactone 38 was obtained as an off-white solid: mp 57.5-60.0°C; IR (CgDg) cm" 1 2965, 1780, 1465, 1430, 1380, 1253, 1195, 862, 695; 2H NMR (200 MHz, CgDg) « 4.65 (t, J = 7.5Hz, 1H), 2.98-2.68 (m, 4H), 2.49-2.32 (m, 4H), 2.09- 1.34 (series of m, 8 H), 1.59 (s, 3H), 1.03 (d, il = 7.4Hz, 3H), 0.99 (d, 2 = 6.1Hz, 3H); 13C NMR (20 MHz, CgDg) ppm 178.99, 144.81, 137.27, 80.68, 78.18, 49.69, 44.26, 41.52, 40.43, 38.83, 35.26, 34.11, 33.92, 27.27, 26.25, 20.12, 16.67 (2 carbons) ; mass spectrum calcd (M+ ) m / e 338.1374, obsd 338.1438. 158 Phenylselenoacetaldehyde. In a 100 mL flask were combined ethanol (50 mL) and phenylsele- nenyl bromide (4.72 g, 20 mmol). SeCHgCHO Ethyl vinyl ether was added with stirring until the color of the reaction mixture changed from brown to yellow. After an additional 15 minf the reaction mixture was quenched with saturated sodium bicarbonate solution and extracted with ether (3x50 mL). The combined organic phases were further extracted with saturated sodium bicarbonate solution (50 mL) and brine (50 mL). After drying d^CO-j), the solvent was removed In vacuo to furnish the crude diethyl acetal (5.62 g). The acetal was subsequently dissolved in ether (125 mL). Then 1 U hydrochloric acid (125 mL) was added and the two-phase system was vigorously stirred for 15 h with a mechanical stirrer. TLC analysis (20% ethyl acetate in petroleum ether) showed reaction to be compete. The layers were separated/ and the aqueous phase was extracted with ether (2x140 ml). The combined organic phases were washed with water (2x150 ml) and brine (150 ml) before being dried (MgS04). After the solvent was 159 removedf the product was purified by short-path distillation in a Kugelrohr apparatus. Phenylselenoacet- aldehyde (3.22 g, 81%): bp 100-130°C/4 mm (lit1 1 2 bp 57°/0.01 mm). Phenylselenoacetaldehyde diethyl acetal: IR (neat) cm ” 1 2960, 2880, 1575, 1474, 1432, 1110, 1050, 725, 680; 1H NMR (60 mHz, CDCl-j) <5 7.67-7.00 (m, 5H) , 4.70 (t, jl = 4Hz, 1H), 3.83-3.20 (m, 4H), 3.08 (d, JL = 6 Hz, 2H) , 1.18 (t, J. = 7Hz, 6 H) . Phenylselenoacetaldehyde: IR (neat) cm " 1 3050, 2965, 2810, 2710, 1710, 1578, 1472, 1434, 730, 680. XH NMR (60 MHz, CDC13) 6 9.50 (t, jl = 6 Hz, 1H), 7.66-7.07 (m, 5H) , 3.48 (d, J[ = 6 Hz, 2H) . (±)-Dihydro-3-methyl-3-vinyl-2(3fl)-furanone (36). In a 50 mL three-necked flask 0 equipped with a mechanical stirrer and septum was prepared lithium diisopropylamide in ether ( 8 mL) from methyllithium (0.90 mL, 1.12 mmol, 1.25 H in hexane) and diisopropyl amine (0.20 mL, 1.22 mmol). The solution was cooled to -40°C before a-methyl-y-butyrolactone (100 mg, 1.00 mmol) was introduced. When the temperature had increased to -20°C, zinc chloride (1.5 mL, 1.02 mmol, 0.68 U in ether) was added. The zinc enolate was permitted to form during 10 min before phenylselenoacetaldehyde (203 mg, 1.02 mmol) was added. After 15 min the reaction mixture was quenched with saturated ammonium chloride solution, washed with ether, and dried (MgSO^) before the solvent was removed under reduced pressure. The aldol inter mediate was immediately dissolved in methylene chloride (3 mL) and transferred to a 25 mL two-necked flask fitted with an addition funnel and septum along with triethyl- amine (0.70 mL, 5.1 mmol). Hethanesulfonyl chloride (0.32 mL, 4.1 mmol) dissolved in methylene chloride (2 mL) was slowly added over 1 h. After 30 min, the reaction mixture was diluted with methylene chloride followed by extraction with 0 . 1 U hydrochloric acid, saturated sodium bicarbonate solution, and brine. The organic layer was dried, and the solvent was removed under reduced pressure. The residual oil was purified by MPLC (silica gel; 10% ethyl acetate in petroleum ether) to afford 36 (42 mg, 34%) as a yellow oil: bp 39-40°C/.l 161 mm; IR (neat) cm" 1 2960, 2920, 1770, 1630, 1445, 1408, 1365, 1180, 1078, 1015, 915; XH NMR (90 MHz, CDCl-j) 6 5.93 (dd, Z = 18 and 10Hz, 1H), 5.23 (d, Z = 10Hz, 1H) , 5.20 (d, Z = 18Hz, 1H), 4.27 (t, jl = 6 Hz, 2H), 2.50-2.00 (m, 2H), 1.43 (s, 3H); mass spectrum calcd (M+-CH 2 C0 2) m/£ 67.0548, obsd 67.0561. Anal. Calcd for C7Hi q °2 : C ' 66,64» H ' 7.99. Found: C, 66.43; H, 8.09. (±) - (3£*, 5R*) -Dihydro-3-methyl-5- [ (6 '£*,7 •£*) - 4' ,5' ,6' ,7'-tetrahydro-6' f7'-diinethylspiro[l,3- dithiolane-2,1'-indanl-7'-y11-3-viny1-2(3B)-furanone (40). Lithium diisopropylamide was prepared from diisopropylamine (0.08 mL, 0.56 mmol) and £- butyllithium (0.35 mL, 0.56 mmol, 1.60 H in hexane) in ether (2 mL) in a 25 mL two-necked flask fitted with a septum. The solution was cooled to -78°C, and a-methyl lactone 38 (127 mg, 0.377 mmol) dis solved in ether (1 mL) was added via syringe. After 1 h 162 of stirring, phenylselenoacetaldehyde (90 mg, 0.452 mmol) was added. The reaction mixture was quenched after 5 h with water, subsequently extracted with ether, dried (MgSO^), and concentrated under reduced pressure. The crude aldol intermediate was immediately dissolved in methylene chloride (3 mL) and transferred to a 25 mL flask equipped with an addition funnel. Triethylamine (0.27 mL, 1.92 mmol) was added dropwise, followed by methanesulfonyl chloride (0.12 mL, 1.55 mmol) dissolved in methylene chloride (2 mL) over 1 h. After 30 min, the reaction mixture was diluted with methylene chloride and washed with saturated sodium bicarbonate solution and water. The aqueous washes were combined and extracted with methylene chloride. The combined organic washes were dried (NajSO^) and concentrated in vacuo. Purification was achieved by MPLC (silica gel; 5% ethyl acetate in petroleum ether; recovered starting material (5 mg) was isolated alongside vinyl lactone 40 (30 mg, 22%), a yellow oil: IR (CC14) cm" 1 2940, 1780, 1640, 1460, 1380, 1125, 685; XH NMR (300 MHz, CDCI3 ) 5 5.78 (dd, jI = 17.4 and 10.5Hz, 1H), 5.17 (d, 3 = 16.8Hz, 1H), 5.16 (d, jI = 10.8Hz, 1H), 4.67 (dd, 3 = 11.3 and 5.4Hz, 1H), 3.39-3.14 (series of m, 4H), 2.42-2.06 (series of m, 8 H), 1.70-1.62 (m, 3H), 1.51 (s, 3H), 1.34 (s, 3H), 1.03 163 (d, jI = 6.7Hz, 3H) ; mass spectrum calcd (M+ ) m/fi 364.1531, obsd 364.1506. (±)- (3B*,5B*)-Dihydro-3-methyl-5-I(6 '£*,7'£*) - 4*,5',6 ',7'-tetrabydro-6',7'-dimethylspiroIl,3-ditholane- 2,l,-indan-7,]-yl]-3-vinyl-2(3fl)-furanone (39). Lactone 39 was prepared in a manner identical to lactone 40. After purification, 39 (28 mg, 23%) was obtained from 37 (111 mg, 0.328 mmol) as a yellow oil: IR (CC14 ) cm” 1 2980, 2940, 1780, 1635, 1440, 1380, 1130, 1020, 925, 690; 1H NMR (300 MHz, CDC13) 5 5.79 (dd, = 17.4 and 10.5Hz, 1H), 5.18 (d, 1 = 17.4Hz, 1H), 5.17 (d, jI = 10.2Hz, 1H) , 4.74 (dd, H = 11.6 and 5.4Hz, 1H), 3.39-3.16 (m, 4H), 2.66-2.03 (series of m, 8 H), 1.73-1.59 (m, 3H), 1.45 (s, 3H), 1.34 (s, 3H), 0.98 (d, H. = 6 .6 Hz, 3H); mass spectrum calcd (M+ ) m/fi 364.1530, obsd 364.1529. 164 (±) -2-(2-Hydroxyethyl)-2-methyl-4-phenyselenyl-3-butenoic acid, y-Lactone (43). Lithium diisopropylamide was formed in ether ( 8 mL) from n- O butyllithium (0.67 mL, 1.07 mmol, 1 . 6 H in hexane) and SePh diisopropylamine (0.15 mL, 1.07 mmol). The solution was cooled to -78°C and lactone 41 (100 mg, 0.374 mmol) was intro duced. After 1 h, the enolate was treated with HMPA (0.15 mL, 0.89 mmol) and methyl iodide (0.08 mL, 1.34 mmol). The mixture was quenched 2.5 h later with water and extracted with ether. The combined organic phases were dried (MgSO^) and freed of solvent. The residue was purified by MPLC (silica gel; 10% ethyl acetate in petroleum ether), which supplied (13 mg, 12%) of 43, a brown oil, as the major product: NMR (60 MHz, CDC^) 6 7.70-7.13 (m, 5H) , 6.67 (d, J[ =16Hz, 1H) , 5.97 (d, J[ = 16Hz, 1H), 4.27 (t, £ = 6 Hz, 2H), 2.40-2.06 (m, 2H), 1.40 (S, 3H). 165 (±) -5- [ (tert-Butyldimethylsilorv) methyl I dihvdr0 - 2 (3JJ) - furanone (44). A 250 mL flask was charged with 0 dimethylformamide (45 mL), imi dazole (13.3 g, 0.195 mol), 3- hydroxymethylbutyrolactone (4.54 g, 39.1 mmol), and i-butyl- dimethylsilyl chloride (6.5 g, 43 mmol) and stirring was maintained at room temperature for 9 h. The reaction mixture was quenched with methanol and partitioned between saturated sodium bicarbonate solution and ether. The organic phase was washed with additional sodium bicarbonate solution and water before being dried (MgS04). The solvent was removed in vacuo. The residue was purified with a Waters Prep 500 HPLC. (silica gel; 10% ethyl acetate in petroleum ether). The silyl ether (2.81 g, 31%) was isolated as a colorless oils IR (CDCI3 ) cm" 1 296Or 2940, 2865, 1775, 1465, 1360, 1255, 1180, 1070, 835; 1H NMR (200 MHz, CDC13 ) 6 4.60- 4.53 (X of ABX, m, 1H) , 3.76 (AB of ABX, Au = 33Hz, = 11.3Hz, = 3.2Hz, jIb x = 3.2Hz, 2H) , 2.60-2.02 (m, 4H) , 0.88 (s, 9H), 0.063 (s, 3H), 0.057 (s, 3H); mass spectrum calcd (M++l) m / & 231.1416, obsd 231.1397. 166 BisImethoxy(thiocarbony1)Jdisulfide. In a 250 mL flask were combined potassium hydroxide ( 1 0 g), water (10 mL), and methanol (20 ? I CHgOCS'ScOCHg mL). The solution was cooled to 0°C with an ice bath, and then carbon disulfide (20 mL) was added. The resultant green reaction mixture was stirred for 1.5 h at 0°C during which time the color changed to yellow. At this point, sodium nitrite (3.0 g) was intro duced followed by acidification to pH 4 with acetic acid. After an additional 0.5 h at 0°C, the mixture was extrac ted with ether (2x75 mL). The organic phases were dried (MgSO^) and treated with charcoal. The solvent was removed in vacuo with the temperature not exceeding 40°C furnishing bisImethoxy(thiocarbonyl) 1 disulfide (2 . 6 g, 7%) as a yellow oil: IR (neat) cm-* 2980, 1730, 1454, 1260, 1173, 1042, 842; *H NMR (60 MHz, CDCI3 ) fi 4.23 (s, 6 H) . 167 (±) - (£) -5- [ (ifiri-Butyldimethy lsiloxy) methyl J -3- ethylidenedihydro-2(3fi)-furanone (45). Method A: Lithium isopropyl- cyclohexylamide was prepared from isopropylcyclohexylamine (0.34 mL, 2.09 mmol) and n- butyllithium (1.27 mL, 2.09 mmol) in ether (3 mL). Lactone 44 (400 mg, 1.74 mmol) dissolved in ether (3 mL) was added via syringe to the reaction solution at -60°C. This temperature was maintained for 1.5 h at which time acetaldehyde (0.10 mL, 1.83 mmol) was added. After stirring for 2 additional hours at -60°C, the reaction mixture was quenched with saturated ammonium chloride solution and extracted with ether. The organic phases were dried (MgSO^) and freed of solvent. The aldol product (419 mg) was isolated as an oil. A portion of the aldol product (200 mg, approximately 0.73 mmol) was combined with methylene chloride (4 mL) and triethylamine (0.51 mL, 3.6 mmol). Methanesulfonyl chloride (0.23 mL, 2.9 mmol) dissolved in methylene chloride (3 mL) was added dropwise over 1 h. After 2 h, the reaction mixture was quenched with saturated sodium bicarbonate solution 168 followed by extraction with ether. The organic phases were subsequently extracted with saturated ammonium chloride solution and brine. The organic phase was dried (MgSO^),. and the solvent was removed in vacuo yielding a brown oil (224 mg). The mesylate was immediately dissolved in benzene (25 mL). 1,8-Diazabicyclol5.4.0]= undec-7-ene (0.48 mL, 3.2 mmol) was added, and the reaction mixture was heated at reflux overnight. The reaction mixture was diluted with ether (60 mL) and washed successively with saturated sodium bicarbonate solution (2x25 mL), saturated ammonium chloride solution (2x25 mL), and brine (25 mL). The ether solution was dried (MgS04) and concentrated under reduced pressure. The ethylidene lactone 45 (62 mg, 29%) was obtained in pure form following purification by MPLC (silica gel; 15% ethyl acetate in petroleum ether). Method B: Lithium diisopropylamide was prepared from n-butyllithium (1.31 mL, 2.17 mmol, 1.65 U in hexane) and diisopropylamine (0.30 mL, 2.17 mmol) in tetrahydrofuran (3 mL). This solution was cooled to -78°C, and lactone 44 (227 mg, 0.985 mmol) dissolved in tetrahydrofuran (3 mL) was added dropwise. After 45 min, bisImethoxy(thiocarbonyl)]disulfide (232 mg, 1.08 mmol) 169 dissolved in tetrahydrofuran (3 mL) was introduced fol lowed by stirring for 2 h at -78°C. Acetaldehyde (0.06 mL, 1.1 mmol) was then added. The mixture was stirred at -78°C for 2 h and at room temperature for 1.5 h. A color change from yellow to brown was noted during the latter period. The reaction mixture was diluted with ether (30 mL) and extracted with sodium bicarbonate solution (20 mL), saturated ammonium chloride solution (20 mL), and brine (20 mL) before being dried (MgSO^). The solvent was removed under reduced pressure, and the crude material was purified by MPLC (silica gel; 15% ethyl acetate in petroleum ether) furnishing the ethylidene lactone 45 (127 mg, 50%) as a yellow oil: IR (CDClg) cm~ 1 2920, 2850, 1770, 1680, 1460, 1330, 1255, 1220, 1120, 1010, 830; NMR (300 MHz, CDC13 ) <5 6.75-6.72 (m, 1H) , 4.62-4.58 (X of ABX, m, 1H), 3.75 (AB of ABX, Au = 36.0Hz, = 11.1Hz, = 3.4Hz, J BX = 3.7Hz, 2H) , 2.82-2.77 (m, 2H), 1.84 (dt, J = 7.0 and 2.0Hz, 3H), 0.85 (s, 9H), 0.05 (s, 3H), 0.04 (s, 3H); l3C NMR (20 MHz, CDCI 3 ) ppm 170.63, 134.54, 127.64, 77.05, 64.72, 27.03, 25.69, 18.15, 15.47, 5.55; mass spectrum calcd (M+-CH3) 241.1260, obsd 241.1298. 170 (±)-(3J5*,5E*)-5-l (.fceri-Butyldimethylsiloxy)methyl] = dihydro-3-methyl-3-vinyl-2(3lI)-furanone (46). In a 10 mL flask was prepared O lithium isopropylcyclohexylamide in tetrahydrofuran (2 mL) from N-butyllithium (0.36 mL, 0.59 mmol, 1.65 JM in hexane) and n- isopropylcyclohexylamine (0 . 1 0 mL, 0.59 mmol). The solution was cooled to -78°C and ethylidene lactone 45 (127mg, 0.495 mmol) dissolved in tetrahydrofuran (2 mL) was added. The enolate was per mitted to form over 1 h, and then HMPA (0.09 mL, 0.49 mmol) and methyl iodide (0.04 mL, 0.59 mmol) were sequen tially added. The mixture was stirred for 5 h at -78°C, after which time it was quenched with water (30 mL) and extracted with ether (2x20 mL). The combined organic phases were subsequently back extracted with saturated ammonium chloride solution and brine. The ether layer was dried (MgS04 ), and the solvent was removed under reduced pressure. The crude material was purified by MPLC (silica gel; 2% ethyl acetate in petroleum ether) furnishing 46 (25mg, 19%) as an oil: IR (CCl^) cm * 2935, 2860, 1785, 1643, 1462, 1257, 1050, 835; 1H NMR 171 (300 MHz, CDCI 3 ) 6 6.03 (dd, J. = 17.5 and 10.6Hz, 1H), 5.19 (d, J = 17.4Hz, 1H), 5.19 (d, il « 10.6Hz, 1H) , 4.57- 4.51 (X of 2 ABX systems, m, 1H), 3.78 (AB of ABX, Ay = 31.5Hz, = 11.3Hz, ■ 4.4Hz, JBX = 3.9Hz, 2H) , 2.19 (AB of ABX, Av = 84.3Hz, = 12.8Hz, = 6 .6 Hz, J BX = 9.0Hz, 2H), 1.38 (s, 3H), 0.88 (s, 9H), 0.08 (s, 3H), 0.07 (s, 3H); 13C NMR (75 MHz, CDCI3 ) ppm 179.34, 139.82, 114.70, 77.28, 63.96, 46.41, 36.09, 25.79, 23.51, 18.27, -5.36, -5.39; mass spectrum calcd (M+-C4 Hg) 213.0947, obsd 213.0939. (±)- (5fi*)-Dihydro-3-thiol-5-I (6'£*,7 '£*)-4•,5■,6',7 *- tetrahydro-6',7'-dimethylspiroII,3-dithiolane-2,1'- indanI-7'-yl]-3-vinyl-2(3fl)-furanone (47). In tetrahydrofuran (5 mL), lithium diisopropylamide was prepared from n-butyllithium (2.03 mL, 3.26 mmol, 1.6 H in hexane) and diisopropylamine (0.46 mL, 3.26 mmol). The solution was cooled to - 78°C, and lactone 34 (479 mg, 1.48 mmol) dissolved in tetrahydrofuran (5 mL) was added. 172 After 1.25 h, bisImethoxy(thiocarbonyl)Jdisulfide (349 mgr 1.63 mmol) was added, resulting in an immediate color change from light yellow to turbid brown. After 2.5 hr acetaldehyde (0.09 mlf 1.63 mmol) was introduced/ and the solution was stirred at -78°C for 3 h and at room temp erature for 1.5 h. The mixture was diluted with ether (80 mL) and extracted with saturated sodium bicarbonate solution (30 mL), saturated ammonium chloride solution (30 mL)r and brine (30 mL). The organic phase was dried (MgS04) and freed of solvent. The residue was purified by MPLC (silica gel? 7.5% ethyl acetate in petroleum ether) which gave both product isomers (47 mg, 8.3%). The more rapidly eluting diastereomer accounted for 33 mg and was obtained as an oil: IR (neat) cm * 2920, 1765, 1675, 1625, 1585, 1440, 1425, 1370, 1270, 1185, 1050, 1000, 945; 2H NMR (200 MHz, CgDg) <5 6.58 (dd, J. » 16.8 and 9.9Hz, 1H) , 5.31 (d, J. ® 16.8Hz, 1H) , 5.08 (d, ,1 = 9.9Hz, 1H), 4.82 (dd, J[ = 10.0 and 6.2Hz, 1H), 3.50 (dd, j; = 8 . 6 and 1.8Hz, 1H), 3.02-2.67 (m, 4H) , 2.34-2.27 (m, 2H), 2.01-1.97 (m, 2H), 1.73-1.23 (series of m, 7H), 1.56 (s, 3H), 0.85 (d, jl = 6.9Hz, 3H). The slower eluting diastereomer accounted for 14 mg and was isolated as a brown solid: mp 101.5-102.5°C; 1H NMR (200 MHz, CgDg) 5 6.47 (dd, J. = 16.7 and 9.8Hz, 1H), 5.29 (d, 1 = 16.7Hz, 173 1H), 5.11 (d, J. = 9.8Hz, 1H), 4.35 (dd, J[ = 10.9 and 5.8Hz, 1H), 3.24 (dd, J. = 12.0 and 8 .8 Hz, 1H), 2.96-2.74 (m, 4H), 2.41-2.31 (m, 2H), 2.07-2.01 (m, 2H), 1.71-1.30 (series of m, 7H) , 1.58 (s, 3H), 0.92 (d, jI = 6 .8 Hz, 3H) ; mass spectrum (on mixture) calcd 382.1095, obsd 382.1201. (+) - <3JB,5£)-3-Ethylidenedihydro-5-1 ( 6 '£,7 'E)-4 • ,5 • , 6 • ,7 '- tetrahydro-6 ', 1 '-dimethylspiro[l,3-dithiolane-2,l indan]-7'-yl]-2(3fl)-furanone and (->-(3£,5£)-3-Ethylidenedihydro-5-I(6'£,7'S)-4'f5',6 *,7'- tetrahydro-6 ',7'-dimethylspiro[l,3-dithiolane-2,l indan]-7'-yl]-2(3fl)-furanone (49). Lithium diisopropylamide was prepared in ether (1 mL) from diisopropylamine (0.13 mL, 0.924 mmol) and n-butyllithium (0.60 mL, 0.924 mmol, 1.55 JH in hex ane) . The solution was cooled to -60°C at which point lactone (-)-35 (250 mg, 0.770 mmol), I“)D -14.5°, dissolved in ether (3 mL) and tetrahydrofuran (2 mL) was added. Enolate formation was permitted to occur over 1.75 h with the temperature of 174 the reaction mixture maintained between -40°C and -60°C. The temperature of the solution was then lowered to -78°C, and freshly distilled acetaldehyde (0.045 mL, 0.808 mmol) was introduced via syringe. The resulting murky solution was stirred for 4.5 h at -78°C. The reaction mixture was was quenched with saturated ammonium chloride solution (15 mL) and extracted with ether (3x20 mL) before being dried (MgSO^) and freed of solvent. The condensation product (280 mg, 99%) was obtained as a viscous oil and used for the next step without further purification: IR (CDC13 ) cm” 1 3600, 3490, 2980, 2940, 2890, 1750, 1628, 1455, 1425, 1365, 1190, 1108, 1065, 1010, 890, 815. A solution of the condensation product, methylene chloride (4 mL), and triethylamine (0.53 mL, 3.80 mmol) was cooled to 0°C, and methanesulfonyl chloride (0.24 ml, 3.04 mmol) dissolved in methylene chloride (3 mL) was added dropwise over 40 min. The reaction mixture was allowed to warm to room temperature with continued stir ring for 2 h. The turbid brown solution was partitioned between saturated sodium bicarbonate solution (10 mL) and ether (40 mL). The organic phase was saved and extracted with saturated ammonium chloride solution (20 mL) and brine (20 mL) before drying (MgSO^). Removal of the 175 solvent in vacuo afforded a dark brown oil which was used immediately in the next step. The crude mesylate was dissolved in benzene (50 mL) containing l,8-diazabicyclo-E5.4.0]undec-7-ene (0.50 mL, 3.32 mmol), and the reaction mixture was refluxed over night. TLC analysis (25% ethyl acetate in petroleum ether) showed only one compound which eluted more rapidly than the starting material. The reaction mixture was diluted with ether (60 mL) and sequentially washed with saturated sodium bicarbonate solution (2x25 mL), satur ated ammonium chloride solution (2x25 mL), and brine (25 mL) before drying (MgS04) and concentration in vacuo. The product was purified by MPLC (silica gel; 25% ethyl acetate in petroleum ether) furnishing ethylidene lactone (-)-49 (130 mg, 49%) as a light yellow oil: Ealp3^ -41.6° (n 0.540, CHCI 3 ); IR (CC14 ) cm” 1 2960, 2920, 2880, 1760, 1685, 1630, 1435, 1332, 1275, 1215, 1130, 993; 1H NMR (300 MHz, CDC13 ) 6 6.73-6.66 (m, 1H), 4.88 (dd, Z = 8 . 8 and 5.7Hz, 1H), 3.38-3.10 (series of m, 4H), 2.77 (qt, Z = 8.7 and 2.3Hz, 1H), 2.39-2.01 (series of m, 6 H), 1.87-1.58 (series of m, 3H), 1.85 (dt, J = 7.1 and 1.9Hz, 3H), 1.47 (s, 3H), 1.00 (d, 1 = 6 .6 Hz, 3H); 13C NMR (75 MHz, CDC13) ppm 171.00, 146.82, 134.48, 133.91, 128.30, 80.47, 77.55, 48.87, 43.73, 41.45, 40.49, 38.73, 34.05, 176 28.51, 27.50, 26.20, 20.54, 16.09, 15.82; mass spectrum calcd (M+ ) id/£ 350.1374, obsd 350.1377. The enantiomer was prepared in like manner from (+)-35 (271 mg, 0.835 mmol), IalD +10.3. The inter mediate condensation product was obtained in 98% (302 mg) yield, and the ethylidene lactone (+)-49 in 38% (109 mg), [ < * ] D 2 2 +36.3° (£ 2.39, CHC13). <-)-<3£,5B)-3-Ethylidenedihydro-5-[(6'£,7\B)-4',5',6 ',7'- tetrabydro- 6 1 ,7'-dimethylspiro[l,3-dithiolane-2rl indanl-7'-ylJ-2(3fl)-furanone and (+)-<3£,5£)-3-Ethylidenedihydro-5-I(6 '£,7'£)-4',5•,6 ',7•- tetrahydro-6 ',7,-dimethylspiroIl,3-dithiolane-2,l*- indan]-7'-yl]-2(3£)-furanone (48). The enantiomers were prepared analogously to their diastere- omers. (-)-Ethylidene lactone 48 (52 mg, 21%) was produced from (+)-lactone 34 (275 mg, 0.849 mmol), (“Jp +5.7°, via the condensation product (264 mg, 84%): IR (CDCl^ cm-* 3610, 3460, 2970, 2940, 1745, 1630, 1445, 1435, 1375, 1265, 177 1195, 1100, 1020, 890, 810; mass spectrum calcd (M+ ) jn/fi 368.1480, obsd 368.1487. (+)-Ethylidene lactone 48 (102 mg, 28%); [a] ^ 3 +27.5° (jc 0.415, CHCl-j), mp 101-104°C, an off-white solid was obtained from the condensation product (387 mg, 94%) which in turn was derived from (-)-lactone 34 (364 mg, 1.12 mmol), [a)D -12.0°: IR (CHC13 ) cm” 1 2935, 1745, 1685, 1460, 1380, 1332, 1277, 1140, 1067, 980; 1H NMR (300 MHz, CDC13) <5 6.76-6.67 (m, 1H) , 4.90 (t, 2 = 7.8Hz, 1H), 3.38-3.05 (series of m, 4H), 2.80-2.70 (m, 1H), 2.45-2.01 (series of m, 6 H), 1.83 (dt, 2 = 7.2 and 1.9Hz, 3H), 1.80-1.57 (series of m, 3H), 1.48 (s, 3H), 1.03 (d, 2 = 6 .8 Hz, 3H); 13C NMR (75 MHz, CDCl-j) ppm 170.99, 145.34, 135.94, 133.90, 128.79, 80.73, 77.59, 49.27, 43.87, 40.89, 40.29, 38.70, 33.96, 29.16, 27.09, 26.02, 19.83, 16.55, 15.62; mass spectrum calcd (M+) m/£ 350.1374, obsd 350.1419. 178 (-) - <3S,5£) -Di.hydro-3-methyl-5- E ( 6 7 '£) -4 • ,5 1 r 6 •,7 '- tetrahydro- 6 ', 7'-dimethylspi roEl,3-dithiolane-2 ,1 '- indanl-7'-yl)-3-vinyl-2(3B)-furanone and (+)-(3B,5S)-Dihydro-3-methyl-5-E (6 '£,7'£)-4•,5• , 6 *,7'- tetrahydro- 6 ■,7'-dimethylspiroE1 ,3-dithiolane-2 ,1 '- indanl-7'-yl]-3-vinyl-2(3B)-furanone (50). In tetrahydrofuran (1 mL) lith ium isopropylcyclohexylamide was prepared from n-butyllithium (0.143 mL, 0.222 mmol, 1.55 H in hexane) and isopropylcyclohexyl- amine (0.036 mL, 0.222 mmol). This solution was cooled to -78°C, and (-)-ethylidene lactone 48 (52 mg, 0.148 mmol) dissolved in tetrahydrofuran (2 mL) was introduced via syringe. The enolate was permitted to form over 3 h. HMPA (0.026 mL, 0.148 mmol) and methyl iodide (0.014 mL, 0 . 2 2 2 mmol) were added sequentially, and stirring was continued at -78°C for 5 h. The reaction mixture was quenched with water (10 mL) and extracted with ether (30 mL). The organic phase was extracted with saturated ammonium chloride solution (10 mL). The combined aqueous phases were back extracted with ether (2x10 mL). The combined organic phases were dried (MgSO^) and freed of 179 solvent. The crude product was purified by MPLC (silica gel; 7.5% ethyl acetate in petroleum ether) yielding (-)- 50 (20 mg, 36%), (alp2 5 -10.0° (£ 1.69, CHCI3 ), as a light yellow oil: IR (CHC13 ) cm" 1 2975, 2935, 1765, 1645, 1460, 1378, 1260, 1103, 1000; 1H NMR (300 MHz, CDC13) 6 6.08 (dd, 0. = 17.5 and 10.6Hz, 1H), 5.20 (d, 1 = 17.4Hz, 1H), 5.20 (d, jI = 10.7Hz, 1H) , 4.85 (dd, il = 11.2 and 5.7Hz, 1H), 3.37-3.14 (series of m, 4H), 2.61 (t, il = 11.9Hz, 1H), 2.42-2.04 (series of m, 6 H), 1.89 (dd, J = 12.7 and 5.7Hz, 1H), 1.75-1.48 (series of m, 3H), 1.51 (s, 3H), 1.36 (s, 3H), 1.04 (d, J = 6 . 6 Hz, 3H); 13C NMR (75 MHz, CDC13 ) ppm 179.52, 145.17, 140.12, 136.19, 114.61, 80.33, 76.82, 49.19, 46.68, 43.01, 41.22, 40.18, 38.61, 33.91, 27.27, 26.14, 26.03, 22.60, 20.05, 16.84; mass spectrum calcd (M+) m / & 364.1530, obsd 364.1524. The (+)-ethylidene lactone 48 (33.2 mg, 0.0941 r \ mmol), (alp +27.5 , was treated analogously and produced (+)-50 (4.1 mg, 12%), (alp2 3 +8.9° (£ 0.32, c h c i 3). 180 (+)-(3B,5fi)-Dihydro-3-methyl-5-l (6 '£,7 ' £ > - 4 ' ,5 ' , 6 ' ,7 •- tetrahydro-6 ',7'-dimethylspiroII,3-dithiolane-2 , 1 indanl-7'-yl]-3-vinyl-2(3fl)-furanone and (-) - <3S, 5£) -Dihydro-3-methyl-5- [ (6 7 •£) -4 • f 5 •, 6 •, 7 •- tetrahydro-6 ',7'-dimethylspiroII,3-dithiolane-2 , 1 '- indan]-7'-yl]-3-vinyl-2(3jJ)-furanone (51). (+)-Lactone 51 (27.8 mg, 51%), [a) D 2 3 +6.3° <£ 1.06, CHCI 3 ) was produced from (+)-49 (53.0 mg, 0.151 mmole), (“Ip2 2 +36.3°, in a fashion analogous to the pre viously reported diastereomers. Likewise, ethylidene lactone (-)-49 (110 mg, 0.314 mmol), [ < * ] D 2 2 -41.6°, yielded (-)-51 (23 mg, 20%), I “Ip2 3 -14.0° (£ 1.07, CHCl^), as a yellow oil: IR (CHCl-j) cm ^ 2965, 2930, 2880, 2840, 1765, 1644, 1456, 1375, 1340, 1198, 1155, 1100, 990, 850; (300 MHz, CDCI 3 ) <5 6.06 (dd, J[ = 17.5 and 10.6Hz, 1H), 5.19 (d, J = 17.3Hz, 1H), 5.19 (d, J = 10.8Hz, 1H), 4.91 (dd, jI » 11.5 and 5.6Hz, 1H), 3.40- 3.16 (series of m, 4H), 2.69 (t, il = 12.1Hz, 1H), 2.42- 2.04 (series of m, 6 H), 1.98 (dd, H = 12.7 and 5.6Hz, 1H), 1.75-1.48 (series of m, 3H), 1.46 (s, 3H), 1.35 (s, 3H) , 1.00 (d, il = 6.5Hz, 3H) ; 13C NMR (75 MHz, CDCI3 ) ppm 181 179.45, 145.54, 140.15, 135.09, 114.52, 80.15, 77.73, 49.42, 46.48, 42.75, 40.81, 40.06, 38.24, 34.00, 27.12, 26.31, 26.26, 22.32, 20.94, 16.09; mass spectrum calcd (M+ ) m/£ 364.1531, obsd 364.1540. (-t-)-Pleuromutilin (I).**2 Pleuromutilin (1), obtained as a crude fermentation product, was purified by chromatography (sil \ ica gel; 50% ethyl acetate in methylene chloride). Subsequent recrystallization from methylene chloride allowed isolation of the pure material as colorless crystals: mp 165-166°C (litlld mp 165-166°C); la^ 2 2 +34.7° (£ 1.03, CHCI3 ) (litlld tcUD 2 5 +33.9° (£ 1.00, CHCI3 )); IR (Nujol) cm - 1 3480, 3380, 2900, 1730, 1450, 1370, 1225, 1210, 1150, 1090, 1025, 1005, 990, 960, 925, 905; 1H NMR (300 MHz, CDCI3 ) 6 6.51 (dd, J = 18 and 11Hz, 1H) , 5.84 (d, J. = 9Hz, 1H), 5.36 dd, J = 11 and 2Hz, 1H), 5.22 (dd 1 = 18 and 2Hz, 1H), 4.11-3.96 (m, 2H), 3.37 (br s, 1H), 2.61- 2.03 (series of m, 5H), 1.83-1.08 (series of m, 10H), 182 1.43 (s, 3H), 1.17 (s, 3H), 0.89 (d, Z = 8 Hz, 3H), 0.71 (d, Z = 7Hz, 3H); 13C NMR (20 MHz, CDC13 ) ppm 216.83, 172.17, 139.03, 117.24, 74.64, 69.79, 61.35, 58.14, 45.45, 44.85, 44.06, 41.87, 36.65, 36.11, 34.47, 30.46, 26.88, 26.46, 24.88, 16.57, 14.81, 11.47; mass spectrum (FAB) (M+ ) 378. Tiamulin (2f). Tiamulin was obtained quantita- OH tively from tiamulin hydrogen fumarate by dissolving the salt into 1 0 % sodium hydroxide solu- 2 tion and extracting with methylene chloride. The combined organic phases were dried (MgSO^) and freed of solvent to provide tiamulin as a light brown oil: IR (neat) cm ^ 3540, 2920, 1715, 1445, 1400, 1370, 1270, 1105, 1000, 970, 720; 2H NMR (300 MHz, CDC13) <5 6.42 (dd, jl = 18 and 11Hz, 1H), 5.69 (d, Z = 9Hz, 1H), 5.28 (dd, J = 12 and 2Hz, 1H), 5.13 (dd, 1 = 17 and 2Hz, 1H), 3.23 (br t, J[ = 7Hz, 1H), 3.11 (s, 3H), 2.61 (s, 4H), 2.47 (q, J[ = 7Hz, 4H), 2.34-2.00 (series of m, 5H), 1.75-1.22 (series of m, 10H), 1.39 (s, 3H), 1.11 183 (s, 3H) , 0.97 (t, J. = 7Hz, 6 H), 0.82 (d, J = 7Hz, 3H) , 0.68 (d, jI = 6 Hz, 3H) . (-)-Pleuromutilone (52).113 Pleuromutilin (1) (5.00 g, 13.2 mmol) was dissolved in methylene chloride (200 mL) and pyridinium chlorochromate (4.2 g , 19 mmol) was added. The mixture was stirred overnight at room temperature. Ether (300 mL) was then added and the mixture stirred to obtain homogeneity. Passage through a short column (Florisil; ether) gave crude pleuromutilone 52 after solvent evaporation. Crystallization from methylene chloride furnished pure pleuromutilone (4.28 g, 8 6 %) as colorless crystals: mp 178.0-179.5°C (lit* * * 5 mp 178-179°C) ; U ] D2* -8.2° (£ 2.45, CHCI3 ) (litlla Eot Ip2 0 - 8 ° (£ 1.038, CHC13)) IR (Nujol) cm- 1 3480, 2900, 1730, 1695, 1455, 1370, 1270, 1230, 1205, 1150, 1080, 1070, 1050, 985, 970, 945, 920, 905; *H NMR (300 MHz, CDCI3 ) & 6 . 6 6 (dd, Z = 17 and 10Hz, 1H), 6.03 (d, 1 = 8 Hz, 1H), 5.36 (d, 1 = 10Hz, 1H), 5.07 (d, Z = 17Hz, 1H), 4.23-4.03 184 (m, 2H), 3.34-3.25 (m, 1H), 2.50-2.02 (series of m, 4H), 1.73-1.16 (series of m, 9H), 1.48 (s, 3H), 1.16 (s, 3H), 1.10 (d, JL = 6 Hz, 3H) , 0.76 (d, J. = 5Hz, 3H) ; 13C NMR (20 MHz, CDCI3 ) ppm 188.47, 186.82, 153.37, 127.49, 110.16, 79.33, 72.05, 64.52, 62.63, 58.26, 51.90, 51.66, 50.39, 49.08, 45.10, 43.16, 39.32, 36.90, 35.20, 28.89, 27.33, 26.46; mass spectrum (FAB) (M+-l) m/fi 375. Tiamulone (53). Tiamulin hydrogen fumarate (5.00 g, 8 . 2 0 mmol) was dissolved in 1 0 % sodium hydroxide solution (100 mL) This solution was 2 extracted with methylene chlor ide (4x50 mL). The combined organic phases were dried (MgSO^) and filtered into a 500 mL flask. Pyridinium chlorochromate (6.00 g, 27.8 mmol) was added, and the reaction mixture was stirred at room temperature overnight. The reaction mixture was extracted with 10% potassium hydroxide solution (2x250 mL), water (250 mL), and brine (2x100 mL). The organic phase was dried (MgSO^) and freed of solvent. 185 Chromium salts were removed by passage of the residue through a short chromatography column (silica gel; methanol). A second chromatography (silica gel; 5% methanol in methylene chloride) furnished tiamulone (53) (3.90 g, 97%) as a pale yellow oil: IR (neat) cm- 1 2970, 2940, 2810, 1730, 1700, 1620, 1455, 1415, 1375, 1330, 1280, 1200, 1150, 1115, 1095, 1060, 1000, 990, 965, 925, 735; 1H NMR (300 MHz, CDCl-j) <5 6.50 (dd, J[ = 17.6 and 10.8Hz, 1H) , 5.87-5.77 (m, 1H) , 5.18 (d, jI = 10.7Hz, 1H) , 4.90 (d, J = 17.5Hz, 1H), 3.11 (br s, 2H), 3.05-2.75 (series of m, 2H), 2.64-2.51 (m, 4H), 2.41 (q, J = 7.1Hz, 4H), 2.18-1.97 (m, 3H), 1.91 (dd, J. = 15.6 and 8.5Hz, 1H), 1.54-1.25 (series of m, 7H), 1.34 (s, 3H), 0.99 (s, 3H) , 0.92 (t, ,1 = 7.4Hz, 6 H) , 0.88 (d, il = 7.1Hz, 3H) , 0.64 (d, jI = 5.8Hz, 3H) . 186 (6£,7E)-7-l (2£, 4S)-5-Ethyltetrahydro-5-hydroxy-4-methyl- 4-vinyl-2-£ury1J-4,5,6,7-tetrahydro-6 ,7-dimethyl-l- indanone (54).**^ Pleuromutilone (52) (500 mg, 1.33 mmol) was dissolved in a 1 0 % solution of potassium hydroxide in ethanol (50 mL), and the solution was refluxed overnight. After cooling, the solution was poured into water (100 mL) and acidified with concentra ted hydrochloric acid. The solution was extracted with dichloromethane (3x75 mL). The combined organic phases were extracted with saturated sodium bicarbonate solu tion, dried (MgS04 ), and freed of solvent. Purification was achieved by column chromatography (Florisil; 10% ethyl acetate in methylene chloride) to afford hemiacetal 54 (351 mg, 83%) as a pale yellow oil: IR (neat) cm- 1 3450, 2950, 1730, 1680, 1620, 1450, 1410, 1370, 1275, 1230, 1185, 1115, 990, 965, 900; NMR (300 MHz, CDC13) i§ 6.1-5 . 8 (m, 1H), 5.1-4.9 (m, 2H), 4.6 and 4.1 (m, 1H), 2.6-0.9 (series of m, 26H). 187 (-)-(3B,5£) -Dihy dr o-3-methy 1-5- [ (4S, 5fi) -4,5,6 ,7- tetrahydro-4,5-dimethyl-3-oxo-4-indanyl]-3-vinyl-2(3fl) - furanone (55). Pleuromutilone (52) (3.00 g, 7.98 mmol) was treated with a 1 0 % solution of potassium hydro xide in ethanol (300 mL) at the reflux temperature overnight. After cooling, the solution was poured into water (600 mL), acidified with concentrated hydrochloric acid, and extracted with methylene chloride (3x500 mL). The combined organic phases were washed with saturated sodium bicarbonate solution, dried (MgSC>4), and freed of solvent. The crude hemiacetal 54, a yellow oil, was immediately dissolved in methylene chloride (150 mL). Pyridinium chlorochromate (10 g, 46 mmol) was added, and the mixture was stirred overnight at room temperature. Ether (600 mL) was added and the mixture was passed through a short column (Florisil; ether). After concentration JLn vacuo, the crude material was subjected to a second chromatography (silica gel; 30% ethyl acetate in petroleum ether) to afford lactone 55 (1.42 g, 60%) as a colorless solid: mp 123-124°C; [°0D2* -16.9 (£ 1.54, 188 CHC13); IR (Nujol) cm” 1 2900, 1760, 1680, 1630, 1455, 1375, 1295, 1280, 1190, 1180, 1115, 1085, 1045, 1010, 985, 925; XH NMR (300 MHz, CDCI3 ) « 6.00 (dd, Z = 1 7 and 9Hz, 1H), 5.15 (d, jI = 17Hz, 1H) , 5.14 (d, Z = 9Hz, 1H) , 5.02 (dd, Z - 10 and 5Hz, 1H), 2.50-2.15 (series of m, 10H), 1.84-1.74 (m, 1H), 1.30 (s, 3H), 1.22 (s, 3H), 0.96 (d, z = 5Hz, 3H); 13C NMR (75 MHz, CDC13) ppm 208.16, 179.54, 175.53, 140.31, 140.05, 114.71, 78.93, 46.37, 40.47, 38.21, 36.43, 35.21, 29.41, 27.38, 26.90, 23.02, 19.00, 16.75. Anal. Calcd for cigH 2 4 ° 3 ! c' 74.97; H, 8.39. Found: C, 74.66; H, 8.32. (-)-(3£,5£)-Dihydro-3-methyl-5-1(6'£, 7 '£)-4•,5 •,61,71 ■- tetrahydr0-6 ',7•-dimethylspiroII,3-dithiolane-2 ,1 '- indan]-7'-yl]-3-vinyl-2(3B)-furanone (50).113 Lactone 55 (100 mg, 0.347 mmol) was dissolved in benzene (25 mL) along with a catalytic amount of p-toluenesulfonic acid ( 1 0 mg). Ethanedithiol (0.20 mL, 2.38 mmol) was added, and the mixture 189 was heated at reflux overnight in a Dean-Stark apparatus. After 24 hr additional ethanedithiol (0.10 ml) was added. No starting material remained after a total of 50 h. The reaction mixture was cooledr poured into benzene (50 mL), extracted with 1 0 % sodium hydroxide solution and water, dried (MgS04 ), and freed of solvent. The residue was purified chromatographically (silica gel; methylene chloride) to yield thioacetal 50 (98 mgf 78%) as a colorless oil which slowly crystallized as an off-white solid upon cooling: mp 75-76°C; ta)D -16.3° (CI^C^) r [a] D 2 5 -6.9° (£ 2.07r CHCI3 ); IR (neat) cm- 1 2960, 2920, 2870, 2830, 1765, 1640, 1450, 1420, 1370, 1330, 1265, 1190, 1090, 915, 845, 810, 730, 700; 1 H NMR (300 MHz, CDCI3 ) 6 6.09 (dd, J = 17.5 and 10.5Hz, 1H), 5.20 (d, J. = 16.7Hz, 1H), 5.20 (d, 1 = 11.3Hz, 1H), 4.85 (dd, 1 = 11.1 and 5.8Hz, 1H) , 3.37-3.12 (series of m, 4H), 2.61 (t, JL = 12.0Hz, 1H), 2.47-2.09 (series of m, 6 H), 1.89 (dd, J. = 12.6 and 5.7Hz, 1H), 1.75-1.43 (series of m, 3H), 1.52 (S , 3H), 1.36 (s, 3H), 1.04 (d, J « 6 .6 Hz, 3H); 13C NMR (20 MHz, CDCI3 ) ppm 179.36, 145.01, 140.03, 136.09, 114.42, 80.14, 77.53, 49.00, 46.46, 42.81, 41.06, 40.02, 38.45, 33.77, 27.10, 25.94 (2 carbons) 22.49, 19.88, 16.66; mass spectrum calcd (M+ ) m/£ 364.1531, obsd 364.1538. 190 Anal. Calcd for ^ 2 QH2 g0 2 S 2 : C, 65.89; H, 7.74. Found: C, 65.71; H, 7.80. (3JS# 5JK) -2-Ethyltetrahydro-3-methy1-5- [ (6 '£,7 '£) - 4',5',61,7'-tetrahydro-6',7'-dimethylspiro[l,3- dithiolane^rl'-lndanl^'-yll-S-vinyl^-furanol (56) .113 Lactone 50 (226 mg, 0.621 mmol) dissolved in ether (10 mL) was cooled to -50°C. A solution of ethyllithium (0.41 mL r 0.62 mmol, 1.5 in ether) was added dropwise. The reaction mixture was stirred at -50°C for 1 hf warmed to -20°Cr and stirred for 0.5 h. At that timef the mixture was cooled to -50°C; ethyllithium (0.20 mL, 0.30 mmol) was again added dropwise. The solution was allowed to warm to -20°C, recooled to -50°C, and the mixture was quenched with saturated ammonium chloride solution (1 mL). After reaching room temperature, the mixture was diluted with water, extracted with ether, dried (MgSO^), and freed of solvent to furnish 56 (220 mg, 90%) as a very unstable yellow oil: IR (neat) cm” 1 3400, 2900, 1700, 1630, 1450, 191 1370/ 1265/ 1195, 1115, 1055, 1020, 995, 980, 960, 900; 1H NMR (300 MHz, CDCI3 ) 6 6.07-5.75 (m, 1H), 5.20-4.93 (m, 2H), 4.64-4.40 (m, 1H), 3.33-2.98 (m, 4H), 2.63- 0.73 (series of m, 23H); mass spectrum calcd (M+-H 2 0) m/£ 376. <3S, 5fi)-Tetrahydro-3-methyl-5-1(6'£,7'£)-4',5',6',7•- tetrahydr0-6 ',7'-dimethylspiroll,3-dithiolane-2 ,1 '- 113 indanl-7'-yl)-3-vinyl-2-furanol (57). A solution of lactone 50 (170 OH mg, 0.466 mmol) in ether (10 mL) was cooled to -20°C, and diisobutylaluminum hydride solution (0.60 mL, 0.60 mmol, 1.0 JH in hexane) was intro duced dropwise. After 30 min, the reaction mixture was cooled to -78°C, water (1 mL) was added, and the temp erature was allowed to warm to room temperature. The mixture was partitioned between ether and water. A minimum quantity of dilute hydrochloric acid was employed to dissolve the aluminum salts. The organic phase was extracted with saturated sodium bicarbonate 192 solution, and water, and dried (MgSO^). Following solvent removal under reduced pressure, lactol 57 (167 mg, 98%) was obtained as a yellow oil: IR (neat) cm- 1 3420, 2900, 1630, 1450, 1415, 1360, 1260, 1000, 905, 800, 720; 1H NMR (300 MHz, CDCl-j) 6 6.00-5.80 (m, 1H) , 5.06-4.93 (m, 2H), 4.81 (br s, 1H), 4.44 and 4.23 (two dd, J = 4.5 and 4.5 Hz, 1H total), 3.31-3.01 (m, 4H), 2.80 (br s, 1H), 2.42-1.93 (series of m, 6 H), 1.89-1.47 (m, 3H), 1.52 and 1.44 (two s, 3H total), 1.37 (dd, Z = 11.2 and 4.5 Hz, 1H), 1.09 and 1.05 (two s, 3H total), 0.96 (d, J = 7.7Hz, 3H); mass spectrum calcd (M+-2) m/£ 364. ( 6 'B, 7 'B>-4',5',6 ',7•-Tetrahydro-6 ',7•-dimethyl-7•- I (2B,4jS) -tetrahydro-5-methoxy-4-methy 1-4-viny 1-2- furyllspiroII,3-dithiolane-2,l'-indanl (58). Lactol 57 (410 mg, 1.12 mmol) and 2,6-lutidine (0.2 mL, 1.7 OCH mmol) were dissolved in methy lene chloride (20 mL) at room temperature. ±-Butyldimethyl- silyl trifluoromethanesulfo- nate (0.5 mL, 2.2 mmol) was added. After being stirred overnight, the reaction mixture contained unreacted starting material. Additional i-butyldimethylsilyl trifluoromethanesulfonate (0.3 mL) was added. When no further change was observed by TLC analysis, methanol (0.5 mL) was added. The mixture was washed with saturated sodium bicarbonate solution, 1 0 % potassium hydrogen sulfate solution, and brine. The organic phase was dried (MgSO^) and freed of solvent. The oil was purified by MPLC (silica gel; 5% ethyl acetate in petroleum ether) to supply 58 (236 mg, 55%) as a colorless oil: IR (neat) cm * 2900, 1630, 1440, 1365, 1270, 1230, 1175, 1075, 990, 900, 845, 805, 770, 725, 670; 2H NMR (300 MHz, CDClj) <5 6.03-5.85 (m, 1H) , 5.07- 4.86 (m, 2H), 4.37-4.25 (m, 2H), 3.36 and 3.31 (two s, 3H total), 3.34-2.99 (m, 4H), 2.64 (t, = 11.3Hz, 1H), 2.45-1.54 (series of m, 9H), 1.47 and 1.43 (two s, 3H total), 1.37 (dd, j[ = 12.3 and 4.9Hz, 1H), 1.07 (d, J = 5.4Hz, 3H), 1.06 (s, 3H); mass spectrum calcd (M ) m/e. 380.1844, obsd 380.1855. 194 2 - 1[(3fir 5£)-Tetrahydro-3-methyl-5-16 'B,7 '£)- 4',5*, 6 ',7'-tetrahydro- 6 ',7'-dimethylspiro[l#3- dithiolane-2,1’-indan]-7'-ylJ-3-viny1-2- fury 11 ozy] ethanol (59). Lactol 57 prepared from lac tone 50 (300 mg, 0.824 mmol) by diisobutylaluminum hydride reduction and aqueous work up was immediately redissolved in benzene (25 mL). Pyridinium tosylate (30 mg) and ethylene glycol (0.2 mL, 3.6 mmol) were added, and the reaction mixture was refluxed in a Dean-Stark apparatus overnight. The reaction mixture was extracted with saturated sodium bicarbonate solu tion, dried (MgS04 ), and freed of solvent. Purifica tion was achieved chromatographically (silica gel; 2 0 % ethyl acetate in hexane) to afford 59 (232 mg, 69%) as a colorless oil: IR (neat) cm * 3420, 2860, 1625, 1410, 1365, 1265, 980, 800, 720, 665; 1H NMR (300 MHz, CDC13 ) 6 6.03-5.82 (m, 1H), 5.05-4.94 (m, 2H), 4.71- 4.31 (series of m, 2H), 3.80-3.45 (series of m, 4H), 3.33-3.02 (series of m, 4H), 2.81 (br s, 1H), 2.58 (t, Z = 11.7Hz, 1H), 2.41-1.97 (series of m, 6 H), 1.79-1.30 195 (series of m, 4h ), 1.43 and 1.41 (two s, 3H total), 1.10 and 1.07 (two s, 3H total), 1.03 and 0.99 (two dr 3. = 6.7Hzr 3H total). 2 - 1 [ <3fi, 5£) -Tetrahydro-3-methy1-5- [ 6 7 •£) - 4',5',6 'r7*-tetrahydro-6 ', 7 '-dinethylspiro(l,3- ditbiolane-2,1'-indanl-7'-yl]-3-viny1-2- fury 1 1 thiolethanethiol (60).113 Lactone 50 (500 mgr 1.37 mmol) dissolved in ether was reduced with diisobutylaluminum hydride to furnish lactol 57 as previously described. The crude lactol was dissolved in benzene (50 mL) with ethanedithiol (0.5 mL, 6.0 mmol) and pyridinium p-toluenesulfonate (50 mg). This mixture was refluxed for 4 h. The reaction mixture was extracted with 10% potassium hydroxide solution (2x40 mL)r dried (MgS04), and freed of solvent to afford 60 (542 mg f 91%) as a yellow oil: IR (neat) cm * 2890, 1620, 1410, 1360, 1260, 1200, 1000, 910, 725; 2H NMR (200 MHz, CDC13) <5 5.96-5.75 (m, 2H), 5.09-4.96 (m, 196 2H), 4.85-4.66 (m, 1H), 4.39-4.22 (m, 1H), 3.32-2.95 (series of m, 4H), 2.83-2.58 (series of m, 4H), 2.49 (tr J. = 11.9 Hz, 1H), 2.41-1.83 (series of m, 6 H), 1.71-1.36 (series of m, 4H), 1.45 and 1.44 (two s, 3K total), 1.15 and 1.09 (two s, 3H total), 1.00 and 0.96 (two d, i = 6 .8 Hz, 3H total). (aJB, 6 *£, 7 '£) - a- [ <2£> -2- (1,3-Di thiolan-2-yl) -2-methyl-3- buteny11-4',5*,6',7'-tetrahydr o-6',7'-dimethylspiro= [l,3-dithiolane-2,1 '-indanl-7'-methanol (61) Lactone 50 (620 mg, 1.70 mmol) dissolved in ether (10 mL) was cooled to -78°C, and a solu tion of diisobutylaluminum hy dride (2.0 mL, 2.0 mmol, 1.0 JH in hexane) was added dropwise. Upon completion of the reaction as indicated by TLC analysis, the reaction mixture was quenched with water (1.0 mL) and allowed to warm to room temperature. Hydrochloric acid (10%) was added, and the mixture was extracted with ether. The combined organic phases were dried (MgS04) and freed of solvent to provide lactol 57 The lactol was immediately redissolved in methyl ene chloride (20 mL). Ethanedithiol (0.4 mL, 4.8 mmol) followed by aluminum chloride (50 mg) were added, and the reaction mixture was stirred at room temperature overnight. The reaction mixture was then extracted with 10% potassium hydroxide solution (2x20 mL), dried (MgSO^), and concentrated under reduced pressure. The residue was purified chromatographically (silica gel; 5% ethyl acetate in petroleum ether) to give alcohol 61 (430 mg, 58%) as a colorless oil which slowly crystal lized on standing: mp 103-105°C; IR (CDClg) cm- 1 3420, 2980, 2940, 2850, 1635, 1460, 1435, 1420, 1380, 1280, 1270, 1250, 1225, 1140, 1100, 1070, 1030, 1010, 960, 920, 860, 825, 800; 1H NMR (300 MHz, CDCl-j) 5 5.88 (dd, Z = 17.4 and 10.8Hz, 1H), 5.08 (d, il = 10.8Hz, 1H), 5.02 (d, Z = 17.4Hz, 1H), 4.66 (s, 1H), 4.35 (br d, Z = 7.9Hz, 1H), 3.69 (br s, 1H), 3.31-3.10 (m, 4H), 3.09- 3.02 (m, 4H), 2.39-1.37 (series of m, 11H), 1.21 (s, 3H), 1.18 (s, 3H), 1.03 (d, z = 6.9Hz, 3H); I3C NMR (20 MHz, CDCI 3 ) ppm 144.09, 143.55, 139.04, 114.28, 76.90, 71.51, 65.05, 50.19, 46.26, 45.87, 41.41, 40.10, 39.80, 38.93, 38.83, 37.81, 34.27, 28.49, 27.13, 24.46, 21.02, 17.47; mass spectrum calcd (M+) m/£ 442.1492, obsd 442.1499. 198 The mixed acetal 60 was also converted to 61 with Lewis acid catalysis. Lactone 50 (580 mgr 1.59 mmol) was reduced with diisobutylaluminum hydride and subse quently treated with ethanedithiol in the presence of a protic catalyst as previously reported to provide 60. The crude mixed acetal was immediately redissolved in methylene chloride (20 mL) and stirred at room tempera ture for 30 h in the presence of aluminum chloride (100 mg). No additional ethanedithiol was introduced. The reaction mixture was extracted with waterf dried (MgS04), and concentrated in vacuo. Following chroma tographic purification as described above, a pale yellow oil (290 mg, 41%) which corresponded to the material isolated above was obtained. 199 <6£,7B)-7-I(IB,35)-3-(1,3-Dithiolan-2-yl)-l-hydroxy-3- methy1-4-pentenylJ-4,5,6 ,7-tetrahydro-6 ,7-dimethyl-l- indanone (62).113 Diisobutylaluminum hydride reduction under the aforemen tioned conditions of lactone 50 (910 mg, 2.50 mmol) pro vided lactol 57, which was OH immediately dissolved in methylene chloride (25 mL). Ethanedithiol (0.60 mL, 7.2 mmol) was added, and the solution was cooled to -78°C. Titanium tetrachloride (0.07 mL, 0.63 mmol) was introduced and the mixture was permitted to warm gradu ally to room temperature. The solution was recooled to -78°C and quenched with water (0.5 mL). Upon warming to room temperature, the mixture was washed with water and 10% potassium hydroxide solution, dried (MgSO^), and concentrated in vacuo. The two products were separated chromatographically (silica gel; 5% ethyl acetate in petroleum ether to elute first component and 25% ethyl acetate in petroleum ether for the second compound). The less polar component proved to be 61 (240 mg, 22%) and the more polar, 62 (430 mg, 47%), a 200 colorless oil: IR (CDCI3 ) cm” 1 3360, 3080, 3050, 2970, 1675, 1620, 1470, 1450, 1440, 1410, 1390, 1380, 1345, 1315, 1265, 1240, 1215, 1165, 1130, 1100, 1050, 1030, 990, 980, 910, 850, 730; 1H NMR (300 MHz, CDCl3 ) « 5 *87 (dd, J = 17.4 and 10.9Hz, 1H), 5.06 (d, J - 10.0Hz, 1H), 5.02 (d, a = 17.2Hz, 1H), 4.65 (s, 1H), 4.20 (br s, 1H), 3.73 (br d, Z = 9.8Hz, 1H), 3.02 (br s, 4H), 2.49-2.27 (series of m, 7H), 1.70-1.45 (series of m, 4H), 1.19 (s, 3H), 1.11 (s, 3H), 0.95 (d, J = 6.4Hz, 3H); mass spectrum calcd (M+-H 2 0) 348.1581, obsd 348.1540. (ctB, 6 7'£)- a - I (2S)-2-m-Dithian-2-yl-2-methyl-3- butenylJ-4',5',6',7'-tetrahydro-6',7 dimethylspi roII,3-dithiolane-2,1• ■-indan]-7'-methanol (63).113 Lactone 50 (310 mg, 0.850 mmol) was reduced to lactol 57 with diisobutylaluminum hydride under the conditions OH previously reported. The crude lactol was immediately 201 dissolved in methylene chloride (10 mL), and 1,3- propanedithiol (0.20 mL, 2.0 mmol) was added. The solution was cooled to -78°C, and titanium tetra chloride (0.04 mL, 0.36 mmol) was introduced. The reaction mixture was stirred overnight gradually warming to room temperature, recooled to -78°C, and quenched with water (0.5 mL). Upon warming to room temperature, the mixture was extracted with water and 10% potassium hydroxide solution, dried (MgSO^) and freed of solvent. Dithiane alcohol 63 (190 mg, 49%) was isolated as a colorless solid following chroma tographic purification (silica gel; 5% ethyl acetate in petroleum ether): IR (CDCl^) cm * 3400, 3050, 2930, 2840, 1630, 1450, 1425, 1380, 1265, 1060, 995, 890, 730, 700; 1H NMR (300 MHz, CDCI3 ) 6 5.88 (dd, J. = 17.4 and 10.8Hz, 1H), 5.14 (dd, 1 = 10.8 and 1.1Hz, 1H), 5.04 (dd, il = 17.4 and 1.1Hz, 1H), 4.39 (t, j I = 5.0Hz, 1H), 4.13 (s, 1H), 3.69 (br s, 1H), 3.34-3.18 (m, 4H), 2.85-2.82 (m, 4H), 2.41-1.75 (series of m, 13H), 1.31 (s, 3H), 1.22 (s, 3H), 1.07 (d, 1 = 6.9Hz, 3H). 202 (6'Br 7'£)-71-[ (2£,4B)-5-112-(Benzylthio)ethyl! thio! = tetrahydro-4-methyl-4-viny 1-2-fury 1 1-4',5',6',7'- tetrahydr o-6', 7' -dimethylspi.ro [1,3-di thiolane-2,11 - indan! (64).313 A solution of freshly prepared triflic anhydride (0.76 mL, S— S^SCHgPh 4.54 mmol) and 2,4,6-collidine (0.60 mL, 4.54 mmol) in methylene chloride (3 mL) was cooled to -60°C, and benzyl alcohol (0.47 mL, 4.54 mmol) was added dropwise. After 30 min at -60°C, collidine (0.15 mL, 1.13 mmol) and alcohol 61 (490 mg, 1.11 mmol) dissolved in a minimum of methylene chloride were added dropwise. The mixture was allowed to reach -35°C at which time TLC analysis showed the reaction to be complete. After cooling to -78°C, methanol (1.0 mL) was used to quench the reaction. Upon reaching room temperature, the mixture was extracted with water, dilute hydrochloric acid, saturated sodium bicarbonate solution, and water. The organic phase was dried (MgS04) and freed of solvent. Purification by column chromatography (silica gel; 2.5% ethyl acetate in petroleum ether) supplied 64 (480 mg, 81%) as an epimeric mixture: IR (neat) cm ^ 3090, 3060, 3040, 2970, 2920, 1640, 1605, 1495, 1455, 1430, 1360, 1260, 1205, 1095, 1070, 1030, 995, 910, 815, 730, 695; 2H NMR (200 MHz, CDC13) 6 7.5-7.2 (m, 5H), 6 .1-5. 8 (m, 1H), 5.2-5.0 (m, 2H), 4.91 and 4.83 (two s, 1H total), 4.5-4.3 (m, 1H), 3.75 (s, 2H), 3.4-3.1 (m, 4H), 2.9-2 . 6 (m, 4H), 2.5-1.4 (m, 11H), 1.53 and 1.52 (two s , 3H total), 1.22 and 1.16 (two s, 3H total), 1.09 and 1.04 (two d, jI = 5.3Hz, 3H total). ( 6 7 '£) -7' - [ (1JB, 3fi) -3- (1,3-Dithiolan-2-yl) -1-methoxy- 3-methyl-4-pentenyll-4',5',6 ',7'-tetrahydro-6 '7 dimethylspiroll,3-dithiolane-2,l'-indanl (65).1^3 Alcohol 61 (516 mg, 1.17 mmol) was dissolved in dry tetrahy- drofuran (10 mL), and the solution was cooled to 0°C. A solution of methyllithium (1.08 mL, 1.51 mmol,1.4 U in hexane) was added dropwise, and the mixture was stirred at 0°C for 15 min. The solution was cooled to -78°C, and methyl iodide (0.10 mL, 1.61 mmol) was introduced. 204 After warming to 0°C, the reaction mixture was extrac ted with saturated sodium bicarbonate solution and water, dried (MgSO^), and freed of solvent. The residue was purified by chromatography (silica gel; 5% ethyl acetate in petroleum ether) to give methyl ether 65 (481 mg, 90%) as a colorless oil: IR (CDCl^) cm”* 2980, 2920, 2840, 1640, 1450, 1425, 1370, 1280, 1200, 1100, 1000, 910, 855, 820, 730; *11 NMR (300 MHz, CDC13) 6 5.84 (dd, H = 17.5 and 10.8Hz, 1H), 5.15 (d, J « 10.8Hz, 1H) , 5.09 (d, jI = 17.5Hz, 1H) , 4.58 (s, 1H) , 3.88 (d, J = 8.7Hz, 1H), 3.38-3.25 (m, 4H), 3.24 (s, 3H), 3.11-3.06 (m, with br s at 3.11, 4H), 2.45-1.40 (series of m, 11H), 1.30 (s, 3H), 1.20 (s, 3H), 1.05 (d, J. = 7.0Hz, 3H) ; mass spectrum calcd (M -C8Hi3s2* m/£ 283.1190, obsd 283.1151; m / & (relative intensity) 283 (3), 239 (39), 217 (70), 185 (23), 179 (45), 163 (33), 159 (60), 145 (29), 131 (33), 125 (40), 123 (41), 105 (100), 99 (47), 97 (53), 91 (48), 79 (36), 77 (37), 65 (32), 61 (58), 58 (47), 55 (38), 53 (36). 205 (ajs,Y Er 4Ef 5E) -4,5,6,7-Tetrahydro-Y -methoxy--a» 4,5- 13 3 trimethyl-3-oxo-a-vinyl-4-indanbutyraldehyde (66). Methyl ether 65 (80 mg, 0.175 o mmol) was dissolved in acetone ( 8 mL). Water (0.5 mL) and methyl iodide (2.0 mL) were added, and the mixture was gently refluxed overnight. After removal of the volatile solvents under reduced pressure, the residue was redissolved in methylene chloride, extracted with water, dried (MgSO^), and freed of solvent. Aldehyde 6 6 (19.4 mg, 38%) was obtained following chromatography (silica gel? 5% ethyl acetate in petroleum ether) as a pale yellow oil: IR (neat) cm” 1 2960, 2920, 2850, 2710, 1725, 1690, 1630, 1460, 1415, 1380, 1280, 1170, 1095, 995, 975, 920, 880, 830, 730; 2H NMR (300 MHz, CDCI3 ) <5 9.24 (s, 1H), 5.72 (dd, £ = 17.5 and 10.6Hz, 1H), 5.20 (d, 1 = 10.5Hz, 1H) , 5.12 (d, jI = 17.6Hz, 1H) , 4.02 (dd, 1 = 11.5 and 2.2Hz, 1H), 2.95 (s, 3H), 2.46-1.65 (series of m, 11H), 1.19 (s, 3H), 1.17 (d, 1 = 7.8Hz, 3H), 1.15 (s, 3H). 206 (6 B»7£) -*4,5,6,7-Tetrahydro-7- [ (IB,3jS)-3- 11- hydroxypropylJ-l-methoxy-3-metby1-4-pentenyl1-6,7- dimethyl-l-indanone (67).113 Aldehyde 6 6 (38.6 mg, 0.131 mmol) was dissolved in ether (5 mL) and cooled to -100°C. Freshly prepared and titrated 0CH3 ethylmagnesium bromide in ether (1 . 0 eguiv) was added. The mixture was stirred and allowed to gradually warm to room temperature. The reaction mixture was extracted with water, dried (MgS04), and freed of solvent. Chromatography (silica gel; 17% ethyl acetate in petroleum ether) separated the two diastereomeric alcohols, both colorless oils, in an overall yield of 17.9 mg (47%). The less polar alcohol, 67a, accounted for 10.1 mg and the more polar, 67b, for 7.8 mg. Spectral data for 67a: IR (neat) cm" 1 3480, 2920, 1690, 1630, 1460, 1410, 1380, 1290, 1190, 1085, 970, 900; XH NMR (300 MHz, CDC13 ) 6 5.97 (dd, J = 17.1 and 10.4Hz, 1H), 5.05 (dd, j; = 11.2 and 1.4Hz, 1H), 4.97 (dd, J. = 17.9 and 1.4Hz, 207 1H), 4.20 (d, J = 8.7Hz, 1H), 3.20 (s, 3H), 3.19-3.13 (m, 1H), 2.48-2.35 (series of m, 6 H), 1.67-1.15 (series of m, 8 H), 1.13 (d, J = 6.9Hz, 3H), 1.10 (s, 3H), 1.04 (S, 3H), 0.97 (t, J = 7.2Hz, 3H); 13C NMR (20 MHz, CDC13) ppm 208.83, 173.89, 144.38, 143.99, 112.44, 83.69, 77.37, 59.22, 44.53, 41.85, 40.95, 39.17, 35.08, 29.46, 27.60 (2 carbons), 24.41, 23.90, 18.85, 17.57, 11.70; mass spectrum calcd (M+-C 3 HgO) m/£ 276.2089, obsd 276.2062. (FAB) (on mixture) (M++l) jd/£ 335. Spectral data for 67b: IR (neat) cm” 1 3470, 2960, 1680, 1625, 1450, 1410, 1370, 1260, 1080, 965, 900, 725; 2H NMR (300 MHz, CDCI3 ) 6 5.73 (dd, il = 17.5 and 10.9Hz, 1H), 5.15 (dd, J[ = 10.8 and 1.3Hz, 1H), 5.04 (dd, H = 17.5 and 1.3Hz, 1H), 4.02 (dd, J. = 8.4 and 1.6Hz, 1H), 3.23-3.19 (m, 1H), 3.13 (s, 3H), 2.45-2.32 (series of m, 6 H), 1.84- 1.59 (series of m, 6 H), 1.25-1.15 (m, 2H), 1.11 (d, H = 6 .6 Hz, 3H), 1.10 (s, 3H), 1.05 (s, 3H), 0.97 (t, J « 7.3Hz, 3H); l3C NMR (20 MHz, CDCl-j) ppm 208.70, 173.57, 144.63, 144.19, 113.71, 83.44, 80.24, 59.29, 45.04, 42.36, 40.76, 39.36, 35.21, 29.52, 27.80, 27.67, 24.86, 24.22, 17.64, 17.45, 11.63; mass spectrum (FAB) (on mixture) (M++l) m/£ 335. 208 Alcohols 67 may also be prepared by reduction of keto enone 6 8 . Keto enone 6 8 (830 mg, 2.50 mmol) was dissolved in methanol (10 mL), and sodium borohydride (100 mg, 2.64 mmol) was added slowly with stirring. After completion of the reaction, water (2 mL) was added, and the volatile material was removed under reduced pressure. The residue was partitioned between dilute hydrochloric acid and methylene chloride. The organic phase was dried (MgSO^) and freed of solvent. Chromatographic purification (silica gel; 20% ethyl acetate in petroleum ether) supplied the alcohols 67 (767 mg, 92%) in a 1:1 ratio as colorless oils. The spectral data are given above. 13 The mass spectra and C NMR data were obtained from reduction reactions while IR and NMR data were obtained from both the Grignard and reduction reactions. 209 (6 £, 7£)-4,5,6,7-Tetrahydro-7-[(l£,3S)-l-methoxy-3- methy1-3-propiony1-4-pentenyl]-6 ,7-dimethy1-1-indanone (68).113 A mixture of the diastereo- meric alcohols 67 (300 mg, 0.897 mmol) was dissolved in methylene chloride (10 mL). Celite (1 g) and pyridinium chlorochromate (360 mg, 1.67 mmol) were sequentially added, and the mixture was stirred overnight at room temperature. Ether (80 mL) was added and the mixture was passed through a plug of Florisil. Concentration in vacuo yielded 6 8 (294 mg, 99%) as a colorless oils *H NMR (300 MHz, CDCI3 ) 6 5.87 (dd, jI = 17.7 and 10.4Hz, 1H), 5.13 (dd, Z = 10.3 and 0.7Hz, 1H), 5.12 (dd, 1 = 16.7 and 0.9Hz, 1H), 3.96 (dd, J = 11.1 and 2.0Hz, 1H), 2.93 (s, 3H), 2.52 (q, Z = 7.3Hz, 2H), 2.46-2.30 (m, 4H), 1.84-1.64 (m, 7H), 1.22 (s, 3H), 1.15 (d, Z = 7.2Hz, 3H), 1.14 (s, 3H), 0.96 (t, J = 7.3Hz, 3H). 210 (-)-Mutilone (6 ). 113 From Pleuromutilone: Pleuromutilone (2.00 g, 5.32 mmol) dissolved in 5% potas sium hydroxide in methanol (100 mL) was heated to the reflux temperature for 2 h. After cooling, the reaction mixture was poured into water and extracted with methylene chloride. The organic phases were washed with saturated sodium bicarbonate solution, dried (MgSO^), and freed of solvent. The residue was purified chromatographically (silica gel; 2 0 % ethyl acetate in petroleum ether) to give mutilone (6 ) (1.48 g, 87%) as a white crystalline solid: mp 158-159°C. From Tiamulone: Tiamulone (53) (7.41 g, 15.1 mmol) was dissolved in 5% potassium hydroxide solution in methanol (160 mL) and heated at the reflux temperature for 2 h. TLC analysis (5% methanol in methylene chloride) indicated that hydrolysis was complete at that time. The reaction mixture was cooled, poured into water (300 mL) and extracted with methylene chloride (4x100 mL). The organic phases were combined# dried (MgSO^), and concentrated Jjq vacuo. The crude material was purified chromatographically (silica gel; 2 0 % ethyl acetate in petroleum ether) yielding mutilone(6)(4.09 g, 85%) as a white solid: mp 157-158°C (litlla mp 154-175°C); U ] D 2 2 -51.3° (£ 4.065, CHCI3 ) (litlla I“1D 2 0 -53° (£ 1.044, CHCl^)); IR (CDCI3 ) cm- 1 3650, 3550, 2985, 2940, 2880, 1740, 1700, 1635, 1455, 1418, 1330, 1290, 1200, 1150, 1110, 1030, 1000, 965; *H NMR (300 MHz, CDCI3 ) 5 6.09 (dd, il = 17.8 and 10.7Hz, 1H), 5.32 (d, J - 17.8H, 1H), 5.32 (d, jI *= 11.2Hz, 1H), 4.70 (d, il * 7.7Hz, 1H) , 3.26 (g, Z = 6.7Hz, 1H), 2.32-2.12 (series of m, 3H), 1.90 (dd, J = 15.4 and 7.8Hz, 1H), 1.74-1.39 (series of m, 9H), 1.37 (s, 3H), 1.11 (s, 3H), 1.06 (d, Z * 6 .6 Hz, 3H), 1.01 (d, Z «= 6 .6 Hz, 3H); 13C NMR (20 MHz, CDC13) ppm 217.13, 215.72, 140.28, 117.16, 67.65, 59.92, 54.49, 45.42, 44.91, 44.33, 42.54, 37.43, 34.69, 30.09, 27.21, 25.30, 24.78, 18.27, 13.48; mass spectrum ions at m / & (relative intensity) 249(3), 193(7), 177(20), 165(17), 164(19), 163(100), 149(5), 137(4), 121(10), 107(6), 93(5), 79(6), 55(7). Anal. Calcd for ^20^30^3° 75.43; H, 9.50. Found: C, 75.38; H, 9.56. 212 The mutilone prepared from tiamulone was identical to that obtained from mutilone by comparison of NMR, 13 C NMR, 1R, and mass spectra. <-)-(3 aS,4 £, 6 fi,8 E, 9 B, 9aB,10B)-6-Ethylhexahydro-8- methoxy-4,6,9,10-tetramethyl-3a,9-propano-3afl- cyclopentacyclooctene-l,5(4fl,6fl)-dione (71). Mutilone (6 ) (6.96 g, 21.9 mmol) and 2 ,6 -di-i-butyl- pyridine (9.5 ml, 44 mmol) were dissolved in methylene och3 chloride (150 mL). Subse quently, methyl trifluoro- methanesulfonate (4.8 mL, 44 mmol) was added, and the mixture was stirred at room temperature. After 48 h, TLC analysis (10% ethyl acetate in petroleum ether) showed no residual starting material. Water (10 mL) was added, and the mixture was stirred for an additional hour. The mixture was extracted with water and saturated sodium bicarbonate solution. The organic phase was dried (MgSO^) and freed of solvent. The residue was purified by column chromatography (silica gel; 5% ethyl acetate in petroleum ether) furnishing white crystals of 71 (7.39 g, 100%): mp 150.5-151.0°C; UJ D 2 5 -38.5° (jc 2.60, CHC13 ); IR (CDCI3 ) cm " 1 2990, 2940, 1740, 1700, 1465, 1418, 1380, 1330, 1290, 1150, 1095, 1060, 995, 960; NMR (300 MHz, CDC13) 6 6.34 (dd, J = 17.6 and 10.8Hz, 1H), 5.31 (d, 1 = 10.8Hz, 1H), 5.12 (d, il = 17.6 Hz, 1H), 3.91 (d, J = 8.5Hz, 1H), 3.29 (s, 3H), 3.18 (q, Z = 6 .6 Hz, 1H), 2.27-2.12 (m, 3H), 1.88 (dd, jl = 15.7 and 8.5Hz, 1H), 1.65-1.38 (series of m, 8 H), 1.36 (s, 3H), 1.19 (s, 3H), 1.05 (d, jl = 6 .6 Hz, 3H), 0.92 (d, J[ = 6.3Hz, 3H) ; l3C NMR (20 MHz,.CDCl3) ppm 216.88, 214.90, 140.48, 117.93, 77.05, 59.35, 55.20, 53.60, 45.43, 45.17, 43.06, 41.91, 37.82, 34.69, 30.22, 27.60, 24.73 (2 carbons), 17.64, 14.44, 13.49; mass spectrum calcd (M+) m/£ 322.2351, obsd 332.2354. Anal. Calcd for C21H32®3: C, 75.86; H, 9.70. Found: C, 75.75; H, 9.74. 214 Spectral data for <3a£,4£,6£,8£,9JB,10£)-2,3,6,7,8,9- Hexahydro-l#8-dimethoxy-4,6,9,10-tetramethyl-6-vinyl- 3a,9-propano-3afi-cyclopentacycloocten-5(4£)-one (70): IR (Nujol) cm” 1 2950, 1690, 1640, 1580, 1460, 1410, 1375 1340, 1260, 1160, 1090, 1030 910, 880, 810; 1H NMR (300 MHz, CDC13 ) 6 6.24 (dd, J = 17.6 and 11.0Hz, 1H), 5.22 (d i = 11.0 Hz, 1H), 5.02 (dd, jl ■ 17.6 and 0.7Hz, 1H), 3.72 (d, jl = 8 .8 Hz, 1H) , 3.45 (s, 3H) , 3.26 (s, 3H) , 3.03 (q, jl = 6 .6 Hz, 1H), 2.45-2.12 (series of m, 6 H), 2.06 (dd, i = 14.8 and 8.9Hz, 1H), 1.89 (dt, J ■ 13.2 and 3.1Hz, 2H), 1.71-1.42 (series of in, 2H) , 1.29 (s, 3H) , 1.10 (s, 3H) , 0.97 (d, £ = 6 .8 Hz, 3H) , 0.96 (d, jl = 6 .6 Hz, 3H). 215 (3&Sr4Sf 8 £r 9£i 93SrlO£) —Hexabydro—4 , 6 / 9 , 1 0 — tetramethyl-8 - [(methylthio)methoxy1-6-vinyl-3a,9- propano-3a£-cyclopentacyclooctene-1,5(4fl, 6 fl)-dione Mutilone (6 ) (1.04 g, 3.26 mmol) was dissolved in a mix ture of dimethyl sulfoxide (50 mL) and acetic acid (10 mL). SCH 0 —/ Acetic anhydride (25 mL) was added, and the mixture was stirred at room temperature overnight. The mixture was extracted with water, and excess dimethyl sulfoxide was removed by distillation under reduced pressure. The residue was purified chromatographically (silica gel; 5% ethyl acetate in petroleum ether) to supply 69 (1.23 g, 99%) as colorless crystals: mp 163-163.5°C; IR (Nujol) cm" 1 2930, 2860, 1730, 1700, 1460, 1415, 1380, 1340, 1310, 1285, 1230, 1150, 1110, 1035; 2H NMR (300 MHz, CDC13 ) 6 6.34 (dd, jl = 17.4 and 10.8Hz, 1H), 5.32 (d, J = 10.8Hz, 1H), 5.15 (d, J - 17.5Hz, 1H), 4.62 (ABq, = 11.2Hz, Au ■ 9.8Hz, 2H), 4.48 (d, J = 8.2Hz, 1H), 3.25 (q, J » 6 .6 Hz, 1H), 2.26-2.11 (m, 3H), 2.23 (s, 3H), 1.89 (dd, J = 16.0 and 8.3Hz, 1H), 1.73- 216 1.40 (series of m, 7H), 1.39 (s, 3H), 1.24-1.12 (m, 1H) , 1.18 (s, 3H) , 1.05 (d, J = 6 .6 Hz, 3H) , 0.98 (d, J[ = 6 .8 Hz, 3H) ; l3C NMR (20 MHz, CDC13) ppm 217.14, 214.64, 139.78, 118.19, 72.96, 72.19, 59.74, 53.73, 45.55, 45.30, 43.13, 41.27, 37.95, 34.82, 30.35, 27.22, 24.86, 24.09, 17.96, 15.79, 14.89, 13.49; mass spectrum calcd (M+-CH 3 SCH2) m/£ 317.2116, obsd 317.2191. Anal. Calcd for C22H34°3S: C ' *>£>.80; H, 9.05. Found: C, 69.65; H, 9.02. (-)-(6 B»7fi)-4,5,6,7-Tetrahydro-7-1 (IB,3 £ ) -l-methoxy-3- methyl-3-propionyl-4-pentenylJ-6,7-dimethyl-l-indanone (68). Mutilone methyl ether 71 (3.70 g, 1 1 . 1 mmol) was combined with a solution of 1 0 % potas sium hydroxide in ethanol (330 wwn3 mL), and the mixture was re fluxed overnight, poured into water (500 mL), and extracted with methylene chloride (4x100 mL). After being dried (MgSO^) and freed of solvent, the residue was purified by HPLC (Waters Prep 500; silica gel; 20% ethyl acetate in petroleum ether) to supply 6 8 (3.13 g, 85%) as a light yellow oil which slowly solidified on standing: mp 50-53 C; [alD -65.5° (£ 2.89, CHCI3 ) ; IR (CDCI3 ) cm- 1 3080, 2980, 2940, 2885, 2840, 1710, 1695, 1640, 1465, 1440, 1425, 1412, 1380, 1350, 1290, 1270, 1240, 1105, 1045, 995, 980, 930, 835, 800; 2H NMR (300 MHz, CDCI3 ) & 5.88 (dd, jl = 17.5 and 10.6Hz, 1H) , 5.13 (d, £ = 11.3Hz, 1H), 5.12 (d, jl = 16.8Hz, 1H), 3.97 (d, £ = 10.9Hz, 1H), 2.94 (s, 3H), 2.53 (q, £ = 7.3Hz, 2H), 2.43-2.21 (m, 4H), 1.92-1.62 (m, 7H), 1.22 (s, 3H), 1.16 (d, £ = 7.2Hz, 3H), 0.97 (t, £ = 7.2Hz, 3H); l3C NMR (20 MHz, CDCI3 ) ppm 211.77, 208.51, 173.63, 143.82, 143.42, 114.35, 82.60, 59.93, 53.22, 41.78, 41.08, 39.61, 35.27, 30.61, 29.52, 27.92, 27.73, 23.96, 19.55, 17.57, 8.50; mass spectrum calcd (M++l) 333.2471, obsd 333.2449. . 218 (6 B, 7B) -7- [ (2B»4fi)-5- (Bromomethy 1) tetrahydro-4-methyl- 4-propionyl-2-furyl)-l-indanone (72) and/or (6 B# 7B)-7-1(2fir 4B)-5-Bromotetrahydro-4-methy1-4- 113 propionyl-2fl-pyran-2-yll-l-indanone (73). B r Keto enone 6 8 (195 mg, 0.588 mmol) was dissolved in methy 72 lene chloride (2 mL) and ethanol (2 mL). Pyridinium hydrobromide perbromide ( 2 0 0 mg, 0.625 mmol) was added, and the mixture was stirred at room temperature and monitored by TLC analysis (20% ethyl acetate in petro leum ether). After 48 h, the mixture was diluted with methylene chloride (20 mL) and extracted sequentially with water, 10% hydrochloric acid (20 mL), and satur ated sodium bicarbonate solution (20 mL) before being dried (MgSO^) and freed of solvent. The residue was purified by chromatography (silica gel; 2 0 % ethyl acetate in petroleum ether) to furnish 72/73 (164 mg, 70%) as a colorless oil: IR (neat) cm * 2970, 2920, 2870, 1700, 1690, 1620, 1455, 1375, 1280, 1090, 1010, 965, 900, 720; 2H NMR (300 MHz, CDCI3 ) 6 4.27 (dd, J = 7.0 and 6.3Hz, 1H), 3.81 (dd, 3. = 10.2 and 6 .6 Hz, 1H), 219 3.35-3.32 (m, 2H), 2.53-1.64 (series of m, 13H), 1.31 (s, 3H), 1.21 (t, JL = 7.2Hz, 3H); mass spectrum (FAB) (M++l) m / & 397. (6£, 7fi)-4,5,6,7-Tetrahydr o-7-1 (1E, 3S) -l-methoxy-3- methyl-3-propionyl-4-pentenyl)-6 ,7-dimethyl-2- 1 1 3 (phenylselenyl)-l-indanone (74). A solution of lithium diiso- propylamide, prepared from diisopropylamine (0.43 mL, 3.1 PhSe mmol) and n-butyllithium (1.9 OCH mL, 3.0 mmol, 1.6 H in hex ane), in tetrahydrofuran was cooled to -78°C. Keto enone 6 8 (508 mg, 1.53 mmol) dissolved in tetrahydrofuran was added and stirring was maintained at -78°C for 1 h. A tetrahydrofuran solu tion containing phenylselenenyl chloride (293 mg, 1.53 mmol) was added and the mixture was allowed to warm to room temperature before being extracted sequentially with water, 1 0 % hydrochloric acid, and saturated sodium bicarbonate solution before being dried (HgSO^) and freed of solvent. Chromatographic purification of the residue (silica gel; 1 0 % ethyl acetate in petroleum ether) gave a mixture of diastereomers 74 (379 mg, 51%) as a pale yellow oil: IR (neat) cm-* 3045, 2960, 2920, 1700, 1690, 1630, 1575, 1470, 1460, 1435, 1370, 1260, 1150, 1085, 1015, 970, 910, 730, 680; *H NMR (300 MHz, CDCI 3 ) 6 7.61-7.55 (m, 2H), 7.30-7.21 (m, 3H), 5.89- 5.78 (m, 1H), 5.17-5.06 (m, 2H), 4.10 (dd, H - 10.3 and 7.0Hz, 1H), 3.78 and 3.72 (two d, J = 10.5 and 9.6 respectively, 1H total), 2.90 and 2.85 (two s, 3H total), 2.62-1.31 (series of m, 11H), 1.27 and 1.24 (two s, 3H total), 1.16 and 1.14 (two d, J = 6.9Hz, 3H total), 1.16 and 1.12 (two s, 3H total), 0.95 and 0.94 (two t, 3. = 7.2Hz, 3H total); mass spectrum (FAB) (M+-l) m/£ 487. Dichlorobis(benzonitrile)palladium(II) . * 1 4 Purified benzonitrile (50 mL) was deoxygenated with a stream of nitrogen for 15 min. CLPdCNCCLH.) 2 6 5 Palladium(II) chloride (500 mg, 2.82 mmol) was added, and the heterogeneous mixture was 221 vigorously stirred while being heated to 100°C. After 2.5 h, the mixture was homogeneous and was cooled to -10°C in a freezer after first adding petroleum ether (50 ml). The precipitate was collected by suction filtration and rinsed with petroleum ether. A second crop was collected from the filtrate after the addition of more petroleum ether and cooling. The orange-brown crystals were dried (20°C/0.3 mm) to afford the desired product (775 mg, 72%): mp 129-130°C (lit11^ mp 131°C). Bis(tri-fl-tolylphosphine)palladium(II) Chloride.115 Purified benzene (62 mL) was deoxygenated by bubbling nitrogen through for 15 min. c|2Pd[p-^Q)3]2 . . t l jr* A positive nitrogen pressure CH3 was maintained as bis(benzo- nitrile)palladium(II) chloride (387 mg, 1.01 mmol) was added. To this brown solution was added tri-.Q-tolylphosphine (615 mg, 2.02 mmol). A yellow precipitate began to appear after 15 min and increased in quantity with continued stirring at room temperature for 2.5 h. The precipitate was 222 collected by vacuum filtration and dried under vacuum at room temperature. The product was recrystallized from methylene chloride with the gradual addition of petroleum ether. The yellow crystalline product was collected by suction filtration/ rinsed with petroleum ether, and dried under high vacuum to furnish 679 mg (8 6 %) of bis(tri-£-tolylphosphine)palladium(II) chloride. (-)-(6B,7fi)-4/5,6/7-Tetrahydro-7-I(IB,3S,4Z)-1-methoxy- 3-methyl-4-(trimethylsiloxy)-3-vinyl-4-hexenyl1-6,7- dimethyl-l-indanone (76). A solution of lithium diiso- ^ propylamide was prepared in tetrahydrofuran (70 mL) from H-butyllithium (3.04 mL/ 4.7 och3 mmol/ 1.55 B in hexane) and diisopropylamine (0.66 mL, 4.7 mmol). This solution was cooled to -78°C and 6 8 (712 mg, 2.14 mmol) dissolved in tetrahydrofuran (10 mL) was added. The bis-enolate was permitted to form over 45 min imparting an orange color to the reaction mixture. 2^.3 Subsequently, HMPA (3.28 mL, 18.8 mmol) was introduced followed 0.5 h later by trimethylsilyl chloride (0.60 mL, 4.71 mmol). The reaction mixture was gradually warmed from -78°C to 20°C over 5 h. The tetrahydro- furan was removed under reduced pressure, and the residue was taken up in pentane (60 mL), and extracted with saturated sodium bicarbonate solution (3x40 mL). The combined aqueous washes were back extracted with pentane (2x20 mL). The pooled organic phases were dried (MgSO^) and freed of solvent yielding a mixture of the mono-and bis-silyl enol ethers. The mixture of silyl enol ethers was dissolved in purified benzene (110 mL) and deoxygenated with a stream of nitrogen for 30 min. The reaction mixture was put under a positive nitrogen pressure, and bis(tri-.e- tolylphosphine)palladium(II) chloride (134 mg, 0.170 mmol) and tributyltin fluoride (1.653 g, 5.35 mmol) were added. The reaction mixture was brought to the reflux temperature and soon turned black. After 5 h the mixture was cooled, diluted with ether (200 mL), and extracted with 1 1} sodium hydroxide solution (125 mL). The aqueous phase was back extracted with ether (2x100 mL). The organic phases were combined, dried (K2 C03), and freed of solvent. The crude material was 224 purified by chromatography (silica gel; 5% ethyl acetate in petroleum ether) to afford (-)-76 (726 mg, 84%); ( “ ) D 2 5 -34.9° (£ 2.42, CC14) : IR (neat) cm" 1 3090, 2975, 2930, 1695, 1663, 1635, 1455, 1412, 1380, 1320, 1250, 1200, 1100, 1005, 978, 905, 840, 788, 760; 1 H NMR (300 MHz, CgDg) <5 6.14 (dd, 1 = 17.8 and 10.6Hz, 1H), 5.17 (d, J[ = 17.4Hz, 1H), 5.16 (d, H = 10.9Hz, 1H), 4.79 (q, J «* 6.7Hz, 1H), 4.32 (dd, J = 7.2 and 2.2Hz, 1H), 3.22 (s, 3H), 2.14-1.75 (series of m, 9H), 1.56 (d, J - 6.7Hz, 3H), 1.56-1.46 (m, 2H), 1.37 (s, 3H), 1.27 (s, 3H), 1.23 (d, J ■ 6.9Hz, 3H), 0.26 (s, 9H); 13C NMR (75 MHz, CgDg) ppm 206.80, 171.04, 157.08, 145.99, 144.26, 112.79, 100.27, 83.58, 59.29, 46.35, 42.99, 42.10, 39.33, 35.17, 29.27, 28.03, 27.31, 24.40, 22.81, 18.04, 12.00, 1.40; mass spectrum m/£ (relative intensity) 293(13), 241(53), 236(16), 209(32), 184(62), 183(50), 169(33), 163(28), 151(25), 73(100); UV (EtOH) X max (e) 239(9750). 225 Spectral data for 11 (1£,25)-1-Ethylidene-2-[(2£)-2- methoxy-2-[(4fi,5fi)-4,5,6,7-tetrahydro-4,5-dimethy1-3- (trimethylsiloxy)inden-4-y1]ethyl]-2-methy1-3- butenylloxy]trimethylsilane (75): IR (CDCI3 ) cm- 1 3100, 2975, iSi(CH,). 1682, 1639, 1460, 1381, 1321, 1255, 1110, 1075, 985, 850; 1H NMR (300 MHz, CgDg) 6 6.12 OCH3 (dd, J = 17.4 and 11.0Hz, 1H) 5.18 (d, jl = 17.3Hz, 1H), 5.17 (d, ,2 = 11.0Hz, 1H), 4.80 (q, J = 6.7Hz, 1H), 4.36 (dd, j; = 7.4 and 1.9hz, 1H), 3.51 (t, J. = 4.5Hz, 1H), 3.24 (s, 3H), 2.21-1.82 (m, 8 H) , 1.55 (d, jl = 6.7Hz, 3H) , 1.53-1.43 (m, 1H), 1.39 (s, 3H) , 1.26 (d, J[ * 5.4Hz, 3H), 1.25 (s, 3H), 0.25 (s, 9H), 0.04 (s, 9H); 13C NMR (75 MHz, CgDg) ppm 184.32, 169.20, 157.14, 146.00, 144.96, 112.86, 100.14, 83.67, 59.19, 46.40, 42.92, 42.32, 40.36, 39.86, 32.57, 28.38, 27.73, 23.93, 22.66, 18.24, 12.00, 1.39, -2.92; mass spectrum calcd (M+ ) m/£ 476.3142, obsd 476.3126. (-) - (6E,7JB) -4,5,6,7-Tetrahydro-7- [ (l£, 3£, 42) -1-methoxy- 3-methy 1-4- (.terl-butyldimethysiloxy) -3-vinyl-4- hexenyll-6,7-dimethyl-l-indanone (77). Lithium diisopropylamide was prepared from diisopropylamine (0.18 mL, 1.32 mmol), and n- butyllithium (0.85 mL, 1.32 mmol, 1.55 E in hexane) in tetrahydrofuran (20 mL). This solution was cooled to -78°C, and the keto enone 6 8 ( 2 0 0 mg, 0.601 mmol) dissolved in tetrahydrofuran ( 2 mL) was added. The yellow enolate was permitted to form over 1 h at which time HMPA (0.92 mL, 5.3 mmol) was added. After 30 min, JL-butyldimethylsilyl trifluoromethanesulfonate (0.30 mL, 1.32 mmol) was introduced. The reaction mixture was permitted to gradually warm to room temperature over several hours. The tetrahydrofuran was removed under reduced pressure, and the residue was dissolved in pentane (20 mL) and washed with saturated sodium bicarbonate solution (3x15 mL). The combined aqueous washes were back-extracted with pentane (3x10 mL). The combined organic phases were dried (MgSO^) and freed of solvent. The crude 227 material was purified by MPLC (silica gel? 10% ethyl acetate in petroleum ether) to furnish the bis-silyl enol ether 78 (111 mg, 33%), I“ ) D 2 5 -30.5° (£ 3.03, CC14), as a colorless oil and the mono-silyl enol ether 77 (134 mg, 50%), I« ] D 2 5 -21.5° (£ 1.62, CC14 ), also as a colorless oil: IR (CC14) cm- 1 3085, 2960, 2935, 2860, 1695, 1662, 1634, 1463, 1381, 1322, 1252, 1103, 1072, 1002, 911, 834; 2H NMR (300 MHz, CDC13) <5 5.93 (dd, a = 17.5 and 10.8 Hz, 1H), 5.08 (dd, a » 10.8 and 1.4Hz, 1H), 5.04 (dd, a = 17.5 and 1.4Hz, 1H), 4.63 (g, a = 6 .8 Hz, 1H), 3.93 (dd, a = 7.5 and 1.7Hz, 1H), 3.06 (s, 3H), 2.43-2.25 (series of m, 6 H), 1.96-1.61 (series of m, 5H), 1.52 (d, a = 6 .8 Hz, 3H), 1.18 (s, 3H), 1.09 (d, 6.9Hz, 3H), 1.08 (s, 3H), 0.97 (s, 9H), 0.16 (s, 3H), 0.15 (s, 3H); 13C NMR (75 MHz, CgDg) ppm 207.03, 171.27, 156.46, 146.82, 144.25, 112.65, 100.12, 83.43, 59.91, 46.36, 42.76, 41.66, 39.13, 35.00, 29.10, 27.85, 27.15, 26.60, 24.24, 23.29, 19.28, 17.82, 11,93, -2.47, -2.60; mass spectrum calcd (M+) m/£ 446.3216, obsd 446.3253. Spectral data for I [l£,2.g)-l-Ethylidene-2-l (2£)-2- methory-2-[(4£,5fi)-4,5,6,7-tetrahydro-4,5-dimethyl-3- (.tfiJLfc-butyldimethysiloxy)inden-4-yl1ethyl1-2-methy1-3- butenyl]oxyJt£x±-butyldimethylsilane (78): IR (CC14 ) cm- 1 3090, 2960, OS) 2930, 2890, 2850, 1675, 1660, 1645, 1470, 1460, 1410, 1380, 1360, 1320, 1250, 1105, 1070, OCH 1000, 975, 910, 830, 675; 1H NMR (300 MHz, CgDg) « 6.13- 6.07 (in, 1H), 5.20-5.08 (m, 2H), 4.78 (q, Z = 6.7Hz, 1H), 4.32 (d, jl = 6.9Hz, 1H) , 4.26 (s, 1H) , 3.21 (s, 3H), 2.27-0.81 (series of m, 12H), 1.59 (d, Z - 6 .8 Hz, 3H), 1.39 (s, 3H), 1.25 (s, 3H), 1.10 (s, 9H), 0.92 (s 9H), 0.22 (s, 3H), 0.14 (s, 3H), 0.05 (s, 3H), -0.06 (s, 3H); mass spectrum calcd (M+ ) m / & 560.4081, obsd 560.4098. 229 (+)-(3£, 4£,4a£, 7a£, 10£)-Tetrahydro-4,10-dimethy1-3- [ (2S)-2-methyl-2-propionyl-3-butenyl]-lB“4,7a-- propanocyclopenta[£]pyran-5(3fl)-one (82). A quartz tube was charged with silyl enol ether 76 (30 mg, 0.0741 mmol) dissolved in methylene chloride (30 mL). The solution was deoxygenated with a stream of nitrogen for 2 0 min, and then the tube was sealed with a stopper. The reaction mixture was irradiated with a 450 W medium pressure Hanovia mercury arc lamp. The progress of reaction was monitored by TLC analysis (10% ethyl acetate in petroleum ether), and after 4 h, little starting material remained. The two observed products were separated by MPLC (silica gel; 10% ethyl acetate in petroleum ether: the slower eluting was hydrolyzed starting material; the faster, a colorless oil, proved to be 82 (13 mg, 53%), IaJD 2 5 +67.0° (£ 1.33, CHCI3 ); IR (CHC13 ) cm - 1 2970, 2885, 1731, 1710, 1460, 1420, 1385, 1265, 1077; 1H NMR (300 MHz, CDCI3 ) 5 5.85 (dd, J = 17.4 and 10.8Hz, 1H), 5.20 (d, J - 10.4Hz, 1H), 5.17 (d, J = 17.5Hz, 1H), 3.82 (dd, J[ = 9.0 and 1.6Hz, 1H), 230 3.66 (ABq, Au = 59Hz, = 8 .6 Hz, 2H) , 2.49 (qd, J = 7.2 and 2.5Hz, 2H), 2.32-1.68 (series of m, 10H), 1.40- 1.30 (mf 2H), 1.31 (s, 3H), 1.15 (d, J[ = 7.1Hz, 3H), ' 0.98 (t, jl = 7.2Hz, 3H), 0.88 (s, 3H) ; 13C NMR (75 MHz, CDCI3 ) ppm 223.09 (s), 213.55(s), 141.74(d), 115.34(t), 83.47(d), 72.65(t), 59.40(s), 53.75(b), 45.97(b), 37.47 (d), 37.39(t), 35.50(t), 34.00(d), 31.40(t), 24.91(t), 24.48(t), 20.74(q), 19.94(t), 19.77(g), 17.02(g), 8.20(g); mass spectrum calcd (M+) m / & 332.2351, obsd 332.2336. 57 ±-Butyldimethylsilyl Trifluoromethanesulfonate. In a 25 mL flask topped with an S-tube and drying tube chj 1-Bu-SiOSOXF, (CaS04) were combined t-butyl- I 2 w dimethylsilyl chloride (3.77 ch 3 g, 25 mmol) and trifluoro- methanesulfonic acid (2.21 mL, 25 mmol). This mixture was heated to 60°C with an oil bath overnight. The S-tube and drying tube were replaced with a distillation head, and the oil bath temperature was increased to 85°C. The product 231 distilled as a colorless liquid (5.64 g, 85%): bp 60- 65°C/5.5 mm (lit5 7 bp 65-67°C/12 mm). (3a£r 4£, 6 S, 8 £, 9JB, 9aE, 10JJ) - 8 - (Jteri-Buty ldiraethylsiloxy) = hexahydro-4,6 ,9,10-tetramethyl-6-vinyl-3a-9-propano- 3aH-cyclopentacyclooctene-l*5(4fl,6fi)-dione (84). Multilone (6 ) (2.00 g, 6.28 mmol), and 2,6-lutidine (3.28 mL, 28.2 mmol) dissolved in tetrahydrofuran (100 mL) were cooled to 0°C with an ice bath. Subsequently, ±-butyldimethylsilyl trifluoro- methanesulfonate (5.20 mL, 22.6 mmol) was introduced and stirring was continued for 5 h. TLC analysis (10% ethyl acetate in petroleum ether) showed the reaction to be complete at that time. The tetrahydrofuran was removed under reduced pressure, and the residue was partitioned between saturated sodium bicarbonate solution (200 mL) and methylene chloride (100 mL). The aqueous phase was further extracted with methylene chloride (2x100 mL). The combined organic phases were dried (MgSO^) and concentrated in vacuo. Silyl ether 232 84, silyl enol ether 83 and lutidine were separated by MPLC (silica gel; 7% ethyl acetate in petroleum ether). The crude silyl enol ether 83 (3.43 g), mp 145- 148°C, was hydrolyzed over a 2.5 h period at room temperature in a mixture of tetrahydrofuran/water/ perchloric acid (90:9:1, 200 mL). The tetrahydrofuran was again removed under reduced pressure. The residue was diluted with saturated sodium bicarbonate solution (175 mL) and extracted with methylene chloride (4x100 mL). The organic phases were combined, dried (HgSO^), and freed of solvent. Purification was achieved using a Waters Prep 500 HPLC (silica gel; 10% ethyl acetate in petroleum ether) to afford silyl ether 84 (2.59 g, 95%) as a white solid: mp 138.0-138.5°C; IR (CCl^) cm" 1 2960, 2865, 1745, 1709, 1635, 1465, 1379, 1330, 1287, 1258, 1052, 1008, 929, 835; 1H NMR (300 MHz, CDCI 3 ) 6 6.48 (dd, J = 17.5 and 10.8Hz, 1H), 5.34 (d, J = 10.8Hz, 1H), 5.15 (d, J = 17.5Hz, 1H), 4.79 (d, J = 8.3Hz, 1H), 3.17 (q, J = 6.7Hz, 1H), 2.18-2.07 (m, 4H), 2.00-1.88 (m, 1H), 1.70-1.50 (m, 7H), 1.38 (s, 3H), 1.15 (s, 3H), 1.03 (d, J = 6.7Hz, 3H), 0.95 (d, 3 = 6 .6 Hz, 3H), 0.89 (s, 9H), 0.18 (s, 3H), 0.16 (s, 3H); 13C NMR (75 MHz, CDC13) ppm 217.19, 214.36, 139.30, 118.20, 67.83, 60.09, 53.64, 45.43, 45.36, 45.14, 233 44.16, 38.14, 34.87, 30.28, 27.23, 26.47, 24.69, 23.92, 19.28, 18.50, 14.42, 13.44, -1.45, -3.67; mass spectrum calcd (M+-C4 Hg) m/£ 375.2356, obsd 375.2361. Anal. Calcd for C26H44°3Si ! C ' 72.17; H, 10.25. Found: C, 72.01; H, 10.36. Spectral data for 83: IR (CC14 ) cm- 1 2933, 2860, 1705, 1640, 1463, 1375, 1295, 1252, 1232, 1048, 831; 1H NMR j (300 MHz, CDC13) 6 6.46 (dd, ,1 = 17.5 and 10.8Hz, 1H), 5.07- 4.95 (m, 3H), 4.57 (br s, 1 H), 3.04 (q, jl = 6.5Hz, 1H), 2.97 (br d, i = 13.7Hz, 1H) , 2.50 (br s, 1H) , 2.11 (dd, J = 15.6 and 8.4Hz, 1H), 1.84-1.35 (series of m, 7H), 1.61 (s, 3H), 1.29 (s, 3H) , 1.17 (d, J = 8 .6 Hz, 3H) , 1.06 (d, ,1 = 6.5Hz, 3H), 0.98 (s, 18H), 0.18 (s, 3H), 0.17 (s, 3H), 0.15 (s, 3H), 0.11 (s, 3H); mass spectrum calcd (M ) jd/£ 546.3924, obsd 546.3927. 234 (+) - (6 £,7fi) -7- [ (IB, 3B) -1- CtejLt-Butyldimethylsiloxy) -3- methyl-3-propionyl-4-pentenyl]-4,5,6 ,7-tetrahydro-6 ,7- dimethyl-l-indanone (85). In a 50 ml flask fitted with a reflux condenser were combined 84 (250 m g f 0.578 mmol) and 1 0 % potassium hydroxide in ethanol (20 mL). The reaction mixture was brought to the reflux temperature. After 2 h, TLC analysis (10% ethyl acetate in petroleum ether) determined that the reac tion was complete. The reaction mixture was diluted with water (40 mL) and extracted with methylene chlor ide (4x20 mL). The organic phases were combined, dried (MgSO^), and freed of solvent. Following purification by MPLC (silica gel; 10% ethyl acetate in petroleum ether), keto enone 85 (187 mg, 75%) was isolated as a white solid: mp 51.0-52.0°C; TaJp2 5 +42.7° (£ 5.13, CHC13 ); IR (CC14 ) cm" 1 2935, 2860, 1715, 1695, 1635, 1462, 1381, 1250, 1085, 919, 830; 1H NMR (300 MHz, CDC13 ) <5 6.02 (dd, 1 = 17.4 and 10.7Hz, 1H), 5.29 (d, jl = 10.7Hz, 1H), 5.21 (d, J. = 17.5Hz,lH), 4.81 (t, jl = 2.2Hz, 1H), 2.50 (q, J = 7.2Hz, 2H), 2.45-2.29 (series 235 of m, 7H) , 1.76-1.68 (m, 4H) , 1.34 (s, 3H) , 1.11 (d, jl = 7.0Hzf 3H), 1.00 (t, jl = 7.2Hz, 3H) , 0.97 (s, 3H) , 0.77 (s, 9H), 0.03 (s, 3H), -0.25 (s, 3H); 13C NMR (75 MHz, CDCI3 ) 214.06, 208.47, 173.38, 144.11, 141.52, 115.65, 73.23, 54.11, 43.75, 42.21, 37.15, 35.19, 31.71, 29.54, 27.66, 26.32, 25.67, 25.48, 20.26, 18.54, 18.31, 8.18, -3.18, -3.89; mass spectrum calcd (M+-C 4 Hg) m / & 375.2355, obsd 375.2351. (6 B»7£)-7- [ (IB,35,42)-1-(ier±-Butyldimethylsiloxy) -3- methy1-4-(tr imethylsiloxy)-3-viny1-4-hexeny1J-4,5,6 ,7- tetrahydro-6,7-dimethyl-l-indanone (8 6 ). Lithium diisopropylamide was OSKCH,). prepared in tetrahydrofuran (20 mL) from n-butyllithium (1.24 mL, 1.92 mmol, 1.55 B in hexane) and diisopropylamine (0.27 mL, 1.92 mmol). This solution was cooled to -78°C, and keto enone 85 (376 mg, 0.869 mmol) dissolved in tetrahydrofuran (5 mL) was added. The bright yellow bis-enolate was permitted to form over 45 min. Then HMPA (0.60 mL, 3.48 mmol) 236 followed by trimethylsilyl chloride (0.24 mL, 1.92 mmol) were added. The reaction mixture was allowed to warm gradually from -78°C to 20°C over 6 h. The tetrahydrofuran was removed in vacuo, and the residue was dissolved in pentane (80 mL) and extracted with saturated sodium bicarbonate solution (2x40 mL). The combined aqueous phases were back extracted with pentane (2x20 mL), and the combined organic phases were dried (MgS04) and freed of solvent. The mono- and bis- silyl enol ethers were separated by MPLC (silica gel; 5% ethyl acetate in petroleum ether) yielding bis-silyl enol ether 87 (139 mg, 29%), and silyl enol ether 8 6 (127 mg, 29%) as light yellow oils. Spectral data for 8 6 : IR (CC14 ) cm" 1 3090, 2930, 1695, 1661, 1634, 1461, 1379, 1315, 1250, 1117, 1070, 910, 831, 670; NMR (300 MHz, CDC13) 6 5.95 (dd, J - 17.5 and 10.7Hz, 1H), 5.14 (d, jI = 10.6Hz, 1H), 5.09 (d, jl = 17.4Hz, 1H), 4.82 (br s, 1H) , 4.64 (q, ,1 = 6 .8 Hz, 1H) , 2.38-2.15 (series of m, 7H), 1.81-1.60 (series of m, 4H), 1.50 9d, I = 6.5Hz, 3H), 1.23 (s, 3H), 1.12 (d, 3 ■ 6.9Hz, 3H), 1.00 (s, 3H), 0.77 (s, 9H), 0.22 (s, 9H), 0.04 (s, 237 3H), -0.24 (s, 3H); 13C NMR (75 MHz, CDCI 3 ) ppm 208.29 172.73, 157.07, 144.48, 144.46, 113.05, 99.85, 73.61, 45.81, 44.03, 43.39, 36.86, 35.16, 29.54, 27.61, 26.41 26.05, 25.38, 22.01, 18.61, 18.40, 11.83, 1.23, -3.16, -3.85; mass spectrum calcd (M+-CH-j) m/£ 489.3220, obsd 489.3211. Spectral data for (4J&, 5£,7£) -4-Ethylidene- 2,2,5,9,9,10,10-heptamethyl-7-[(4£,5B)-4,5,6,7- tetrabydro-4,5-di«ethyl-3-(trimethylsilozy)inden-4-yll- 5-vinyl-3,8-dioxa-2,9-disilaundecane (87): *H NMR (300 MHz, CDC13> 6 OSKCHjJj 6.03-5.89 (m, 1 H), 5.28-5.06 (m, 3H), 4.86-4.82 (m, IB), 4.64 (q, J » 6.7Hz, 1H), 2.64- ‘CHj^SIO 2.05 (series of m, 4H), 1.80- 1.59 (m, 2H), 1.50 (dd il = 6.7 and 1.3Hz, 3H), 1.22 (s, 3H), 1.11 (br d, J » 7.0Hz, 3H), 1.01-0.95 (m, 6 H), 0.77 (s, 9H), 0.22 (s, 9H), 0.02 (s, 9H), -0.24 (s, 3H), -0.28 (s, 3H). 238 (6B» 7fi)-4,5,6,7-Tetrahydro-7-£(IB,3B)-l-methoxy-3- methyl-3-12-(phenylselenyl)propionyl1-4-penteny1]-6,7- dimethyl-l-indanone (88). Silyl enol ether 76 (30 mg, Phs* 0 0.074 mmol) was dissolved in dry benzene (2 mL). Pheny.l- selenenyl chloride (14 mg, OCH, 0.074 mmol) dissolved in o benzene (2 mL) was introduced via syringe. The reddish color disappeared nearly instantaneously leaving a yellow solution. TLC analysis (1 0 % ethyl acetate in petroleum ether) showed the reaction to be complete after 20 min. The solvent was removed in vacuo, and the product was purified by MPLC (silica gel; 12% ethyl acetate in petroleum ether) providing 8 8 (25 mg, 69%) as an oil: IR (neat) cm”^ 3045, 2910, 1690, 1620, 1577, 1437, 1365, 1200, 995, 912, 785, 739; 1H NMR (300 MHz, CDCl-j) 6 7.54-7.24 (m, 5H, -SeCgH5), 6.21-5.83 (m, 1H, -CH=C), 5.22-5.09 (m, 2H, -C=CH2), 4.12-3.93 (m, 1H, H-C-OMe), 3.77-3.62 (m, 1H, H-C-Se), 3.11 and 2.94 (two s, 3H total, -OCH3), 2.60-2.04 (m, 7H), 1.90-1.52 (m, 4H), 1.46 and 1.41 (two d, jl = 6.9Hz, 3H total, Se-CH-Cfl-j), 1.37 and 1.30 239 (two s, 3H total), 1.22-1.09 (m, 6 H, methyl groups on cyclohexane); *3C NMR (75 MHz, CDCl-j) ppm 211.68, 208.68, 208.55, 173.53, 173.46, 143.87, 143.66, 143.30, 143.03, 136.17, 135.86, 128.85, 128.80, 128.49, 128.31, 114.48, 114.29, 82.76, 82.16, 59.89, 59.49, 59.39, 53.21, 53.16, 41.98, 41.81, 41.46, 41.01, 40.26, 39.70, 35.20, 35.16, 29.56, 29.49, 27.85, 27.67, 27.62, 27.43, 24.49, 24.00, 23.93, 19.22, 18.81, 17.59, 17.40; mass spectrum calcd (M+) m/£ 488.1830, obsd 488.1883. (6B#7B)-7-[ (lB»3S)-3-[2-Bromopropionyl]-l-methoxy-3- methyl-4-penteny 1 ]-4,5,6 ,7-tetrahydro-6 ,7-dimethyl-l- indanone (89). Silyl enol ether 76 ( 8 8 mg, 0.217 mmol) dissolved in tetrahydrofuran ( 6 mL) was 0 c h 3 cooled to 0°C with an ice bath. N-Bromosuccinimide (41 mg, 0.23 mmol) was added, and stirring was continued for 20 min. At that time, TLC analysis (10% ethyl acetate in petroleum ether) indicated the reaction to be complete. The reaction mixture was poured into a combination of brine (40 mL) 240 and saturated sodium bicarbonate solution (40 mL), and extracted with ether (4x40 mL). The combined organic phases were dried (MgS04) and freed of solvent. The crude material was purified by MPLC (silica gel; 23% ethyl acetate in petroleum ether) delivering 89 (76 mg, 85%) as a light yellow oil: IR (CDClg) cnT* 3100, 2980, 2940, 1719, 1695, 1648, 1465, 1446, 1425, 1412, 1383,1200, 1100, 1000; 1H NMR (300 MHz, CDC13) 6 6.04 (dd, il = 17.5 and 10.5Hz, 1H), 5.22-5.16 (m, 2H, C=CH2), 4.72-4.69 (m, 1H, Me-CHBr), 4.11 (dd, J = 11.0 and 3.0 Hz, 1H, H-C-OMe) and 3.95 (d, J[ = 10.3Hz, 1H, H-C-OMe), 3.08 and 2.91 (two s, 3H total, -OCH-j), 2.48- 2.28 (series of m, 9H), 1.83-1.74 (m, 2H), 1.71 and 1.69 (two d, {L = 6.7Hz, 3H total, Br-C-CH3 ), 1.39 and 1.28 (two s, 3H total), 1.16 and 1.13 (two s, 3H total), 1.12 (d, 1 = 6.9Hz, 3H); 13C NMR (75 MHz, CDC13) ppm 208.60, 208.28, 205.59, 204.64, 173.65, 173.50, 143.69, 143.38, 142.70, 114.72, 114.43, 82.20, 81.97, 59.44, 59.32, 53.14, 52.62, 42.57, 41.71, 41.39, 40.55, 39.54, 38.93, 35.17, 35.13, 29.54, 29.46, 27.84, 27.64, 27.40, 24.46, 23.88, 21.78, 21.11, 19.66, 18.74, 17.59, 17.24; mass spectrum calcd (M+-CH 4 OBr) m / & 299.2011, obsd 299.2020; m/£ (relative intensity) 241 299(2), 249(98), 247(100), 215(21), 169(20), 168(21), 163(42), 107(26), 91(27), 81(96). (-) — (4JS) -4- [ (2£) -2-Methoxy-2- [ (6 '£,7 '£) -4 • ,5 •,6 *,7 1 ■- tetrahydro-6',7'-dimethylspiro Cl,3-dithiolane-2, 1' - indan]-7'-yl]ethyl]-4-methyl-5-hexen-3-one (90). A 250 mL flask equipped with a Dean-Stark apparatus was charged with benzene (100 mL), p-toluenesulfonic acid mono hydrate (150 mg), 1,2- ethanedithiol (0.76 mL, 9.0 mmol), and keto enone 6 8 (750 mg, 2.26 mmol). The reaction mixture was brought to the reflux temperature and monitored daily by TLC analysis (10% ethyl acetate in petroleum ether). Additional aliquots (0.76 mL) of ethanedithiol were added daily until the reaction was complete (4 d). The reaction mixture was then diluted with ether (100 mL), extracted with 1 1] NaOH (5x75 mL), dried (MgSO^), and freed of solvent. The mono- and bis-dithioketals were separated by MPLC (silica gel; 3% ethyl acetate in petroleum ether). The bis-dithioketal 242 91 (256 mg, 23%) was obtained as a viscous brown oil while mono-dithioketal 90 (255 mg, 28%), [aJD^ -17.3° (£ 4.00, CHC13), was isolated as an off-white crystalline solid: mp 102.5-106.0°C. Spectral data for 90: IR (CC14) cm- 1 2980, 2925, 2840, 1711, 1633, 1463, 1374, 1195, 1042, 917; 2H NMR (300 MHz, CDC13) 6 5.87 (dd, 3 = 17.5 and 10.6Hz, 1H), 5.12 (d, 3 = 16.8Hz, 1H), 5.11 (d, 3 = 11.0Hz, 1H), 3.93 (d, 3 = 10.6Hz, 1H), 3.44-3.06 (series of m, 4H), 3.04 (s, 3H), 2.76- 2.27 (series of m, 7H), 2.12-1.97 (m, 3H), 1.72-1.46, (m, 3H), 1.34 (s, 3H), 1.20 (s, 3H), 1.14 (d, 3 = 7.1Hz, 3H) , 0.98 (t, 3 = 7.3Hz, 3H) ; 13C NMR (75 MHz, CDCI 3 ) ppm 211.51, 144.23, 143.65, 137.73, 114.07, 83.85, 77.21, 60.41, 53.33, 49.72, 44.14, 42.91, 42.70, 40.56, 37.85, 34.09, 30.45, 27.80, 27.37, 21.30, 19.64, 18.38, 8.56; mass spectrum calcd (M+) m / & 408.2156, obsd 408.2130. 243 Spectral data for (S'BrT'fi)-?'-!(lE,3S)-3-(2-Ethyl-l,3- dithiolan-2-yl)-l~methoxy-3-methyl-4-pentenyl]- 4',5',6',7'-tetrahydro-6',7'-dimethylspiro II,3- dithiolane-2rl'-indanl (91): IR (CC14 ) cm" 1 2975, 2928, 2880, 1460, 1370, 1275, 1097, 917; mass spectrum m/fi (relative intensity) 378(3), OCH 376(12), 319(12), 316(12), 260(22), 259(100), 237(12), 225(33), 205(61), 191 , 179(81), 169(84), 163(48), 157(37), 149(24), 145 , 95(27), 91(28), 81(30), 57 (87). Hydrolysis of Bis-dithiolane (91). The bis-dithioketal 91 (152 mg, 0.313 mmol) was combined with acetone ( 8 mL), water (1.7 mL), sodium carbonate (83 OCH3 mg, 0.78 mmol), and. methyl iodide (3.4 mL, 54 mmol). The 244 mixture was heated at the reflux temperature for approximately 72 h. At that time, TLC analysis (10% ethyl acetate in petroleum ether) indicated the lack of starting material. The volatile substances were removed in vacuo and the residue was partitioned between methylene chloride and saturated sodium bicarbonate solution. The organic phase was dried (MgSO^) and freed of solvent. Chromatographic purification (silica gel; 13% ethyl acetate in petroleum ether) afforded 6 8 (48 mg, 46%) whose NMR and IR spectra were identical to those of the materials previously prepared via either the degradative or synthetic route. (6£,7£)-3a,7a-Epoxyhexahydro-7-[(lS,3£-l-methoxy-3- methyl-3-propionyl-4-pentenyll-6f7-dimethyl-l-indanone (92). Enone 6 8 (400 mg, 1.20 mmol) was dissolved in methanol (24 mL), and the solution was cooled to 0°C with an ice bath. Then 30% hydrogen 245 peroxide ( 8 mL, 78 mmol) was introduced followed by the addition of 4 U sodium hydroxide solution (4 mL) over 5 min. The mixture was stirred and permitted to gradu ally rise to room temperature. TLC analysis (20% ethyl acetate in petroleum ether) was used to follow the progress of reaction. Additional portions of hydrogen peroxide (4 mL) were added until no further progress was observed by TLC analysis. The reaction mixture was diluted with water (40 mL) and extracted with methylene chloride (3x40 mL). The combined organic phases were dried (MgSO^) and freed of solvent. The remaining enone 6 8 (36 mg, 9%) was separated from the desired product by MPLC (silica gel; 17% ethyl acetate in petroleum ether) to yield the epoxide mixture 92 (312 mg, 75% or 82% based on recovered starting material) in a 15 to 85 ratio as a white solid: mp 69°-74°C; I“JD^ -19.0° (£ 3.05, CHCI3 ) ; IR (CDCI3 ) cnf1 3095, 2985, 2940, 2885, 2840, 1740, 1705, 1633, 1460, 1410, 1375, 1052, 978, 860, 788; XH NMR (300 MHz, CDCI3 ) 6 5.87 (dd, 1 = 17.4 and 10.8Hz, 1H), 5.16 (d, J = 10.6Hz, 1H), 5.12 (d, 1 = 17.3Hz, 1H), 3.92 (br d, Z = 10.8Hz, 1H), 3.15 and 3.02 (two s, 3H total, -OCH 3 , 15:85), 2.59-1.62 (series of m, 11H), 1.31-1.12 (m, 2H), 1.22 (s, 3H), 1.18 (s, 3H), 1.04 (d, £ = 6 .8 Hz, 3H), 0.98 246 (t, jI = 7.2Hz, 3H) ; 13C NMR (75 MHz, CDC13 ) ppm 211.98, 211.73, 210.75, 209.94, 143.42, 143.07, 114.64, 114.38, 82.85, 81.66, 71.65, 68.71, 68.23, 67.09, 60.00, 59.35, 53.23, 52.71, 41.77, 40.25, 39.79, 39.49, 38.58, 37.80, 35.05, 32.85, 32.61, 30.56, 30.49, 30.40, 26.93, 25.85, 25.38, 24.75, 24.22, 23.88, 19.65, 19.57, 19.38, 17.37, 16.50, 8.36, 8.31; mass spectrum calcd (M+ ) m/£ 348.2301, obsd 348.2303. (6fi, 7fi)-7-1(IB,35,4£)-4-(Trimethylsiloxy)-l-methoxy-3- methyl-3-viny1-4-hexenyl]-3a,7a-epoxyhexahydro-6,7- dimethyl-l-indanone (93). In tetrahydrofuran (10 mL) lithium diisopropylamide was formed from u-butyllithium (0.50 mL, 0.81 mmol, 1.6 in 0 C H 3 hexane), and diisopropylamine (0.11 mL, 0.81 mmol). This solution was cooled to -78°C, and ketone 92 (128 mg, 0.367 mmol) dissolved in tetrahydrofuran (2 mL) was added. The yellow enolate was formed over 1 h. Sub sequently, HMPA (0.26 mL, 1.5 mmol) and trimethylsilyl 247 chloride (0 . 1 0 mL, 0.81 mmol) were added resulting in immediate disappearance of the yellow color of the reaction mixture. Stirring was continued for 5 h while the temperature slowly rose to 10°C. The tetrahydro- furan was removed in vacuo. The residue was taken up in pentane (40 mL) and extracted with saturated sodium bicarbonate solution (3x20 mL). The aqueous phases were back-extracted with pentane (2x10 mL). The combined organic phases were dried (MgSO^) and freed of solvent. MPLC (7% ethyl acetate in petroleum ether) was used to separate recovered starting material (55 mg, 43%) from silyl enol ether 93 (49 mg, 32% or 56% based on recovered starting material): IR (CC14) cm-* 3090, 2970, 2935, 1740, 1660, 1632, 1450, 1408, 1378, 1320, 1250, 1105, 1073, 1005, 975, 913, 841; *H NMR (300 MHz, CDC13 ) 6 5.91 (dd, Z = 17.5 and 10.8Hz, 1H), 5.09 (d, Z = 11.1Hz, 1H), 5.05 (d, Z = 17.9Hz, 1H), 4.66 (q, Z * 6 .6 Hz, 1H), 3.91 (br d, Z = 4.6Hz, 1H), 3.29 and 3.10 (two s, 3H total, -0CH3 15:85), 2.34-1.53 (series of m, 9H), 1.50 (d, Z - 6.5Hz, 3H), 1.36-1.06 (m, 2H), 1.16 (s, 3H), 1.11 (s, 3H), 0.97 (d, Z = 6 .8 Hz, 3H), 0.20 (s, 9H); 13C NMR (20 MHz, CDC13 ) ppm 209.79, 208.96, 156.19, 145.40, 143.42, 143.10, 114.67, 114.42, 112.69, 100.49, 82.92, 82.54, 81.77, 71.36, 248 6 8 .6 8 , 58.14, 45.87, 42.04, 41.91, 41.34, 39.93, 37.95, 35.08, 33.22, 32.97, 32.71, 30.67, 30.54, 26.96, 26.01, 25.69, 24.92, 24.47, 24.03, 22.49, 20.38, 19.75, 19.55, 17.64, 11.76, 8.50, 1.28; mass spectrum calcd (M+) jm/£ 420.2696, obsd 420.2704. (4fi)-4-[(2£)-2-I(4fi,5£)-3a,7a-Epoxy-3a,4,5,6,7,7a- hexahydro-4,5-dimethyl-3-(trimetbylsiloxy)inden-4-ylJ- 2-methoxyethylJ-4-methyl-5-hexen-3-one (95). Anamolously, the silyl enol ether of the cyclopentanone was once produced under reac tion conditions previously (CH,),SI0 OCH reported to form silyl enol ether 93. Epoxy ketone 92 (140 mg, 0.402 mmol) furnished 29 mg of 95 (17%) and recovered starting material (38 mg, 27%) : 1H NMR (300 MHz, CgDg) 6 5.75 (dd, i = 17.4 and 10.7Hz, 1H), 5.05 and 5.04 (two d, 3. = 17.5Hz, 1H total, C=CH), 4.95 and 4.94 (two d, £ = 2.4Hz, 1H total, C=CH), 4.44 (t, 1 = 2.4Hz, 1H), 3.99 (br d, jl = 10.1Hz, 1H), 3.55 and 3.09 (two s, 3H total, -OCHj, 7:93), 2.63-1.67 (series of m, 249 11H), 1.37 3H), 1.07 (d, 1 = 7.0Hz, 3H), 0.16 (6JB, 7fi) -7- [ (lfi, 3B, 4J[) -4- (±£i±-Butyldimethylsiloxy) -1- methoxy-3-methy1-3-viny1-4-hexeny11-3a,7a- epoxyhexahydro-6,7-dimethy1-1-indanone (94). Lithium diisopropylamide was produced in tetrahydrofuran (10 mL) from a-butyllithium (0.48 mL, 0.735 mmol, 1.5 B in Och hexane) and diisopropylamine (0.11 mL, 0.770 mmol). The solution was cooled to -78°C, and ketone 92 (122 mg, 0.350 mmol) dissolved in tetrahydrofuran (2 mL) was added. The enolate was allowed to form over 40 min. Subsequently, HMPA (0.24 mL, 1.40 mmol) and £- butyldimethylsilyl trifluoromethanesulfonate (0.18 mL, 0.770 mmol) were added. Stirring was continued for 8 h during which time the temperature of the reaction mixture was permitted to rise to room temperature. The mixture was concentrated under reduced pressure. The residue was taken up in pentane (40 mL) and extracted with saturated sodium bicarbonate solution (3x20 mL). The aqueous phases were back-extracted with pentane (2x10 mL). The combined organic phases were dried (MgSO^) and freed of solvent. Purification was achieved by MPLC (silica gel; 5% ethyl acetate in petroleum ether) which furnished 94 (18 mg, 11%) as a light yellow oil: IR (CC14 ) cm” 1 3080, 2960, 2920, 2850, 1745, 1660, 1630, 1460, 1405, 1375, 1320, 1255, 1200, 1100, 1085, 1005, 910, 830; 1H NMR (300 MHz, C6 D6) 6 6.03 (dd, Z » 17.6 and 10.7Hz, 1H), 5.11 (dd, = 17.5 and 1.2Hz, 1H), 5.09 (dd, Z = 10.7 and 1.2Hz, 1H), 4.73 (q, ,1 = 6 .8 Hz, 1H) , 4.24 (dd, Z ~ 6 . 8 and 1.7Hz, 1H), 3.09 and 3.06 (two s, 3H total), 2.21 (dd Z = 14.8 and 7.0Hz, 1H), 2.15-1.66 (series of m, 10H) 1.56 (d, 2 = 6 .8 Hz, 3H), 1.32 (s, 3H), 1.29 (s, 3H), 1.07 (s, 9H), 1.04 (d, Z = 7.0Hz, 3H), 0.21 (s, 3H), 0.20 (s, 3H); mass spectrum calcd (M+-C4 Hg) m/£ 405.2461, obsd 405.2459. 251 Attempted Reformatsky Ring Closure. Zinc dust (516 mg, 7.90 mmol), cuprous bromide (38 mg? 0.26 mmol) and diethylaluminum chloride (0.32 mL, 0.32 mmol, 1 U in hexane) were combined in tetrahydrofuran (1 mL) and stirred at room temperature for 45 min. a-Bromoketone 89 (65 mg, 0.158 mmol) dis solved in tetrahydrofuran (5 mL) was added over 4.5 h with a syringe pump. After being stirred overnight at room temperature, the mixture was quenched with pyri dine and filtered through Celite. The solvent was removed In vacuo, and the residue was partitioned between ether and 10% hydrochloric acid. The organic phase was dried (MgSO^) and freed of solvent. Purifi cation was achieved by MPLC (silica gel; 10% ethyl acetate in petroleum ether) which afforded keto enone 6 8 (17 mg, 32%). This material matched the keto enone previously prepared. REFERENCES 1. Devon, T.K.; Scott, A.I. "Handbook of Naturally Occurring Compounds"; Academic Press: New York, 1972; Vol. II. 2. For a review of metabolites produced by basidiomycetes see: Ayer, W.A.; Browne, L.M. Tetrahedron 1981, 22, 2199-2248. 3. Kavanagh, F.; Hervey, A.; Robbins, W.J. Proc. Natl. Acad. Sci■ USA 1951, 22, 570-574; 1952, 21, 555-560. 4. Knauseder, F.; Brandi, E. 2. Antibiotics 1975, 21, 125-131. 5. Michel, K.H.; Dorman, D.E.; Occolowitz, J.L. £ u £x . Chemother. Inf.ec±. I2ls. Proc. Int. £nngr. £hfimo.ther. nth 1979, 1, 479-480. 6. Berner, H.; Vyplel, H.; Schulz, G.; Stuchlik, P. Tetrahedron 1983, 21, 1317-1321. 7. Hogenauer, G. in "Mechanism of Action of Antibacterial Agents," Kahn, R.E., Ed. Springer- Verlag, New York, 1979, pp. 344-360 and references cited therein. 8. (a) Berner, H.; Schulz, G.; Schneider, H. Jlatrahfidrnn 1980, 21, 1807-1811. (b) Ibid. 1981, 22, 915-919. (c) Berner, H.; Schulz, G.; Fischer, G. Monatsh. Chem. 1981, 112. 1441-1450. (d) Berner, H. ; Vyplel, H.; Schulz, G.; Stuchlik, P. Tetrahedron 1983, 21, 1317-1321. (e) Berner, H.; Vyplel, H.; Schulz, G.; Schneider, H. Tetrahedron 1983, 21, 1745-1748. (f) Schulz, G.; Berner, H. Tetrahedron 1984, 41, 905-917. (g) Berner, H.; Vyplel, H.; Schulz, G. Ucnatsch. jCimm. 1983, 111, 501-507. 252 253 (h) Berner, H.; Vyplel, H.; Schulz, G.; Stuchlik, P. Tetrahedron 1984, 42, 919-923. (i) liiid. Monatsh. Chem. 1983, 111, 1125-1136. 9. Anchel, M. J. Biol. Chem. 1952, 122, 133-139. 10. (a) Arigoni, D. Gazz. Chim. Ital. 1962, 92. 884-901. Arigoni, D. Pure Appl. Chem. 1968, 12, 331-347. (b) Birch, A.J.; Cameron, D.W.; Holzapfel, C.W.; Rickards, R.W. Chem. Ind. (London) 1963, 374- 375. Birch, A.J.; Holzapfel, C.W.; Rickards, R.w. Tetrahedron 1966, Suppi. 2, Part n , 359- 387. 11. (a) Naegeli, P. Ph.D. Thesis ETH, Zurich, 1961. (b) Buzzolini, M. Ibid. 1966. (c) Bonavia, G. Ibid. 1968. (d) Hasler, H. Ibid. 1979. 12. Dobler, M.; Durr, B.G. Cryst. Struct. Comm. 1975, 4, 259-264. 13. For a review of allylic pyrophosphate metabolism see: Cane, D.E. Tetrahedron 1980, 22, 1109- 1159. 14. Kahn, M. Tetrahedron Lett. 1980, 4547-4548. 15. (a) Gibbons, E.G. 2. Am. Chem. Soc. 1982, 124, 1767-1769. (b) Ibid. 2. Org. Chem. 1980, 42, 1540-1541. 16. Our efforts in this area have been made possible by the generosity of the following individuals in supplying us with quantities of pleuromutilin and/or tiamulin: Dr. Russ Buchman (SDS Biotech), Dr. R. Nagarajan (Eli Lilly Company), and Dr. Heinz Berner (Sandoz Forschungsinstitut). 17. Mukherjee, S. 2. Ind. JCham. Sac* 1962, 2 1 , 347- 352. 18. Hauser, F.M.; Progany, S.A. Synthesis 1980, 814- 815. 19. Piers, E.; Britton', R.W. ; DeWaal, W. Can. 2. Chem. 1969, 42, 4307-4312. 254 20. Evans, D.A.; Sims, C.L.; Andrews, G.C. I. Am. Chem. Soc. 1977, 55, 5453-5461. 21. Trost, B.M.; Nishimura, Y.; Yamamoto, K.; McElvain, S.S. 5. Am. Chem. Soc. 1979, 101. 1328-1330. 22. Ryu, I.; Murai, S.; Hatayama, Y.; Sonoda, N. Tetrahedron Lett. 1978, 3455-3458. 23. Hiyama, T.; Shinoda, M.; Nozaki, H. Tetrahedron Lett 1978, 771-774. 24. (a) MacAlpine, G.A.; Raphael, R.A.; Shaw, A.; Taylor, A.W. ; Wild, H.-K. J[. Chem. Soc.. Chem. Commun. 1974, 834-835. (b) Karpf, M. ; Dreiding, A.S. Helv. Chim. Acta 1976, 55, 1226-1232. (c) Hiyama, T.; Shinoda, M.; Saimoto, H.; Nozaki, H. Bull. Chem. Soc. Jpn. 1981, 54., 2747- 2758. 25. Claesson, A.; Olsson, L.-I.; Bogentoft, C. Acta Chem. Scand. 1973, 2 1 , 2941-2944. 26. (a) Meskins, F.A.J. Synthesis 1981, 501-522. (b) Van Allen, J.A. firs. 5ynth., Coll. JLol- IX 1963, 21-22. (c) Mackenzie, C.A.; Stocker, J.H. J. Org. Chem. 1955, 25, 1695-1701. (d) Schwenk, E.; Fleischer, G.; Whitman, B. I. A m . Chem. Soc. 1938, 55, 1702-1703. 27. Tsunoda, T.; Suzuki, M.; Noyori, R. Tetrahedron Lett. 1980, 1357-1358. 28. Evans, D.A.; Truesdale, L.K.; Grimm, K.G.; Nesbitt, S.L. I. Am. Chem. Soc. 1977, 55, 5009- 5017. 29. (a) Allen, C.F.; Kalm, M.J. £rg. Synth. Coll. Vol. IX 1963, 616-620. (b) McMurry, J.E. Organic Reactions? 1976, 24., 187-224. (c) Dean, P.D.A. 5. Chem. Soc. 1965, 6655. (d) Sheehan, J.C.; Davies, G.D. 5* Qrg. Chem. 1964, 25, 2006-2008. 255 (e) Liotta, D.; Sunay, U.; Santiesteban, H.; Markiewicz, W. J. Org. Chem. 1981, 4iL, 2605- 2610. (f) Jung, M.E.; Blumenkopf, T.A. Tetrahedron Lett. 1978, 3657-3660. (g) Taber, D.F.; Saleh, S.A.; Korsmeyer, R.W. 2. Chem. 1980, 41, 4699-4702. 30. Dunn, M.S.; Smart, B.W. Org. Synth. Coll. Vol. IY 1963, 55-58. 31. Paquette, L.A.; Doehner, R.F. 2. Org. Chem. 1980, 41, 5105-5113. 32. Cairns, T.L. J. Am. Chem. Soc. 1941, £1, 871- 872. 33. Pirkle, W.H.; Hoekstra, M.S. J. Org. Chem. 1974, 3 2 , 3904-3906. 34. Pirkle, W.H.; Hauske, J.R. 1 . Org . Chem. 1977, 42, 2781-2782. 35. (a) Ratcliffe, R . ; Rodehurst, R. J. Org. Chem. 1970, 35., 4000-4002. (b) Collins, J.C.; Hess, W.W. JlLg. Synth. 1972, 5 1 , 5-10. 36. Roush, W.R. 2. Am* Chem. fLoc. 1978, 100. 3599- 3601. 37. Paquette, L.A.; Schostarez, H . ; Annis, G.D. J. Am. Chem. Soc. 1981, 103. 6526-6527. 38. Evans, D.A. In "Asymmetric Synthesis", Morrison, J.D., Ed.; Academic Press, Inc.: New York, 1983; Vol. 3, Chapter 1. 39. (a) Clive, D.L.J.; Russell, C.G.; Suresh, C.S. 2* Org. Chem. 1982, 4Z, 1632-1641. (b) Clive, D.L.J.; Russell, C.G. 3 . Chem. £££•, Chem. Commun. 1981, 434-436. 40. Kowalski, C. J. ; Dung, J.-S. Am* Chem. Soc. 1980, 7950-7951. 41. (a) Ohnuma, T.; Hata, N.; Fujiwara, H.; Ban, Y. 2. Org. Chem. 1982, 4Z, 4713-4717. 256 (b) Hudrlik, P.F.; Kulkarni, A.K. J. Am. Chem. £cc. 1981, 6251-6253. (c) Koppel, G.A.; Kinnick, M.D. J. Chem. Soc.. Chem. Commun. 1975, 473. 42. Martinez, G.R.; Grieco, P.A.; Williams, E.; Kanai, K.; Srinivasan, C.V. J. Am. Chem. Soc. 1982, 1 M , 1436-1438. 43. Recently a procedure was published where nickel boride reduced a vinyl phenyl selenide in the presence of a vinyl sulfone: Back, T.G. J. Chem. jSee., .Chem. Commun. 1984, 1417-1418. 44. Taniguchi, M.; Koga, K.; Yamada, S. Tetrahedron 1974, M / 3547-3552. 45. (a) Narasaka, K.; Uchimura, T. Chem. Lett. 1982, 57-58. (b) Torii, S.; Okamoto, T.; Kadono, S. Chem. Lett. 1977, 495-498. 46. Tanaka, K.; Tamura, N. ; Kaji, A. Chem. Lett. 1980, 595-598. 47. (a) Yamamoto, K.; Tomo, Y.; Suzuki, S. Tetrahedron Lett. 1980, 2861-2864. (b) Larson, G.L.; Fuentes, L.M. J. Am* Chem. Soc. 1981, 1Q1, 2418-2419. (c) Grieco, P.A.; Wang, C.-L.; Burke, S.D. J. Chem. Soc.. Chem. Commun. 1975, 537-538. 48. Fujita, T.; Suga, K.; Watanabe, S. Yukaqaku 1976, 25, 480-484. 49. (a) Corey, E.J.; Chaykovsky, M. J. Am. Chem. Soc. 1965, SI, 1353-1364. (b) Ibid. Org. Synth. 1969, 4 9 . 78-80. (c) Bertini, F.; Grasselli, P.; Zubiani, G. jl. Chem. Soc.. Chem. Commun. 1969, 1047-1048. (d) Hart, P.A.; Sandmann, R.A. Tetrahedron Lett. 1969, 305-306. (e) Takaki, K.; Yasumura, M.; Negoro, K. Anqew. Chem. Int. £d. Engl. 1981, 25, 671-672. 50. Yamamoto, K . ; Tomo, Y. Chem. lt£hi. 1983, 531- 534. 257 51. The investigations reported in this chapter section were accomplished by Dr. Philip C. Bulman-Page while on a NATO Postdoctoral Fellowship of the Science and Engineering Research Council held at The Ohio State University during 1981-1983. 52. (a) Takanor S.; Yonaga, M.; Chiba, K.; Ogasawara, K. Tetrahedron Lett. 1980, 3697-3700. (b) Nicolaou, K.C.; Magolda, R.L. J. Org. Chem. 1981, 41, 1506-1508. 53. (a) Caruso, A.J.; Polonsky, J.; Rodriguez, B.S. Tetrahedron Lett. 1982, 2567-2570. (b) Schmidt, R.R.; Abele, W. Angew. Chem. Int. Ed. Engl. 1982, 21, 302-303. 54. Roush, W.R.; D'Ambra, T.E. J. Org. Chem.. 1980, 41, 3927-3929. 55. Bulman-Page, P.C.; Roberts, R.A.; Paquette, L.A. Tetrahedron Lett. 1983, 3555-3558. 56. Mawhinney, T.P.; Madson, M.A. 2. Org. £h£m., 1982, 42, 3336-3339. 57. Corey, E.J.; Cho, H. ; Rucker, C.; Hua, D.H. Tetrahedron Latt. 1981, 3455-3458. 58. Lemieux, R.K.; Kondo, T. Carbohydr. Res. 1974, 21, C4-C6. 59. For example: (a) Fujita, C.; Nagao, Y.; Kaneko, K. Chem. Pharm. Bull. 1978, 2 1 , 3743-3751. (b) Smith, R.A.J.; Hannah, D.J. Synth. Commun. 1979, 2, 301-311. (c) Ho, T.-L.; Wong, C.H. Can. J. Chem.. 1972, 11, 3740-3741. (d) Ho, T-L.; Ho, H.C.; Wong, C.M.; 2. Chem. Soc.. Chem. Commun. 1972, 791. (e) Reece, C.A.; Rodin, J.O.; Brownlee, R.G.; Duncan. W.G.; Silverstein. R.M. Tetrahedron. 1968, 24, 4249-4256. 60. (a) Corey, E.J.; Erickson, B.W. 1. Org. Chem. 1971, 2 1 , 3553-3560. (b) Cain, E.N.; Welling, L.L. Tetrahedron Lett. 1975, 1353-1356. 258 61. (a) Pojer, P.M.; Angyal, S.J. Aust. J. Chem. 1978, 31, 1031-1040. (b) Takano, S.; Hatakeyama, S.; Ogasawara, K. jl. Chem* £&£., Chem. Commun. 1977, 68. (c) Ho, T.-L.; Wong, C.M. Synthesis 1972, 561. (d) Oishi, T.; Kamemoto, K.; Ban, Y. Tetrahedron Lett. 1972, 1085-1088. 62. Stevens, P.G. £. Am- Chem* Soc. 1932, 2 A , 3732- 3738. 63. (a) Corey, E.J.; Bock, M.G. Tetrahedron Lett. 1975, 2643-2646. (b) Yamada, K . ; Kato, K.; Nagase, H.; Hirata, Y. Tetrahedron Lett. 1976, 65-66. (c) Pojer, P.M.; Angyal, S.J. Tetrahedron Lett. 1976, 3067-3068. (d) Schwindeman, J.A.; Magnus, P.D. Tetrahedron Lett. 1981, 4925-4928. (e) Suzuki, K.; Inanaga, J. ; Yamaguchi, M. Chem. Lett. 1979, 1277-1278. 64. (a) Boar, R.B.; Hawkins, D.W.; McGhie, J.F.; Barton, D.H.R. J. Chem. £&£., Perkin Trans. 1 1973, 654-657. (b) Chittenden, G.J.P. Carbohydr. Res. 1970, 15., 101-109. (c) Jewell, J.S.; Szarek, W.A. Tetrahedron Lett. 1969, 43-46. 65. Arnarp, J.; Lonngren, J. Acta Chem. Scand. £ 1978, 2 1 , 465-467. 66. Djerassi, C.; Scholz, C.R. J. Am* Chem. Soc. 1948, 2 1 , 417-418. 67. Mukaiyama, T. Organic Reactions 1982, 2JL, 203- 331. 68. (a) Urabe, H . ; Takano, Y. ; Kuwajima, I. J. Am* Chem. Soc. 1983, 105. 5703-5704. (b) Kuwajima, I.; Urabe, H. J. Am* Chem. £&C. 1982, l M r 6831-6833. 69. (a) Heathcock, C.H.; Buse, C.T.; Kleschick, W.A.; Pirrung, M.C.; Sohn, J.E.; Lampe, J. J. Org. Chem. 1980. 15, 1066-1081. 259 (b) Ireland, R.E.; Mueller, R.H.; Willard, A.K. J[. Am. Chem. Soc. 1976, M , 2868-2877. (c) Ireland, R.E.; Thaisrivongs, S.; Vanier, N.; Wilcox, C.S. J[. Org. Chem. 1980, A l t 48-61. 70. Narasaka, K.; Soai, K. ; Aikawa, Y.; Mukaiyama, T. Bull. Chem. Soc. Jpn. 1976, 42, 779-783. 71. Murata, S.; Suzuki, M.; Noyori, R. J. Am. Chem. Soc. 1980, 122, 3248-3249. 72. (a) Alexis, A.; Chapdelaine, M.J.; Posner, G.H.; Runquist, A.W. Tetrahedron Lett. 1978, 4205- 4208. (b) Mukaiyama, T.; Banno, K.; Narasaka, K. 2. A m . Chem. Soc. 1974, 22, 7503-7509. (c) Smith, A.B.; Guaciaro, M.A.; Schow, S.R.; Wokulich, P.M.; Toder, B.H.; Hall, T.W. 2. Am* Chem. Soc. 1981, 103. 219-222. 73. Ishida, A.; Mukaiyama, T. Bull. Chem. £Q£. 2pn* 1977, 22, 1161-1168. 74. Taylor, M.D.; Minaskanian, G.; Winzenberg, K.N.; Santone, P.; Smith, A.B. 2» Chem. 1982, A l t 3960-3964. 75. Banno, K.; Mukaiyama, T. M I . Chem. £&£. 2pn- 1976, A h 1453-1454. 76. Patterson, I. Tetrahedron Lett. 1979, 1519-1520. 77. (a) Boyer, J . ; Corriu, R.J.P.; Perz, R.? Reye, C. il. Chem. Soc.. Chem. Commun. 1981, 122-123. (b) ibid. Tetrahedron 1983, 22, 117-122. 78. Nitz, T.J.; Paquette, L.A. Tetrahedron Lett. 1984, 3047-3050. 79. (a) Termont, D.; DeClerco, P.; De Keukeleire, D. ; Vandewalle, M. Synthesis. 1977, 46-48. (b) Van Audenhove, M.; De Keukeleire, D.; Vandewalle, M. Tetrahedron Lei£. 1980, 1979- 1982. 26Q 80. (a) Oppolzer, W. Acc. Chem. Res. 1982/ 1I>, 135- 141. (b) Tietzef L.-F. ; Reichert/ U. Anaew. Chem. 1980/ 92/ 832-833. 81. (a) Termontz D.; De Keukeleire/ D.; Vandewalle/ M. jI. £hem. £nc., jEfirkin Trans. 1 1977/ 2349- 2353. (b) Pirrung/ M.C. J. Am. Chem. Soc. 1979, 101. 7130-7131. (c) Margaretha, P. Tetrahedron 1973, 21, 1317- 1319. 82. (a) Cantrell, T.S.? Haller, W.S.; Williams, J.C. i. Ox#. Chem. 1969, 21, 509-519. (b) Corey, E.J.; Bass, J.D.; Le Mahieu, R.; Mitra, R.B. J. Am. Chem. Soc. 1964, M , 5570- 5583. 83. Kosugi, H.; Sekiguchi, S.; Sekita, R.; Uda, H. Bull. Chem. Soc. Jpn. 1976, 49. 520-528. 84. Kueh, J.S.; Mellor, M.; Pattenden, G. J. Chem. Soc.. Chem. Commun. 1978, 5-6. 85. (a) Wolff, S.; Ayral-^Kaloustian, S.; Agosta, W.C. jJ. £ 1 3 . Chem. 1976, 41, 2947-2950. (b) Salomon, R.G.; Sachinvala, N.D.; Raychaudhuri, S.R.; Miller, D.B. J[. Am. Chem. SSl£. 1984, 1 M , 2211-2213. (c) Koft, E.R. ; Smith, A.B. jJ. Am. Chem. Soc. 1984, 1 M , 2115-2121. 8 6 . (a) Cargill, R.L.; Dalton, J.R.; Morton, G.H.; Caldwell, W.E. Synth. 1984, £2, 118-124. (b) Fetizon, M.; Lazare, S.; Pascard, C.; Prange, T. j. Chem. £ j q q . , Perkins Trans. 1 1979, 1407-1410. 87. (a) Ariel, S.; Askari, S.; Scheffer, J.R.; Trotter, J.; Walsh, J.E. j. Am* Chem. Soc. 1984, 106. 5726-5728. (b) Byrne, B.; Wilson, C.A.; Wolff, S.; Agosta, w.c. j. Chem. £&£., Perkins Trans. 1 1978, 1550- 1560. (c) Wolff, S.; Schreiber, W.L.; Smith, A.B.; Agosta, W.C.; j. Am. Chem. £qc. 1972, 9 1 , 1 1 9 1 - 7806. 261 (d) Gloor, J.? Schaffner, K. Helv. Chim. Acta 1974, 5 1 , 1815-1845. 88. Beyerman, H.C.; Bontekoe, J.S. Reel. Trav. Chim. Pays-Bas 1962, £1, 691-698. 89. Chaudhary, S.K.; Hernandez, 0. Tetrahedron Lett. 1979, 95-98. 90. Salomon, R.G.; Ghosh, S. Org. Synth. 1984, 52., 125-131. 91. Hart, D.J. Science 1984, 2 2 1 , 883-887 and relevant references cited therein. 92. (a) Hart, D.J.; Chuang, C.-P. J. Org. Chem. 1983, 1782-1784. (b) Stork, G . ; Mook, R . ; Biller, S.A.; Rychnovsky, S.D. J. Am- Chem. Soc. 1983, 105. 3741-3742. 93. Danishefsky, S.; Yan, C.F. Synth. Commun. 1978, £, 211-218. 94. (a) Blanco, L . ; Amice, P.; Conia, J.M. Synthesis 1976, 194-196. (b) Reuss, R.H.; Hassner, A. J. Qrg. Chem. 1974, 1 1 , 1785-1787. 95. Bachi, M.D.; Frolow, F.; Hooraert, C. 1 . Q l q . Chem. 1983, 4£, 1841-1849. 96. Kuivila, H.G. Synthesis 1970, 499-509. 97. (a) Hart, D.J., The Ohio State University, personal communication, 1984. (b) Stork, G . ; Sher, P.M. 1. Am* Chem. Soc. 1983, I S i l , 6766-6767. 98. (a) Stork, G . ; Mook, R. J. Am* Chem. Soc. 1983, 105. 3720-3722. (b) Stork, G.; Baine, N.H. J. Am* Chem. £g£. 1982, 1 M , 2321-2323. 99. Zembayashi, M.; Tamao, K.; Kumada, M. Synthesis 1977, 422-423. 262 100. (a) Williams, J.R.; Sarkisian, G.M. Synthesis 1974, 32-33. (b) Masuoka, N.; Kamikawa, T. Tetrahedron Lett. 1976, 1691-1694. (c) Harayama, T.; Cho, H.; Inubushi, Y. Tetrahedron LahJt. 1977, 3273-3276. 101. (a) Trost, B.M.; Murayama, E. 2.. Am* Chem. fLoc. 1981, 1 0 2 , 6529-6530. (b) Ihid. Tetrahedron Lett. 1982, 1047-1050. 102. (a) Sutherland, J.K. Chem. Soc. Rev. 1980, S., 265-280. (b) Amupitan, J.; Sutherland, J.K. J. Chem. iLoo-, Cham. .Commun. 1980, 388-399. 103. Maas, D.D.; Blagg, M.; Wiemer, D.F. J. Org. Chem. 1984, A l , 853-856. 104. Maruoka, K.; Hashimoto, S.; Kitagawa, Y.; Yamamoto, H. ; Nozaki, H. j. Am- Chem. £££. 1977, 00, 7705-7707. 105. Tsuji, J.; Mandai, T. Tetrahedron Lett. 1978, 1817-1820. 106. Nishiyama, H.; Sakuta, K.; Xtoh, K. Tetrahedron Lett. 1984, 2487-2488. 107. Baraldi, P.G.; Barco, A.; Benetti, S.y Pollini, G.P. ; Polo, E.; Simoni, D. il. Org. Chem. 1985, 00, 23-29. 108. (a) Trost, B.M.; Verhoeven, T.R. jl. Am. Chem. Soc. 1979, 1 0 1 , 1595-1597. (b) Ibid. 1977, 02r 3867-3868. (c) Ibid. 1980, 1 0 2 , 4743-4763. (d) Trost, B.M.; Warner, R.W. 0. Am* jCham* Soc. 1982, 1M , 6112-6114. (e) Kitagawa, Y.; Itoh, A.; Hashimoto, S.; Yamamoto, H.; Nozaki, H. O. Am* Chem. Soc. 1977, 00, 3864-3867. 109. Earl, R.A.; Townsend, L.B. £rg. Synth. 1981, £0, 81-87. 110. "Aldrich Catalog/Handbook of Fine Chemicals 1984- 1985"; p. 724. 263 111. Stowell, J.C.; Keith, D.R.; King, B.T. flr.g. Synth. 1984, £ 2 , 140-148. 112. Baudat, R.; Petrzilka Helv. Chim. Acta 1979, 62. 1406-1410. 113. The experimental reported here was accomplished by Dr. Philip C. Bulman-Page while on a NATO Postdoctoral Fellowship of the Science and Engineering Research Council held at The Ohio State University during 1981-1983. 114. Kharasch, M.S.; Seyler, R.C. ? Mayo, F.R. J[. A m . Chem. Soc. 1938, ££, 882-884. 115. Neilan, J.P.; Laine, R.M.; Cortese, N.; Heck, R.F. J. £ r s . Chem. 1976, 41, 3455-3459.