llllIIIIIIII-Illllivl,

This is to certify that the
dissertation entitled

THE REGlO-AND STEREOSELECTIVE SYNTHESlS OF 2,3,5-

TRISUBSTITUTED TETRAHYDROFURANS V|A CYCLlZATION OF

EPOXY DIOLS AND TOTAL SYNTHESIS OF THE PROPOSED
STRUCTURE OF MUCOXlN

presented by

Radha Sridhar Narayan

has been accepted towards fulfillment
of the requirements for the

Ph. D.

 

degree in Chemistry

 

 

 

 

   

Major Professor’s Ignature

’[Zég 2.0051

Date

MSU is an Affirmative Action/Equal Opportunity Institution

  

 

 

 

 

 

 

 

   

Ll RARY
Michigan State
University

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/ClFlC/DateDue.p65-p.15

——-_..__. -

 

_————
fl—‘h

 

THE REGIO-AND STEREOSELECTIVE SYNTHESIS OF 2,3,5-TRISUBSTITUTED
TETRAHYDROPURANS VIA CYCLIZATION OF EPOXY DIOLS AND TOTAL
SYNTHESIS OF THE PROPOSED STRUCTURE OF MUCOXIN
By

Radha Sridhar Narayan

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Chemistry

2004

 

ABSTRACT
THE REGIO—AND STEREOSELECT IVE SYNTHESIS OF 2,3,5-TRISUBSTITUTED
TETRAHY DROFURANS VIA CYCLIZATION OF EPOXY DIOLS AND TOTAL
SYNTHESIS OF THE PROPOSED STRUCTURE OF MUCOXIN
By
Radha Sridhar Narayan

This dissertation describes the development of a method for the stereoselective
synthesis of 2,3,5—trisubstituted tetrahydrofurans (THFs), and the application of this
method towards the total synthesis of mucoxin — a nonclassical annonaceous acetogenin.
The synthesis also features a novel cyclization of a 1,2,5 triol system resulting in the
formation of a 2,5-disubstituted THF.

Our interest in the synthesis of variously substituted THFs stems from the recent
discovery of arachidonic acid tetrahydrofuran diols (AA-THF diols) -— a novel class of
secondary metabolites of AA. A total of 24 regio— and stereoisomeric THF diols can be
formed from arachidonic acid. The chemical synthesis of these metabolites was
undertaken in order to access them as single compounds for further biological studies.
Our method involves the acid promoted cyclization of epoxy diols containing directing
groups with different electronic properties. Depending upon the choice of the directing
groups and the acid promoter, several regio- and stereoisomeric THF diols could be
accessed from a common precursor. These studies are described in Chapter I.

Chapter II includes a survey of the structure, classification and biological activity

of annonaceous acetogenins. Representative total syntheses of several members of this

family of natural products are also discussed.

The later chapters discuss the application of our epoxy diol cyclization
methodology towards the total synthesis of mucoxin. Mucoxin, the first example of an
annonaceous acetogenin containing a hydroxylated THF ring, has shown highly potent
cytotoxic activity of against human tumor cell lines. The synthesis of the left hand
portion of mucoxin is described in Chapter III. The core THF diol unit was constructed
using a thiOphenyl directing group in the epoxy diol cyclization. Further, preliminary
studies on the coupling of the left and right hand fragments are also discussed.

The completion of the total synthesis is described in Chapter IV. The disubstituted
THF ring in mucoxin was constructed using a novel orthoester mediated cyclization of
1,2,n triols. The butenolide ring was introduced using the previously known thiophenyl
lactone, to complete the synthesis. However, the spectral data for the synthetic material
did not match that reported for the natural product. After analyzing data for both natural
as well as synthetic compounds, and conformational analysis using molecular mechanics,

we have proposed an alternative structure for the natural product.

To Sridhar

ACKNOWLEDGEMENTS

This dissertation would be incomplete without the mention of all those who
contributed to my successful graduate career. Babak has been an enthusiastic and fun
advisor to work with. Since I am his first graduate student, we both have gone through
five years of learning and teaching together. I thank him for being extremely kind and
supportive throughout. He always let me be independent and explore my own ideas. His
encouragement and confidence in me has helped me achieve goals that at times seemed
out of reach.

I would like to thank Profs. Maleczka, Wulff, and Weliky for serving on my
committee and their advise on several occasions. I also thank Prof. Tepe for his help with
my postdoctoral proposal and Prof. Hollingsworth for — among other things — a generous
gift of 2-deoxy-D—ribose.

All the past and present group members have been my good friends. I especially
thank Chryssoula and Courtney for their help and friendship. Jennifer has been a great
colleague and neighbor to work with. I enjoyed her company working late hours in the
lab and numerous interesting discussions we had. I also thank Qifei, Rachael, Montserrat,
Ben, Marina, Tao and Jun for the fun time we shared.

Dr. Daniel Holmes has been extremely helpful with the NMR spectroscopy. The
crucial NMR analysis toward the end would not have been possible without his assistance
and active interest.

I extend my very special thanks to Prof. William Roush at the University of

Michigan for his help on numerous occasions throughout my graduate career.

 

 

 

I am grateful for the support of my parents and the rest of the family, which has
been crucial to my success. Their love and faith has helped me overcome all hurdles and
pursue my goals tirelessly.

Finally, I thank my husband Sridhar for his constant encouragement every step of
the way. His suggestions and advice have helped a great deal. I truly appreciate his

unconditional love and support.

vi

t
V.
I
iii
‘H
.1.
-\
3x
.
i!
q
.
A
l
V
F
l
.~
.\
.
,
,
v|
.
.
q
Q
.
l
1

 

 

TABLE OF CONTENTS

List of Tables
List of Figures
List of Schemes

Key to Symbols and Abbreviations

CHAPTER I
Method development for the stereoselective synthesis of 2,3,5—trisubstituted
tetrahydrofurans
A. Introduction
1. Novel metabolites of arachidonic acid
2. Stereoselective synthesis of 2,3,5 trisubstituted THFS - a brief review
B. Regio- and stereoselective synthesis of 2,3,5 THFs via cyclization of methylene
interrupted epoxy diols
1. Method design
2. Background on regiocontrol in cyclization of epoxy alcohols
3. Method development
C. A novel method for the oxidative cleavage of olefins
D. Experimental
1. Experimental section for synthesis of 2,3,5 trisubstituted THFs
2. Experimental section for the oxidative cleavage of olefins
E. References

CHAPTER II
The annonaceous acetogenins: structure, biological activity and total syntheses
A. Historical background
B. Structure and prOposed biogenesis
C. Biological activity
I. In vitro Studies
2. In vivo studies
3. Activity against multidrug resistant (MDR) cells
4. Pesticidal activity
D. Mechanism of action
E. Structure—activity relationships
F. Classical vs. nonclassical acetogenins
G. Total synthesis of the annonaceous acetogenins
1. Multiple intramolecular Williamson etherification strategy
2. Epoxide cascade strategy
3. Biomimetic ‘naked carbon skeleton’ strategy
4. Step-growth oligomerization strategy

vii

xiv

xvi

67
67
68
71
72
73
73
74
75
77
80
8 l
84
88
89
9O

5. Sequential, modular strategy

91
6. Miscellaneous 94
H. References 100

CHAPTER III

Synthesis of the left hand fragment (C12-C34) of mucoxin and preliminary studies on its
coupling with the right hand fragment

105
A. Introduction 105
B. Retrosynthesis 108
C. Evaluation of the proposed intermolecular regio- and stereoselective epoxide
opening strategy 1 19
I. Design and synthesis of chiral allylic alcohol 111—3 119
a) Synthesis of a model allylic alcohol 145
b) Determination of the enantiomeric excess and the absolute configuration of
diol III-S9 . 148
2. Synthesis of vinylic epoxide III-4 152
3. Attemped intermolecular epoxide opening 157
D. Experimental section 172
E. References 21 1
CHAPTER [V
Total synthesis of the prOposed structure of mucoxin 219
A. Revised strategies for the coupling of left— (C13-C37) and right-hand (Cl—C12)
fragments of mucoxin 219

1. Evaluation of coupling strategies involving organozinc additions

222
2. Conventional organometallic addition using chelation control to couple the two

halves of mucoxin 229

B. Completion of the total synthesis of the proposed structure of mucoxin 244
C. Comparison of spectroscopic data and conclusions 253
D. Experimental section 280
E. References 319
APPENDIX 324

viii

LIST OF TABLES

Table H: 0304 — Oxone® mediated cleavage of complex olefins 38

Table II-l: Relative tumor growth inhibition (EDS0 mg / mL) for representative

acetogenins compared to adriamycin 72
Table II-2: LT50 values for German cockroach fifth instars 74
Table IIIoI: Optimization of the Sharpless asymmetric epoxidation of III-64 133
Table 111-2: Cyclization of 111-5 under various conditions 139
Table 111-3: Synthesis of model allylic alcohol 111-140 148
Table 111-4: Preliminary attempts at Optimization of the coupling of III-4 and 111-140 158
Table IV-l: Optimization of the chelation controlled addition 232
Table IV-2: Mosher’s ester analysis of IV-99 and IV-100 251
Table IV-3: Mosher’s ester analysis of IV-lOl and IV-102 253
Table IV-4: Mosher’s ester analysis of IV-105 and IV-106 255
Table IV-5: Comparison of 1H NMR chemical shifts of biS«THF portions (CS—C17) of

natural mucoxin vs. IV-117 261
Table IV-6: Comparison of 1H chemical shifts of bis-THF portions (C8-C17) of natural

mucoxin vs. IV-122 264

ix

 

LIST OF FIGURES

Figure I-I: Pathways of arachidonic acid metabolism 1
Figure 1-2: Pr0posed biosynthesis of AA-THF-diols 3
Figure 1-3: Regio- and stereoisomers of AA-THF—diols 5
Figure 1-4: Intramolecular oxymercuration strategy for the construction of 2,3,5
trisubstituted THFs 7
Figure I-5: Roush’s three component coupling approach to trisubstituted THFs 8
Figure I-6: Iodoetherification of alkene diols to stereoselectively access hydroxylated
THFs 8
Figure 1-7: Intramolecular iodoetherification of C6 allyl pyranosides used by Mootoo 9
Figure 1-8: Sugimura’s B-silyl cation cyclization tactic 9
Figure I-9: Cyclization pathways of methylene interrupted epoxydiol 1-29 11
Figure l-lO: Synthetic scheme to access enantiopure epoxy diols 11
Figure [—1 1: Conventional Baldwin vs. Warren’s hybrid nomenclature for epoxide ring
opening 12
Figure 1-12: The first report of epoxy alcohol cyclization to construct THF ring by Kishi
13
Figure I-l3: Nicolaou’s strategy for endo over exo selectivity in epoxide ring opening 14
Figure 1—14: Cyclizations of trans vinylic epoxides 14
Figure I-15: Cyclizations of cis vinylic epoxides 15
Figure I-l6: Hirama’s viz-allyl palladium cyclization strategy 16
Figure I-17: Use of catalytic antibodies to achieve endo selective epoxide opening 16
Figure I-18: Mukai’s alkynyl epoxide cyclization via cobalt complexation 17
Figure I-19: Proposed in situ deprotection - cyclization of epoxy diol 13
Figure [-202 Cyclic ethers derived from epoxy sulfide I-100 via episulfonium
intermediate

28
Figure 1-21: Warren’s phenylthio polyol cyclization strategy for synthesis of THFs and

THPS

29

Figure 1222 Possible equilibration between activated epoxy sulfide [-100 and the

corresponding episulfonium ion 33
Figure 1-23: Isomeric THF diols available from a common precursor 1-93 34
Figure 1242 Oxidative cyclization of linoleic acid to produce THF diols 36
Figure “-1: Uvaricin - the first acetogenin isolated from U varia accuminata

(Annonaceae) 68
Figure “-2: Generic structure of a binuclear acetogenin 68
Figure 113: Classification and representative structures of acetogenins 69
Figure II-4: Proposed biosynthetic pathways for two main classes of acetogenins 71
Figure II-S: Some acetogenins that showed high in vivo cytotoxicity profiles ‘73
Figure II-6: Annonin I 75

Figure II-7: Model of bis—THF acetogenins interacting with complex 1 in mitochondrial
membrane (Ref. 35) ' 76
Figure II-8: NADH-oxidase inhibitory potencies of bullatacin analogs 79

Figure "-9: The first total synthesis of an acetogenin, (+)—(36—epi)-ent-uvaricin 82
Figure “-10: Trost’s synthesis of (+)-squamocin K (key retrosynthetic disconnections) 84
Figure “-1 1: Trost's synthesis of (+) - squamocin K 85
Figure "—12: Marshall's stereoselective 8152' addition approach to oxygenated THF

precursors 86
Figure ”-13: Marshall’s synthesis of bullanin 87
Figure II-I4: Hoye's synthesis of (+)-parviflorin 88
Figure II-IS: Syn and anti oxidative cyclizations of hydroxy olefin 89
Figure II—16: McDonald's biomimetic oxidative cyclization strategy 90
Figure II-l7: Casiraghi's iterative vinologous aldol reaction strategy 91
Figure II-18: Proposed mechanisms for metal mediated oxidative cyclization of hydroxy

olefins 92
Figure II-19: Sinha and Keinan’s library synthesis of bis - THF core units 93
Figure 1120: Koert's modular strategy to construct bis - and tris - THF system 94
Figure 11-21: Jacobsen's synthesis of muconin 95
Figure II—22: Tanaka's stereodivergent strategy for construction of adjacent bis-THF

systems 97
Figure II-23: Evans‘ synthesis of mucocin 98
Figure III-1: Mucoxin 105
Figure III-2: Mucoxin: retrosynthetic analysis 108
Figure III-3: Grubbs’ tandem olefin metathesis - hydrogenation protocol 109
Figure III-4: Common tactics used for regiocontrol in intermolecular epoxide opening

reactions 111
Figure III-5: Sharpless’ protocol for C3 selective epoxide ring opening of 2,3 epoxy

alcohols 112
Figure III-6: Hirama’s conditions for regio-and stereoselective addition of aromatic

alcohols to highly functionalized vinyl epoxides 1 13
Figure III-7: Trost’s strategy for 1,2 addition of alcohols to vinylic epoxides 1 14

Figure III-8: Trialkyl stannanes proved inefficient as electrophiles in Trost’s studies 114
Figure III—9: Trost’s two-component catalyst system for asymmetric allylic alkylation of

alcohols 115
Figure III-IO: A representative example of regio-and stereoselective ring opening of
sugar derived oxiranes l 15
Figure 111-1 1: Mioskowski’s conditions for stereoselective SN2 addition of alcohols to
vinyl epoxides 1 16
Figure III-12: Lautens' protocol for 8N2 substitution of vinylic epoxides by alcohols
under mild conditions 116
Figure III-l3: Jacobsen’s strategy to construct the THF ring of muconin 118

Figure III-14: Isomeric THF diols available from a common epoxy diol precursor 119
Figure III—15: Stereochemical similarities and differences between the target THF unit

111-3 and an available precursor III-56 120
Figure III—16: A route to transform III-57 to the target allylic alcohol III-3 120
Figure III-l7: Proposed synthesis of the left hand (Cl3-C34) fragment of mucoxin 121
Figure III-18: Schlosser’s B—oxido ylide route to trans alkenols 124
Figure III-l9: Curran’s self-oxidizing protecting group 129

xi

 

Figure III-20: An endo selective epoxide opening of Ill-108 to generate 3-hydroxylated

trisubstituted THF 111-109 134
Figure III-21: Cyclization of an epoxy sulfide derived from 2-deoxy-D-ribose (Chapter 1)

via episulfonium ion formation 135
Figure III-22: Stereoisomeric THF diols originating from trans alcohol III-64 136
Figure III-23: Cis-vinylic epoxide may exhibit reduced endo-selectivity during

intramolecular cyclization reaction 137
Figure III-24: Payne like equilibration of epoxy sulfide III-121 under acidic conditions

139

Figure III-25: Rayner’s conditions for intermolecular trapping of episulfonium ions 140
Figure III-26: Possible route for cyclization of epoxy sulfides under acidic conditions;

endo / exo notation is relative to epoxide. 141
Figure III-27: Comparison of structures of epoxy sulfides III-S and 111-29 142
Figure III-28: 1,2 vs. 1,3 Chelation control in addition of vinyl magnesium bromide to

aldehyde 111-139 147
Figure III-29: Mnemonic device for Sharpless asymmetric dihydroxylation reaction as

applied to trans olefin III- 102 149

Figure III-30: A positively helical system comprises of two interacting chromophores
twisted in a clockwise direction going from the front to the back chromophore 150

Figure III-31: ECCD Spectrum of III-146 in MeCN 152
Figure III-32: Design of an epoxy sulfide substrate for regioselective ring opening by
alcohols 163
Figure III-33: Regio-and stereoselective alkyl group transfer to epoxy sulfides 165
Figure III-34: Cyclic sulfates and sulfites as epoxide surrogates 168
Figure III-35: 5N2 displacement of allylic electrophiles with alkoxides 169

Figure IV-l: Original regio- and stereoselective intermolecular epoxide opening strategy

219
Figure IV—2: General representation of the revised strategy 220
Figure IV—3: Design of the new synthetic strategy 222
Figure IV-4: Chelation controlled vs. Felkin-Anh transition state for reduction of ketone
IV-18 227
Figure IV-S: Revised stepwise strategy to assemble fragments IV-8, IV-32 and IV-35
230
Figure IV-6: Sharpless’ mechanism for vanadium catalyzed epoxidation of allylic
alcohols 235
Figure IV-7: Kishi’s transition state analysis to explain the diastereoselectivity observed
in directed epoxidation of bis-homoallylic alcohols 236

Figure IV-8: Application of Kishi’s T.S. models to bis-homoallylic alcohol IV-50 237
Figure IV-9: Representative examples of Shi asymmetric epoxidation of cis olefins 233
Figure IV-lO: Proposed radical intermediate during oxygen transfer step in Jacobsen

epoxidation 239
Figure IV-l 1: Sharpless’ protocol for stereospecific conversion of vicinal diols into

epoxides 240
Figure IV-12: Proposed one pot cyclization of triols (IV-76) to the corresponding cyclic

hydroxy ethers (IV-78) 241

xii

 

Figure IV-13: Assembly of the real aldehyde (IV-86) and partially functionalized right

hand piece IV-87 244
Figure IV-14: Empirical mnemonic device for the asymmetric dihydroxylation reaction
246
Figure IV-15: Application of the asymmetric dihydroxylation mnemonic to olefin IV-85
247
Figure IV-l6: nOe correlations in IV-101 and IV-105 containing trans and cis di-

substituted THF rings respectively 254
Figure IV-l7: Mucoxin: synthetic and originally proposed structures 258
Figure IV-18: Comparison of partial 1H NMR spectra of the natural mucoxin and IV-117

260
Figure IV-19: Comparison of partial 1H NMR Spectra of natural mucoxin and IV-122

263
Figure 1V-20: HRMS fragmentation pattern of the tris-TMS derivative of mucoxin.

(* = observed peak) 265
Figure IV-21: nOe correlations in the two synthetic diastereomers 265
Figure IV—22: Intramolecular hydrogen bonding in mucoxin as proposed by McLaughlin

266
Figure IV-23: Truncated stereoisomeric bis-THF analogs of proposed structure of

mucoxin 267
Figure IV-24: Low energy conformations of cis—threo isomer IV-123 268
Figure IV-25: Karplus equation plot for vicinal oxygenated systems 269
Figure IV-26: Low energy conformations of cis-erythro isomer IV-124 271
Figure IV-27: Low energy conformations of trans-threo isomer IV-125 273
Figure IV-28: Low energy conformations of trans—erythro isomer IV-126 274
Figure IV—29: Jimenezin: proposed structure (IV-125) vs. real structure (IV-126) 275
Figure IV-30: Possible alternative structure of mucoxin 276
Figure IV-31: Synthesis of hydroxy THF (C12-C34) portion and its union with iodide

IV-87 via chelation controlled addition 277
Figure IV—32: Completion of the total synthesis 278
Figure IVH33: Summary of structure proof of synthetic material (IV-117) 279

xiii

 

LIST OF SCHEMES

Scheme I-l: Spontaneous S-exo cyclization of free epoxy diol 19
Scheme 1-2: Synthesis of acetonide protected epoxy diols 19
Scheme 1-3: Various acids screened for deprotection — cyclization of [-89 and 1-91 20
Scheme 14: Synthesis of silyl protected epoxy diols 21
Scheme I-5: Preparation of epoxy diols with different pendant groups 23
Scheme 16: Acid catalyzed cyclization of epoxy alcohols [-94 and 1-95 24
Scheme I—7: Cyclization of epoxy diols containing electron withdrawing and neutral
pendant groups 25
Scheme 1-8: Cyclization of vinylic epoxy diol 26
Scheme [9: Absence of equilibration between vinyl THF [-109 and THP [-110 under the
cyclization conditions 27
Scheme I-lO: Epoxy sulfide cyclization 30

Scheme I-1l: Cyclization of [-100 in polar and nonpolar media using different acids 31
Scheme I-12: Absence of equilibration between phenylthio THFS I-123 and [-124

(products prior to acetylation) under the cyclization conditions 32
Scheme 1-13: Cyclization of diastereomeric epoxy sulfide [-125 33
Scheme H4: The 0504 — Oxone® method for the oxidative cleavage of olefins 37

Scheme I-15: Plausible mechanism of 0304 - Oxone® mediated cleavage of olefins 39

Scheme III-1: Proposed intermolecular epoxide opening strategy 118
Scheme III-2: Alkyne zipper reaction strategy 123
Scheme III-3: Propargylic ester strategy 123

Scheme III-4: Application of Schlosser’s method to synthesize trans alcohol 111-62 125
Scheme III-5: Iodide alkynylation route 127

Scheme III-6: Synthesis of trans homoallylic alcohol III-62

128

Scheme III-7: Attempted use of Curran’s self—oxidizing protecting groups in our system
129
Scheme III—8: Synthesis of the differentially protected triol III-104 131
Scheme III-9: Selective deprotection of the PMB group in 111-104 131
Scheme III-10: Synthesis of allylic alcohol III-64 132
Scheme III-11: Use of the Hata reagent to install the thiophenyl pendant group 134

Scheme III-12: BF3-0Et2 mediated cyclization of the epoxy sulfide III-5 using previously
optimized conditions 138

Scheme III-13: Cyclization of three different epoxy sulfides under the same conditions

142

Scheme III—14: Another attempt to improve the endo selectivity in the cyclization of 111-5
143
Scheme III-15: Preparation of the aldehyde 111-135 144
Scheme III-16: Synthesis of a model aldehyde III-139 146
Scheme III-l7: Synthesis of dibenzoate derivatives of the diol 111-103 for ECCD analysis

151
Scheme III-18: preparation of the three component coupling partners, 111-6 and 111-8 153

xiv

 

Scheme III-19: Synthesis of bromomethylacrylic acid 111-149 154

Scheme III-20: Synthesis of vinylic epoxide 111-4 156
Scheme III-21: A trial intermolecular ring opening of the vinylic epoxide III-4 using
Mioskowski’s conditions 157

Scheme III-22: Synthesis of simplified model vinylic epoxides and an allylic alcohol 159
Scheme III-23: Further optimization studies on the ring opening using model systems 160

Scheme III-24: Application of Lautens’ conditions to model systems 160
Scheme III-25: Screening of various acid catalysts for 8N2 opening of the model epoxide
161

Scheme III-26: Attempted epoxide opening reactions using a tributyl tin ether 163
Scheme III-27: Synthesis and acid catalyzed intermolecular coupling reaction of an

epoxy sulfide with an alcohol nucleophile 164
Scheme III-28: Attempted preparation and reaction of a trialkoxy aluminum with the

epoxy sulfide III- 189 166
Scheme III-29: Attempted alkoxy group transfer to the epoxy sulfide III-189 167

Scheme III-30: Attempted preparation and ring opening of cyclic sulfates and sulfites 168
Scheme III-31: Attempted preparation and displacement reactions of allylic triflate and

tosylate 170
Scheme IV -1: Synthesis of the model iodide 223
Scheme IV-2: Attempted organozinc additions to aldehyde lV-8 224
Scheme IV-3: Synthesis of ketone IV-18 via organozinc addition to acid chloride IV-l7

225
Scheme IV-4: Attempted addition of epoxy iodide IV-20 to acid chloride IV-l7 via the

the organozinc reagent 226
Scheme IV-S: Attempted hydride reduction reactions of ketone IV-18 223
Scheme IV-6: Model studies on HWE olefination approach 228
Scheme IV-7: Synthesis of the requisite homoallylic halides 230
Scheme IV-8: Synthesis of bis-homoallylic alcohol IV-50 233
Scheme IV-9: Feasibility studies of the new strategy described in Figure IV—5 234
Scheme IV-IO: One pot cyclization of a model triol IV-82 243
Scheme lV-ll: Synthesis of the real bis-homoallylic alcohol (IV-85) 245

Scheme IV-12: Application of triol cyclization method to the real system 247
Scheme IV-13: Chiral alcohols (IV-85 and IV-98) used in Mosher’s ester analysis 249
Scheme IV—l4: Synthesis of Mosher’s esters of IV-85

250

Scheme IV-15: Synthesis of or-SPh lactones IV-lll and IV-112 256
Scheme IV-16: Completion of the total synthesis of proposed structure of mucoxin

(IV-117) 257

Scheme IV—l7: Synthesis of C36 epimer of IV-117 25g

Scheme IV-18: Synthesis of (8,9-epi) IV-117 262

XV

AA

Ac
AcOH
acac

Bn
BOC-ON
BuzBOTf
CI

CSA

6

D

DDQ
DEAD
DET
DIAD
DIBAL-H
DIPT
DMAP
DMF
DMP
de

dr
ECCD
ec

EE
equiv.
13120
EtOAc
EtOH

0
D

h

HMPA
HRMS
Hz

Im

lPr

IR

J
KHMDS
L

KEY TO SYMBOLS AND ABBREVIATIONS

arachidonic acid

acetyl

acetic acid

acetoacetate

benzyl
2-(tert-butoxycarbonyloxyimino)-2-phenylacetonitrile
dibutylboron trifluoromethanesulfonate
chemical ionization

camphorsulfonic acid

chemical shift (parts per million)

dextro (denotes configurational relationship with
(R)-(+)—g1yceraldehyde)
2,3-dichloro—5,6—dicyano-1,4-benzoquinone
diethyl azodicarboxylate

diethyl tartrate

diisopropyl azodicarboxylate
diisobutylaluminum hydride

diisoprOpyl tartrate
4-(dimethylamino)pyridine
dimethylformamide

Dess—Martin periodinane (oxidation reagent)
diastereomeric excess

diastereomeric ratio

exciton coupled circular dichroism
enantiomeric excess

l-ethoxyethyl

equivalent(s)

diethyl ether

ethyl acetate

ethanol

gram(s)

hour

hexamethylphosphoramide

high resolution mass spectrometry

Hertz

imidazole

isopropyl

infrared spectrum

coupling constant

potassium hexamethyldisilylazide

levo (denotes configurational relationship with

(S)—(—)-glyceraldehyde)

xvi

 

LAH
LDA
M
mCPBA
MeCN
MeOH
Ms

MS

m/z
NBS
NCS
NMR
OAc
0"”
FCC
PMB
PPT S
pTSA
R

S

SAD
SAE
TBAF
TBAI
TBDPS
TBS
TBHP
TES
TFA
TBSOTf
THF
THP
TMS
TMSOTf
TPAP
TsOH

lithium aluminum hydride

lithium diisopropylamide

molar (concentration)
4-chloroperbenzoic acid

acetonitrile

methanol

methane sulfonate

mass spectrometry

mass to charge ratio
N—bromosuccinamide
N—chlorosuccinamide

nuclear magnetic resonance

acetate

trifluoromethanesulfonate
pyridinium chlorochromate
para—methoxybenzyl

pyridinium p-toluenesulfonic acid
para-toluenesulfonic acid

rectus (Cahn-IngoId-Prelog system)
sinister (Cahn-Ingold-Prelog system)
Sharpless asymmetric dihydroxylation
Sharpless asymmetric epoxidation
tetrabutylammonium fluoride
tetrabutylammonium iodide
t—butyldiphenylsilyl
t-butyldimethylsilyl

t-butyl hydroperoxide

triethylsilyl

trifluoroacetic acid
t-butyldimethylsilyl trifluoromethanesulfonate
tetrahydrofuran

tetrahydropyran

trimethylsilyl

trimethylsilyl trifluoromethanesulfonate
tetrapropylammonium perruthenate
para-toluenesulfonic acid

xvii

["71
11.:

 

 

 

CHAPTER I

METHOD DEVELOPMENT FOR THE STEREOSELECTIVE SYNTHESIS OF

2,3,5-TRISUBSTITUTED TETRAHYDROF URAN S

A. Introduction

1. Novel metabolites of arachidonic acid

Arachidonic acid (AA) is a C20 polyunsaturated fatty acid found in

phosphatidylinositol and other phospholipids as a C2 ester of glycerol. AA is stored in a

O

//

LTA4
(leukotriene)

Prostacyclin
synthase

HQ 0

"")\/\/\COOH
\~ /

HO
OH

6-exo PGF,”
(prostacyclin)

Figure [-1

B
lipoxygenasrx 6 5

COOH

l-1, arachidonic acid
A j cyclooxygenase

Cl)” ”“WCOOH
O- /

OH
PGH2

o
HOOCW
OH OH

PGE2
(prostaglandin)

0

5,6 EET

P-450 epoxygenase

Thromboxane
synthase

OH
..i~\:/\/\/COO
/O\/\/\A H
HO / ,
OH

TX82
(thromboxane)

: Pathways of arachidonic acid metabolism

 

variety of cell membranes and as a response to physiological or pathological stimuli, is
released into the cells by hydrolytic cleavage of phospholipids.1 Once liberated,
depending on the parent cell type, AA is metabolized via one of the three pathways (A, B
or C, Figure I-I). Each pathway involves a class of enzymes that oxidatively metabolize
AA.2 Cyclooxygenases (path A) and lipoxygenases (path B) are responsible for formation
of prostanoids (which include prostacyclins, prostaglandins, and thromboxanes) and
leukotrienes, respectively. Each class of metabolites is comprised of a large number of
compounds (collectively called as eicosanoids) with great diversity of structures and
functions.3 Figure I-l shows only a representative structure of each class. In fact, in
humans, AA is the most important precursor of prostaglandins and related secondary

metabolites. Eicosanoids have profound physiological effects including the onset of pain

and fever, regulation of blood pressure and blood clotting, control of sleep/wake cycle

and inflammatory response.4'6 Due to this, a large body of pharmacological research has

targeted the enzymes and receptors involved in AA metabolism—7’10

Of all the AA metabolic pathways, P—450 epoxygenase route (path C, Figure [-1)
is least scrutinized. Known metabolites along this path include regioisomeric AA
monoepoxides (such as 5, 6 EET, Figure I-1) and the corresponding diols formed by the
action of epoxide hydrolases.“'13 Though less explored, these metabolites have also been
shown to possess important biological activities. 5,6 EET, for example, is a potent

stimulator of prolactin release and an effective vasodilator.”15 11,12 DHET ~ 11,12 diol

of AA (not shown) is Na“/K+ ATPase inhibitor.13

 

Recently, a novel class of AA metabolites —- termed as arachidonic acid
tertahydrofuran diols (AA-THF—diols) has been discovered.16 It has been proposed that
AA-THF—diols, (box in Figure [-2) are formed along the P-450 epoxygenase path as
depicted in Figure [-2.17 Since monoepoxides and the corresponding diols of AA are well

precedented, it is conceivable that diepoxides (and even higher order epoxides) and their
hydrolyzed products may be formed via the same metabolic path. Accordingly,

Moghaddam et al. found that when monoepoxides of AA (I-2 and [-3, Figure I-2) were

 

exposed to clofibrate‘ treated mouse liver microsomes, mixtures of regioisomeric
diepoxides (I-4) were generated.17 Also, treatment of synthetically prepared

regioisomeric diepoxides of AA with the microsomes resulted in formation of the

corresponding AA-THF diols via cyclization of adjacent diepoxides (I-7 and [-8). These

 

 

 

 

 

P-45 0 0 [01 O O
H 0 92/\/=\,R1 + R2\/:\/L\/R‘ ——> Fig/W91
epoxygenase l-2 1-3 1.4
epoxide
epoxide hydrolase
hydrolase
HO OH O 7
RZMRI
Ho OH HO OH l-‘I
nay—{flm + 92mm —> o Ho OH
l-5 1-6 PM“
L I-8 J
OH OH
Hz 0 O R,
5 2 R1 + R22 5
3OH Ho3

AA-TH F -diOlS
R1 = CQH4COQH R1 = 05H8002H R1 = CgH12C02H
R2 = C9H14CH3 R2 = CsH1OCH3 R2 = C3H5CH3

 

 

 

Figure [-21 Proposed biosynthesis of AA—THF-diols

 

 

' Clofibrate is an inducer of P-450 epoxygenase and epoxide hydrolase.

3

_—___

 

novel structures generated in vitro, were identified by comparison of their GC/MS
fragmentation with that of synthetic samples AA-THF diols prepared via mCPBA
epoxidation and subsequent acidic hydrolysis of AA. Based on these in vitro studies, a
plausible biochemical route to AA-THF-diols was proposed (Figure 1-2).

Later, it was also shown that the proposed AA-THF-diols are biosynthesized in

viva.'6 Lipids isolated from liver extracts of clofibrate treated mice were derivatized to

their or-pentafluorobenzyl esters, which were then transformed to the corresponding TMS
ethers to facilitate GC/MS analysis. Comparison of the mass fragmentation of these
derivatives with similar derivatives of synthetically prepared AA-THF—diols confirmed
their presence in the liver extracts.

Our primary interest in AA-THF-diols stems from their interesting biological

activity.16 When rat pulmonary alveolar epithelial cells were incubated with

AA-THF-diols, a rapid increase in intracellular Ca+2 ion concentration was observed (as
detected by fluorescence measurements). This finding is significant in view of the crucial
role of intracellular Ca+2 ion levels in controlling physiological processes such as signal
transduction, protein phosphorylation and cell homeostasis. Interestingly, in the same
assays, AA did not show any detectable Ca”2 influx, while AA-diepoxides showed a
limited degree of potency, possibly due to their slow hydrolysis to AA-THF—diols. These
preliminary studies prompted us to initiate a program to further investigate biological
activity of AA-THF-diols, conduct SAR studies and delineate their precise mode of
action at the molecular level.

The primary bioassays (via’e supra) were carried out using reigo- and

stereoisomeric mixtures of AA—THF—diols. As shown in Figure 1-2, from three pairs of

4

adjacent diepoxides of AA, six regioisomeric AA-THF-diols would be produced. Since
the starting epoxides are cis, only two configurations about the THF ring, namely, all-cis
and 2,3-cis-5-trans are possible. Taken together, twenty four different regio- and
streoisomers of AA-THF-diols can exist (Figure IV-3, enantiomers not shown). Our
proposed biological studies in this area, required access to these THF diols as regio- and

l stereodefined Single compounds. During earlier studies,17 it was found that isomeric

1 mixture AA-THF-diols (obtained via epoxidation and subsequent acid catalyzed
cyclization of AA) could be separated only into two fractions, viz., all—cis and 2,3 cis-S-
trans stereoisomers (Figure IV-3). The separation was carried out using HPLC and the

fractions were not amenable to any further purification. We therefore decided to access

OH OH
_ __ _ _ ' ,O
COgH WCO2H
OH OH
OH OH
OH OH
OH oH
__ o _ ’ .0 _
COzH WCOZH
OH OH
OH OH
HOQC _. O __ HOZCW
OH OH
OH OH
O * ,.o _ _
__ COZH WCOQH
OH OH
OH OH
OH OH
all-cis 2,3-cis-5-trans

 

Figure I-3: Regio- and stereoisomers of AA-THF-diols

regio- and stereoisomerically pure compounds by way of chemical synthesis.

2. Stereoselective synthesis of 2,3,5 trisubstituted THFs — a brief review

Total synthesis of AA-THF—diols would also allow access to structurally diverse

OH analogs such as unnatural stereoisomers or variants containing modified

, aliphatic appendages to facilitate SAR studies. We felt that a straightforward

way to exercise regiocontrol in the total synthesis of AA-THF—diols would

be to first construct the THF diol core represented by general description [-9. The

functional group handles (X and Y) would then be elaborated to install the desired side

chains. In this way, unnatural analogs containing modified side chains would be easily

accessed. The THF-diol fragments of type 1-9 when constructed in enantiopure forms

should lead to the corresponding AA-THF-diols and / or analogs in regio- and
stereodefined manner.

Thus, attention was focused on Stereoselective synthesis of the trisubstituted THF-
diol intermediates. In order to introduce stereodiversity in the synthesis, we were looking
for a versatile route that will allow access to all possible stereoisomers of 1-9 in a quick
and efficient manner. Stereoselective synthesis of 2,5 disubstituted THFs is an
extensively studied area due to their presence in polyether antibiotics, annonaceous
acetogenins and other medicinally and biologically relevant natural products containing

such THF moietiesm'21 Trisubstituted THFs, on the other hand are relatively less

explored motifs. Methods for stereoselective construction of 3-hydroxy-2,3,5 substituted
THFs have appeared in the last few years. Representative syntheses of such trisubstituted

THF are described below.

Landais and co-workers used B-hydroxyhomoallylicsilanes (I-10, Figure 1—4) for
mercury mediated electrophilic cyclization to construct 2,3,5 trisubstituted THFs in good

diastereoselectivities.22 The stereocontrol in the ring closure step arose from the

preferential equatorial disposition of the silicon substituent in the chair like transition

RMeZSi RMeZSi

1. Hg(OAC)2 lCa003
W _> 0 Br

OH 2. KBI', 800/0

 

lV-10 dr = 92 : 8 IV-12
i OH
RMeZSi ) .
.E,
lV-11

Figure 1-4: Intramolecular oxymercuration strategy for the construction of 2,3,5

trisubstituted THFs

state (I-ll). Stereospecific conversion of the C-Si bond to the CO bond allowed access
to the corresponding all-cis hydroxytetrahydrofuran.

Roush has developed a highly convergent three component coupling strategy for
stereoselective construction of 2,3,5 trisubstituted THFs via the net [3+2] cycloaddition
of allyl silanes with aldehydes (Figure 1.5).23 Chiral allylsilanes (1-13) obtained via
allylboration of the corresponding aldehyde (not shown) are treated with the second
aldehyde in presence of a Lewis acid to furnish trisubstituted THF units in high
diastereoselectivities. The THF product arises through trapping of the developing positive

charge on the silicon-bearing carbon by the aldehyde oxygen, concomitant with a 1,2

silyl migration. In case of BF3°OEt2 coordinated aldehyde, the reaction proceeds via
synclinal transition state I-l4, in which steric interactions between R and BF3 are

minimized leading to the 2,5—cis THF I-lS. On the other hand, in presence of chelating

7

Lewis acids such as SnCl4, I- 16 1S proposed to be the lowest energy pathway producing

2,5-trans THF [-17 as the major diastereomer. Fleming—Tamao oxidation to access
hydroxy THFS was demonstrated on silyl substituted THFS (substrates similar to [-15 and

I-17) in the same report.

 

 

 

PDMBQSI
BnOCHZCHO R HS'Mezph
0 "31:3 ———>B w/Jxan
BF3-OEt2 53% 0%:
>12 :1 FEM TESO
SIMEPh l-1 5
Bno ; l'14
TESO PhMeZSi
l_13 H SiMeZPh
BnOCH2CHO RH:—-s{icun
O“
snc147 700/0 . noTESO OBn
> 20 1 1
MS l-17

Figure 1-5: Roush’s three component coupling approach to trisubstituted THFS

The cyclization of alkene diols such as [-18 (Figure [-6) by way of
iodoetherification has been reported by Guindon and coworkers as a general method to
prepare the corresponding 2,3,5 trisubstituted THFs (I-21) with complete
diastereoselectivity.24 In the cyclization of I-18, two transition states [-19 and I-20 were
invoked to explain the observed 2,3—trans selectivity (I-21). Alternative transition states

involving the opposite face of olefin (and thus leading to 2,5-cis isomer) are disfavored

H l

 

_ H H :_ Me
Me \ COZE. 12,NaH003 3791 Me o 2 ooze:
OH OH Me THF, 86% ”j 0025‘ 3
“I 1' Me OH
H 8 HQ l-20 l-21

Figure 1-6: lodoetherification of alkene diols to stereoselectively access hydroxylated

THFS

 

due to A 1,3-strain between the allylic hydroxyl and olefin methyl substituent. The

overall 2,3,5 stereochemical relationship depends upon the configuration of the

participating carbinol center.

Intramolecular iodoetherification approach was also used by Mootoo and co—

workers for cyclization of C6 allylated pyranoside substrates (I-22 and I-24, Figure
1-7).25 Ether ring closure is accompanied by pyranoside opening under the reaction
conditions. Diastereoselectivity of the cyclization was found to be dependent upon
configuration of the allylic carbinol center.

BnO OBn

 
  
 
  

 

 

H H
0 0013113 lDCP 1W0“)
BnO‘“ 83% BnO
l-22 1-23
(only cis)
BnOh‘ OBn
H b
O OCPHS lDCP 'AWCHO
, 89% 1..
Brio“ BnO
[-25
"24 (cis/ trans: 3 :2)

Figure I-7: Intramolecular iodoetherification of C6 allyl pyranosides used by Mootoo

In the total synthesis of (—-)-trans-kumausyne, the trisubstituted hydroxy THF

core (I-28, Figure [-8) was constructed via BF3-OEt2 promoted allylsilane addition to

substituted glyceraldehyde (I-26).26 Intermediate B-silyl cation (I-27) is trapped by

CH0 1 ,, SiMe3 1" O,
R \(i- ”810493 R ' J 1 "
O O b 0 ~., SlM83
‘1-,,EF3-OEt2 0

O -.
73% 1 "' F1
1-28

l-26 l-27 (single diastereomer)

 

 

Figure I-8: Sugimura’s B-silyl cation cyclization tactic

9

 

ir'ir

mi 1

 

internal oxygen nucleophile resulting in thermodynamically more stable 2,5-trans THF
(I-28)

27-32

Although the above mentioned and other related methods afford 3—hydroxy-

2,3,5-trisubstituted THFS in good diastereoselectivities and yields, they suffer from lack
of versatility. In most strategies, the stereoselectivity is substrate derived rather than
reagent derived. Depending upon the chirality of existing stereocenter(s) in the substrate,
a specific diastereomer is obtained. Thus, an inherent limitation on these methods is the
inability to provide various stereoisomeric THFS starting from a common precursor.
Clearly, these methods were unsuitable to quickly access our requisite trisubstituted THF-

diols scaffolds in a stereodivergent manner.

B. Regio- and stereoselective synthesis of 2,3,5 THFS via cyclization of methylene

interrupted epoxy diols

1. Method design

Upon re-examination of the proposed biosynthesis of AA-THF-diols (Figure [-2),
we thought that cyclization of methylene interrupted epoxydiol systems such as [-29
(Figure [-9) would serve our purpose. Pathways a and b lead to the trisubstituted THFS
with desired relative disposition of hydroxyl groups while c would result in a THP ring
formation. Design of the epoxydiol (I-29 with elements to achieve regiocontrol in the
cyclization, should lead to regio-and stereoisomerically complementary THFS [-30 and

[-31‘ from a common precursor. Oxygenated stereocenters in the epoxydiol substrate

 

 

' Generation of THF [-30 involves inversion at C2 whereas that of THF-[~31 involves

inversion at C1. The hybrid exo / endo nomenclature is explained later in the same
section.

10

 

 

a
x o OH HO 50H
2 O
Ho HO 2 OH
b
l-30 (5-exo) l-29 l-31 (S-endo)

(5-ex0 / 6-endo)

1-32 0(6- endCO)
(6-exo / 7-endo)

Figure I-9: Cyclization pathways of methylene interrupted epoxydiol [-29

would be established using Sharpless asymmetric dihydroxylation33 and epoxidation34

protocols, which are known to be highly stereoselective, reliable and efficient methods to
oxidatively functionalized olefins. Moreover, such an approach would be highly versatile
since by appropriate choice of the chiral ligands and the olefin geometry all possible
stereoisomers of I-29 can be easily accessed. A route to synthesize the requisite
epoxydiols is outlined in Figure 1-10 ([-37 is a diol protected version of I-29). Thus, with
design for stereodefined synthesis of the epoxydiol portion of I-29 in hand, we needed to
devise appropriate control elements (for example, nature of protecting groups (P) or the

pendant group (Y) in I-37) to realize regioselectivity in the cyclization event.

H AD- mix-13 X)\l/\ll/H steps X,H/\/\/QH_____. L-( (-+) -DET
x/Vfig/
[0-35

OH O
l-33 1-34

OP 0 steps XWY
OH
X JY\I>\/
OP

1-36 10-37

Figure I-10: Synthetic scheme to access enantiOpure epoxy diols

11

Baldwin’s empirical rules for ring closure have served to explain and reliably

predict regioselectivites in cyclization reactions.”36 In case of Opening of three-

membered rings to form cyclic structures, (I-38), the rules lie between those for
tetrahedral and trigonal systems and the em mode is generally favored.
em Applying Baldwin’s rules to epoxydiol [-29 (Figure 1-11 (1eft)), path a being a
A

Y

x
\hd/ S-exo (I-33) closure is expected to be favored over path b involving a S-endo
e o

"33 cyclization. On the other hand, according to Warren’s modified hybrid
nomenclature, path b would be labeled as S—exo / 6—endo closure (Figure I-11 (right)).37

This terminology originates from viewing the ring closure from two different
perspectives. Ignoring the C4—O bond ([-41) the ring closure can be classified as 5-ex0
since the rupturing bond (CS-O) is exo to the incipient five-membered ring. However, if
C4-C6 bond is disregarded the cyclization (I-42) resembles a 6-ena’0 closure. Whether
this hybrid ring closure terminology is just a matter of semantics or it has actual effects

on the outcome of cyclization remains unclear from Warren’s studies. In the

OH l-29

|-4O

<9

OH
l-31 (S—endo)

 

 

l-41

 

01'

l

l-31

 

"29 5 i— m
r39 I-3o (5-exo) a ,x'0 JOSE/EV
OH OH x 2 1%] city

Roz“
_._.X
2

(S-exo / 6-endo)

 

L

Figure I-ll: Conventional Baldwin vs. Warren’s hybrid nomenclature for epoxide ring

opening

12

present discussion, the conventional Baldwin’s nomenclature is used for clarity.

2. Background on regiocontrol in cyclization of epoxy alcohols

Regioselective cyclization of epoxy alcohols has been extensively exploited for
construction of cyclic ethers widely found in biologically relevant natural products.”21
Application of this approach to obtain small (5-7 membered) cyclic ethers was first used
by Kishi in the total synthesis of lasalocid A (Figure I—12).38‘39 Basic hydrolysis of epoxy
acetate I-43 and treatment of the resultant epoxy alcohol with acetic acid afforded the
cyclized product [-44 via 5-ex0 mode. Interestingly, the desired product was actually
hydroxy THP ring (I-47), which is the disfavored 6-end0 ring closure product of 1-43.

Thus, the hydroxy THF ([-44) was isomerized to the hydroxy THP ([-47) via hydrolysis

of oxonium intermediate I-46.

 

 

steps
MeO
H O O
O 2 0 -"Et
. ,..Et 1. MsCl, Py o 03

o = o A ; ‘ _‘ o i o

Et H E 0H 2. aq. acetone H O gt 9 Et H E!

AQCO3, 65‘70 Et
MS 1-46 1-47

Figure I-12: The first report of epoxy alcohol cyclization to construct THF ring by Kishi

About a decade later, Nicolaou reported a strategy for activation of endo epoxide
ring opening pathway over the exo counterpart (Figure I-l3).40'41 By placement of a It
System next to the epoxide, incipient carbocation at the proximal epoxide carbon ([-49,
path a) is stabilized due to conjugation of the electron deficient orbital with the It orbitals.

The partial positive charge at the distal carbon ([-51, path b) on the other hand, enjoys no

13

such extra stabilization. Thus, the endo opening path (a) leading to THP I-50 is preferred
over the exo mode (b) leading to THF 1-52.
Accordingly, trans epoxide I-53 (Figure I-l4) containing a vinyl appendage

afforded the corresponding 6—end0 product 1-54 with complete regioselectivity and
o H
30 ”:5 ”5” $155 350
..50 HQ I-51I-52

Figure I-13: Nicolaou’s strategy for endo over exo selectivity in epoxide ring opening

excellent yields, whereas the trans alkyl epoxide exclusively produced the S—exo product
(1-55). Both cyclizations proceeded with complete stereochemical inversion at the
reacting carbon. In case of oxepane generation from trans vinyl epoxy alcohol 1-56, the
endo selectivity was slightly reduced. However, the selectivity could be improved by

using a chlorinated vinyl group, possibly due to better stabilization of the positive charge.

0.1 eq.

CSA ' H0 CSA Ho,
R HO CH Cl 9 HO CH Cl g o i O
2 2 2 2 R H 0HH
l-53 l-54 l-55 l-56 l-57 l-58
R = CHQCHQCOQMe o : 100 R = CH2=CH2 82 : 13
94% 75%
R = CH=CH2 100 : 0 R = (E)CH=CHCI 92 : 8
95% 75%

 

Figure l-14: Cyclizations of trans vinylic epoxides

This strategy however, was not successful in case of cis epoxides. C is~vinyl
epoxide I-59 (Figure I—lS) furnished the corresponding endo (1-60) and em (1-61)

Products with almost no selectivity (THP : THF = 44 : 56).41 A slight improvement in the

14

ratio was achieved again by using chlorinated vinylic substituent. For larger oxepane
rings, the selectivities further depleted. In case of unsubstituted vinyl appendage, oxepane
I-63 was obtained as a 1:1 mixture of cis and trans isomers (not shown). Thus, this

technique failed to rcgioselectively produce cis THPs and oxepanes.

 

 

95%/Sq Ho J“) 33.)?“ Ho
0O I] co 0&7
RHO CH2012 R H o 6HH RHO CH2C|2 R H 0 cm 0
l-59 l-60 l-61 I-62 l-63 1.54
R = CH2=CH2 44 . 56 R = CH723=ch2 50 : 50
95%
a = (5)53:ch 7e : 24 R = (5)351:ch 58 1 32

 

Figure I-15: Cyclizations of (‘is vinylic epoxides

After Nicolaou’s reports, several other strategies to achieve endo selectivity in
epoxide opening were published. Hirama, in 1990, developed palladium catalyzed
stereospecific cyclization of hydroxy epoxides (Figure 1-16).42 Trans (I-65) and cis (I-69)
epoxy silyl ethers afforded the corresponding cis and trans THPs (I-67 and I-72,
respectively) in excellent yields and stereoselectivity. It was proposed that TBAF
treatment of the starting epoxy silyl ether generates ammonium alkoxide species, which
is a good nucleophile in subsequent palladium catalyzed allylic etherification. Both,
generation of the mallyl palladium species as well as the ring closure involve complete

stereochemical inversion

15

 

COgEt

TBDPSO I 1.TBAF EK320 REL \ COgEt V0025:
o —- [If O;
o
2. Pd(PPh3)4 SUN @59
CHCI3; 90% 9
l-65 l-66 I-67 l-68

 

> 99 : 1 (l-67 : l-68)

 

EtOZC
TBDPSO I 1.TBAF H [53/63 @0028... VCOQEI
O N 8020 ‘~.’ +QOH
2. Pd(PPh3)4 Bu4N @C’e
CHCI3; 89% 99
l—69 l-70 l-71 l-72

2 : 98 (l-67 : l-68)
Figure I-l6: Hirama’s n-allyl palladium cyclization strategy

thus leading to observed diastereoselectivities.
Lerner and Janda demonstrated the utility of catalytic antibodies to facilitate
chemically disfavored transformations by achieving forbidden 6-end0 route in

intramolecular epoxide opening reactions (Figure I-l7).43 Trans epoxide (I-73) was

regioselectively cyclized to the THP (I-75) using monoclonal catalytic antibodies raised
against N—oxide I-76. The antigen (I-76) closely mimics the TS (I-74) along the 6-end0
ePoxide opening path and thus produced antibodies that facilitated organization of the
reaction geometry to prefer THP formation. Also, in the process racemic epoxide I-73
Was resolved producing one enantiopure hydroxy THP (I-75). This elegant technique,

however is substrate specific and thus cannot be used as a general method in organic

 

 

 

 

Ar 5_
0 HO 0“
6-endo ,‘ U f“: >
HO Ar 6* ‘0 Ar/\\“ 0 Ar
i-i l-76
l-73 l-74 I-75
(racemic) (T.S.) (chIral)

Figure I-l7: Use of catalytic antibodies to achieve endo selective epoxide opening

16

 

 

synthesis.

Mukai and co-workers developed C02(CO)8 mediated cyclization of alkynyl

epoxy alcohols to favor the 6-ena’0 Opening (Figure I-l8).44 The strategy involved initial

formation of a cobalt complex of the epoxy alkyne (I-78). The complexed epoxide in
presence of a Lewis acid underwent ring opening to produce the olefin intermediate (1-
79) via anchimeric assistance of the antiperiplanar C-Co bond. Attack of the hydroxyl
group onto the available face of the olefin led to the corresponding THP (1-80) with net

retention of configuration at the propargylic carbon.

 

 

 

 

 

 

pA
O
O 1.002(00)8
H OH\
0“ § A 2-BF3'OEt2 (OC)300— —Co(Co)3 (OC)acd- —CO(CO)3
SIM63 86o/o ®
SlMea SIM93
‘— _..i
|.77 I-78 ”9
H OH
0 CAN
0 /CO(CO)3 0 §
(OC)300/ SiMe3 SIMe3
l-8'l
"30 91 :9 (trans: cis)

Figure I-18: Mukai’s alkynyl epoxide cyclization via cobalt complexation

From the above discussion it may be stated that epoxide ring opening by an
internal hydroxy nucleophile usually prefers the era route, the selectivity however can be

Channeled along the endo pathway by use of vinylic or alkynyl directing groups.

3. Method development
T0 our knowledge, all studies in the context of regiocontrol in intramolecular epoxide

Opening have involved systems containing a single hydroxyl group available for

17

 

nucleophilic attack and hence only two competing (vide supra) paths in the cyclization
event. Our epoxy diol system I-29 (Figure I—9) presents an added level of complexity in
that there are two endo (b and c) and an em path (a) available.’ The S-em path being the
most preferred, should be easily accessible. On the other hand, even if the system is
designed to promote erzdo cyclization, the relative preference between S-endo and 6-emlo
processes would be hard to predict if both the hydroxyls are equally available for
cyclization. Thus, selectively accessing either of the two endo routes appeared
challenging due to their competition with each other in addition to the more preferred 5-

am pathway.

 

OH OH X C Y
OP ' X 0 2 S O Y 5 1
"37 I-30 (S-exo) I-31 (S-endo) l-32 (6-end0)

Figure I-19: Proposed in situ deprotection - cyclization of epoxy diol

From the outset, to avoid spontaneous cyclization of the free epoxy diol (vide
infra), we decided to synthesize protected epoxy diol systems (1-37, Figure I-l9)
containing suitable control elements (such as protecting group P and directing
functionality Y). The goal was to optimize conditions that would accomplish one pot diol

deprotection and regio- and stereoselective cyclization reactions.

 

 

Although, in principle, a 4—ex0 pathway is also possible, it is almost never encountered.

18

 

 

 

OH 0 OH OH
Ph3P=CHC02Et WV TBDPS-CI W
OH THF, 90 °C; 92% 0H DMF, rt OH
(E:Z=5:1) 72%
l-82 1-83 "34
a o
DEAL-H “fit/m 1. 319(5) B6gg/SH2CI2 Z
. TBDPSO , \ OH ' 4' TBDPSOV" o 082
532.0 C OH 2. 8201, pyridine 082
60 °C;85°/o
"85 I‘86

Scheme I-l: Spontaneous S-exo cyclization of free epoxy diol

 

Since the critical issue to be addressed was regiocontrol in the proposed
cyclization reactions, we decided to quickly access the requisite epoxy diol substrate
from commercially available Z-deoxy-D-ribose (1-82, Scheme I-l). Wittig olefination of

I-82 using (carbethoxymethylene)triphenylphosphorane afforded (1,3 unsaturated ester
I-83 in good (5 : 1) diastereoselectivity.45 After silyl protection of the primary hydroxyl
group (72%) and subsequent DIBAL-H reduction (95%) the corresponding triol (I-85)

was isolated as a single diastereomer. mCPBA epoxidation of 1-85 directly produced the

corresponding cyclized product via S-exo route as expected, which was characterized as

OH Me20(OMe)2 0 Z
CSA, 4 A MS W DIBAL-H
TBDpSO/\/K/\/CO2Et \ COZEI m

. > TBDPSO ,
. -/ Ergo, 0 °C
89%

 

OH acetone, r1 0
[.34 80% |-37

O o TsCI, EI3N
/\/'\/\/\ mCPBA, CH2CI2 AMA Me3N.HCI
TBDPSO - / \ OH Ar TBDPSO e./ 2 CH
C) 0.5M NaH003 O CHQCiQ, 0 00
n; 730/0 8970
I-88 I-89

 

 

M 70 PhSNa. DMF M 70
TBDPSO OT TBDPSO

53/ s 0 °C to rt 65/ 2 SP”
60%
l-QO l-91

 

Scheme I-2: Synthesis of acetonide protected epoxy diols

19

 

THF I-86 (1 : 1 mixture of isomers) after perbenzoylation.

Next, protected epoxy diols 1-89 and I-91 were examined in order to evaluate the
possibility of controlling the regioselectivity of cyclization. Based on simple molecular
models, it appeared that the C5 oxygen of acetonide (I-89 and 1-91) might be sterically
less hindered and hence more available for the nucleophilic attack. In that case, the
corresponding S-endo product would be obtained preferentially. Also due to neighboring
group participation of the phenylthio group' in I-91, C2 might be selectively activated
over C3 toward nucleophilic attack leading to endo cyclized product(s). The acetonides
were accessed by protection of the diol functionality prior to epoxidation However, all
attempted in situ acetonide cleavage — epoxide opening reactions of 1-89 and I-91
(Scheme 1-3) using various protic and Lewis acids promoters resulted in either

decomposition or recovery of the starting materials.

PTSA
H0104
0%?) CSA
TBDPSO/MY ——————-> no cyclized product
6 BF3'OEI2
l-89, v = OH BC'a
l-91, Y = SPh amberlyst
Ti(o‘Pr)4

Scheme I-3: Various acids screened for deprotection — cyclization of I-89 and I-91

We next turned to the more easily cleaved trimethylsilyl groups to protect the diol
functionality (Scheme I-4). Accordingly, the available diol I-84 was protected as bis-

TMS ether I-92. During the silylation reaction, it was critical to maintain a 1:1
stoichiometry of TMSCI and Et3N to avoid intramolecular Michael addition of the

hydroxyl group on to the (1,6 unsaturated ester to produce the corresponding THF ring.

 

 

‘ This phenomenon is discussed in more detail later in this section.

L

 

20

 

DIBAL-H reduction of I-92 afforded allylic alcohol L915 (90%). In order to simplify
analysis of cyclization products we decided to prepare diastereomerically pure epoxides

I-94 and 1-95 using the Sharpless asymmetric epoxidation.

OTMS

 

 

 

 

 

OH TMS-Cl, lmid DIBAL-H
TBDPSCMCOZE‘ TBDPSO , \ COZEI
i DMAP, THF GTMS E120, 0 OC
0“ 45 °C, 75% 90%
l-84 I-92
D-(—)-DET (5 eq.)
onus Ti(O‘Pr)4 (3.6 eq.) owns 0
reopso , \ OH 4' TBDPSO , OH
OTMS tBIJOOH, 4 A MS OTMS
[.93 CHQCiQ, -20 °C [-94
73°/o; > 98°/o d6
L-(—)-DET (5 eq.)
OTMS Ti(O'Pr)4 (3.6 eq.) OTMS o
TBDPSO , \ OH 4’ reopso , i i OH
OTMS tBIJOOH. 4 A MS OTMS
I-93 CH20I2. -20 °C l-95

55%; 84% de
Scheme 1-4: Synthesis of silyl protected epoxy diols

The SAE reaction proved tricky due to the acid sensitivity of the TMS protecting
groups in the substrate. When standard catalytic conditions46 (10 mol% Ti(O‘Pr)4, 12

mol% DET) were utilized, the epoxidation was not complete even after prolonged
reaction times (24 - 48 h). In addition, products arising from silyl deprotection were
observed, probably as a result of prolonged exposure to the Lewis acidic conditions. On
the other hand, when I-93 was treated with 1 equiv. of Ti(O‘Pr)4 and 1.2 equiv. of DIET,
the starting olefin was completely consumed within a few hours. Unfortunately, the yield
of the desired epoxide was only about 30%, and considerably larger amounts of silyl
deprotected products were recovered. After considerable optimization, we found that the

epoxidation could be efficiently promoted using super—stoichiometric quantities 0f

21

 

 

 

 

reagents (3.6 equiv. Ti(O‘Pr),, 5 equiv. DET).47 Under these conditions epoxide I-94 was

obtained as a single diastereomer in 73% yield (in case of D-(—-)-DET). We believe that
the short reaction time (2 h) was crucial in suppressing the silyl deprotection pathway that
plagued our earlier attempts. Under similar conditions, L-(+)-DET gave lower (55%)
yield of epoxide I-94, with a diastereomer ratio of 92 : 8.

Using silyl protected epoxy diol systems, we hoped to be able to control the
regioselectivity of cyclization by varying electronic properties epoxide pendant groups.
Accordingly, derivatives I-96 through I-100 were prepared via standard transformations
(Scheme I-S). Oxidation of epoxy alcohol I-94 using usual protocols such as Swern,
SO3°Py and Dess—Martin periodinane reactions afforded the desired aldehyde I-96 in low

(up to 40%) along with TMS cleaved by products. After some experimentation we found
that by buffering the DMP reaction with pyridine,48 the yield could be increased to 90%.

Aldehyde I-96 was treated with the ylide generated from methyltriphenylphosphonium
bromide to generate vinylic epoxide 1-97 in moderate yield. Subsequent catalytic

hydrogenation of I-97 provided straightforward access to alkyl substituted epoxide I-98.

22

 

 

 

 

6 0mg 0 1 WP. pyr OTMS 0 CH0 PhsPCHBBr
reopso :5 2 OH ' TBDPSO : *—_j—-———*
OTMS CHQCIQ, 90°70 OTMS BU‘EJ, Eigo
l-94 I-96 55“
reopso ; \ reopso ,
OTMS H2. EtOAC OTMS
l-97 60% 1-98
orms o M92804 CNS 0
TBDPSO , OH —————> reopso , OMe
OTMS LiHMDS, THF orMS
l'94 75°/o I-99
OTMS o Bu3P, Ph232 OTMS o
TBDPSO , OH reopso ._ sen
OTMS Et3N; 85% OTMS
l-94 l-‘IOO

Scheme I-S: Preparation of epoxy diols with different pendant groups

O—Methyl epoxy alcohol 1-99 was obtained in optimal yields by methylation of 1-94 with
LiHMDS / (CH3)ZSO4. Other conditions such as LiHMDS / CH3I, and Nail / CH3I lead
to side products arising from removal of TMS groups and subsequent O-methylation of
the secondary hydroxyl groups. Finally, thiophenylmethyl substituted epoxide I-100 was
accessed by treatment of epoxy alcohol 1-94 with the Hata reagentw‘s0

Since epoxide ring opening is usually more facile under acidic than basic conditions, we
examined acid mediated silyl deprotections of the epoxy diols. We anticipated that the
regioselectivity in cyclization of epoxy diol I-94 (Scheme I-6) would be dictated by

35.36

optimal alignment of the newly forming and rupturing bonds and destabilization of

the partial positive charge on C2 due to electron withdrawing hydroxyl pendant group.

Both the controlling factors would lead to nucleophilic attack on to C3. Exposure of 1-94

to BF3’OEt2Sl (Scheme I-6) cleanly produced THF I-101 as single diastereomer, which

23

 

was characterized by COSY experiments as the expected 5-ex0 product after

peracetylation to I-102. Also, lack of nOe correlations in I-102 across the THF ring
suggested trans relation between H3 and H6, in agreement with complete stereochemical
inversion at C3. The same results were obtained when deprotection-cyclization of I-94
was triggered by aqueous acetic acid.52 The diastereomeric epoxide (I-95) after similar

acid treatments (A and B, Scheme I—6) also efficiently afforded the corresponding S-exo
product with inversion of configuration at C3. Thus, the stereochemical relationship
between the diol and the epoxide was inconsequential to regio- and stereochemical
outcome of the cyclization reaction and two stereochemically complementary THF diols

(I-101 and I-103)' were accessed.

 

 

 

 

A OH OAc
:. ACQO, Py ?
OTMS O 8 o HO 0":3 2 OH R0 O~.,3 2 OAC
MK 6 /° 60 °C 88°/
90 6i 2 OH ‘—'1 €15 ' o 55
onus H0 AcO
|_94 B I-101 l-102
R = TBDPS 850/0 A BFa‘OEtQ (6 eq.), E120, 0 cc to n
BAcOH :HZO :THF (6 :3 : 1), O°Ctort
OTMS OH OAC
(O ACQO, Py
5 J _ 1
ROWOH LEE—p R0 0 3 2 OH 60 00 —= R0 0 3 2 OAC
ems .- ,.
HQ‘ 5 (68% two steps) Aco‘ 5
1-95 I-103 1-104

Scheme 1-6: Acid catalyzed cyclization of epoxy alcohols I-94 and [~95

Along similar lines, methoxy substituted epoxy diol I-99 produced the
corresponding S-exo product (I-105). In this case, however cyclization under protic

conditions was more efficient than using Lewis acid promoter (Scheme I-7). Substrate

uh

. Although I-101 and I-103 are tetraols, the primary hydroxyl groups are considered as functional
group handles. The cyclization products of all the epoxy diols under consideration would be
referred to as THF or THP diols.

24

 

 

I-96 was designed to obtain S-exo THF diol with a more versatile functional group handle
(an aldehyde appendage). Unfortunately, under both Lewis and Bronsted acidic
conditions, most of the starting material decomposed and only small amounts of the
desired S-exo product (I-106) were obtained. Interestingly, the em product, after
acetylation was isolated as bicyclic acetoxy acetal I-107 generated by intramolecular

addition of CS-OH to the aldehyde functionality.

or A QAC
/\)J\”"/S<<p/r\ 60% HO O.“ OMe Ai) BF3°OEt2 (6 eq.). Et20.0°C tort
m ii) A020, Py., 60 °c
Act):5

 

 

 

 

 

HO 6; OMe-fi
arms 2 Bi)ACOH1H202THF(6:3:1),0°Ctort
[-99 ‘ B l-105 ii) A020, Py., 60 OC
R = TBDPS 750/o
A
. OH
ows o r H
RO/MCHO — 17 A, ROWH —- 0' "'OAC
onus 2 :5 0 R0 0 OAC
8 HO H
R 1-96 .____.
= TBDPS o
20 /° l-106 ' I-107
A
OTMS "——’ QAc
OTMS 55
I 98 B AcO
" l-108
R = TBDPS 786/0

Scheme I-7: Cyclization of epoxy diols containing electron withdrawing and neutral

pendant groups

Alkyl substituted epoxide 1-98, would serve to evaluate Warren’s suggestion that
conventional intramolecular endo Opening of three—membered rings might have an exo

component that is likely to influence regioselectivity in their cyclization reactions (vide
supra).37 Since both epoxide carbons (I-98) are electronically similar, the major product

should be the result of most favored alignment of the forming and breaking bonds. As

shown in Scheme I—7, irrespective of the nature of acid promoter, 5~ex0 product I-108

25

 

 

0‘

 

 

was obtained as a single regio- and stereoisomer. Our results with I-98 clearly indicate
that the hybrid terminology has no particular advantage in terms of predictability of the
regiochemical outcome of the cyclization, at least in case of epoxides. Also, I-98
electronically resembles the epoxy diol intermediates in the proposed biosynthesis of
AA-THF—diols (I-7 and LS, Figure I-2). Complete 5-ex0 selectivity in cyclization of I-98
suggests that non-enzymatic in vivo cyclization of L7 and 1-8 likely proceeds exclusively

via S-exo path to generate regioisomeric the AA-THF—diols.

OAc

 

 

 

A ROW
82% 4., A i) BF3-OEt2 (6 eq.), Ergo, 0 °C to rt
OTMS 0 OAc Ii) Ac20, Py., 60 °C
R0 4 \___ l-109 Bi) ACOH2H20:THF(6:3:1),O°Ctort
75mg 3 ii) A020, Py., 60 °C
l7 R070 Ci)10%HCI:THF(9:1),O°Ctort
'9 \ ii) Acgo, Py., 60 °C
3 or c R — TBDPS
Aco“'6 4"0Ac ( ‘ )
80%
MW

Scheme 1-8: Cyclization of vinylic epoxy diol

Using vinyl epoxide I-97 (Scheme I-8), we hoped to direct the cyclization along
the endo pathway via stabilization of developing positive charge at C3.“41 In BF3°OEt2

promoted deprotection—cyclization, the S-erzdo product (I-109) was obtained as a single
diastereomer with complete regioselectivity. While this was initially attributed to faster
formation of a five~membered ring relative to a six—membered ring, curiously in polar
protic media the 6-endo product (I-110) was selectively obtained. Moreover, the strength
of protic acids did not affect the selectivity (B and C, Scheme I-8).

Semiempirical calculations (PM3 force field, Spartan V5. 1.3) indicated I-110 (6-
endo THP product prior to acetylation, Scheme I-9) to be slightly more stable (ca. 2 Kcal

/ mol) than 1-109 (S—endo THF product). This raised the possibility that I-109 is initially

26

produced in aqueous acidic conditions and is then isomerized to the THP(I-110). To test
this proposition, I-109 was exposed to 10% HCI in THF. However, no isomerization of

the THF to I-110 was observed and the former was quantitatively recovered. Similarly,

 

the THP (I-110) did not isomerize to the THF when treated with BF3OOEt2. Thus, clearly,

formation of both the rings under the reaction conditions was irreversible. These

experiments suggest that the cyclization reactions may be operative under kinetic rather

than thermodynamic control.

 

 

OH
ROW 10% HCI X _ ROW
TV ‘.
4"?) H A BF3°OEt2 HO‘ 5 4 OH
F109 i-110

Scheme I-9: Absence of equilibration between vinyl THF 1-109 and THP I-110 under the

cyclization conditions

Thus, it is likely that the TMS groups are cleaved at comparable rates (and hence
equally available for nucleophilic attack) and in a nonpolar medium, S-endo TS. is the
lower energy path while in a polar protic medium, the 6-end0 TS. is energetically
favored. Another possible scenario is that the two silyl groups are cleaved at different
rates in different media, i.e., under nonpolar conditions (A, Scheme I—8) C6-OTMS (I-97)
is cleaved faster and therefore is readily available for the ring closure as compared to
C7-OTMS and vice versa. Though factors controlling the regioselectivity are not clear at
present, we were nevertheless able to access THF diol 1-109, which is regio- and
stereochemically distinct from those prepared earlier (1-101 and 1-103 Scheme 1—6). In

addition THP diol 1-110 was also obtained in complete regio—and diastereoselectivity.

27

 

TMSO H

 

 

- O TMSO OH (LA)
7 H+ / LA 2 1
R0 6; 2 SPh —_——_’ R0 5; n’ 6.)
OTMS H OTMS H stph
l-100, R = TBDPS l-113
HO HO 6 0
R0 6 5 O.,,2 SPh FiO 6 Olga SPh go 6 5 o 1 R0 5 1
- ,w u, HO“. 2 SPh
"OH HO 5 OH ; 2 SPh ‘2
OH OH
l-114 I-115 I-116 l-117

Figure I-20: Cyclic ethers derived from epoxy sulfide I-100 via episulfonium

intermediate

We next turned to explore cyclization reactions of epoxy diol I-100 containing
phenylthio directing group (from now on referred to as epoxy sulfide). It was anticipated
that under acidic conditions, the epoxy sulfide would generate an episulfonium ion
(I-113, Figure I-20) via neighboring group participation of the phenyl thio groups}55 The
two potential sites for nucleophilic attack in I-113 are C1 and C2. Depending upon which
nucleophile (CS—OH or C6-OH) participates in the cyclization, four regioisomeric
products (I-114 through 1-117) can be generated. I-114 and I-115 are the S-endo and 6-
endo cyclization products with respect to the original epoxy sulfide I-100, and S-exo and
6~exo with respect to the episulfonium ion 1-113. For consistency in the discussion 1-114
and I-115 will be referred to as S-endo and 6—end0 products. THP 1-116 and oxepane
I-117 arise from cyclization reactions at C1, which are endo paths with respect to the
episufonium intermediate (1-113).

If conditions could be found that would favor the formation of I-114 and / or

1-115 over 1-116 and l or I-11'7, this strategy would constitute a novel tactic to achieve

28

 

 

endo over exo selectivity in epoxide opening reactions. An added advantage of this
technique is the stereochemical outcome of the cyclization. During the formation of I-114
and 1-115 from the epoxysulfide (I-100) via I-ll3, the configuration at C2 is retained as a
result of double inversion. Thus, stereochemical relation between C2 and C3 in the two
cyclic ethers is complementary to that if a single inversion were involved (for example by

using vinyl epoxide substrates). Moreover, thiophenylmethyl group in I-114 and I-115

56.57

can be easily transformed into an aldehyde functionality, which is a versatile

functional group handle.

 

 

 

F Ph fl
é, PhS
. OH 4 '—’ o
l 40 °C, 48 h HO I
OH OH 90% OH
L HO
F118 I-1 19 I-1 20

Figure 1-21: Warren’s phenylthio polyol cyclization strategy for synthesis of THFS and

THPs

Warren has used phenylthio polyol systems to construct THF and THP rings

through episulfonium intermediates.58 A representative example is shown in Figure
121.5%} Prolonged exposure of triol I-118 to acid at 40 °C afforded THF I-120 as a

single regioisomer. I-120 was shown to be the thermodynamic product and the THP
formed by attack of the primary hydroxyl group was converted to I-120 under the
reaction conditions. Intermediacy of an episufonium ion was clearly established due to
migration of the phenylthio group. Under the reaction conditions, attack at the more
substituted carbon of the episulfonium ion was usually observed. A similar method for

stereOSpecific synthesis of THFS using phenylthio diols has been developed by

29

 

1‘.
_ a
,. .
...
.
.
-
4
t
I
I
V.»
5‘.
..

 

 

 

 

 

 

 

 

 

Williams.62 However, the use of a phenylthio directing group in the intramolecular

opening of epoxides containing internal nucleOphiles in the context of heterocycle

synthesis is not known.

 

 

TMSO O AcO QAc
ROMSPh fl FiO 6 0.,2 SPh + ROW?”
OTMS 5.
OAc AcO‘
l-100, R = TBDPS I-121 I-122
conditions yie|d(%) product A i) BFg-OEtQ (6 eq.), Ergo, 0 °C to n
ii) Acgo, Py., so °c
A ‘55 °”'Y' ‘21 Bi) AcOH 2H20:THF(6:3:1).0°C tort
B 75 l-121 :l-122(30 :70) '3) A020 PM. 60 °0
C I) 10% HCI :THF (9 : 1), 0 °C to rt
0 74 I-121 : I-122 (3 :97) ii) A020, Py., 60 °c

 

Scheme l-10: Epoxy sulfide cyclization

Scheme l-lO depicts the results of the epoxy sulfide (I-100) cyclization under the

standard conditions (A, B and C) used earlier. Gratifyingly, the S-endo product l~121 was
obtained with using BF3 in CHZCI2 via nucleophilic attack at C2 was obtained by the
liberated C5 hydroxyl. The structure was determined by COSY analysis of the
peracetylated derivative (I-121, Scheme l-lO). Also, 1D NOESY experiments indicated a
cis relationship between H2 and H3, which confirmed a net retention of configuration at

C2. Therefore, it was concluded that I-121 is formed via the corresponding episulfonium
intermediate (1-113, Figure l-20).

However, a mixture of regioisomeric THFS was obtained upon treatment of l-lOO
with aqueous acetic acid and the products were characterized as {-121 and [~122 (30:70)

using the same NMR techniques. I-122 is the S-exo product originating from the parent

30

 

 

epoxy sulfide (by ring opening at C3) without the involvement of the episulfonium ion.
Surprisingly, stronger protic acid (10% HCl, C, Scheme l-10) almost completely reversed
the regioselectivity to afford the 5-ex0 T HF diol I-122 as the only isolated product.‘ Also
treatment of I-100 with PPTS in EtOH led to the formation of I-122 as the major product
(l-121 : I-122 = 5:95 as determined by GC). Aqueous PPTS mediated cyclization also

afforded I-122 as the major product, although the reaction was much slower (50%

 

 

complete in 24 h).
m conditions 90 AC0 5 o. 2 5 o. 9:0 SPh
”0 6, 2 SP“ ——-—-—* 6 " SPh + Fro/\Lv/VV
OTMS "’OAc A00?
l-100, R = TBDPS I-121 l-122
conditions yield(%) I-121 :l-122
H2804, EtOH 81 8 :92
pTSA, EtOH 83 6 :94
ACOH. EtOH 82 8 : 92
HF-Py, EtOH 95 9 : 91
PPTS, EtOH 73 3 : 97
pTSA, CHZCIZ 95 95 :5
TFA, CHZCIZ 89 9o : 10
HF'Py, CHQCIZ 80 78 :22

 

Scheme I-ll: Cyclization of 1-100 in polar and nonpolar media using different acids

Scheme l-ll shows data collected by Dr. Meenakshi Sivakumar.63 Simultaneous
deprotection - cyclization reaction of I-100 was carried out using various acids in polar
(EtOH) and nonpolar (CHZCIZ) solvents. The results indicate that irrespective of the acid

strength, the direct cyclization of the epoxy sulfide by S-exo path was favored in polar

 

 

' The 97 : 3 ratio of I-122 : 1-121 in C (Scheme l-lO) was determined by GC.
31

 

.\i. .
. .y. a
_ . ,.
v. A.
. .o .
u up! -|\i-
. y»

 

\ .

i .

 

medium whereas in a protic environment, cyclization of the episulfonium at C2 was the
preferred route.

Semiempirical calculations (Spartan 5.3.1, AMl force field) showed I-124
(Scheme I-12) to be lower in energy (ca. 2 Kcal / mol) than I-123. However, I-123 when
treated aqueous acids does not isomerize to I-124 thereby ruling out the possibility that
under polar protic conditions, the S-enda product was initially formed and then slowly

equilibrated to I-123. Similarly 1-123 in presence of BFfOEtZ did not convert to I-124.

These experiments indicated that the cyclization reactions are operative under kinetic

control as in the case of the vinyl epoxide .

 

0” aq.AcOH O 9:
. oo ., ' son
R0 6 0.,2 SPh or 101 HCI; ROW
-. ‘ /\ 52
’OH BF3'OEi2 HO
F123 I-124

Scheme I-12: Absence of equilibration between phenylthio THFs 1-123 and I-124

(products prior to acetylation) under the cyclization conditions

One possible explanation to the preferential formation of the S—em product in
polar protic media is that the episulfonium formation is prevented due to solvation of the
Phenylthio group. It is conceivable that the lone pair on sulfur atom is engaged in
hydrogen bonding with the protic solvent and hence is not available for anchimeric
assistance to generate the episulfonium ion. The extent of such hydrogen bonding and
hence extent of episufonium formation is likely to depend on overall polarity of the

medium.

32

 

 

 

TMSO +O,H(LA) TMSO OH(LA)
2 1

RO 6: 2 SPh —‘-————‘= R0 6, +
OTMS OTMS H S‘ph
l-100, R = TBDPS l-113

QH OH
O. '- 2 O.
ROW/Vsph ROWSPh
5. .,
Ho“ 'OH
l-124 I-123

Figure I-22: Possible equilibration between activated epoxy sulfide I-100 and the

corresponding episulfonium ion

Another likelihood is that activated epoxy sulfide I-100 and the episulfonium
(I-113) exist in a dynamic equilibrium. Both species are trapped by different hydroxyl
groups and the relative rates of intramolecular trapping might be different. This rate
difference could be either due to difference in inherent reactivities of I-100 and I-113 or
in the availability of the nucleophiles. The availability of nucleophiles may in turn
depend upon the relative rates of silyl deprotection and the steric environment. Overall,
the nature of the reaction medium may influence the availability of the nucleophiles and

therefore relative rates of cyclization of I-100 vs. I-113.

 

 

TMSO 0 A00 OAc
o
ROWSPh A, B 01’ C HO 6 2 SPh + Rowsph
OTMS 3
OAc AcO
I-125, F1 = TBDPS l-126 1-127
conditions yield(%) H26 2 I-127 A i) BF3°OE12 (6 eq.), E120, 0 °C to rt
, ii) Ac20. Py.. 60 °C
A 65 (97-3) Bi)AcOH2H20:THF(6:3:1),0°Ctort
B 75 (20 :80) ii) A020. Py-. 60 °C
C i) 10% HCI :THF (9 : 1), 0 °C to rt
0 74 (5 :95) ii) Ac20, Py., 60 °C

 

Scheme I-13: Cyclization of diastereomeric epoxy sulfide I-125

33

w‘.

I.
.11. -.

 

We also investigated the cyclization reactions of diastereomeric epoxy sulfide
I-125, and obtained results similar to that in case of I-100. Under Lewis acidic conditions
(A, Scheme l-l3) the S-endo product (I-126) was obtained with excellent regioselectivity
whereas using 10% HCl, the selectivity was reversed to access the 5-exo product (I-127).
Since the outcome of the cyclization is independent of the epoxide stereochemistry, THF
diols I-121 and I-126, which are regio- and stereochemically distinct from the earlier
THF diols (I-101, I-103 and I-lll), could be obtained. Thus, using diastereomeric epoxy
sulfide systems 1-100 and I-125, a total of five different THF diols could be accessed.
To summarize, as part of our research program directed toward the total synthesis and
X11102» biological studies of recently discovered metabolites of arachidonic acid

OH (AA-THF-diols), we needed to access THF diol cores represented by 1-9 in

stereodefined forms. In the present study, a versatile strategy for regio- and

stereoselective synthesis of such THF diols has been developed.64

 

 

OTMS
no , \ OH
OTMS
I-93. R =-. TBDPS
1. D-(-) DET 1. L-(+) DET
2. functional group 2. functional group
' interchange interchange
OTMS O Y OTMS_.\Q Y
F10 , no , i
orMs OTMS
l-128 I-129
or ACOH or AcOH or AcOH
OR OH OH
.5 OR ED}, OH
HO OH HO OH
I-101 M 1 1H l-123 l-103 l-130

Figure I-23: Isomeric THF diols available from a common precursor 1-93

34

 
 

 

 

 

Our approach is based on the regiocontrolled cyclization of methylene interrupted
epoxy diols such as I-128 and I-129 (Figure I-23) that contain a directing group (Y)
adjacent to the epoxide. Substrates I-128 and I-129 were accessed in enantiopure forms
by Sharpless asymmetric epoxidation of the allylic alcohol (I-93) derived from 2-deoxy—
D-ribose. Depending upon the electronic properties of the pendant group (Y) and nature
of the acid promoter, five regio- and stereoisomerically complementary THF diols
(Figure l-23) were selectively accessed from a single precursor (I-93). Since the
cyclization in all cases was completely Stereospecific, the remaining isomers can be
obtained simply by changing the relative configuration between the diol and the epoxide.
With the enantiomerically pure THF diol cores in hand, we intend to pursue the total
synthesis of THF-diols derived from arachidonic acid (vide supra).
This method has been also applied to the total synthesis of mucoxin — a non

classical acetogenin. These studies are the subject of Chapters III and IV.

C. A novel method for the oxidative cleavage of olefins

In the course of related work on the synthesis of 2,3,5-tisubstituted

tetrahydrofuran diols, a coworker Benjamin Travis attempted to prepare these compounds
via the direct oxidative cyclization of 1,4-dienes such as methyl linoleate (Figure l-24).(’5

He found that treatment of 1,4-dienes with catalytic 0804 in the presence of

stoichiometric amounts of various co—oxidants such as KMnO4 and Oxone® (potassium
peroxymonosulfate) produced the requisite THF diols. However, the yields did not attain
useful levels. The poor yield of the desired products in these reactions was attributed to
competing over-oxidation of the olefin functionalities, resulting in C—C bond cleavage.
Thus, carboxylic acids were obtained as the major side products.

35

 
 

 

 

OH

 

(CH ) O
M9020’ 2 7 CSHH
HO
l-132
\ / 0504, Oxone +
002”“ DMF n ,
30% OH
1-131 Meozc. \ O 05H11
(CH2)7
OH
l-133

Figure 1-24: Oxidative cyclization of methyl linoleate to produce THF diols

Upon further investigation, it became apparent that this Oxone® mediated
oxidative cleavage of olefins was a fairly general process and was, in effect, an
alternative method for the ozonolysis of olefins. Such a method would be useful, since

safety concerns preclude the use of ozonolysis for large-scale processes. In fact, serious
. . . . 66-6 . .
accrdents due to explosrons have been reported in some instances. 8 Ozonolysrs IS also

operationally difficult, requiring specialized equipment for the generation of ozone.
Therefore, a safer and simpler chemical alternative is desirable.

The traditional alternative to ozonolysis involves the dihydroxylation of olefins,
followed by the cleavage of the resulting vicinal diols. The dihydroxylation is effected
using high-valent oxides of transition metals such as manganese, osmium, ruthenium or
tungsten. The 1,2-diols are then cleaved using the Lemieux—Johnson protocol or its

69—73

variants. However, the direct oxidative cleavage of olefins without the intermediacy

74-76

of 1,2-diols is not as common. It should be noted that the oxidative cleavage of

osmate esters — which are precursors to 1,2-diols during dihydroxylation — has been

previously observed, although the process has not been optimized for the direct cleavage

36

 

 

 

 

 

0.01 eq. 030,,

 

R3 4 eq. Oxone HO>= R3
,——-<__ 4' O + O
91 92 DM F, rt 9' =<n2
80 - 95%

Scheme I-14: The 030,, - Oxone® method for the oxidative cleavage of olefins

of olefins.77

Preliminary optimization of the 0304 - Oxone® cleavage reaction was done by B.
Travis, who demonstrated that simple alkyl and aryl olefins (such as stilbene, styrene,
cyclohexene, and l-decene) smoothly underwent cleavage to the corresponding
carboxylic acids. My contribution to this study was the evaluation of the scope of this
method, especially on more functionalized olefins, and on olefins with different
substitution patterns. We were also interested in exploring the effect of electron density
on reactivity of olefins toward oxidative cleavage.

Table l-l shows the scope of this oxidative cleavage method on a wide variety of
olefin substrates.78 All mono- and disubstituted olefins underwent cleavage smoothly to

afford the corresponding carboxylic acid or ketone products (entries 1—5). In case of tri-
and tetrasubstituted olefins (entries 7-9), low yields of the desired products were
obtained under standard conditions. The major products in these cases were the 1,2-diols,
presumably as a result of slow cleavage of the osmate ester intermediates. This problem
was overcome by the addition of 4 equivalents of solid NaHCO3 to the reaction mixture.
The addition of bicarbonate likely helps reduce the acidity of the medium, thus slowing
down the osmate ester hydrolysis pathway.

Cyclohexenone (entry 6) afforded pentanedioic acid, most probably via an

intermediate a-keto carboxylic acid, which undergoes decarboxylation under the

37

Entry Substrate Product Yield

 

1 Fro/W R=H ROMCOZH R=H. 85%
7 R=AC 5 R=AC, 93%)
O
2 <:>—< ----- CH
.j , R=CHO, 34%
OH ’08
O
3 INNW tut-W 800/0
an ll’OBn
O

4 E \ O @0H 91%

N02 0
5 O \ O OAOH 950/0
0211
O O
5 (Z HOZCMCOZH 92 /°
co H co H
7 \ 2 O 2 82%

O
8 80%
6 Wow
0

Ph
9 fl!“ ©)\ 85 A.

O

10 d _: “Ozcwcozti 67%
O O
‘ ' o

recovered SM

12

11

Table I-l: OsO4 -— Oxone® mediated cleavage of complex olefins

oxidative reaction conditions. Baeyer—Villiger type oxidative cleavage of or-dicarbonyl

79.80

compounds by peroxo reagents has been previously reported and is likely to be

operating in the oxidation of enones. In case of nootkatone (entry II), more electron rich
and sterically available exo olefin reacted preferentially. Lastly, alkyne (entry 12) proved

to be immune to oxidative cleavage and was recovered unscathed.

38

C)
0504 [0] ‘KO‘ "’{O

i
C) (3
0‘0

 

 

iDé ‘DKDSE;f—.“\\\
R R —““———"
H H H0230?
R R R H
1-134 1-135 I-136
o o - ~
O\"~ of“ Oxrdative
62/0535? 0303H Cleavage A O [O] i
7 JL “r n OH
R H
R R
F137 F138 F139

Scheme 1-15: Plausible mechanism of 0304 — Oxone“; mediated cleavage of olefins

We believe that 1,2 diols may not be intermediates in this reaction path for two
reasons. First, the oxidative cleavage proceeds efficiently under anhydrous conditions,
which would not promote hydrolysis of the osmate ester. Second, styrene glycol when
subjected to the reaction conditions was recovered quantitatively.

A plausible mechanism of this oxidative cleavage process is depicted in Scheme
I~15. Oxone is thought to be involved at three different stages — (i) oxidation of the initial
osmate ester (I-135) to Os(VIII) species I-136, (ii) oxidative cleavage of I-136 to the
aldehyde I-l38, and (iii) independent oxidation of the aldehyde to carboxylic acid I-139.

Thus, a general, simple and mild method for the generation of carboxylic acids

and ketones directly from olefins was established.78 The optimized conditions involved

the treatment of the starting olefins with 0.01 equivalents of 0804 and 4 equivalents of
Oxone® in DMF (Scheme I-14). These reactions were typically complete within three to
four hours at room temperature, and yields were typically high (80 — 95%). Further
mechanistic studies on this reaction and its extension to prepare aldehydes and esters by

similar C~C cleavage of olefins is being explored by B. Travis and other co-workers.

39

 

(a

. .

 

If
'i‘.

 

 

D. Experimental

General Procedures
All reactions were carried out in flame-dried glassware under an atmosphere of

dry nitrogen or argon. 4 A molecular sieves were dried at 160 °C under vacuum prior to
use. Unless otherwise mentioned, solvents were purified as follows. THF and EtzO were
either distilled from sodium benZOphenone ketyl or used as is from a solvent purification
system. CHZCIZ, toluene, CH3CN and Et3N were distilled from CaHZ. DMF, diglyme, and
DMSO were stored over 4 A mol. sieves and distilled from CaHz. All other commercially
available reagents and solvents were used as received.

1H NMR Spectra were measured at 300, 500 or 600 MHz on a Varian
Gemini-300, a Varian VXR—SOO or a Varian Inova-6OO instrument respectively. Chemical

shifts are reported relative to residual solvent (5 7.27, 2.50 and 4.80 ppm for CDCl3,

(CD3)ZSO and CD3OD respectively). 13C NMR spectra were measured at 125 MHz on a
Varian VXR-SOO instrument. Chemical shifts are reported relative to the central line of
CDCl3 (5 77.0 ppm). Infrared Spectra were recorded using a Nicolet IR/42 spectrometer

FT-IR (thin film, NaCl cells). High—resolution mass spectra were measured at the

University of South Carolina, Mass Spectrometry Laboratory using a Micromass VG~705

mass spectrometer. Optical rotations were measured on a Perkin—Elmer polarimeter
(model 341) using a l mL capacity quartz cell with a 10 cm path length.

Analytical thin layer chromatography (TLC) was performed using Whatman glass

plates coated with a 0.25 mm thickness of silica gel containing PF254 indicator, and

compounds were visualized with UV light, potassium permanganate stain, p-

40

‘1

 

 

 

 

anisaldehyde stain, or phosphomolybdic acid in EtOH. Chromatographic purifications
were performed using Silicycle 60 A, 35-75 pm silica gel. All compounds purified by
chromatography were sufficiently pure for use in further experiments, unless indicated
otherwise. GC analysis was performed using HP (6890 series) GC system containing
Altech SEE-54, 30 m x 320 mm x 0.25 mm column. Analytical and semi—preparative

HPLC normal phase separations were performed using HP 1100 series HPLC system.

1. Experimental section for synthesis of 2,3,5 trisubstituted THFS

OH TBDPS-CI /\/C'):/\/00 E
Ho/\/K/\/CO2Et fl» TBDPSO \ 2 t

: DMF, 11 i
OH 72%) OH

l-83 I'84
To a solution of 1-8345 (8.2 g, 0.04 mol) in DMF (30 mL), imidazole (3.0 g, 0.044
mol) and t-butylchlorodiphenylsilane (12 g, 0.044 mol) were added at room temperature.
The mixture was stirred at room temperature for 3 h, after which time the reaction was
quenched by adding 11,0 and diluted with ethyl acetate. The layers were separated and

the aqueous layer was extracted with ethyl acetate (3x100 mL). The organic layers were
combined, dried over NaZSO4, filtered and concentrated. The E and Z isomers ( approx.

5 : 1 ratio) were separated by flash column chromatography (ethyl acetate / hexanes = 20

/ 80). The purified E isomer I-84 was obtained as a yellow oil (72% yield).

Data for 1-84: 1H NMR (500MHz, CDC13) 6 7.65-7.63 (m, 4 H), 7.45—7.37 (m, 6 H),
6.99-6.92 (m, 1 H), 5.87 (dt, J = 15.7, 1.4 Hz, 1 H), 4.17 (q, J = 7.07 Hz, 2 H), 3.80-3.79
(m, 3 H), 3.60-3.58 (m, 1 H), 2.60 (br-s, 1 H), 2.47—2.43 (m, 1 H), 2.37-2.32 (m, 1 H),
2.15 (hrs, 1 H), 1.27 (t, J = 7.07, 3 H), 1.06 (s, 9 H); 13c NMR (125 MHz, (386),) a
166.5, 145.1. 135.7, 132.9, 130.3, 128.1, 124.2. 73.5, 71.6, 64.8, 60.5, 36.1, 27.1, 19.4,

41

14.5; IR (neat, thin film), 3461, 3973, 2932, 2859, 1968, 1899, 1830, 1719, 1655, 1472,

1428, 1393, 1370, 1267, 1167, 1113, 1044, 824, 741, 702 cm"; HRMS (CI) calcd for

C25H34OSSi, 460.2519 m/z (M+ NH 4f“; observed, 460.2550 m/z.

 

“km E TMS-CI, Imid OTMS CO E
\ 2 1 \ 2 1
TBDPSO i DMAP, THF TBDPSO i
9“ 45 °C, 75% OTMS
1-84 1-92

To a solution of I-84 (0.5 g, 1.13 mmol) in THF (5 mL), imidazole (308 mg, 4.52
mmol), chlorotrimethylsilane (0.57 mL, 4.52 mmol) and cat. dimethylaminopyridine
were added and the mixture was refluxed for 4 h. The reaction was cooled to room

temperature, diluted with ethyl acetate and filtered. The precipitate was washed with
ethyl acetate (200 mL). The filtrate was washed with H20 and brine, dried over NaZSO4,

filtered and concentrated. The crude product was purified by flash column
chromatography (ethyl acetate / hexane = 5/95) to isolate 1-92 as a colorless oil (75%
yield).

Data for 1-92: 1H NMR (500MHz, CDCI3) 6 7.67-7.65 (m, 4 H), 7.43-7.35 (m, 6 H),
6.99-6.93 (m, 1 H), 5.81 (d, J: 14.2 Hz, 1 H), 4.18 (q, J: 7.1, 2 H), 3.90—3.87 (m, l H),
3.75-3.72 (m, l H), 3.62-3.52 (m, 2 H), 2.41—2.26 (m, 2 H), 1.28 (t, J = 7.1, 3 H), 1.05 (s,
9 H), 0.07 (s, 9 H), 0.04 (s, 9 H); 13C NMR (125 MHz, CDC13) 5 166.6, 147.3, 135.9,

133.6, 130.0, 128.0, 123.3, 72.7, 65.7, 60.2, 35.1, 27.1, 19.4, 14.5, 0.6, 0.5; IR (neat, thin

film) 3086, 2957, 2896, 2859, 1982, 1893, 1824, 1722, 1657, 1474, 1429, 1368, 1318,

1252, 1113, 982, 841, 745,702 cm"; HRMS (c1) calcd for C,,H,Oo,st,, 587.3044 m/z

(M+ H)+; observed, 587.3030 m/z.

42

OTMS DiBAL-H OTMS

reooso , \ 002E! “*8 O 0°C reooso , \ on
' 2 ' "TM
OTMS 90% o 3
1-92 1-93

A solution of L92 (2 g, 3.4 mmol) in EtZO (15 mL) was cooled to 0°C. To this, a
solution of DIBAL-H (1.0 M in hexane, 13.6 mL) was added. The reaction was
continued at 0°C and it was complete after 30 min. The reaction was quenched by adding
saturated aqueous solution of NaoK tartrate (25 mL) and diluted with ether (50 mL). To
this biphasic mixture, glycerol (0.7 mL) was added and the mixture was stirred
vigorously for 8 h. The layers were separated and the aqueous layer was extracted with
ether (2x50 mL). The organic layers were combined, dried over Nast4, filtered and
concentrated. Purification after flash column chromatography led to 1-93 (1.66 g, 90%
yield) as a colorless oil.

Data for 1.93; ‘H NMR (500MHz, CDC1,) 6 7.67-7.64 (m, 4 H), 7.41-7.34 (m, 6 H),
5.66-5.64 (m, 2 H), 4.07 (d, J = 4.6 Hz, 2 H), 3.76—3.71 (m, 2 H), 3.64 (dd, J = 10.6, 5.7
Hz, 1 H). 3.52 (dd, J = 10.4, 6.1 Hz, 1 H), 2.22-2.19 (m, 2 H), 1.04 (s, 9 H), 0.08 (s, 9 H),
0.01 (8,9 H); 13C NMR (125 MHz, CDCI3) 6 135.9, 133.7, 131.3, 130.6, 129.8, 127.9,
73.8, 65.9, 64.1, 35.3, 27.1, 19.4, 0.7, 0.6; IR (neat, thin film) 3349, 3073, 2957, 2859,
1962, 1900, 1824, 1474, 1429, 1250, 1113, 972, 841, 702 cm"; HRMS (CI) calcd for

C29H4804513, 545.2939 m/z (M+ H)+-, observed, 545.2927 m/z.

43

D-(—)-DET (5 eq.)

 

OTMS Ti(o‘Pr),(3.6 eq.) OTMS O
TBDPSO , \ OH —= TBDPSO , OH
OTMS tBuOOH, 4 A MS OTMS
1-93 0142012, —20°c 1-94

73%; 98% do
To a round bottom flask charged with powdered, preactivated mol. sieves (50
mg), CH2012 (2 mL) was added and cooled to -30°C. To this, Tudor), (0.4 mL, 0.132
mmol) was added followed by addition of D-(—)—DET (0.32 mL, 0.184 mmol in 1 mL

CHzClz). This mixture was stirred at -30°C, under N2 for 30 min after which time a

solution of the allylic alcohol I-93 (0.2 g, 0.368 mmol in 2 mL CHZCIZ) was added
dropwise (over 30 min) to the reaction. This mixture was held for 45 min. at -20°C and
t»BuOOH (0.50 mL, 0.184 mmol) was added to the reaction. Stirring was continued at

~20°C for 2 h and quenched by adding saturated solutions of NaZSO4 (0.32 mL) and

NaZSO3 (0.6 mL) and diluted with 10 mL ether. The mixture was stirred vigorously at
room temperature for 3 h (yellow paste was formed in the reaction) and refrigerated
overnight. The paste was diluted with anhydrous E120 (200 mL) and celite was added to
it. This mixture was filtered on a celite pad using a sintered funnel. The yellow residue
was further washed with anhydrous ether (200 mL) when it turned granular. The filtrate
was concentrated and the crude product was purified by column chromatography (ethyl
acetate / hexanes = 10 / 90). The epoxide 1-94 was obtained as a colorless oil (152 mg,
73% yield).

Data for 1.94: [(1]D20'2 + 35.6 (c 1.0, CHC13); 1H NMR (500MHz, CDC13) o 7.65-7.63 (m,
4 H), 7.42735 (m, 6 H), 3.96-3.93 (m, 1 H), 3.88-3.86 (m, 1 H), 3.78-3.74 (m, l H),

3.60-3.55 (m, 2 11), 352.349 (m, 1 H), 3.05 (dt, J = 5.9, 2.2 Hz, 1 H), 2.84-2.82 (m, 1

44

H), 1.96-1.90 (m, 1 H), 1.57-1.48 (m, 2 H), 1.04 (s, 9 H), 0.06 (s, 9 H), 0.05 (s, 9 H); 13C

NMR (125 MHZ, CDC13) 0 135.8, 133.6, 129.9, 127.9, 71.7, 65.7, 61.9, 58.4, 54.2, 34.6,

27.1, 19.4, 1.2, 0.4; IR (neat, thin film) 3418, 3071, 2957, 2864, 1962, 1893, 1824, 1590,

1472, 1428, 1252, 111.1, 841, 747. 702 cm"; HRMS (CI) calcd for C29H4805813.

561.2888 m/z (M+ HY; observed, 561.2881 m/z.

OTMS

TBDPSO , \ OH

OTMS
1-93

D-(—)-DET (5 eq.)

 

Ti(o‘Pr), (3.6 eq.) ows O
3’ TBDPSO , i ? OH

t88001-1, 4 A MS 6mg

CHZCIZ, -20 °C (.95

55%; 84% de

1-95 (114 mg, 0.02 mmol) was prepared from allylic alcohol I-93 (200 mg, 0.37

mmol) following the same procedure as for 1-94 using L—(+)-DET.

Data for 1-95: [61],,”2 -21.8 (c 0.73 CHC13); 1H NMR (500MHz, CDC13) 6 7.66-7.65 (m,

4 H), 7.42-7.35 (m, 6 H), 4.06—4.04 (m, 1 H), 3.90-3.88 (m, 1 H), 3.78 (dt, J = 6.4, 2.2

HZ, l H), 3.61—3.57 (m, 1 H), 3.51 (d, J = 2.7, l H), 3.49 (d, J = 2.3 Hz, 1 H), 3.06—3.03

(m, 1 H), 2.89 (m, 1 H), 1.85-1.80 (m, 1 H), 1.67 (5 (br), 1 H), 1.43 (ddd, J = 14.4, 7.2,

2.6 Hz, 1 H), 1.04 (s, 9 H), 0.1 (s, 9 H), 0.06 (s, 9 H); 13C NMR (125 MHz, CDCI3) 6

135.8, 133.5, 129.9, 127.9, 71.1, 65.3, 62.0, 59.4, 54.0, 34.1, 27.1, 19.3, 0.5, 0.4; IR

(neat, thin film) 3430, 3073, 2957, 2859, 1967, 1900, 1821, 1590, 1474, 1429, 1252,

1113, 841, 743, cm'l; HRMS (C1) calcd for C29H4805513, 561.2888 m/z (M+ H)+;

observed, 561.2872 m/z.

OTMS
TBDPSO ,
OTMS
I-94

O

DMP, py. OTMS 0 CH0
_—_————’ TBDPSO ;
CHQCIQ; 900/0 (DI-MS
1-96

45

Pyridine (50 1.1L) was added to a mixture of Dess-Martin Periodinane (45 mg, 0.09

mmol) in CH2C12 (1.5 mL). To this, a solution of 1-94 (45 mg, 0.08 mmol) in 1.5 mL

CHZCI2 was added and the reaction was stirred at room temperature for 1 h after which
time it was diluted with ether (15 mL). The reaction was quenched by adding satd.
NaHCO3 (5 mL) containing Na28203 (2.5 g) and the mixture was stirred for 5 min after
which ether (15 mL) was added and the layers were separated. The ether layer was
washed with H20 (15 mL), dried over NaZSO,, filtered and concentrated. The product
was purified by column chromatography (ethyl acetate / hexanes = 5 / 95) to furnish the
aldehyde I-96 as a colorless oil (90% yield).

Data for 1-96: 1H NMR (500MHz, CDC13) 8 8.96 (d, J = 6.4 Hz, 1 H), 7.65-7.35 (m, 10
H), 3.99-3.96 (m, l H), 3.75 (dt, J = 6.3, 3.3 Hz, 1 H), 3.56 (dd, J = 10.6, 6.6 Hz, 1 H),
3.51 (dd, J: 10.6, 6.0 Hz, 1 H), 3.32 (dt, J = 5.8, 1.8 Hz, 1 H), 3.04 (dd, J = 6.3, 1.8 Hz,
1 H), 2.02-1.96 (m, 1 H), 1.57-1.53 (m, 1 H), 1.04 (s, 9 H), 0.05 (s, 9 H), 0.04 (s, 9 H);
l3C NMR (125 MHz, CDC13) 6 198.6, 135.8, 133.5, 130.0, 127.9, 76.7, 71.2, 65.6, 59.2,

55.1, 34.0, 27.1, 19.4, 0.5, 04; IR (neat, thin film) 3073, 2959, 2932,2859, 1968, 1893,

1824, 1732, 1474, 1429, 1390, 1252, 1113, 843, 743, 702 cm“; HRMS (CI) calcd for

99114605813, 559.2731 m/z (M+ H)+; observed, 559.2721 m/z.

OTMS

 

OTMS o PhaPCHSBr o
TBDPSO/MCHO TBDPSO , \
6TMS BULi, E120 OTMS
55%
1-96 1-97

A mixture of methyltriphenylphosphonium bromide (206 mg, 0.58 mmol) in THF

(2 mL) was cooled to 0°C. To this, butyllithium (0.48 mmol, 0.13 mL of 0.25M solution

46

in hexanes) was added and stirred for 30 min. during which time the solution turned
yellow and clearer. This ylide solution was added to a precooled (0°C) solution of I-96

(90 mg, 0.16 mmol) in THF (2 mL). The reaction was warmed to rt and stirred for 6 h

and quenched by adding H20 (10 mL) and diluted with ethyl acetate (20 mL). The
organic layer was washed with NH4C1 (10 mL). The aqueous layer was extracted with

ethyl acetate (2x20 mL). The organic layers were combined, dried over Nay-504 and

concentrated. The crude product was purified by column chromatography (ethyl acetate /
hexanes = 1/99) to yield the vinyl epoxide I-97 as a colorless oil (178 mg, 55% yield).

Data for 1.97: 1H NMR (500MHz, CDC13) 6 7.65-7.63 (m, 4 H), 7.45—7.34 (m, 6 H),
5.58-5.51 (m, l H), 5.42 (dd, J = 17.4, 1.5, 1 H), 5.28-5.22 (m, 1 H), 3.974394 (m, 1 H),
3.75 (dt, J = 6.3, 3.4 Hz, 1 H), 3.58 (dd, J = 10.5, 6.3 Hz, 1 H), 3.52 (dd, J = 10.6, 6.2 Hz,
1 H), 3.03 (dd, J = 7.6, 2.1 Hz, 1 H), 2.95-2.92 (m, 1 H), 1.99-1.93 (m, 1 H), 1.50—1.45
(m, 1 H), 1.04 (s, 9 H), 0.06 (s, 9 H), 0.05 (s, 9 H); 13C NMR (125 MHz, CDC13) 6 136.2,
135.8, 133.6, 129.8, 127.8, 119.1, 71.8, 65.7, 58.8, 35.0, 27.1, 19.4, 1.2, 0.5; IR (neat,
thin film) 3073, 2959, 2859, 1962, 1887, 1818, 1591, 1429, 1252, 1113, 841, 741, 702
cm"; HRMS (CI) calcd for C30H4304Si3, 557.2939 m/z (M+ H)+; observed, 557.2934

m/z.

OTMS OTMS

/\J\/<Cl)/\ 10% Pd/C /\/K/<CID/\
reopso , \ ——"——* TBDPSO ,
OTMS H2. EtOAC OTMS
1-97 60% I-98

10 % Pd-C (4 mg) was added to a solution of I-97 (40 mg, 0.072 mmol) in ethyl

acetate (2 mL) and the mixture was stirred under H2 atmosphere at room temperature for

47

1.5 h. The reaction was filtered through a celite pad and the residue was washed with
ethyl acetate. The filtrate was concentrated and the crude product was purified by flash
column chromatography (ethyl acetate / hexanes = 1/99) to furnish alkyl epoxide I-98
(60% yield) as a colorless film.

Data for 1-98: 1H NMR (500MHz, CDC13) 6 7.68-7.66 (m, 4 H), 7.44-7.36 (m, 6 H).
3.96-3.93 (m, 1 H), 3.78 (dt, J = 5.8, 3.5 Hz, 1 H), 3.63 (dd, J = 6.2, 10.4 Hz, 1 H), 3.53
(dd, J: 6.2, 10.6 Hz, 1 H), 2.83—2.80 (m, 1 H), 2.60 (dt, J = 5.5, 2.2 Hz, 1 H), 1.92 (ddd,
J = 14.2, 7.5, 5.3 Hz, 1 H), 1.62-1.45 (m, 3 H), 1.06 (s, 9 H), 0.99 (t, J = 7.5 Hz, 2 H),
0.09 (s, 9 H), 0.07 (s. 9 H); 13C NMR (125 MHz, CDC13) 6 135.6, 133.4, 129.7, 127.7,

71.8, 65.5, 59.9, 56.4, 35.1, 26.9, 25.1, 19.2, 9.9, 0.4, 0.3; IR (neat, thin film) 3076,

3961, 1736, 1429, 1250, 113, 841, 742, 702 cm".

 

OTMS OTMS
o M62804 0
TBDPSO ; OH ¢ TBDPSO i OMe
OTMS LiHMDS. T HF OTMS
[.94 750/0 [~99

A solution of I-94 (51 mg, 0.09 mmol) in THF (0.7 mL) was cooled to 0 °C. To
this solution (CH3)ZSO4 (50 (LL, 0.52 mmol) and LiHMDS (140 (1L of 1.0 M solution in
THF) were added. After 2 h, the reaction was diluted with ethyl acetate (20 mL) and
washed with HZO (2x15 mL). The aqueous layer was extracted with ethyl acetate (2x20
mL). The organic layers were combined, dried over NaZSO4, filtered and concentrated.

The product 5e was purified by flash column chromatography (ethyl acetate / hexanes = 5

/95) as a colorless oil (75% yield).

48

 

.‘)

 

Data for 1.99; 1H NMR (500MHz, CDC13) 6 7.66-7.63 (m, 4 H), 7.42-7.34 (m, 6 H),
3.94—3.91 (m, 1 H), 3.77—3.73 (m, 2 H), 3.61-3.48 (m, 3 H), 2.93 (dt, J = 5.9, 1.9 Hz, 1
H), 2.79-2.77 (m, 1 H), 1.93-1.87 (m, 1 H), 1.53-1.49 (m, 1 H), 1.51 (s, 3 H). 1.04 (s, 9
H), 0.11 (s, 9 H), 0.06 (s, 9 H); 13c NMR (125 MHz, CDC13) 6135.8, 133.6, 129.8,

127.8, 71.9, 65.8, 63.3, 58.6, 54.5, 34.9, 27.1, 19.4, 0.6, 0.5; IR (neat, thin film) 3073,

2957,2859, 192, 1893, 1824, 1589, 1474, 1429, 1252, 1113, 843,747, 702 cm“.

 

OTM OTMS
S 0 80313, P1662 o
TBDPSO , OH *7 TBDPSO , son
ems Et3N; 85% ems
1.94 1-100

To a solution of diphenyl disulphide (60 mg, 0.275 mmol) in triethyl amine (0.2
ml). tributyl phosphine (63 (LL, 0.275 mmol) was added, stirred for 5 min. and cooled to
0° C. To this, a solution of the epoxyalcohol I-94 (50 mg, 0.09 mmol) in triethyl amine
(0.2 ml) cooled to 0° C was added dropwise. The reaction was allowed to warm to room
temperature and stirred for 4 h. The reaction was diluted with ether (20 mL) and washed
with H20 (2x15 mL). The aqueous layer was extracted with ether (2x20 mL). I-100 was
purified by flash column chromatography (ethyl acetate / hexanes = 5 / 95) as a yellow oil
in 85% yield.

Data for 1-100: 1H NMR (500MHz, CDC1,) 6 7.66-7.63 (m, 4 H), 7.42—7.33 (m, 8 H),

7.27-723 (m, 2 H), 7.18-7.15 (m, 1 H), 3.89-3.86 ((11, J = 7.6.4.2 Hz, 1 H). 3.75-3.72 (dt,
J = 6.0, 3,6 Hz 1 H), 3.57 (dd, J = 10.6, 6.0 Hz, 1 H), 3.48 (dd, J = 10.6, 6.2 Hz, 1 H),
3.07 (dd, 13.8, 5.2 Hz, 1 H), 2.95 (dd, 13.9, 5.3 Hz, 1 H), 2.87-2.83 (m, 2 H),1.82-1.77

(m. 1 H), 1.52—1.47 (m, 1 H), 1.03 (s, 9 H), 0.06 (s, 9 H ), 0.05 (s, 9 H); 13C NMR (125

49

no,
\

l'.

11:

.13.

 

 

I.»

MHz, CDC13) 6 135.9, 133.6, 130.3, 129.9, 129.2, 127.9, 126.8, 71.7, 65.8, 57.5, 57.1,
36.7, 34.9, 27.1, 19.4, 0.6, 0. IR (neat, thin film) 3073, 2957, 2859, 1856, 1831, 1712,

1574, 1473, 1427, 1391, 1113, 941, 841, 741,cm“; HRMS (CI) calcd for C35H5204SSi3,

653.2972 m/z (M+ H)+; observed, 653.2969 m/z.

 

OTMS OTMS
\Q BU3P, 1311282 HQ
TBDPSO , ' OH *7 TBDPSO , ' son
OTMS EtaN; 83% OTMS
1-94 l-125

1-125 was prepared following the same procedure as that for I-100.

Data for I-125: 1H NMR (500MHz, CDC13) 6 7.65-7.62 (m, 4 H), 7.42—7.32 (m, 8 H),
7.28-7.25 (m, 2 H), 7.20-7.17 (m, 1 H), 3.98 (dt, J = 5.0, 2.5 Hz, 1 H), 3.73 (dt, J = 6.3,
2.4 Hz, 1 H), 3.47-3.45 (m, 1 H), 3.07 (dd, 13.9, 5.3 Hz, 1 H), 2.97 (dd, 13.9, 5.8 Hz, 1
H), 2.88 (dt, 2.0, 5.5 Hz, 1 H),l.82-1.77 (m, 1 H), 1.21 (ddd, 18.3, 7.9, 2.9 Hz, 1 H), 1.02
(s, 9 H), 0.08 (s, 9 H ), 0.07 (s, 9 H); 13C NMR (125 MHz, CDC13) 6 135.9, 133.6, 129.8,
129.2, 127.9, 126.9, 71.7, 65.3, 57.9, 57.2, 36.9, 34.3, 27.1, 19.3, 0.6, 0.5. IR (neat, thin
film) 3176, 2957, 2859, 1956, 1831, 1587, 1474, 1429, 1250, 1113, 943, 841, 74l,cm“;

HRMS (CI) calcd for C35H5204SSi3, 653.2972 m/z (M+ H)”; observed, 653.2965 m/z.

General Procedure for BF3-E110 Mediated Epoxide Opening Reactions:

A solution of the epoxide (0.088 mmol) in anhydrous EtzO (1 mL) was cooled to
0 °C. BF3°Et20 (0.616 mmol) was added to this solution at 0 °C. The reaction was
allowed to warm to the room temperature for 1 h. The reaction was quenched by adding
H20. The mixture was diluted with ethyl acetate (10 mL) and washed with NaHCO3

50

(satd., 5 mL). The aqueous layer was extracted with ethyl acetate (2x10 mL). The
organic layers were combined, dried over NaZSO4, filtered and concentrated. The crude

product was subjected to acetylation without purification.

General Procedure for Protic Acid Mediated Epoxide Opening Reactions:

A solution of the epoxide (0.1 mmol) in THF (0.5 mL) was cooled to 0 °C.
Aqueous protic acid (AcOHzH20:THF (6:321) or 10% HCI : THF (9 : l) 3 mL) was
added to the THF solution at 0 °C and the reaction was allowed to warm to room

temperature for 3 h, after which time the reaction was diluted with ethyl acetate and

neutralized by adding satd. NaHCO3 solution. The aqueous layer was extracted with ethyl

acetate (2x15 mL). The organic layers were combined, dried over N32504, filtered and

concentrated. The crude product was subjected to acetylation without purification.
General Procedure for the Acetylation Reaction:

The crude cyclization product (0.11 mmol) was dissolved in pyridine (0.5 mL).
Acetic anhydride (0.66 mmol) was added to the solution and the mixture was stirred at 60
°C for 4 h. The reaction was cooled to room temperature, diluted with ethyl acetate (15
mL) and washed with 10% HCI (2x10 mL). The aqueous layers were combined and
extracted with ethyl acetate (2x15 mL). The organic layers were combined, dried over

NaZSO4, filtered and concentrated. The crude product was purified by flash column

chromatography (hexanes/ethyl acetate).

51

Spectral data for cyclization products:

OTMS QAC

O O. ' OAc
AorB 73
TBDPSOMOH ______. romeo/W
1-94 I-102

A i) BF3-0E12 (6 eq.), Eigo, 0 °C to rt
ii) ACQO, Py., 60 °C, 75% (two steps)

Bi) AcOH : H20 :THF (6:3 : 1),0°Ctort
ii) ACZO, Py., 60 °C, 72% (two steps)

Data for 1.102: (ot]D'-’°-2 +469 (c 1.7, CHC13); ]H NMR (500MHz, CDC13) 6 7.66-

7.63 (m, 4 H), 7.43-7.35 (m, 6 H), 5.34 (dt, J = 4.5, 2.1 Hz, 1 H),5.15-5.11 (m, 1 H), 4.6
(dd. J = 12.1, 2.7 Hz, 1H), 4.31 (dt, J = 7.8, 4.5 Hz, 1 H), 4.154. 10 (m, 2 H), 3.72 (dd, J
= 11.0, 3.3 Hz, 1 H), 3.68 (dd, J = 11.1, 4.2 Hz, 1 H), 2.50—2.44 (m, 1 H), 2.05 (s, 6 H),

2.02 (s, 3 H), 1.91 (ddd, J = 13.7, 4.4, 2.9 Hz, 1 H), 1.03 (s, 9 H); 13c NMR (125 MHz,

CDC13) 6 170.9, 170.3, 135.8, 133.3, 130.1, 128.0, 85.1, 76.3, 72.7, 64.9, 63.2, 34.9,
27.0, 21.3, 21.2, 21.0, 19.4; IR (neat, thin film) 3070, 2932, 2859, 1984, 1903, 1744,

1429, 1370, 1237, 1113, 824, 743, 704 cm"; HRMS (FAB) calcd for C,,H,,O8St,

543.2415 m/z (M+H)+; observed, 543.2390 m/z.

OTMS OAc
“‘9 A or B TBDPSO O 3 2 OAc
TBDPSO , ' OH -—-——> ‘
OTMS Acd~5
1.95 MM

A i) BF3'OE12 (6 eq.), E120, 0 °C to rt
ii) A020, Py., 60 °C, 80% (two steps)
Bi) AcOHszozTHF(6 :3:1),0°Ctort
ii) A020, Py., 60 °C 76% (two steps)

Data for 1.104: [61020-2 +15.8 (c 0.77, CHC13); ‘H NMR (500MHz, CDC13) 6

7.66-7.63 (m, 4 H), 7.42-7.34 (m. 6 H), 5.34-5.33 (m, 1 H), 5.09 (dt, J = 5.9, 2.8 Hz. 1

H), 4.3 (dd, J = 12.2, 3.0 Hz, 1 H), 4.19 (dt, J = 10.2, 6.1 Hz, 1 H), 4.12-4.08 (m, 1 H),

52

4.00-3.98 (m, 1 H) 3.74 (dd, J = 11.0, 3.9 Hz, 1 H), 3.62 (dd, J = 11.0, 4.6 Hz, 1 H), 2.09-

2.02 (m, 2 H), 2.05 (s, 3 H), 2.03 (s, 3 H), 2.00 (s, 3 H), 1.03 (s, 9 H); 13C NMR (125
MHz, CDC13) 6 170.8, 170.6, 170.2, 135.8, 133.3, 130.0, 128.0, 85.5, 76.3, 72.8, 64.2,

63.1, 35.5, 27.0, 21.3, 21.0, 19.4; IR (neat, thin film) 3073, 2932, 2859, 1975, 1906,

1746, 1429, 1370, 1237, 1113, 862, 802, 743, 704 cm"; HRMS (FAB) calcd for

C29H3908Si, 560.2680 m/z (M+NH4)+; observed, 560.2694 m/z.

OTMS O 1 OAc

O. - M
romeo/Movie ___.A or B reoosowo e
OTMS .

A00; 5

1-99 l-105

A i) BF3'OE12 (6 eq.), E120, 0 °C to 11
ii) ACQO, Py., 60 °C, 60% (two steps)
Bi)AcOH:HZO:THF(6 :3 : 1).O°Ctort
ii) A020, Py., 60 °C 75% (two steps)

Data for 1.105; [61],,”2 +318 (c 1.0, CHC13); 1H NMR (500 MHz, CDC13) 6 7.66-
7.63 (m, 4 H), 7.42—7.35 (m, 6 H), 5.31 (dt, J = 6.4, 2.7 Hz, 1 H), 5.09 (m, 1 H), 4.32 (dt,
J = 7.9.4.7 Hz, 1 H), 4.09 (m, 1 H), 3.72 (dd, J = 11.0, 3.6 Hz, 1 H), 3.66 (dd, J = 11.0,
4.4 Hz, 1 H), 3.61 (dd, J = 10.9, 3.2 Hz, 1 H), 3.56 (dd, J = 10.9, 5.6 Hz, 1 H), 3.35 (s, 3
H), 2.45-2.40 (m, 1 H), 2.07 (s, 3 H), 2.05 (s, 3 H), 1.90 (ddd, J = 13.9, 4.7, 3.0 Hz, 1 H),
1.03 (s, 9 H); ”C NMR (125 MHz, CDC13) 6 170.9, 170.5, 135.8, 133.4, 130.0, 128.0,

84.8, 76.3, 73.6, 71.8, 64.9, 59.5, 34.6, 27.0, 21.3, 19.4; [R (thin film) 3073, 3017, 2932,

2859, 1968, 1900, 1736, 1590, 1471, 1429, 1372, 1235, 1113, 1055, 762, 704 cm";

HRMS (C1) calcd for C28H3807Si, 513.2309 m/z (M-H)’; observed , 513.2306 m/z.

53

 

 

 

 

OTMS O H
MCHO A or B 0' "'OAc
TBDPSO 6i 2 TBDPSO O OAC
OTMS H

1.95 H 07

A i) BF3-OE12 (6 eq.), EtQO, 0 °C to rt
ii) A020, Py., 60 °C, 17% (two steps)
Bi)AcOH:H20:THF(6:3:1),0°Ctort
ii) Ac20, Py., 60 °C 20% (two steps)

Data for 1.107: [(1:1020‘2 +456 (c 0.9, CHC13); 1H NMR (500MHz, CD61.)
6 7.62—7.59 (m, 4 H), 7.43-7.34 (m, 6 H), 6.00 (d, J = 6.8 Hz, 1 H), 4.66 (dd, J = 6.8, 1.6
Hz. 1 H), 4.57 (m, 1 H), 4.52-4.50 (m, 1 H), 4.34 (m, 1 H), 3.68 (dd, J = 11.2, 3.8 Hz, 1
H). 3.43 (dd, J = 11.2, 6.6 Hz, 1 H), 2.08 (s, 6 H), 2.06-2.11 (m, 2 H),1.02 (s, 9 H); 13c
NMR (125 MHz, CDC13) 6 170.3, 169.6, 135.7, 133.1, 130.1, 128.0, 92.2, 82.3, 76.2,

74.2, 64.3, 33.9, 27.0, 21.7, 19.4; IR (neat, thin film) 3070, 2932, 2859, 1968, 1896,

1744, 1429, 1370, 1235, 1113, 897, 824, 758, 704 cm"; HRMS (FAB) calcd for

(32711340751, 537.171 1 m/z (M+K)+; observed, 537.1732 m/z.

OTMS O 0 OAc
~.3
TBDPSOW fl. TBDPSOW
OTMS A C0:5
1-98 I-108

A i) BF3-OE12 (6 eq.), Et20, 0 °C to rt
ii) A020, Py., 60 °C, 80% (two steps)
Bi)AcOH 2H202THF(6 :3 : 1),0°Ctort
ii) A020, Py., 60 °C 78% (two steps)

Data for I-108: [6],)?"-2 +219 (c 0.3, CHC13); 1H NMR (500MHz, CDC13) 6 7.67
7-62 (m, 4 H), 7.42-7.34 (m, 6 H), 5.32-5.30 (m, 1 H), 4.95 (ddd, J = 8.3, 6.6, 4.0 Hz, 1
H). 4.18—4.14 (m, 1 H), 3.72 (dd, J = 11.1, 3.5 Hz, 1 H), 3.68 (dd, J = 11.0.4.3 Hz, 1 H),
2.45-2.39 (m, 1 H), 2.05 (s, 6 H), 1.86 (ddd, J = 13.7, 5.7, 3.5 Hz, 1 H), 1.73 (ddd, J =

13
14317.5. 3.9 Hz, 1 H), 1.58-1.54 (m, 1 H), 1.03 (s. 9 H). 0.89 (t. J = 7.5 Hz. 3 H) C

54

 

NMR (125 MHZ, CDC13) 6 171.0, 170.7, 135.8, 133.4, 130.0, 128.0, 84.4, 80.0, 76.3,
76.0, 65.0, 34.6, 30.0, 27.0, 24.3, 21.3, 19.4, 9.6; IR (neat, thin film) 3071, 2928, 2857,

1975, 1887, 1740, 1590, 1462, 1429, 1370, 1242, 1113, 1020, 801, 741,702 cm".

OTMS 1. BFa'OEt2 (5 64) OAc

o
4 Et 0, 0 °C to rt
reopsoM 2 2 TBDPSOW
OTMS

2. Ac20, Py., 60 C ai'OAc

1-97 0
R = TBDPS 60 A (two steps) I-109

 

Data for 1409: [61],,”2 —12.0 (c 0.3, CHC13); lH NMR (500MHz, CDc1,)

6 7.65—7.63 (m, 4 H), 7.42-7.34 (m, 6 H), 5.82-5.75 (m, 1 H), 5.27-5.23 (m, 1 H), 5.20-
5.16 (m, 1 H), 5.13-5.10 (m, l H), 4.96 (m, 1 H), 4.34-4.30 (m, 1 H), 3.81 (d, J = 5.3 Hz,
1 H), 2.07-2.03 (m, 1 H), 2.05, (s, 3 H), 2.02 (s, 3 H), 1.95-1.91, (m, 1 H), 1.03 (s, 9 H);
13C NMR (125 MHz, CDC13) (1 170.7, 170.3, 136.0, 135.8, 133.5, 130.0, 127.9, 116.5,
84.8, 78.8, 74.8, 63.3, 33.4, 27.0, 21.3, 21.2, 19.4; IR (neat, thin film) 3072, 2932, 2858,
1746, 1590, 1474, 1429, 1374, 1235, 1113, 860, 823, 734, 704 cm'l; HRMS (FAB) calcd

for C28H36068i, 535.1918 m/z (M+K)+; observed, 535.1912 m/z.

OTMS TBDPSO 7 O
M B 01' C \
TBDPSO 7, \

 

OTMS 3 80% AcO“'6 4"‘0Ac
1-97 -
R = TBDPS ' "O

B i) AcOH :HQO :THF (6:3 : 1),0°Ctort
ii) A020, Py., 60 °C

C i) 10% HCl :THF (9 : 1), 0 °C to rt
ii) Ac20, Py., 60 °C

Data for 1.110; [a]DZO'2—12.O(c 0.3, CHC13); 1H NMR (500MHz, CDC13) 6 7.69-
7.63 (m, 4 H), 7.41-7.32 (m, 6 H), 5.81-5.75 (m, 1 H), 5.35-5.32 (m, 1 H), 5.23-5.20 (m,

1 H). 4.70 (ddd, J = 11.2, 9.5, 4.8 Hz, 1 H), 3.79-3.71 (m, 3 H), 3.43 (ddd, J = 9.7, 4.5,
55

2.2 Hz, 1 H), 2.58 (dt = 9.7, 4.5, 2.2 Hz, 1 H), 1.99 (s, 3 H), 1.93 (s, 3 H), 1.56-1.50 (m, 1
H), 1.02 (s, 9 H) ”C NMR (125 MHz, CDC13) 6 169.8, 169.6, 135.9, 134.9, 133.8, 129.8,

127.8, 118.2, 80.5, 79.9, 69.9, 66.6, 63.4, 35.1, 26.9, 21.2, 21.1, 19.5; IR (neat, thin film)

3037, 2959, 2932, 2859, 174-4, 1474, 1428, 1374, 1235, 1115, 995, 825, 798, 740, 706

cm"; HRMS (Cl) calcd for C28H36068i, 497.2359 m/z (M+H)+; observed, 497.2377 m/z.

 

TMSO H O 1. 813082 (6 eq.) AcO
‘ Et 0, 0 °C to rt .
TBDPSO 5, 2 son 2 4. TBDPSO o 5 072 son
OTMS H 2. A020, py., 60 °C 40%
l-100 65 /0 (two steps) 1-121

Data for 1-121: [61],,”2 - 37.5 (c 0.8, CHC1,); 1H NMR (500MHz, CDC13) 6 7.63-

7.61 (m, 4 H), 7.43-7.33 (m, 8 H), 7.26-7.23 (m, 2 H), 7.19-7.15 (m, 1 H), 5.33-5.31 (m,
1 H), 5.10-5.07 (m, 1 H), 4.37 (dt, J = 9.0, 5.8 Hz, 1 H), 4.05-4.02 (m, 1 H). 3.77-3.72
(m, 2 H), 3.13 (dd, J = 13.5, 5.7 Hz, 1 H), 2.17-2.12 (m, 1 H), 2.05-2.00 (m, 3 H), 1.99 (s,
3 H), 1.95 (s, 3 H), 1.02 (s, 9 H) ”C NMR (125 MHz, CDCl,) 6 170.1, 170.0, 135.5.
133.1, 130.1, 129.8, 129.0, 127.7, 126.6, 80.3, 76.4, 74.9, 74.4, 62.7, 34.9, 32.8, 26.7,
21.0; IR (neat, thin film) 3073, 2932,2859, 1956, 1900, 1744, 1588, 1474, 1429, 1373,

1230, 1113, 951, 823, 741, 704 cm"; HRMS (C1) calcd for C,,H,Oo,,ss1, 593.2393 m/z

(M+H)+; observed, 593.2383 m/z.

 

TMSO ’fio 1.10%HCI:THF(9:1) O Qficsph
TBDPsoMsph *' TBDPSOW
ems H 2. A620, py.,60°C g

74% (two steps) AcO
1-100 I-122

56

11...

Data for 1.122; [611,20-2 - 37.5 (c 0.8, 0H01,)-, 1H NMR (500MHz, 0D01_,) 6 7.66-
7.60 (m, 4 H), 7.42-7.33 (m, 8 H), 7.27-7.23 (m, 2 H), 7.18-7.14 (m, 1 H), 5.29 (dt, J =
6.8, 2.5 Hz, 1 H), 5.12 (dt, J = 7.6, 3.4 Hz, 1 H), 4.32 (dt J = 7.8, 4.5 Hz, 1 H), 4.06 (m,
1 H), 3.70 (dd, J -= 11.0, 3.6 Hz, 1 H), 3.65 (dd, J = 11.1, 4.2 Hz, 1 H), 3.38 (dd, J = 14.3,
3.4 Hz, 1 H), 3.07 (dd, J = 14.3, 7.5 Hz 1 H), 245-239 (m, 1 H). 2.00 (s, 3 H), 1.88 (s, 3
H), 1.85-1.84 (m, 1 H), 1.03 (8,9 H); 130 NMR (125 MHz, 0D01,) 6 170.9, 170.3, 136.3,
135.8, 133.3, 130.2, 130.0, 129.1, 128.0, 126.5, 84.9, 79.2, 73.8, 64.9, 35.6, 34.7, 27.0,
21.3; IR (neat, thin film) 3073,2932, 2859, 1962, 1891, 1742, 1588, 1472, 1428, 1370,

1239, 1113, 1026, 823,740,702 cm"; HRMS (01) calcd for C33H4006581, 621.2706 m/z

(M+ C2H5)+; observed, 621.2702 m/z.

TMSO H 1. BF3-OEt2 (6 eq.) A00

0 Et 0 0 °C to rt 0
TBDPSO/MSPh 2 . 4' TBDPSO\/'6\5<—f\sph
OTMS H 2. ACQO, py., 60 °C

0 OAC
l-125 70 /o (two steps) I-126

 

Data for I-126: [61],,”2 + 35.6 (c 1.0, 01101,); 1H NMR (500 MHz, CDCI3) 6

7.66-7.65 (m, 4 H), 7.44-7.35 (m, 8 H), 7.27-7.16 (m, 3 H), 5.25-5.22 (m, 1 H), 5.07 (dt,
J = 7.0, 4.5 Hz, 1 H), 4.13 (dt, J = 7.7, 4.9 Hz, 1 H), 3.95 (ddd, J = 8.0, 5.8, 3.9 Hz, 1 H),
3.81 (d, J = 4.4 Hz, 1 H), 3.12 (dd, J = 13.7, 5.8 Hz, 1 H), 3.02 (dd, J = 13.7, 8.0 Hz, 1
H), 2.33-2.27 (m, 1 H), 2.01, (s, 3 H), 1.96 (s, 3 H), 1.89-1.85 (m, 1 H), 1.02 (s, 9 H); 130
NMR (125 MHz, 0D01,) 6 170.5, 170.4, 136.0, 135.8, 133.6, 130.1, 129.9, 129.2, 127.9,
126.7; IR (neat, thin film) 3074, 2932,2859, 1962, 1900, 1742, 1588, 1473, 1428, 1373,

1242, 1113, 953, 823, 741, 702 cm"; HRMS (CI) calcd for C33H4006831, 593.2393 m/z

(M+ H)+; observed, 593.2377 m/z.
57

2. Experimental section for the oxidative cleavage of olefins

General Procedure for the Oxidative Cleavage of Mono and Disubstituted Olefins

(condition B):

The olefin (1 eq) was dissolved in DMF (0.2 M), and 0804 (0.01 eq, 2.5% in

tBuOH) was added and stirred for 5 min. Oxone® (4 eq) was added in one portion and

the reaction was stirred at RT for 3 h or until the solution becomes colorless. This
usually marks the completion of the reaction, which was verified by TLC or GC. NaZSO3

(6 eq w/w) was added, to reduce the remaining Os(VIII), and stirred for an additional
hour or until solution became dark brown / black. EtOAc was added to extract the

products and 1N HCI was used to dissolve the salts. The organic extract was washed
With 1N HCI (3x) and brine, dried over Na2504, and the solvent was removed under

reduced pressure to obtain the crude product. Products were purified by silica gel column

chromatography.

General Procedure for the Oxidative Cleavage of Tri and Tetrasubstituted Olefins

(condition B):

The olefin (1 eq) was dissolved in DMF (0.2 M), and 0804 (0.01 eq, 2.5% in

tBuOH) was added and stirred for 5 min. A solid mixture of Oxone® (4 eq) and NaHCO3

(4 eq) was then added in one portion and the reaction was stirred at RT for 3 h or until

solution becomes colorless. This usually marks the completion of the reaction, which

58

u.‘

 

was verified by TLC or GC. NaZSO3 (6 eq w/w) was added, to reduce the remaining
Os(VIII), and stirred for an additional hour or until solution became dark brown / black.
EtOAc was added to extract the products and 1N HCI was used to dissolve the salts. The
organic extract was washed with 1N HCl (3x) and brine, dried over NaZSO4, and the

solvent was removed under reduced pressure to obtain the crude product. Products were

purified by silica gel column chromatography.

Spectral data:

Spectral properties of nonanoic acid (Table I-l, entry 1), p-methylbenzoic acid, [2-
nitrobenzoic acid, adipic acid, benzoic acid (entries 4—7), acetophenone and 3R-
methyladipic acid (entries 9 and 10) match those reported by Aldrich and comparison to

authentic samples.

(100% m ACOAWCOW

7
93%

1H NMR (CDCI3, 300 MHz): 64.02 (t, 2H, J=6.9 Hz), 2.32 (t, 2H, J=7.4 Hz),
2.02 (s, 3H), 1.56-1.61 (m, 4H), 1.29 (bs, 8H); 130 NMR (CDC13, 75 MHz): 6179.6,

171.4, 64.5, 33.9, 29.0, 28.9, 28.8, 28.4, 25.7, 24.5, 20.9; IR (neat, thin film) 3455, 2931,

2856 1739, 1737, 1242 cm“; LRMS (70 eV, El) m/z 199 [M—HZOI+, 157 [M-OACF’.

MW Condition A ,,,,, O__<O

’OH "on
R = H, 44%
R = CHO, 340/0

(1R, 2R, 5R)—2-Acetyl—5-methyl cyclohexanol (R = H):

59

1H NMR (CDC13, 300 MHz): 63.80 (ddd, 1H, J=4.4, 9.6, 11.1 Hz), 2.27 (ddd,
1H, J=3.6, 9.6, 12.9 Hz), 2.17 (s, 3H), 1.91-2.00 (m, 2H), 1.68-1.74 (m, 1H), 1.38-1.52
(m, 1H), 1.22-1.27 (m, 1H), 0.91-1.03 (m, 1H), 0.92 (d, 3H, J=6.3 Hz); 130 NMR
(CDCI3, 75 MHz): 6212.9, 70.4, 585,422,340, 31.1, 29.2, 27.5, 22.0; IR (neat, thin

film) 3417, 2952, 2927, 2869, 1705 cm"; LRMS (70eV, El) m/z 156 Mi 138 lM-HZO]+,
95 [M-HzO—C(O)Me |+.
(1R, 2R, 5R)—2-Acetyl-5—methyl cyclohexanyl formate (R = CHO):

1H NMR (CDC13, 300 MHz): 6 7.95 (s, 1H), 5.06 (ddd, 1H, J=4.4, 9.6, 11.2 Hz),
2.59 (ddd, 1H, J=6.9, 8.9, 14.5 Hz), 2.15 (s, 3H), 2.1 1-2. 13 (m, 1H), 1.93 (qd, 1H, J=3.9,
6.9 Hz), 1.68-1.77 (m, 1H), 1.50-1.62 (m, 1H), 1.27-1.41 (m, 1H), 0.87-1.06 (m, 4H); 130
NMR (CDC13, 75 MHz): 6 209.4, 160.3, 73.1, 55.2, 39.3, 33.3, 30.0, 29.3, 27.8, 21.7; IR
(neat, thin film, cm") 2952, 2929, 2869, 1728, 1178; LRMS (70 eV, 131) m/z 185
1M+H]*, 149 [M-H0o2Hr; HRMS [M+H]+ Calcd for 0,0H16o3: 184.1099 m/z .

Observed 184.1095 m/z.

“0% Condition A . 0‘6

, 80% ,
OBn OBn

‘H NMR (CDC13, 300 MHz): 6 7.21-7.32 (m, 5H). 4.56 (d, 1H. J=11.3 “214.37

(d, 1H, J=11,3 Hz), 3.6 (dt, 1H, J=6, 10.4 Hz), 2.53 (ddd, 1H, J=3.8, 10.1, 12.6 Hz), 2.16
(s, 3H), 2.12-2.19 (m, 1H), 1.75 (qd, 1H, J=3.6, 10.2 Hz), 1.64-1.70 (m, 1H), 1.25-1.52

(m, 2H), 0.93 (d, 211, 1:33 Hz); 130 NMR (CDCI3, 75 MHz): 6212.3, 138.5, 128.2,

60

 

 

 

 

 

 

 

 

 

 

127.6, 127.4, 79.1, 70.9, 56.6, 39.4, 33.5, 30.9, 27.7, 22.1; IR (neat, thin film) 2950,

2927, 2867, 1739, 1712 cm]; LRMS (70 eV, El) m/z 228 [M-HZOF, 140 [M-OBnF;

HRMS 1M+H1+ Calcd. for 0,6szoz; 246.1620 m/z. Observed 246.1631 m/z .

0
6 Condition B /U\/\/\
———“_—’ C02H

80 %

‘H NMR (CDCI3, 300 MHz): 62.41-245 (m, 2H), 2.31-2.36 (m, 2H), 2.11 (s,
3H), 1.56-1.62 (m, 4H); ”C NMR (CDCI3, 75 MHz): 6208.8, 179.1, 43.2, 33.7, 29.9,
240,229; IR (neat, thin film) 3455, 2939, 1714 cm"; LRMS (70 eV, El) m/z 144 M”,

126 WHO 1*.

O 0
Condition A m
g : 60% i -
, - o

1H NMR (CDCI3, 300 MHz): 62.41-245 (m, 2H), 2.31-2.36 (m, 2H), 2.11 (s,
3H). 1.56-1.62 (m, 4H); 130 NMR (CDC13, 75 MHz): 6208.8, 179.1, 43.2, 33.7, 29.9,
24.0, 22.9; IR (neat, thin film) 3455, 2939, 1714 cm"; LRMS (70 eV, 131) m/z 144 M+,

126 [M-H,or.

61

E. References

10.

11.

12.

l3.

14.

15.

16.

17.

18.

19.

Voet, D.; Voet, J. 6.; 2nd ed.; J. Wiley & Sons: New York, 1995, pp 704-713.
Johnson, M.; Carey, F.; McMillan, R. M. Essays Biochem. 1983, 19, 40.

Needleman, P.; Turk, J.; Jakschik, B. A.; Morrison, A. R.; Lefkowith, J. B. Annu.
Rev. Biochem. 1986, 55, 69.

Hayaishi, O. J. Biol. Chem. 1988, 263, 14593.
Samuelsson, B. Science 1983, 220, 568.

Samuelsson, B.; Dahlen, S. E.; Lindgren, J. A.; Rouzer, C. A.; Serhan, C. N.
Science 1987, 23 7, 1171.

Samuelsson, B. Drugs 1987, 33, 2.

Weissman, G. Sci. Am. 1991, 264, 84.

Ford-Hutchinson, A. W. Trends Pharmacol. Sci. 1991, 12, 68.
Taylor, G. W.; Clarke, S. R. Trends Pharmacol. Sci. 1986, 7, 100.

Capdevila, J. H.; Karara, A.; Waxman, D. J.; Martin, M. V.; Falck, J. R.;
Guenguerich, F. P. J. Biol. Chem. 1990, 265, 10865.

Oliw, E. H.; Guengerich, F. P.; Oates, J. A. J. Biol. Chem. 1982, 257, 3771.

Schwartzman, M. L.; Balazy, M.; Masferrer, J.; Abraham, N. G.; McGiff, J. C.;
Murphy, R. C. Proc. Natl. Acad. Sci. U. S. A. 1987, 84, 8125.

Proctor, K. G.; Falck, J. R.; Capdevila, J. C irc. Res. 1987, 60, 50.
Cashman, J. R.; Hanks, Dr, Weiner, R. I. Neuroendocrinology 1987, 46, 246.
Hammock, B. D., Personal communication.

Moghaddam, M. F.; Motoba, Ks, Borhan, B.; Pinot, F.; Hammock, B. D. Biochim.
Biophys. Acta 1996, 1290, 327.

Koert, U. Synthesis 1995, 115.

Cardillo, (3.; Orena, M. Tetrahedron 1990, 46, 3321.

62

 

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

Boivin, T. L. B. Tetrahedron 1987, 43, 3309.

R30, A. 8.; Paknikar, S. K.; Kirtane, J. G. Tetrahedron 1983, 39, 2323.
Andrey, O.; Landais, Y. Tetrahedron Lett. 1993, 34, 8435.

Micalizio, G. C.; Roush, W. R. Org. Lett. 2000, 2, 461.

Guindon, Y.; Soucy, F.; Yoakim, C.; Ogilvie, W. W.; Plamondon, L. J. Org.
Chem. 2001, 66, 8992.

Seepersaud, M.; Blumenstein, M.; Mootoo, D. R. Tetrahedron 1997, 53, 5711.
Osumi, K.; Sugimura, H. Tetrahedron Lett. 1995, 36, 5789.

Larsen, C. H.; Ridgway, B. H.; Shaw, J. T.; Woerpel, K. A. J. Am. Chem. Soc.
1999, 121, 12208.

Shan, W. R; Wilson, P.; Liang, W.; Mootoo, D. R. J. Org. Chem. 1994, 59, 7986.

Semmelhack, M. F.; Epa, W. R.; Cheung, A. W. H.; Gu, Y.; Kim, C.; Zhang, N.;
Lew, W. J. Am. Chem. Soc. 1994, 116, 7455.

Tamaru, Y.; Hojo, M.; Kawamura, S.; Sawada, S.; Yoshida, Z. J. Org. Chem.
1987,52, 4062.

Abeywickrema, A. N.; Beckwith, A. L. J.; Gerba, S. J. Org. Chem. 1987,52,
4072.

Mukaiyama, T.; Hayashi, M.; lchikawa, J. Chem. Lett. 1986, 1157.

Johnson, R. A.; Sharpless, K. B. in Catalytic Asymmetric Synthesis; 2nd ed.;
Oiima, 1., Ed; Wiley—VCH: New York, 2000, pp 357-398.

Johnson, R. A.; Sharpless, K. B. in Catalytic Asymmetric Synthesis; 2nd ed.;
Ojima, 1., Ed; Wiley-VCH: New York, 2000, pp 231-280.

Baldwin, J. E. J. Chem. Soc., Chem. Commun. 1976, 734.

Baldwin, J. E.; Cutting, J.; Dupont, W.; Kruse, L.; Silberman, L.; Thomas, R. C.
J. Chem. Soc., Chem. Commun. 1976, 736.

Mcintyre, 8.; Warren, S. Tetrahedron Lett. 1990, 31, 3457.

Nakata, T.; Schmid, G.; Vranesic, B.; Okigawa, M.; Smithpalmer, T.; Kishi, Y. J.
Am. Chem. Soc. 1978, 100, 2933.

63

 

 

 

 

39.

41.

42.

43.

45.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

Nakata, T.; Kishi, Y. Tetrahedron Lett. 1978, 2745.

Nicolaou, K. C. ,,Prasad C. V. C.; Somers, P. K. ,Hwang, C. K. J. Am. Chem. Soc.
1989, 111, 5330.

Nicolaou, K. C. ,Prasad, C. V. C. S,omers, P. K.; Hwang, C. K. J. Am. Chem. Soc.
1989, 111, 5335.

Suzuki, T.; Sato, 0.; Hirama, M. Tetrahedron Lett. 1990, 31, 4747.
Janda, K. D.; Shevlin, C. G.; Lerner, R. A. Science 1993, 259, 490.
Mukai, C.; Sugimoto, Y.; Ikeda, Y.; Hanaoka, M. Tetrahedron 1998, 54, 823.

Baskaran, S; Vasu, J. ,Prasad, R. ,Kodukulla, K.;Trivedi, G. K.; Chandrasekhar,
J. Tetrahedron 1996, 52,4515.

Gao, Y. ,Hanson, R. M. ,,Klunder J. M. ,Ko, S. Y.; Masamune, H.; Sharpless, K.
B. J. Am. Chem. Soc. 1987, 109, 5765.

Reed, L. A. ,Ito, Y. ,,Masamune S.; Sharpless, K. B. J. Am. Chem. Soc. 1982,
104, 6468.

Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277.

Nakagawa, 1.; Hata, T. Tetrahedron Lett. 1975, 1409.

Nakagawa, 1.; Aki, K.; Hata, T. J. Chem. Soc., Perkin Trans. 1 1983, 1315.
Pettersson, L.; Frejd, T. J. Chem. Soc., Chem. Commun. 1993, 1823.
Roush, W. R.; Russorodriguez, S. J. Org. Chem. 1987, 52, 598.

Gill, D. M. ,Pegg, N. A.; Rayner, C. M. J. Chem. Soc., Perkin Trans. 1 1993,
1371.

Gill, D. M.; Pegg, N. A.; Rayner, C. M. Tetrahedron 1996, 52, 3609.
Miyauchi, H.; Nakamura, T.; Ohashi, N. Bull. Chem. Soc. Jpn. 1995, 68, 1731.
De Lucchi, O.; Miotti, U.; Modena, G. Org. React. (N. Y.) 1991, 40, 157.
Padwa, A.; Gunn, D. E., Jr.; Osterhout, M. H. Synthesis 1997, 1353.

FOX. D. J.; House, D.; Warren, S. Angew. Chem, Int. Ed. 2002, 41, 2462.

64

 

 

 

 

 

 

59.

61.
62.

63.

65.

66.

67.
68.
69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

House, D.; Kerr, F.; Warren, S. Chem. Commun. 2000, 1779.

Fox, D. J.; House, D.; Kerr, F.; Warren, S. Chem. Commun. 2000, 1781.

House, D.; Kerr, F.; Warren, S. Chem. Commun. 2000, 1783.

Williams, D. R.; Phillips, J. G.; Barner, B. A. J. Am. Chem. Soc. 1981, [03, 7398.
Sivakumar, M.; Borhan, B. Tetrahedron Lett. 2003, 44, 5547.

Narayan, R. S.; Sivakumar, M.; Bouhlel, 13.; Borhan, B. Org. Lett. 2001, 3, 2489.
Travis, B.; Borhan, B. Tetrahedron Lett. 2001, 42, 7741.

Dorofeev, S. B.; Eletskii, A. V.; Smirnov, B. M. Dokl. Akad. Nauk. Sssr. 1981,
257, S92.

Ogle, R. A.; Schumacher, J. L. Process SafProg. 1998, 17, 127.
Koike, K.; Inoue, G.; Fukuda, T. J. Chem. Eng. Jpn. 1999, 32, 295.
Lee, D. G.; Chang, V. S. J. Org. Chem. 1978, 43, 1532.

lshii, Y.; Yamawaki, K.; Ura, T.; Yamada, H.; Yoshida, T.; Ogawa, M. J. Org.
Chem. 1988, 53, 3587.

Henry, J. R.; Weinreb, S. M. J. Org. Chem. 1993, 58, 4745.

Antonelli, E.; D'Aloisio, R.; Gambaro, M.; Fiorani, T.; Venturello, C. J. Org.
Chem. 1998, 63, 7190.

Sato, K.; Aoki, M.; Noyori, R. Science 1998, 281, 1646.

Kaneda, K.; Haruna, S.; Imanaka, T.; Kawamoto, K. J. Chem. Soc., Chem.
Commun. 1990, 1467.

Albarella, L.; Giordano, F.; Lasalvia, M.; Piccialli, V.; 5108, D. Tetrahedron Lett.
1995, 36, 5267.

Yang. H; Zhang, C. J. Org. Chem. 2001, 66,4814.
Sharpless, K. B.; Akashi, K. J. Am. Chem. Soc. 1976, 98, 1986.

Travis, B. R.; Narayan, R. S.; Borhan, B. J. Am. Chem. Soc. 2002, 124, 3824.

65

 

79.

80.

Hassall, C. H. Org. React. (N. Y.) 1957, 9, 73.

Panda, R.; Panigrahi, A. K.; Patnaik, C.; Sahu, S. K.; Mahapatra, S. K. Bull.
Chem. Soc. Jpn. 1988, 61, 1363.

66

 

CHAPTER II

THE ANNONACEOUS ACETOGENINS: STRUCTURE, BIOLOGICAL

ACTIVITY AND TOTAL SYN THESES

A. Historical background

The plant family annonaceae has proven to be a rich source of natural products
possessing wide variety of biologically and medicinally valuable properties.”
Traditionally, extracts of many species belonging to this family have been used in folk
medicines such as insecticides, fungicides, antiparasitics, antimalarials, emetics,
antitumor agents and as a cure for snake bites.5 Before the early 1980’s, phytochemical
Studies on these medicinal plants mostly involved the isolation of numerous secondary
metabolites including isoquinoline alkaloids, polyphenols, carbohydrates, lipids, proteins,
aromatic compounds, essential oils and terpenes.6 However, no reports of systematic
pharmacological studies aimed towards delineating the bioactive components of these

traditional folk therapeutics had been reported. In 1982, Jolad and co-workers discovered

uvaricin — a novel antitumor agent, from ethanol root extracts of Uvaria accuminata (a
member of the Annonaceae family).7 Uvaricin was isolated using bioactivity-gmded

fractionation and was shown to possess high in vivo potency as an antilukemic (P-388)
agent. Uvaricin demonstrated an activity of 157% test / control (T/C) at 1.4 mg / kg in the
PS test system. Activity in the PS test system is defined as an increase in the survival of
treated animals over that of controls resulting in a T / C of 125%. In the same report,

Jolad and colleagues disclosed the gross chemical structure of uvaricin — determmed

67

using 1H and 13C NMRs, IR and mass spectroscopic fragmentation pattern. Thus,
structurally, uvaricin was shown to be a C34 fatty acid derivative bearing a terminal
unsaturated lactone ring and two adjacent tetrahydrofuran (THF) rings flanked by
hydroxyl groups along the long aliphatic chain (Figure 11- l)

0‘“ 0‘“
OH OH

Figure II-l: Uvaricin — the first acetogenin isolated from Uvaria accuminata

(Annonaceae)

Since the discovery of uvaricin, its novel structure and promising bio-activity
triggered a large body of research in this area which has lead to isolation, structure

elucidation and in some cases, biological studies of over 400 related compounds — now
- 2. .8-11 .
termed as the annonaceous acetogemns. 5 The acetogenlns have emerged as a new

class of highly potent bioactive compounds, which quite likely were the active

ingredients of traditional folk medicines that originated from the Annonaceae family.

B. Structure and proposed biogenesis

poiymethylene spacer

 

 

 

 

 

 

 

 

 

 

 

1 O _
termmal
HO n D \ o y-methyI-y-Iactone
OH *
oligonuclear 0
core
p
O
unbranched
HO {Hg/\m hydrocarbon chain

 

 

 

 

 

 

Figure II-2: Generic structure of a binuclear acetogenin

68

Structurally, the Annonaceous acetogenins are polyunsaturated C32 or C34 fatty

acid derivatives that are combined with a 2-propanol unit and cyclized to generate THF
0r THP unit(s) along the long hydrocarbon chain (Figure “-2).12 The most commonly
found structural features include (a) a long aliphatic chain bearing a terminal methyl-
substituted a,B—unsaturated y-lactone moiety; (b) Up to three THF rings along the

hydrocarbon chain; (c) oxygen functionalities such as hydroxyl or acetoxy groups,

 

17,18-bis-epi-goniocin (tri-TH F) 2,4-cis-isomurisolenin (ketolactone
substituted mono-THF classical)

OH OH

      

cis-goniodonin (y-lactone modified- Muconin (nonclassical-THP)

mon-THF classical)

OH

     

15 OH

Mucoxin (nonclassical-
hydroxylated THF)

Figure “-3: Classification and representative structures of acetogenins

69

ketones, epoxides and /or double bonds.

Figure "-3 depicts representative structures of acetogenins with the class name
indicated in parentheses. The annonaceous acetogenins are most conveniently classified10

(based on the number and type of the cyclic ether core structure) into two broad groups,
viz., a) classical acetogenins — acetogenins comprising of none to three THF units along
with a y~lactone terminus. This class is subdivided as mono-THF, adjacent bis-THF,
nonadjacent bis-THF, tri-THF, non-THF ring containing acetogenins. Each subclass is
then further divided according to the nature of the terminal lactone ring. b) Nonclassical

acetogenins -— structurally, they can be further divided into two broad groups, viz., i) THP

ring containing acetogenins: mucocinl3 (Figure II-3) was the first nonclassical acetogenin

with a THP ring nonadjacent to a THF ring. A few more examples of this class of

acetogenins (not shown) are muconinl4 (THP ring adjacent to a THF ring), pyranicinl5

(mono-THP ring), jimenezin16 (hydroxylated THP ring adjacent to a THF ring). ii)

Hydroxylated THF ring containing acetogenins: mucoxin” (Figure 11-3) was the first

nonclassical acetogenin reported’ containing a hydroxylated THF ring. Two other

. . . . . 18
examples are - gomotriocmlO and donna1en1n.

It has been proposed that the THF, THP and epoxide rings in acetogenins arise

from isolated double bonds via epoxidation / cyclization events as shown in Figure 11-

3.4 . . . . .
4. The erythro / threo relationship between the carbmol centers and THF r1ngs arises

from the olefin geometries while the trans / cis relationship across the THF rings is a

' Four acetogenins were previously reported to possess hydroxylated THF rings,” however their

2
structures were proved erroneous and have been corrected.
70

.i—ui—
—— _ ——n

“w ‘W HO

lepoxidation lepoxidation

m/L;::::l7/F—\i3;m' OH /L;::::l7/F_\lA/F—\:_§
OH OH OH OH OH

adjacent - bis THF ring nonadjacent - bis THF ring

 

 

Figure “-4: Proposed biosynthetic pathways for two main classes of acetogenins

consequence of the facial selectivity of epoxidation event. Discovery of precursors with

appropriately positioned double bonds (for example, muridienins and chatenaytrienins -

proposed precursors for mono and bis-THF acetogenins respectively) along with semi

synthesis of THF units from double bond containing acetogenins have supported the

hypothesis that acetogenins are derived from C-32 (lacceroic) and C-34 (ghedoic) fatty

acids after enzyme mediated coupling with a three—carbon unit and epoxidation /

cyclization sequence.l2 Detailed biosynthetic studies on acetogenins have been thwarted

partly because of practical difficulties in establishing plant tissue cultures.

H

C. Biological activity

The annonaceous acetogenins exhibit high levels of pesticidal, antibacterial,

antimalarial, and antiparasitic properties. More interestingly, these compounds have been

shown to be highly cytotoxic antitumor agents in both in vitro as well as in vivo testing.

In addition, recently, some members of this family have been shown to possess the ability

71

to combat resistance in multidrug resistant (MDR) tumor cells and in pesticide-resistant

insectslg'z' Some bioactivity studies are briefly described below:

1. In vitro studies

In vitro assays have revealed some of the acetogenins to be among the most
potent cytotoxic agents known to date. In a recent review, Marshall et al. have
summarized22 (Table II-l) relative tumor growth inhibition abilities of representative

acetogenins compared to adriamycin (an anticancer drug currently in clinical use).

0

o o“

   

 

 

 

 

 

 

 

 

 

 

 

Sylvaticin Longitolicin
Compound Human tumor lines (cell culture)
A-549 (lung) MCF-7rbreast1 111291661611)
Asimocin 3.1 x 10‘12 29 x 10'12 <10’12
Bullanin 3.4 x 10‘l4 3.2 x 10'14 4.8 x 10'12
Sylvaticin <10‘8 3.8 x 10‘5 1,6 x 10"1 \
Longifolicin 1.1 x 10'6 1.2 x 10‘5 — \
L Adriamycin 2.4 x 104 1.0 x 10'2 l 3.8 x 10‘2 1

 

 

 

Table “-1: Relative tumor growth inhibition (ED50 mg / mL) for representative
acetogenins compared to adriamycin

These acetogenins were proven to be more potent than adriamycin in preliminary

assays. Besides being highly potent, acetogenins have also been shown to exhibit

72

selective cytotoxicity to tumor cells over other innocent cells.19 In addition, they have

also exhibited selectivities among various cancerous cell lines.”25

2. In vivo studies

Although extensive animal testing of acetogenins has not been done, the studies
so far have been promising. Early on, uvaricin (157% T/C at 1.4 mg/kg, vide infra)7,
rollinones (147% T/C at 1.4 mg/kg, Figure II-S) and asimicin (124% T/C at 25 pig/kg,
Figure Il-5)5 were shown to possess in vivo activity against 3PS (murine lymphocytic
leukemia). Bullatacin (Figure “-5) effective at 50 rig/kg against L121O (murine

leukemia) in normal mice was 300 times more potent than paclitaxel.26 Both bullatacin

and bullatalicin (effective at 1 mg/kg) were almost equivalent to cisplatin.3

   

O Bullatacin

Asimicin

O O 8
OH . OH O O
Rolltnone 0

Figure “-5: Some acetogenins that showed high in vivo cytotoxicity profiles

3. Activity against multidrug resistant (MDR) cells
The effect of acetogenin treatment on MDR tumor cells is demonstrated by
Oberlies and co-workers’ in vitro studies on wild—type and adriamycin resistant human
mammary adenocarcinoma cells (MCF-7/wt and MCF-7/Adr respectively)?“21 After
treatment of both the cell lines with bullatacin (1.0 ug/mL) for 48-h, MCF—7/wt cells

showed regrowth (comparable to the vehicle treated control) when fed fresh media; while

73

MCF-7/Adr cells did not. Thus, bullatacin was cytostatic to wild type cells but cytotoxic

to the drug resistant cells.

4. Pesticidal activity

Several classes of synthetic compounds such as chlorinated hydrocarbons,
organophosphates, carbamates etc. have been used as insecticides against cockroaches.
Their repeated use has resulted in development of resistance, which calls for new
insecticides with novel mode of action. Alali and co-workers conducted comparative

studies of dietary toxicities of acetogenins vs. conventional insecticides on insecticide

. . . 3 -
—-res1stant as well as suscepnble stralns of cockroaches.2 The acetogenms showed lower

or comparable LT50 values (number of days before death of 50% of the population at a

dose of 1000 ppm) in both the strains (Table “-2). Also, in a yellow fever larva assay 44

acetogenins were found to be highly potent.27

 

 

 

 

 

 

 

 

Compound Active Jwax (susceptible Muncie
type ingredient strains) (resistant strain)
parviflorin 0.8 1.2
natural
asimicin 2.3 3.8
chlorpyrifos 0.5 4.2
synthetic
L hydramethylnon 10.4 6. 1

 

 

 

Table “-2: LT50 values for German cockroach fifth instars

74

D. Mechanism of action

It has been proposed that acetogenins exert their bioactivity via inhibiton of two
target proteins: a) NADH-ubiquinone oxidoreductase (Complex 1) which is a membrane

26.28.29

bound protein in mitochondrial electron transport system. b) ubiquinone-linked

NADH oxidase present in the plasma membrane of only tumor cells.30 Both of these
binding events result in ATP deprivation, which eventually leads to apoptosis. Weiss and

128 found that annonin I (Figure II-6) was an excellent inhibitor of

co-workers in 199
Complex 1 (IC50 = 0.8 nM/mg for insecticidal mitochondria; IC50 < 0.1 uM/mg for bovine

heart muscle mitochondria). Using ESR spectroscopy they demonstrated that electron

transport from Complex I to ubiquinone is interrupted in the presence of annonin I.

OH OH OH 0

Figure “-6: Annonin I

Complex I has also attracted attention due to its implication in several diseases

including idiopathic Parkinson’s disease, maturity onset diabetes, stroke-like episodes
and Huntington’s disease.“ Some acetogenins (rolliniastatin-l and —2) were found to be
more potent inhibitors of complex I than piericidin-A (the most potent Complex I
inhibitor previously reported).32 Thus, it is of great interest to study the interactions

between Complex 1 and acetogenins in greater detail. While the precise mode of
complexation of acetogenins with their target proteins at molecular level is not known,
some general proposals have been put forth: a) Shimada et al. studied acetogenin

conformations within artificial liposomal membranes using techniques such as 1H

75

/-— Membrane /water interface

 

 

 

 

 

 

B Complex 1

 

 

 

 

 

 

 

a! THF-ring moiety with flanking OH groups
. Lactone-ring moiety
0: Phospholipid

Figure “-7: Model of bis-THF acetogenins interacting with complex I in mitochondrial

membrane (Ref. 35)

Intermolecular nOe’s and d1fferential calor1metr1c scanmng data.33 4 They proposed that

acetogenins containing mono-, adjacent and nonadjacent bis-THF units had the THF
rings residing at the glycerol head region of phosphatidylcholine serving as hydrophilic

anchor at the membrane interphase. The y—lactone ring diffuses in the membrane interior
to bind with the target site (Figure “-7).35 Depending upon the length of the linker unit

(alkyl chain that connects the lactone to the THF core) the lactone penetrates the lipid
bilayer to different depths. Thus, a given acetogenin molecule with specific spacer length
can adapt to the geometry of only a specific cell type, hence the observed selectivity in its

mode of action. b) Miyoshi and co-workers36 through their studies of 22 representative

acetogenins using submitochondrial particles as well as quantum chemical calculations

76

(MNDO-AMI), proposed that the stereochemistry across the THF rings is not crucial to
the activity and that the alkyl spacer connecting the y-lactone ring and THF core must be
of optimal length and flexibility for high potency. c) Studies on ion (Ca+2 and Mg”)
complexation abilities of acetogenins have shown that the complexation depends upon

the relative configuration of acetogenin molecule and nature of the ion. Considering that
. . . . . ’ 3
NADH -— ub1qu1none ox1doreductase 18 an Iron cluster enzyme, some researchers7 have

proposed iron-mediated complexes between acetogenins and the protein targets.

E. Structure-activity relationships

Because of a lack of understanding of the exact mode of binding of acetogenins

with the target proteins, researchers have not been able to design systematic SAR studies.

15.21.27.32,36,38,39 A1311 Ct al in

However, based on the work of several investigators, a

recent review“), suggested some generalizations on the SARs of the annonaceous

acetogenins which are summarized below. 1) The general order of potency of acetogenins
is: bis-adjacent THF > bis—nonadjacent THF > mono-THF > non-THF. The ring size
(THF vs. THP) and stereochemistry about the rings is practically inconsequential to the
potency and selectivity. 2) 01,13-unsaturated y-lactone is an essential feature and any
structural modifications lead to diminished activity. 3) The spacer length (distance
between the THF ring core and the lactone ring) is critical to the potency. For example,
Iii-carbon chain in mono- and bis—THF compounds is optimum. 4) Three hydroxyl
groups (two flanking the THF core and third somewhere along the long hydrocarbon

chain) are responsible for optimal polarity and topology needed for most effective

77

binding. Beyond four hydroxyl groups activity decreases significantly. 5) In general, a
ketone functionality instead of a hydroxyl group reduces the activity.
More recently, Miyoshi and co-workers have reported the first SAR study using a

series of synthetic acetogenin analogs, which were designed to delineate structural
features critical to acivity.36 Bullatacin is one Of the most active inhibitors of Complex 1.
Miyoshi et al. synthesized simplified analogs of bullatacin (Figure 118) and tested them
for NADHooxidase inhibition. The results (summarized in Figure [18) clearly indicated
that the inhibitory activity was completely lost when the THF core and the terminal
lactone ring were decoupled (G-J). Also, when the two ring moieties were used in

combination in various molar ratios, no synergistic enhancement of activity was observed

78

Inhibitor

1050 (nm)

 

 

 

A, m=10, n=9, Fl‘ = H2 = H

B, 171:7, n=6, R1: R2: H

C, "1:10, 0:9, R1: H, R2: COCH3

D, m=10, n29, R1: COCH3, R2: OOCH3

 

H, 11:1
1, 11:4
J, 11:10

12110.1)

A1.2 (10.1)
81.9 ($0.1)
C 2.0 (1:02)
018 (:2)

1.6 (10.1)

1.2 (a; 0.2)

4500 (:1.- 300)

H, >20,000
I, >20,000
J. 6200 (:1: 400)

Figure 11-8: NADH-oxidase inhibitory potencies of bullatacin analogs

(data not shown). Among the other modifications — bis-acetogenin (A), bis-acetogenin

With shorter linkers (B), reduced bis-acetogenin (E) and bis-acetogenin with inverted

lactone configuration (F) did not exhibit any perturbation in activity. However,

acetylation of the hydroxyl groups flanking the THF core (C and D) did result in slightly

79

reduced potency. Thus, it was concluded that the THF (with two flanking hydroxyl
groups) and lactone ring systems must be linked together for optimum activity. Since
variations in other functional groups did not lead to any significant change in enzyme
inhibition, the critical structure features of bullatacin or any further insights into precise
mode of binding remain undiscovered.

In separate studies reported earlier, the cytotoxicity of bullatacin against

carcinoma cells decreased significantly (about lOé-fold) upon saturation of the double
bond in the (LB-unsaturated y-lactone.”‘40 Curiously, in Miyoshi’s studies (vide supra)
analog E (reduced bis-acetogenin) did not show depletion in inhibitory activity compared

to bullatacin or analog A. Thus, whether or not the cytotoxicity profile of acetogenins

correlates to the inhibitory potency remains unclear.

F. Classical vs. nonclassical acetogenins

The annonaceous acetogenins due to their highly potent, selective cytotoxicity
and pesticidal activities especially against drug resistant tumor cells and insects are
increasingly being looked at as new generation antitumor therapeutics and pesticides.
Classical acetogenins have been and continue to be investigated in areas spanning
isolation, purification, structure elucidation, semi and total synthesis, bioactivity testing
and studies on mechanism of action. In recent years, nonclassical acetogenins with

unique structural features have emerged.‘0 Novel structures that offer new synthetic

challenges and promising bioactivity have prompted total syntheses of some of the THP

containing nonclassical acetogenins. In some cases, the originally proposed structure was

revised after the total synthesis.41 To our knowledge, however, none of the hydroxylated

80

THF containing nonclassical acetogenins have been synthesized or studied in any further

detail.

G. Total synthesis of the annonaceous acetogenins

Due to excellent biological and medicinal activities along with unique structural
features, the annonaceous acetogenins have attracted the attention of several synthetic
groups over the last two decades. Acetogenins, though found in a large number of plant
species, exist only in minute amounts as complex mixtures of related isomers. As a result,
the isolation and purification process is often tedious. On an average, about 10-20 mg of
material can be obtained form 15 kg of stem bark, which requires multistep seperation
involving partition extraction and chromatography on several different columns followed

by repetitive HPLC.42 Moreover, since acetogenins are often waxes or gums, their

structure elucidation using X-ray crystallography is not possible. Thus, total synthesis has
played an important role in this field of research. Synthetic materials have been obtained
in sufficient amounts for confirmation (in some cases revision) of proposed structures,
establisment of relative absolute configurations and for biological testing. In addition,

total synthesis has provided expeditious routes to obtain unnatural stereoisomers and

other simplified structural analogs of the natural products to gain insights into SARs.35'43

Acetogenins embody adjacent or nonadjacent polyether rings, which in the early years of
discovery were unique and challenging structural features from a synthetic point of view.
This triggered the development of several elegant methods to synthesize such polycyclic
substituted ether units and useful chiral building blocks. Thus acetogenins have served to

advance synthetic chemical methodologies.

81

In 1991, Hoye and coworkers reported the total synthesis of (+)«(36-epi)-ent-

uvaricin — the first of any member of the acetogenin family (Figure II-9).4""“15 This classic

 

 

 

   

   
  
 

O 1.Acetone,H" 1 01.1 ONa
Etc .011 2. LAH ¢ 'EOK Mom“ SAE
1 2. NaBH , MeOH
E‘O 0“ 3159" W O 3. MsCl,4E13N 2598'”
o ".1 4. Nal, acetone "-2 then DBU, 40 °C °
80% 4. DIBAL, Et20, 0 °C
15-30°/o
HO,“ OH OH 1 DMF, CSA
Megco, rt
1. TsCl (1 eq.), E13N . 2. A020, Py, rt
DMAP. 0 °c (CoHtolzCuU 3. MeOH, pTSA

A
'—

 

 

2.Amberlyst-15 4. TBDPSCI. DMAP

   

 

 

 

MeOH n Eth
30-40% HO OT, 09H“, 5. TsCl, DMAP, Py
11
"-5 "'5 6. TBAF, rt
H Pd(PPh3)4, 0111. Eth

1.Ll Z TMS ‘ rt, 36°/o

BF3,OEt2,-78°C O 0
2. TBAF 1M

CsHts 6 SP“

ACO

 

HO

1. Rh(PPh)3CI
H2

2. Oxone, MeOH
H20, 0 °C
3. PhMe, reflux
A00

"-10 (+)e(36-ep1)-ent-uvaricin
Figure “-9: The first total synthesis of an acetogenin, (+)-(36-epi)-ent-uvaricin
Synthesis involved a bi-directional approach to secure the bis-THF core of the molecule. _

The synthetic scheme is described in Figure “-9. Starting from (+)-diethyl tartrate II-l

derived diiodide II-2, E, E-bis allylic alcohol “-3 was obtained using Weiler dianion

82

alkylationi‘6 Sharpless asymmetric epoxidation of “-3 furnished the bis-epoxydiol “-4.

The two ends of C2 symmetric diol “-4 were distinguished by formation the
monotosylate, which was subjected to one-pot acid promoted acetonide cleavage, epoxide
opening reaction, to provide the C15 — C24 bis-THF core “-5. Alkylation of the tosylate
"-5 using excess lithium dinonylcuprate furnished intermediate II-6, which after
protective group manipulations was transformed into epoxide II-7. Lithium
trimethylsilylacetylide opening of epoxide II-7 provided alkyne “-8, which was coupled
to the vinyl iodide “-9 using the Sonogashira protocol. Enyne reduction, oxidation of
sulfide and thermal elimination of the resultant sulfoxide produced compound “-10 (in
total twenty eight steps), which after Mosher’s ester analysis and spectroscopic
comparison with the natural product was assigned to be a diastereomer of natural uvaricin
differing only at C36 stereocenter (II-10, Figure II-9).

After Hoye’s initial report, a large number of syntheses of natural acetogenins as
well their analogs have appeared in the literature. A few recent syntheses are cited

4754

here. Several reviews dedicated to the synthetic approaches have also been

publishedzz‘SS‘57 From a synthesis design point of view, acetogenins can be divided into
four well-defined domains, viz., the oligonuclear cyclic ether core, terminal y-methyl-y-
lactone moiety, an acyclic alkyl chain connecting the two cyclic domains and a long
unbranched hydrocarbon chain often containing oxygen functionalities. Several elegant
routes to construct and couple oligonuclear cyclic ether core and the terminal lactone unit
have been described in the total synthesis literature. The long hydrocarbon chain can be

easily incorporated using routine chemical transformations at an early or later stage in

83

synthesis. In most syntheses, the oligo-cyclic ether fragment is constructed first and then
is coupled to the terminal y—methyl-y-lactone ring with the appropriate linker.

The following sections describe representative total syntheses of the annonaceous
acetogenins. Since our own efforts have dealt with method development for highly regio-
and stereoselective synthesis of substituted THF rings, the synthetic strategies are
described focusing on the construction of cyclic polyether cores; synthesis of the terminal
lactone with an appropriate spacer and completion of the total synthesis is mentioned
briefly in some cases. The classification is based on strategies used for the construction of

the oligonuclear cyclic ether fragments.

1. Multiple intramolecular Williamson etherification strategy

Trost designed a versatile strategy to synthesize structurally related acetogenins
using intramolecular double Williamson etherification protocol to construct the bis —THF
core.58 Synthesis of one of the members, (+)-squamocin K is described in Figures “-10

and “-11.

      
  
  

    

 
  

   

 

 

 

 

HO“,
HO“. . 7 ><
"/‘OH '
multiple J‘OMS Asymmetric
intramolecularfi M597 Dihydroxylatio?
Williamson HO“, .
etherification ><
HO 9
9 9 9
"-11 (+)-squamocin K "-12 "-13

Figure Il-10: Trost’s synthesis of (+)-squamocin K (key retrosynthetic disconnections)

The total synthesis scheme in a forward sense is depicted in Figure “-11.

Standard functional group manipulations of known bis-homoallylic alcohol “-14

84

provided the Julia olefination precursors II-l7 and “-18. The desired E olefin II-13,

albeit obtained in a moderate selectivity (E : Z = 3 : 1), preferentially reacted in

subsequent asymmetric dihydroxylation reaction which obviated the need for seperation.

The bis-mesylate II-19 upon acetonide deprotection and exposure to base underwent

intramolecular displacement reaction to yield the bis—THF system “-20. Finally, the

butenolide ring was efficiently introduced using a ruthenium-catalyzed Alder—ene

1. ADmix-p
MGSOZNHZ
71°/o

  

 

 

0“ 11-14
94%

1. BuLi, then
82C!

2. (M6)2C(OM9)2

 

2. 5% Na(Hg)

 

1. Amberlyst
MeOH

2. t-BuOK
61%

 

 

 

   

><“”" ><O"'
0 H2, Pd/C O
HO 98% HO
1- Pth, '2 Doss-Martin
75 /o Periodinane
2. PhSOZNa 98%
94%
O.,,

    

 
   

 
 
   

 

Phso2 ".17
O

HO:,_ HO“. O
OH
1. EtOZC—=—_:—QH
Opnu(cOD)c1T

2. H2, Rh(PPh)3Cl
7 "'OH 7 "'OH

 

"-20 "-1 1

Figure II-ll: Trost's synthesis of (+) — squamocin K

59.60

protocol.

Thus, the total synthesis of (+)-squamocin K (II-11) was completed in 15

steps along the longest linear sequence. Since all the oxygenated stereocenters were

85

'

 

established using Sharpless asymmetric dihydroxylation reaction, by merely varying the
ligands and the role of oxygen functionalities (as electrophile and nucleophile during
etherification process), in principle, a variety of diastereomers can be easily accessed.

Marshall and co-workers in their elegant studies have shown that chiral

. . . - t
(Jr-oxygenated stannanes (II-21, Figure II-12) undergo stereospeCIfic rearrangement6| 6'

to produce y—oxy allylmetallic species (tin II-22 or indium “-23 depending on the
reagents), which upon addition to aldehydes provide syn or anti diols, II-24 and "-25
respectively. BF3-OEt2 mediated addition of y-oxy allyl stannane II-22 proceeds via an
acyclic transition state while y-oxy allyl indium species II-23 forms a cyclic transition
state to afford the corresponding diols as single diastereomers. This asymmetric allylation
of suitably functionalized aldehydes in combination with intramolecular Williamson

etherification reaction has been effectively used in the total syntheses of several

 

 

 

 

 

 

 

 

 

 

.BFa
”fl 0
s .SnBu BF3'OE12 s 0 OR .
/ " 3 ——————’ \ .. H OMOM
OMOM -78 °C, 5 min Bu3Sn OMOM 352' addition l H
"-21 "-22 H SnBu
3
1,3 rearrangement
InCl3 anti - diol acyclic transition state
-78 °C
__ n - dial
0 0” 11-24 3"
U 9' H l MOMO l ,0X OR
\R/\/OMOM ln-. ._ _
i \ —* 01/ '\OMOM "4,5
Clzln \ \—
H H 0R1
"'23 cyclic transition state

Figure "-12: Marshall's stereoselective 5152' addition approach to oxygenated THF

pI'CCUI'SOl'S

86

acetogenins.54 The synthesis of bullanin (II-33) is shown in Figure Il-l3. BF3-OEt2
mediated SEZ’ addition of chiral allyl stannane II-26 to enantiopure y,O-bis-alkoxy

aldehyde II-27 resulted in corresponding syn diol which was subsequently converted to
the tosylate II-28. Upon revealing the terminal aldehyde, II-29 was treated with chiral Ot-
oxy-allyl stannane II-30, this time in presence of InBr3 to afford the corresponding anti —
diol. Tosylation of the anti diol led to the bis-tosylate II-31. One-pot silyl deprotection /
tosylate displacement reaction of II-31 furnished the bis-THF scaffold 32. Sonogashira
coupling of alkyne II-32 with appropriate vinyl iodide followed hydrogenation and
global deprotection provided the natural product bullanin II-33. Either enantiomer of syn
(II-24) and anti (II-25) adducts is accessible by appropriate choice of chiral (it-alkoxy

allylic stannanes. In addition, the carbinol stereocenters in aldehyde of type II-27 are

OHCMOB” ores MOMO

 

 

 

 

 

Bu 1. reed ores
é \ “-27 , BUWOB”
”380 Buss” OMOM BF3.OE12, -78 °C, 77% Opt 5R2 062
- . , , 86°/
11 26 2 T820 py ° "-28 H1 = Ts; n? = TBS
BU3SFL,~ \
1- H2 ores MOMO 1. MOMO 3\\
Pd-C ? 7 "-30
BUWCHO 4,
2_ FCC (39‘ (592 OR2 lnBr3 EtOAC,
(62%) 1 2 92%
11-29 R = Ts; R = TBS 2. ngo, py
94%
QTBS MOMO QMQM
I i i . -. \
Bu \ . - \ 3% TBAF 8” 5 ”O ’O 3\\
(36‘ OR2 0620113 57% TBSO OMOM OMOM

11-32
11-31 R‘, R3 = Ts; R2 = res

1JMO
/

u n 3 O

Pd(0) 4% BU ." "I
__ 5 O O 11

' OH OH OH 0

Steps 11-33 bullanin

O

 

Figure II-13: Marshall’s synthesis of bullanin
87

established using Sharpless asymmetric dihydroxylation. Thus, this approach has proved

to be quite versatile due to availability of a strereodiverse pool of chiral building blocks.

2. Epoxide cascade strategy

In 1996, Hoye disclosed a highly efficient approach to build C-_, symmetric bis-

THF units in acetogenins using an ‘inside out’ epoxide cascade reaction (Figure II—l4).64

Bis-epoxy diol “-35 was synthesized in high yield and entiomeric purity from bis-allylic
diol “-34, which in turn was readily prepared from commercially available all trans
1,5,9-cyclododecatriene. The bis-epoxy diol intermediate (II-36, not isolated) generated
after Sharpless asymmetric dihydroxylation reaction spontaneously cyclized in an ‘inside
out’ fashion to form the bis—THF unit II-37. The corresponding C2 symmetric bis-
epoxide “-38 was desymmetrized by using trimethylsilyl acetylide as the limiting

reagent. Subsequent routine transformations including Sonogashira coupling protocol to

 

   

 

 

 

 

   

   

 

 

OH OH i—OTBDPS ‘— OTBDPS
O
| :10 1.L.i : TMS
(0.5 eq.)
SAE 1. TBDPSCI BFa-OEtz
L'(+)‘DET 2. AD-mix-B ”’0 2-1-1 : 071115
97% ee, 3, TFA (excess)
87% yield 1 85% BF3.OE12
O):
~ OTBDPS 0 3.192003, MGOH
OH OH TBDPSO
"-34 11-35 __ "-36 J 11-37 "-38
Not isolated
O
T880 0
l \
311-40 47

 

CzHis Pd(PPh3)QC|2, Cul, Et3N, 82%
steps

   

\1' NVW
HO

Figure II-l4: Hoye's synthesis of (+)-parviflorin
88

couple vinyl iodide II-40 with alkyne II-39 produced (+)-parviflorin (II-41) in a concise

manner involving longest linear sequence of 14 steps.

3. Biomimetic ‘naked carbon skeleton’ strategy

Townsend and Basak have hypothesized that polyether natural products might

arise from a cascade of hydroxy-directed syn-oxidative cyclizationsés'66 rather than the

traditional anti opening of polyepoxy alcohols (Figure II-IS).67 Syn cyclization involves

direct etherification of the olefin moiety, while anti cyclization involves intermediacy of
an epoxy alcohol, which undergoes intramolecular epoxide opening. The two events
result in formation of cyclic ethers bearing complementary stereocenters at the points of

cyclization

 

Rz syn oxidative mm
fl=< 4- R .
O 13' Ot-FE

 

R
OH 8' RE cyclization
hydroxy directed
epoxidation
Flz anti cyclization HZ
R C -‘ 4 R O“. "I
OH R" 0 '65 F1' 0415

Figure II-lS: Syn and anti oxidative cyclizations of hydroxy olefin

Inspired by this proposal, McDonald and co-workers have explored metal
mediated syn-oxidative polycyclization reaction of hydroxypolyenes.68 A representative

Optimized result in shown in Figure “-16. Upon exposure to dichloroacetyl perrhenate
reagent, hydroxytriene II-42 oxidatively cyclized to afford a mixture of products II-43
and “-44 in a ratio favoring (4 :1) the expected all syn selective cyclized product II-43
over the trans, trans, cis (syn, syn, anti cyclization) product "-44. The stereochemistry of

tris~THF core of “-43 matches that of an acetogenin goniocin II-45. Although several
89

 

R -. __ —— —"
HO H C121125
"-42
(ClgCHCOflRGOg
(0120H00120, 01-12012
63%38 :39: (4 : 1)

 

 

 

  

H H
\ .‘O ' OH
H - OH + R '
0121125 0 O (312st
"-43 all trans product "-44 trans, trans, cis product

Figure Il-l6: McDonald's biomimetic oxidative cyclization strategy

stereogenic centers are set during the oxidative cyclization reaction, this approach has not
been popular for use in total synthesis. One of the reasons could be that efficiency and
selectivity of cyclization decreases with increase in polyene chain length. This has been
partly attributed to possible chelation effects of earlier formed ether rings with
alkoxyrhenium intermediates as the cyclization proceeds. Along similar lines, in presence

of oxygen or any other chelating functionality elsewhere in the molecule, stereochemical
outcome of cyclization is not predictable as indicated by Sinha’s studies69 (not shown) on
goniocin synthesis. Also, depending upon the double bond substitution pattern, the

reactive conformation in cyclization event is not always predictable. Thus, the overall

complexity of the approach has prevented its extensive use in total synthesis of

acetogenins.

4. Step-growth oligomerization strategy

Casiraghi has used a vinylogous aldol condensation reaction of heterocyclic silyloxy
dienes (Figure II-17) to prepare oligomeric THF systems found in acetogenins?”71 Initial
condensation of the diene with chiral aldehyde “-46 in the presence of BF3°OEt2,

afforded the adduct “-47 after hydrogenation. After suitable functional groups
90

manipulations (Figure lI-l7), 2-acetoxytetrahydrofuran Il-49 was obtained. The next
condensation step involved Lewis acid mediated anomeric activation of “-49 followed
by C-glycosidation type coupling thus generating only the trans isomer across the THF
ring. However both erythro and threo isomeric products II-50a and II-50b were formed
in equal proportions (separable prior to hydrogenation). Repetition of the same sequence
(shown only for II-50a) yielded bis-THF units as a diasteromeric pair (II-51 and II-52).

This approach has also not been used in total synthesis design probably because of the

inevitable formation of mixtures.

 

 

 

 

 

 

O 0 0 OH OH
/\)LH 1. \ / ‘SlFia MO 1.aq.AcOH Rifo
O . +- O
~83 .888 so 28
“-46 ' 2' 11-47 3- a 4 11-48
OTBS R8310 0
1.TBS-Cl KEG—7‘ 1. U
OA 4
2 DIBAL-H c TBSOTf
' 2. hydrogenation
3' ACZO "'49 ll-50a : ll-50b = (1 :1)
ores OTBS
O, O O o, ...O O
+
"-508 “-5013
Reiterate
ores I ‘ ores 1
0,, ‘ 0., . 30 O 0!. ' or. O O
"-51 "-52

Figure II-l7: Casiraghi‘s iterative vinologous aldol reaction strategy

5. Sequential, modular strategy

In general acetogenins containing adjacent bis-THF core are the most potent of

all.10 A step towards systematic SAR studies of this class of acetogenins would be

91

development of an expeditious way to construct libraries of stereoisomeric bis - THF

cores with appropriate functional group handles for further elaboration.

vans
__ hro
RM Re207 9, [2+2] R 9””
.—.——_——.> O" {,0 ———-—> - ’0 ....___,____.. H
OH H, R 115‘s R 0 Files R ., l
9 ’OH
_ 015
Q___ 0 threo
R \ VO(acac)2 R O | O ‘H, All/R1
______.. 3 ;. .__. R _“——*
OH R O ' 03 Fl 90) 1 R a
l 1'BUOOH QC EU 0 OH

Figure “-18: Proposed mechanisms for metal mediated oxidative cyclization of hydroxy

olefins

Sinha and Keinan have developed modular strategies for such library synthesesi3

using a combination of the following chemical transformations: a) metal mediated

stereospecific oxidative cyclization of 4-alkenols (Figure Il—18) —- ReZO7 mediated
cyclization-[2'74 generated syn while VO(acac)2 formed anti oxidative products. Thus, by

appropriate choice of the metal oxidant, two diastereomeric THFs could be obtained from
a single hydroxy olefin. b) Sharpless asymmetric dihydroxylation and c) Mitsunobu
inversion of chiral alcohols.

Figure “-19 depicts a small library synthesis of bis-THF cores (II-S6-II-63) by
combined use of the above-mentioned protocols. The starting chiral unsaturated hydroxy-
lactone 11-53 was prepared from the corresponding olefin precursor (not shown) via
asymmetric dihydroxylation reaction. Thus, all four stereoisomers of II-53 were equally
accessible. Treatment of “-53 with RezO7 or VO(acac)2 generated corresponding trans
(II-54) or cis (II-55) mono-THF products in high (90%) diastereoselectivity. Reiteration

of the sequence along with Mitsunobu inversion of the secondary hydroxyl stereocenter

92

  
   

‘ c VO(acac)2,TBHP, CH2012

jg d i) 4-nitrobenzoic acid, DEAD
-‘ PPh3, ii)aq. KOH iii) 3N HCI
e R8207, H5106, CH2012

a R9207, Iutidine, CH2012
0 H2, Lindlar's catalyst

 

OH
\ C H
"-53 10 21H “-55
//H21C10|H(321 10

        

         

v. C1QH21HO C10H21 HO C10H21HO‘“ C10H21HOW C10H21HOW ClOH21HOV' CtOH21HO‘“ C10H21

"-56 "-57 "-58 “-59 "-60 “-61 "-62 "-63
Figure “-19: Sinha and Keinan’s library synthesis of bis - THF core units

afforded eight isomeric bis — THF units (II-56 to II-63). In a similar manner, the
remaining 56 isomers were synthesized and some of them were used in total syntheses of
asimicin, bullatacin, trilobacin, rolliniastatin and solamin.

Koert and co-workers have used another modular strategy (Figure 11-20) to

sequentially assemble oligo-THF units-’5’76 Their approach involves the stereoselective

addition of a Grignard reagent of type II-65 or its organozinc counterpart II-70 to
enantiopure mono-THF aldehydes such as II-64. The Grignard addition proceeded via a
chelation controlled transition state to generate the adduct “-66 in high
diastereoselectivity. On the other hand, Lewis acid mediated organozinc addition
afforded the Felkin—Ahn product II-71, also in very high diastereoselectivity. Each of the
adducts II-66 and Il-71 were transformed to the corresponding bis-THF units II-68 and

II-73 via the intermediacy of epoxy alcohols, II-67 and [1-72.
93

Bng/—_\;‘—\
_/ 5 {l d O m
'u -, .- 1. ACOH, H O, THF
TBDPSO O o "'65 x i O o 2 a

 

 

_ TBDPSO H06
"-64 CuBr.SM92 "-66 2. MsCl, py
$2; 92 :8 3. K2003, MeOH
O

 

 

AcOH repeat
TBDPSO ., ., . TBDPSO .
’o 9 O o

HO O 700/0 OH OTBDPS OH
"-67 "-68

"-69
H IZn/‘H
/J},\< "-70 OX0

. 1. AcOH, H20, THF
TBDPSO

 

 

 

o b TBDPSO "'0 H05 4
35,032 u 71 X 2. MsCl. py
"'64 ds = 95 : 5 ' a. K2003, MeOH
54°/o
AcOH M repeat
0 HO ’0 530/, o o o o o
"—72 "-73 OH OTBDPS "-74 OH

Figure II-20: Koert's modular strategy to construct bis — and tris — THF system

Reiteration of the same sequence provided higher THF units “-69 and “-74
following the same mechanism. Since all possible stereoisomers of reactants were
available, a series of stereoisomeric THF systems could be generated. However, a
limitation of this strategy is that the level of diastereoselection in both, the chelation-
controlled and the Felkin—Ahn addition depends on whether the chirality of the

organometallic species is matched or mismatched with respect to the facial selectivity of

the aldehyde.

6. Miscellaneous

Jacobsen and co-workers synthesized muconin — a THP ring containing

nonclassical acetogenin — using an Ireland—Claisen rearrangement and ring closing

metathesis as key transformations to construct the THF-THP core (Figure Il—Zl).77

94

 

1. TEMPO, NaOCl QAcozH

 

 

 

 

 

H25C12\;/\OH 2. MgBFQOEtg H25C12N
MOMo - fi" i _
"_75 H2C—CHMgBr MOMO f OTMS
3. NaH, lCHgCOQNa "'76
; 68% H23C11 / O
l ,o
t MemoH ti
9 \ OPBB
OH 0 J
‘ TMSO + H 1.Cr-(S,S)-salen " "-80
OPBB T;
\ OJV 2. TFA o OPBB
"-77 We "-78 recrystallization ".79

55%, 99% ee

E4020 / steps \ \ / 1. Schrock's catalyst
H25C12' O“. O OF’BB a H25C12 0v. 0 OPBB fl.

 

 

MOMO "-81 T836 "-82 2. H2 / Pd-C
O
\\ PhS
4i O
TBSO
H25012m0i‘l "-84 4
. O 0

steps

 

resc') "-83
Figure II-21: Jacobsen's synthesis of muconin

The chiral building blocks, viz., diol “-75 and dihydropyran II-79 were obtained
from inexpensive, racemic materials, commercially available in bulk quantities. Thus,
racemic tetradecene oxide (not Shown) upon hydrolytic kinetic resolution (HKR) protocol

earlier developed in their laboratories, using chiral Co (S,S)-salen complex afforded the
enantioenriched diol “-75 (> 99% ee) in good yields.78 Also, asymmetric hetero-Diels-
Alder reaction of diene II-77and dienophile II-78 catalyzed by Cr—(S,S) salen furnished
the dihyropyran “-79 (> 99% ee after recrystallization) in acceptable yields.79

Esterification of chiral acid II-76 with alcohol II-79 set the stage for the Ireland—Claisen
rearrangement, which generated intermediate II-81. Transformation of carboxylic acid
II'81 to bis-allyl ether “-82 and the subsequent RCM reaction installed the THF-THP

scaffold II-83 (after hydrogenation of the RCM product). The butenolide ring was

95

 

ll-S-l

m:

Ti:

incorporated in the final stages via organozinc mediated addition of the terminal alkyne
“-84 to the aldehyde generated from "-83. Muconin II-85 was thus synthesized in over
thirty-six steps.

Tanaka recently reported a straightforward, versatile strategy for construction of

adjacent bis-THF units (Figure II-22).80 Starting a-tertahydrofuranic aldehydes of type
II-86 were readily accessible using a method developed in the same laboratory.81 Zinc

mediated asymmetric alkynylation82 of “-86 with alkyne "-87 using (lS,2R)-N-

methylephedrin (NME) as a chiral auxillary, provided alkynol “-88 in excellent
diastereoselectivity which was manipulated in two different ways (a and b). Path 3
involved transformation of the 1,2 diol functionality in “-88 to epoxyalcohol II-89 which
spontaneously cyclized in a S-exo-tet mode to yield the bis-THF unit II-90. Along path b,
the roles of oxygen functionalities were switched. Thus, intramolecular Williamson
etherification of tosylate "-91 lead to diastereomeric bis-THF core II-92. Since antipodes

of all the chiral materials were available, various diastereomeric bis—THF units could be

96

 

 

Ph
Ph
9%0 b 94
H C fl é/VH'W / i O a
2512; 0“ CH0 ”5&va H25012W

 
      

' Zn OT 3 g . .
OTBS "-86 (1S(, 23)2-NME OTBS HO a ;. trig/Pd-C OTBS 6U
Eth, 97% "-88 - ”SCVPY "-89
. 3. K2003
dr > 32.1 Me OH
1. H2, Pd'C, Et3N 570/0
2. p-TsCl, py b
3. H2, Pd-C
4. NaH, 70%
_ H
HZSCIZM M H25C12M/OH H25012 . O“. , .
Ores OH Ores Ors Ores OH
"-92 "-91 "-90

Figure II-22: Tanaka's stereodivergent strategy for construction of adjacent bis-THF

systems

accessed efficiently. In principle, this approach can be further extended to construct
oligomeric THF cores.
Evans has utilized the temporary silicon-tethered (TST) ring closing metathesis

(RCM) method developed earlier in their laboratories,83 for the synthesis of mucocin

(Figure 11.23).52 The appropriately functionalized THP and THF fragments “-96 and

"-98 respectively, were obtained from a common chiral epoxide II-93. THP ”-96 was
Synthesized using highly diasteroselective, reductive bismuth tribromide mediated
CYclization protocol (II-95 to II-96) developed in their laboratories. Cobalt (II) catalyzed
oxidative cyclization to construct trans THF II-98 also proved highly stereoselective.
Fully functionalized fragments “-96 and II-99 were tethered by treatment of "-96 with

excess iPrZSiClz, washing off the excess reagent and then introducing II-99 in the same

97

 

 

ores eier3

0., Hon. t - HO,
M Steps—' ' mteps O BUMeZSIH 41A
\
\ HO ' 93% 0 \

 

 

 

 

 
 

 

 

OH HO ; 2 ;
"-93 1, pMBOH 11-94 OPMP 9 "-95 0PMP (dsz19:1) 9 "-96 (amp
DIAD, PPh3
2' WMgBr +
CuCN, 72%
OH Co(modp)2. 02 steps
/ —¢ . OH ———> .
/ / v '/
teuOOH 83% 0 5
PMBO _ ’ p - HO -
ll 97 (ds 219:1) MBO II 98 "‘99
1. 68, 'PrQSiClg
.3: CHQCIQ, lmld.
TBSOCI “”0 0 then 71, 74%
1. HF / MeCN 2. Grubb's
1 9 5 0 catalyst, 83%
2. TsNHNH2 0.8,/0 "-1 oo
NaOAC lpr/ ‘Ipr

 

9 OH OH O
"-101 mucocin

Figure II-23: Evans' synthesis of mucocin

pot. RCM reaction of the tethered product (not shown) furnished fully assembled
intermediate II-100, which after cleavage of the silyl tether and enyne reduction provided
muconin II-101. This strategy being highly convergent, offers avenues for structural
diversity in the two cyclic ether units to be coupled.

In conclusion, the annonaceous acetogenins have proven to be one of the most
potent classes of cytotoxic antitumor agents. More interestingly, they have shown high
Potency against multidrug resistant tumor cells and pesticide resistant insects. In spite of
the promising biological activity, this class of natural products remains under-explored in
area of lead deveIOpment for pharmacological applications. Synthetic chemists can
contribute to this area by design and development of rapid syntheses and hi gh-throughput

screening of libraries of constitutional and stereoisomers of acetogenins and their

98

synthetic analogs. Our studies on the synthesis of mucoxin —— a novel nonclassical

acetogenin are described in chapters III and IV.

99

 

H. References

10.

ll.

12.

l3.

l4.

Cave, A.; Figadere, B.; Laurens, A.; Cortes, D. in Progress in the Chemistry of

Organic Natural Products; Herz, W., Kirby, G. W., E., M. R., Steglish, W.,
Tamm, C., Eds; Springer-Verlag: New York, 1997; Vol. 70, pp 81-287.

Zeng, L.; Ye, Q.; Oberlies, H.; Shi, G.; Gu, Z.-M.; He, K.; McLaughlin, J. L. Nat.
Prod. Rep. 1996, I3, 275.

Gu, Z.-M.; Zhao, G.-X.; Oberlies, N. H.; Zeng, L.; McLaughlin, J. L. Rec. Adv.
Phytochem. 1995, 29, 249.

Fang, X. P.; Rieser, M. J.; Gu, Z. M.; Zhao, G. X.; McLaughlin, J. L. Phytochem.
Anal. 1993, 4, 27.

Rupprecht, J. K.; Hui, Y. H.; McLaughlin, J. L. J. Nat. Prod. 1990, 53, 237.

Leboeuf, M.; Cave, A.; Bhaumik, P. K.; Mukherjee, B.; Mukherjee, R.
Phytochemistry 1982, 21, 2783.

Jolad, S. D.; Hoffmann, J. J.; Schram, K. H.; Cole, J. R.; Tempesta, M. S.; Kriek,
G. R.; Bates, R. B. J. Org. Chem. 1982,47, 3151.

Zafra-Polo, M. C.; Gonzalez, M. C.; Estornell, E.; Sahpaz. S.; Cortes, D.
Phytochemistry 1996, 42, 253.

Zafra-Polo, M. C.; Figadere, B.; Gallardo, T.; Tormo, J. R.; Cortes, D.
Phytochemistry 1998, 48, 1087.

Alali, F. Q.; Liu, X.-X.; McLaughlin, J. L. J. Nat. Prod. 1999, 62, 504.

Johnson, H. A.; Oberlies, N. H.; Alali, F. Q.; McLaughlin, J. L. in Biologically
Active Natural Products: Pharmaceuticals; Cutler, S. J ., Cutler, H. G., Eds; CRC

Press: Boca Raton, FL, 2000, pp 173-183.

Gleye, C.; Raynaud, S.; Hocquemiller, R.; Laurens, A.; Fourneau, C.; Serani, L.;
Laprevote, 0.; Roblot, F.; Leboeuf, M.; Fournet, A.; De Arias, A. R.; Figadere,

B.; Cave, A. Phytochemistry 1998, 4 7, 749.

Shi, G. E.; Alfonso, D.; Fatope, M. 0.; Zeng, L.; Gu, Z. M.; Zhao, G. X.; He, K.;
Macdougal, J. M.; Mclaughlin, J. L. J. Am. Chem. Soc. 1995, 117, 10409.

Shi, G.; Kozlowski, J. F.; Schwedler, J. T.; Wood, K. V.; MacDougal, J. M.;
McLaughlin, J. L. J. Org. Chem. 1996, 61, 7988.

100

15.

I6.

17.

18.

I9.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

Alali, F. Q.; Rogers, L.; Zhang, Y.; McLaughlin, J. L. Tetrahedron 1998, 54,
5833.

Chavez, D.; Acevedo, L. A.; Mata, R. J. Nat. Prod. 1998, 61, 419.

Nonfon, M.; Lieb, F.; Moeschler, H.; Wendisch, D. Phytochemistry 1990, 29,
1951.

Jiang, Z.; Chen, R.—Y.; Chen, Y.; Yu, D.-Q. J. Nat. Prod. 1998, 61, 86.

Oberlies, N. H.; Jones, J. L.; Corbett, T. H.; Fotopoulos, S. S.; Mclaughlin, J. L.
Cancer Lett. 1995, 96, 55.

Oberlies, N. H.; Croy, V. L.; Harrison, M. L.; McLaughlin, J. L. Cancer Lett.
1997.115, 73.

Oberlies, N. H.; Chang, C. J.; McLaughlin, J. L. J. Med. Chem. 1997, 40, 2102.
Marshall, J. A.; Hinkle, K. W.; Hagedorn, C. E. Isr. J. Chem. 1997, 37, 97.

Alali, F. Q.; Kaakeh, W.; Bennette, G. W.; McLaughlin, J. L. Econ. Entomol.
1998, 91, 641.

Hopp, D. C.; Zeng, L.; Gu, Z. M.; McLaughlin, J. L. J. Nat. Prod. 1996, 59, 97.

Hopp, D. C.; Zeng, L.; Gu, Z. M.; Kozlowski, J. F.; McLaughlin, J. L. J. Nat.
Prod. 1997, 60, 581.

Ahammadsahib, K. 1.; Hollingworth, R. M.; Mcgovren, J. P.; Hui, Y. H.;
Mclaughlin, J. L. Life Sci. 1993, 53, 1113.

He, K.; Zeng, L.; Ye, Q.; Shi, G. E.; Oberlies, N. H.; Zhao, G. X.; Njoku, J.;
McLaughlin, J. L. Pestic. Sci. 1997, 49, 372.

Londershausen, M.; Leicht, W.; Lieb, F.; Moeschler, H.; Weiss, H. Pestic. Sci.
1991, 33, 427.

Lewis, M. A.; Amason, J. T.; Philogene, B. J. R.; Rupprecht, J. K.; Mclaughlin, J.
L. Pestic. Biochem. Physiol. 1993, 45, 15.

Morre, D. J.; Decabo, R.; Farley, C.; Oberlies, N. H.; Mclaughlin, J. L. Life Sci.
1994, 56, 343.

Weiss, H.; Friedrich, T.; Hofhaus, G.; Preis, D. Eur. J. Biochem. 1991, I97, 563.

101

32.

33.

34.

35.

36.

37.

38.

39.

41.

42.

43.

45.

47.

48.

49.

Esposti, M. D.; Ghelli, A.; Ratta, M.; Cortes, D.; Estornell, E. Biochem. J. 1994,
301, 161.

Shimada, H.; Grutzner, J. B.; Kozlowski, J. F.; McLaughlin, J. L. Biochemistry
1998, 37, 854.

Shimada, H.; Kozlowski, J. F.; Mclaughlin, J. L. Pharmacol. Res. 1998, 37, 357.

Kuwabara, K.; Takada, M.; Iwata, J.; Tatsumoto, K.; Sakamoto, K.; Iwamura, H.;
Miyoshi, H. Eur. J. Biochem. 2000, 267, 2538.

Miyoshi, H.; Ohshima, M.; Shimada, H.; Akagi, T.; Iwamura, H.; McLaughlin, J.
L. Biochim. Biophys. Acta 1998, I365, 443.

Hoppe, R.; Flasche, M.; Scharf, H. D. Tetrahedron Lett. 1994, 35, 2873.

Gu, Z. M.; Zeng, L.; Fang, X. P.; Colmansaizarbitoria, T.; Huo, M.; Mclaughlin,
J. L. J. Org. Chem. 1994,59, 5162.

Landolt, J. L.; Ahammadsahib, K. I.; Hollingworth, R. M.; Barr, R.; Crane, F. L.;
Buerck, N. L.; McCabe, G. P.; McLaughlin, J. L. Chem-Biol. Interact. 1995, 98,

1.

Mikolajczak, K. J.; Madrigal, R. V.; Rupprecht, J. K.; Hui, Y. H.; Liu, Y. M.;
Smith, D. L.; McLaughlin, J. L. Experentia, 46, 324.

Takahashi, S.; Maeda, K.; Hirota, S.; Nakata, T. Org. Lett. 1999, I, 2025.

Zhao, G. X.; Miesbauer, L. R.; Smith, D. L.; Mclaughlin, J. L. J. Med. Chem.
1994, 37, 1971.

Keinan, E.; Sinha, A.; Yazbak, A.; Sinha, S. C.; Sinha, S. C. Pure Appl. Chem.
1997, 69, 423.

Hoye, T. R.; Suhadolnik, J. C. J. Am. Chem. Soc. 1985, 107, 5312.

Hoye, T. R.; Hanson, P. R.; Kovelesky, A. C.; Ocain, T. D.; Zhuang, Z. P. J. Am.
Chem. Soc. 1991, 113, 9369.

Huckin, S. N.; Weiler, L. J. Am. Chem. Soc. 1974, 96, 1082.
Dixon, D. J.; Ley, S. V.; Reynolds, D. J. Angew. Chem, Int. Ed. 2000, 39, 3622.
Dixon, D. J.; Ley, S. V.; Reynolds, D. J. Chem. Eur. J. 2002, 8, 1621.

Hu, T.-S.; Yu, Q.; Wu, Y.-L.; Wu, Y. J. Org. Chem. 2001, 66,853.
102

50.

5].

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

65.

66.

67.

68.

69.

70.

Harcken, C.; Bruckner, R. New J. Chem. 2001, 25, 40.
Hoppen, S.; Baurle, S.; Koert, U. Chem. Eur. J. 2000, 6, 2382.

Evans, P. A.; Cui, J.; Gharpure, S. J.; Polosukhin, A.; Zhang, H.-R. J. Am. Chem.
Soc. 2003, 125, 14702.

Jiang, H. Ph. D. Thesis, University of Virginia, 2000.

Marshall, J. A.; Piettre, A.; Paige, M. A.; Valeriote. F. J. Org. Chem. 2003. 68,
8290.

Koert, U. Synthesis 1995, 115.
Hoppe, R.; Scharf, H. D. Synthesis 1995, 1447.

Casiraghi, G.; Zanardi, F.; Battistini, L.; Rassu, G.; Appendino, G. Chemtracts
1998, 11, 803.

Trost, B. M.; Calkins, T. L.; Bochet, C. G. Angew. Chem, Int. Ed. 1997, 36, 2632.

Trost, B. M.; Shi, Z. P. J. Am. Chem. Soc. 1994, 116, 7459.

Trost, B. M.; Muller, T. J. J.; Martinez, J. J. Am. Chem. Soc. 1995, 117, 1888.
Marshall, J. A. Chem. Rev. 1996, 96, 31.

Marshall, J. A.; Welmaker, G. S.; Gung, B. W. J. Am. Chem. Soc. 1991, 113, 647.
Marshall, J. A.; Hinkle, K. W. J. Org. Chem. 1995, 60, 1920.

Hoye, T. R.; Ye, Z. X. J. Am. Chem. Soc. 1996, 118, 1801.

Townsend, C. A-; Basak, A. Tetrahedron 1991,47, 2591.

Koert, U. Angew. Chem, Int. Ed. 1995, 34, 298.

Cane, D. E.; Celmer, W. D.; Westley, J. W. J. Am. Chem. Soc. 1983, 105, 3594.
Towne, T. B.; McDonald, F. E. J. Am. Chem. Soc. 1997, 119, 6022.

Sinha, S. C.; Sinha, A.; Sinha, S. C.; Keinan, E. J. Am. Chem. Soc. 1997, 119,
12014.

Zanardi, F.; Battistini, L.; Rassu, G.; Pinna, L.; Mor, M.; Culeddu, N.; Casiraghi,
G. J. Org. Chem. 1998, 63, 1368.

103

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

Casiraghi, G.; Zanardi, F.; Rassu, G. Pure Appl. Chem. 2000, 72, 1645.
Tang, S. H.; Kennedy, R. M. Tetrahedron Lett. 1992, 33, 5303.

Tang, S. H.; Kennedy, R. M. Tetrahedron Lett. 1992, 33, 5299.
Kennedy, R. M.; Tang, S. Tetrahedron Lett. 1992, 33, 3729.

Koert, U.; Stein, M.; Harms, K. Tetrahedron Lett. 1993, 34, 2299.
Koert, U.; Wagner, H.; Pidun, U. Chem. Ber. 1994, 127, 1447.

Schaus, S. E.; Branalt, J .; Jacobsen, E. N. J. Org. Chem. 1998, 63, 4876.

Tokunaga, M.; Larrow, J. F.; Kakiuchi, F.; Jacobsen, E. N. Science 1997, 277,
936.

Schaus, S. E.; Branalt, J.; Jacobsen, E. N. J. Org. Chem. 1998, 63, 403.

Maezaki, N.; Kojima, N.; Tominaga, H.; Yanai, M.; Tanaka, T. Org. Lett. 2003,
5, 1411.

Maezaki, N.; Kojima, N .; Asai, M.; Tominaga, H.; Tanaka, T. Org. Lett. 2002, 4,
2977.

Frantz, D. E.; Fassler, R.; Tomooka, C. S.; Carreira, E. M. Acc. Chem. Res. 2000,
33, 373.

Evans, P. A.; Murthy, V. S. J. Org. Chem. 1998,63, 6768.

104

CHAPTER III

SYNTHESIS OF THE LEFT HAND FRAGMENT (C 12-C34) OF MUCOXIN AND
PRELIMINARY STUDIES ON ITS COUPLING WITH THE RIGHT HAND

FRAGMENT

A. Introduction

The annonaceous acetogenins are C32 or C34 fatty acid derivatives originating
from the plant family annonaceae found in tropical and sub-tropical regions. In recent
years, this class of bioactive compounds has captured the attention of researchers in the
chemical, biological and medicinal sciences due to their high potency (sub—nanomolar
1CSO values) and selective cytotoxicity profiles against a variety of human tumor cell lines

including multi—drug resistant tumor cells (Chapter II)” Classically, the acetogenins

comprise of one or more 2,5—disubstituted THF rings along the long fatty acid chain.

 

Figure III-l: Mucoxin

More recently, some acetogenins — now termed as nonclassical acetogenins —
containing THP or hydroxylated 2,3,5-trisubstituted THF rings have been isolated.6 In
addition to their biological activities, the novel structural features of nonclassical

acetogenins have aroused the interest of synthetic chemists.7'10

105

Mucoxin (Figure III-1) is one of the nonclassical acetogenins isolated by

McLaughlin and coworkers in 1996 from the bioactive leaf extracts of Rollinia mucosa. H

In vitro cytotoxicity assays against a panel of six human tumor cell lines showed mucoxin
to be more potent and seleCtive against MCF-7 (breast carcinoma) cell lines (ED50 = 3.7 x
10‘3 rig/mL) than adriamycin (ED50 = 1.0 x 10’2 pg/mL). The isolation procedure for
mucoxin involved activity directed open column fractionation using brine shrimp

lethality test and at later stages purification by 1H NMR—monitored repetitive reverse and
normal phase HPLC techniques. As is often the case with acetogenins,12 after such

rigorous purification procedures, only 1.8 mg of mucoxin was isolated. Due to a limited
supply of the natural sample, only the constitution and the relative configuration of the

seven oxygenated stereocenters of the bis—THF core (CS-C17, Figure III-1) of mucoxin
were established.11 Although, no attempts were made to determine the absolute

stereochemistry of C8-C17 portion of mucoxin, the absolute configuration at C36 was

assigned to be S based on the observation that over 400 acetogenins isolated to date have
been shown to possess S configuration at that carbon.6

Mucoxin is the first nonclassical acetogenin possessing a hydroxylated THF ring
(C13-C17, Figure III-l).’ This structural feature adds an element of complexity in the
design of its total synthesis. Several chiral building blocks used to construct 2,5-
disubstituted THF units have originated from the earlier total syntheses of classical
acetogenins (Chapter 11). However, it may not be possible to use these building blocks ‘as

is’ to construct the hydroxylated 2,3,5 tri—substituted THF ring of mucoxin. Also,

m

- . . . 3 .
Four acetogenins were prevrously reported to possess hydroxylated THF rings;l however their

2
structures were proved erroneous and have been corrected.

106

straightforward and efficient modifications of such existing building blocks to
incorporate the ring hydroxyl group are not readily apparent.

Mucoxin attracted our attention as a synthetic target for several reasons. We
envisioned that the hydroxylated THF fragment of mucoxin could be easily accessed

using our method for regio- and stereoselective construction of hydroxylated, tri-
substituted THF units (Chapter 1).'4 Thus, it would serve to test the viability and

generality of our method in a total synthesis setting. A major focus of our laboratories is
on developing straightforward and versatile methods to synthesize THF units with
various substitution patterns, using regiocontrolled cyclization reactions. Thus, the right-
hand portion (Cl-C12) of mucoxin also attracted our attention as a possible avenue for
new method development to install the 2,5-disubstituted THF ring. In light of the fact that
the proposed structure of mucoxin contains unusual elements (a hydroxylated THF ring),
previously unknown in acetogenins, it becomes important to confirm the proposed
structure, establish the relative and absolute configuration as well as further explore the
bioactivity profile. Total synthesis would provide the material necessary for such
investigations on mucoxin. Structure—activity relationship studies on classical acetogenins

have indicated that adjacent bis—THF acetogenins with three free hydroxyl groups possess
the most potent cytotoxic and pesticidal properties.6 Mucoxin presents itself as an

interesting case study since it embodies all the above-mentioned features but at the same
time possesses a unique disposition of one of the hydroxyl groups. Also, as in any other
total synthesis, an appropriately designed synthetic strategy would provide an expeditious

access to unnatural constitutional and stereoisomers of mucoxin. Finally, the

107

”Y.;;

(D

intermediates generated during the total synthesis could serve as truncated analogs of

mucoxin that can be employed for SAR analysis to delineate essential pharmacophores.

4444 .—_—.—— ..__.—__._ __ 44

B. Retrosynthesis

Most acetogenin total syntheses reported so far involve first, construction of the
polycyclic ether core bearing suitable functional group handles, followed by sequential
elaboration to install the long hydrocarbon side chain and the terminal y-lactone with an
appropriate linker.12‘]5'l7 From the outset, we sought a more convergent approach that
involved coupling the right (Cl-C12) and the left (C13-C34) hand fragments of mucoxin
in fully elaborated forms. Our strategy, in the form of a retrosynthetic analysis, is

summarized in Figure III-2.

Tandem

7 O A)
m RCM- hydrogenation

gmucoxin III-1

991 RfFfegio- and 9R1 0*" O
16 13 ‘3 + 1
14». 11 \‘0 \ selective 14‘: / ‘0 O
OH,

 

 

epoxide 0R1
Ill-2 opening Ill-3 Ill-4
OH O
3 SP“ reso
16 i ’O ¥—1 + 1/\/\/I + 8' O
OR \‘ /
III-5 Ill-6 Ill-7 Ill-8

Figure III-2: Mucoxin: retrosynthetic analysis

Since the absolute stereochemistry of the C8-C17 bis-THF core of mucoxin is

unknown, we Opted to synthesize enantiomer III-1 (Figure III-2). Our retrosynthesis

108

began by two simultaneous fragmentations of the natural product along the C2-C35 and
the C IO-Cll bonds.
Grubbs has recently reported a ‘one pot’ tandem olefin metathesis hydrogenation

sequence to directly obtain reduced metathesis products (Figure III-3).18 After

completion of the metathesis reaction at 40 °C, hydrogen was introduced in the same
reaction vessel which generated the active hydrogenation catalyst RuHCl(H2)(PCy3)2.
Upon increasing the temperature to 70 °C, hydrogenation took place cleanly to afford the
corresponding saturated products. Figure III-3 shows examples of this protocol relevant
to our total synthesis. The hydrogenation occurred readily in case of bisallyl ether III-9 at
atmospheric pressure, whereas higher pressure (100 psi) was needed to obtain the lactone
III-12 due to the steric and electronic factors. Inspired by this report, we decided to
construct the C9-C 12 THF and the terminal 01,13 unsaturated y-lactone rings of mucoxin

using a tandem double RCM / hydrogenation sequence of the precursor III-2 (Figure

 

 

III-2).
PCY3
o 1. cat. Ill-13 0 0,11
J k 4' O Cl'RunPh
2. H2, 1 atm. 3
Ill-9 40 °C to 70 °C . ""10 PC”
100% (GC yield) ""13

.3: m "-._
N N
0 0 Q T Q
- Cl,
\rrlko 1.cat.lll14 A U ova“;
. pcy Ph
W 2. H2, 100 p51 3

Ill-11 40 °C to 70 °C Ill-12 Ill-14
100% (GO yield)

 

Figure III—3: Grubbs’ tandem olefin metathesis - hydrogenation protocol

To finish the retrosynthetic analysis, 2,3,5-trisubstituted THF unit III-3 would be
obtained via regio- and stereoselective cyclization of epoxy-diol III-5 using the
methodology described in Chapter 1.14 Finally, vinylic epoxide III-4 would be

109

19.20

synthesized using a Knochel type three component coupling reaction of alkynyl

iodide III-6, 1,4 diiodobutane III-7 and allylic bromide III-8 (Figure III-2).
We planned to assemble the advanced intermediate III-2 via a regio-and
stereoselective intermolecular opening of the vinylic epoxide Ill-4 by the allylic alcohol

III-3. Intramolecular epoxide opening by a hydroxyl nucleophile has been extensively
studied and utilized to prepare medium sized cyclic ether units.“22 In fact, it is probably

the most commonly used tactic to install multiple cyclic ether segments, as demonstrated

in the elegant total syntheses of polyether natural products including marine toxins such

23

as brevetoxins, ciguatoxins, maitotoxin and simpler annonaceous acetogenins. In

contrast, the intermolecular version of the process using alcohols as external nucleophiles

has been investigated to a much lesser extent.

To the best of our knowledge, intermolecular epoxide ring opening by means of

alcohols has remained unused as a strategy in complex total synthesis settings. This could

24-27

be attributed to several factors. First, alcohols, in general are poor nucleophiles and

- . . . -3 . .
epox1des, inherently are not very reactive electrophiles.28 ‘0 Thus, thelr union often needs

l-3

harsh conditions such as the use of alkoxides at elevated temperatures3 3 or epoxide

34-37

activation using strong Lewis or Bronsted acids which could be incompatible with

sensitive functionalities present elsewhere in the reacting partners. Moreover, even under
such forcing conditions, a large excess of alcohol is required to drive the reaction to
completion. Secondly, most intermolecular epoxide opening reactions thus far have

involved simple alcohols like MeOH, EtOH, benzyl alcohol and phenol that can be used

as solvents.38’4l

110

This, clearly, is not viable from a total synthesis standpoint since complex alcohol
coupling partners most likely will not be available in such large quantities. Also, another
potential limitation on the use of complex alcohols as nucleophiles is that as the alcohol
gets sterically hindered, its nucleophilicity is likely to drop further. Finally, it is more

difficult to achieve high levels of regio— and stereocontrol in intermolecular fusion of an

. . . . 4 3
alcohol and epOdee (vzde supra) as compared to its lntramolecular counterpart. 2‘4

11M" H B1 NU

 

 

 

3 O . _. 37 2
A R1/<12/\OH N-u- ' W0 ;\Nu R,/Y\OH
(‘3 H OH
Ill-1 5 III-16 III-17
OH
R, \ R
Nu Nu 0 Nu 0”
B 01' RKQ/VR leR
OH 1.2 add. 1,4 add. Nu
R III-18 Ill-19
a, \
Nu
III-20

Figure III-4: Common tactics used for regiocontrol in intermolecular epoxide opening

reactions

Having mentioned the difficulties in epoxide ring opening by external alcohol
nucleophiles, it would be in order to point out a few literature reports that have dealt with
the issue. Two commonly used techniques to realize regiocontrol in intermolecular
epoxide opening reactions are outlined in Figure III-4. 2,3 Epoxy alcohols (part A,
III-15) form bidentate chelates with metal centers (III-16) which leads to selective
activation of C3 of the epoxide toward a nucleophilic attack.“44 In case of vinylic
epoxides of type III-18, under acidic conditions, the epoxide carbon adjacent to the

olefinic moiety is selectively activated toward nucleophilic attack due to resonance

lll

stabilization of partial positive charge. Depending upon the nature of transition metal

activator or Lewis acid catalyst the corresponding 1,4 (III-l9) or 1.2 (III-20) addition
products can be 0btained.‘7'4‘42‘45‘46 Representative examples of these strategies,

specifically, in the context of alcohol nucleophiles are described below.

 

W0” (excess) OM
Ti(O‘Pr)4. reflux ‘7 W0“
mln. 90°/ OH
o 30 ° Ill-22
WOH 100/ 1 regioselectivity
2

A O
OH (excess) _ W0”

Ti(O‘Pr)4, reflux F OH
18 h 88% Ill-.23 . .
100/1 regloselectlwty

Figure III-5: Sharpless’ protocol for C3 selective epoxide ring opening of 2,3 epoxy

alcohols

In 1985, Sharpless and co-workers developed a procedure for highly

regioselective opening of 2,3 epoxy alcohols using stoichiometric Ti(O‘Pr)4 as a chelating
agent“ (Part A, Figure III-4). 2,3 epoxy-l-hexanol III-21 (Figure III-5) when treated

With allyl alcohol produced the corresponding C3 ring opened product III-22 in excellent
yields and regioselectivity. The Same transformation using bulkier iPrOH was sluggish
and took prolonged heating for completion. This efficient protocol although widely used,
is restricted to sterically unhindered alcohols. It should also be noted that only the alcohol
nucleophiles had to be used in large excess at elevated temperatures whereas other
nucleophiles including azidCS, cyanides, thiophenols, and amines reacted efficiently at
ambient temperature.

Vinylic epoxide substrates of type III-18 (Part B, Figure III-4) have been used

more frequently as alkylating agents for alcohols under transition metal catalyzed or

112

 

Lewis acidic conditions. Hirama and co-workers showed that the densely functionalized

cyclopentadiene monoepoxide (III-25, Figure III-6) could be regio-and stereoselectively
opened by the azatyrosine III-24 using CsF as an activator in good yields.“ This method

though attractive due to the functional group tolerance and reactant stoichiometry, was
COzMe

EEO...

CSF, DMF N \

l
60 °C, 70h Cl
66%

   

Figure III-6: Hirama’s conditions for regio-and stereoselective addition of aromatic

alcohols to highly functionalized vinyl epoxides

applicable to very specific aromatic alcohols. Extension of this protocol to other aromatic

or aliphatic alcohols has not been reported.

Most transition metal mediated nucleophilic additions to vinylic epoxides have
been known to produce 1,4 addition products (III-19, Figure III-4).43 However, Trost, in
1988 reported Pd(0) catalyzed regioselective 1,2 addition to vinylic epoxides (Figure
"17).24 The trick was to use a tin ether, which formed an intermediate ‘ate’ complex

thereby tethering the nucleophile to the epoxide prior to attack (III-28 and III-29). The

nucleophile was then delivered selectively at the carbon adjacent to the vinyl group.

113

m

0. O ’— O
S \ 9
sit/C} \

n’
H9C4/ ‘04 H9

 

 

 

 

Ph/\.-r"’<T/\ #- O ( H 6
(dbalang-CHCls Hjx 4— +1., 4,:
Ill-27 PPh3, 77% F, l" {:1)—x};
0+
/Pd L L / \L J
Ill-28 Ill-29
OH OH
Ph/M Ph \
O\/\OH OwOH
""30. erythro Ill-31, threo
(ca. 1 :4)

Figure III-7: Trost’s strategy for 1,2 addition of alcohols to vinyl ic epoxides

Thus vinylic epoxide III-27, when treated with a cyclic stannane in presence of
P(1(0), gave diastereomeric 1,2 addition products III-30 and III-31 exclusively in 77%
yield. However this methodology is faced with several limitations and is far from being
0f general applicability. First, the stereochemical fidelity of the starting epoxide is lost
during the reaction. It was shown that irrespective of the epoxide geometry, the threo ring
opened product was always obtained as the major product. To explain the results, Trost
Proposed that eclipsing interactions of the R group with allyl fragment would destabilize
conformation III-28 (Figure III-7) relative to III-29. Secondly, the use of cyclic
stannanes was essential. Trialkyl stannanes (III-33, Figure III-8) produced the desired
products in low (up to 25%) yields and the reaction was accompanied by rearranged
products. Thus, complex mono alcohols did not participate effectively.

OCH3
10.1.13... p.110, 0V0
7650 + b 435:...

Ill-32 Ill-33 Ill-34

 

Figure III-8: Trialkyl stannanes proved inefficient as electrophiles in Trost’s studies

114

 

More recently, Trost has also developed trialkyl borane mediated, Pd(0) catalyzed
allylic alkylation of alcohols using racemic vinylic epoxides (Figure III-9).46 The

proposed mechanism again involves tethering the nucleophile to the epoxide via a borate

 

 

 

(S'C4H913B A _ O
M RAOH 4e NOH o ‘. O r T; 0
Pd(0),lll-37 CV” P‘NH ”ll-(A,
Ill-35 upto 94% ee Ill-35 111.37 thP
41 %-85% yields ‘1

 

 

 

 

Figure III-9: Trost’s two-component catalyst system for asymmetric allylic alkylation of

alcohols

complex before ether bond formation. Nevertheless, this method is also restricted to
primary alcohol pronucleophiles and has been applied only to terminal vinylic epoxides.

Another class of epoxides that participate in intermolecular ring opening is sugar derlved

 

 

1, 2 oxiranes.
HO HO
FIG—g3 DMDO Ro—gs AgBF4 RFQRZ—Ofi
RO / Acetone HO O Fi’OSnBu3 VCR
NHBUa
Ill-38 Ill-39 "'40 AQ‘L’W
no
upto 60% 90%
BU3SHOOR.
Ill-41

Figure III-10: A representative example of regio—and stereoselective ring opening of

sugar derived oxiranes

Danishefsky has extensively studied glycosidation reactions of glycosyl donors

such as III-39 using a variety of promoters.“54 The sequence outlined in Figure III-10,

involves AgBF4 promoted O-glycosidation using stannylated glycosyl acceptors.

115

Danishefsky has invoked a stannyl group transfer to the C2 oxygen of the donor III-40 to
reason the observed (Jr—selectivity in the glycosidation reaction. The driving force in such

an epoxide activation event is the generation of an oxonium ion intermediate.

Mioskowski and Lautens, independently reported their studies on vinyl epoxide

. . 42.55 .
ring opening by alcohols about the same tlme, whlch were most relevant and

promising to us for use in our total synthesis. A representative example from
Mioskowski’s studies is shown in Figure III-11.55 Out of a wide range of Lewis acids

screened, BF3°OEt2 worked most efficiently for the regio— and stereoselective opening of

III-42 with a variety of alcohols. The hindered secondary alcohol III-43, having an

O BF3- OEt2 HO O
4 3 (1 mOio/o) 4
._ 3
H‘9C{A\_+ +><K CH2CI2.rt Hafiz—L“
70% Ill-44
threo : erythro = 3 : 2

 

Ill-42 Ill-43
E : Z: 3 : 2

Figure III-11: Mioskowski’s conditions for stereoselective 8N2 addition of alcohols to

vinyl epoxides

adjacent neopentyl center, also participated efficiently. The ring Opening occurred
exclusively at C3 of the epoxide III-42 and the diastereomeric ratio was conserved in the

product III-44. Also, the reaction was found to be tolerant to different solvents (benzene,

EtzO) and temperatures (-78 °C to reflux).

 

OR
0 cat. [Rh(CO)2Cl]2 :
RZ/WR1 > Rz/WRl
ROH 62 °/o-94 % OH
Ill-45 cyclic or Ill-46 >20 : 1 diastereo -
and regioselectivity

acyclic
Figure III-12: Lautens' protocol for 5N2 substitution of vinylic epoxides by alcohols
under mild conditions

116

Lautens used a rhodium(I) catalyst to promote the intermolecular epoxide opening
reaction by alcohol nucleophiles (Figure III-12).42 A variety of epoxides containing
functional groups like esters and silyl ethers elsewhere, reacted efficiently to afford the
corresponding 5N2 products with inversion of configuration at the reactive carbon. The
alcohol nucleophile however had to be used in excess (106q.) and only simple unhindered
alcohols were examined. Nonetheless, this method seemed promising because of the mild

conditions utilized. We also thought that it might be possible to recycle any excess

alcohol necessary to promote the reaction.
Encouraged by Mioskowski’s and Lautens' studies as well as our own experience
in the regioselective epoxide ring opening area, we decided to attempt an intermolecular
coupling of the functionalized allylic alcohol III-3 and the vinylic epoxide III-4 units
(Figure III-2) in the total synthesis of mucoxin. This strategy was particularly attractive to
us because of (i) functional group tolerance of the coupling reaction thereby allowing the
convergent assembly of advanced intermediates, ii) possible avenues for introducing
diversity in terms of the size of the ring (THF and THP) to be installed and (iii) the
stereogenic centers (in principle, all four stereoisomers of vinylic epoxide of type III-4

could be easily accessed using Sharpless asymmetric epoxidation reaction of the

appropriate cis or trans allylic alcohol precursor).

ll7

 

 

 
    

in 1. Ill-51 m
H25C12 0,. O OPee 1125012 0‘. O OH
3 2. H2 / Pd-C 5
T850 "'47 T880 Ill-48
PhS O
\
\
YQ/ steps 4; O
N resO
F3C .1 Ill-49
F::1C§—O’.'l’1°\
Me 0 Ph
Me7( Me Me
F3C 01:3 H25C12
, . 7 .
Ill-51 H8 H6 OH
Schrock's metathesis 111-50 muconin
catalyst

 

 

 

 

 

 

 

 

Figure III-13: Jacobsen’s strategy to construct the THF ring of muconin

To our knowledge, only Jacobsen and coworkers have used a RCM protocol to
install the THF ring of an acetogenin.7 In their synthesis of muconin (III-50, Figure
Ill-13), a lengthy, multi-step route was used to access the precursor III-47 (the total
synthesis is described in more detail in Chapter 11). One might anticipate that ring
opening of an appropriate vinyl epoxide fragment by a suitable allylic alcohol (Scheme
III—l) would provide a quick entry to substrates like III-52. If achieved under mild
conditions, this type of regio- and stereoselective epoxide Opening, coupled with RCM,

would offer a versatile, expeditious and efficient strategy to assemble THF and THP rings

Y
H. O "9‘
OH

Ill-52

in acetogenins.

OH

RM + MR, —_..

Scheme III-l: Proposed intermolecular epoxide opening strategy

118

C. Evaluation of the proposed intermolecular regio- and stereoselective epoxide

opening strategy

1. Design and synthesis of chiral allylic alcohol III-3

During the course of our earlier work (Chapter 1), five regio-and stereoisomeric
THF diols (III-54—III-58, Figure III-l4) were accessed using the 2-deoxy-D-ribose
derived epoxy diol system III-53 as the common precursor. Depending upon the epoxide
stereochemistry, the choice of the pendant functional group X (III-53) and the acid
promoter, all five THF diols (III-54 to III-58) were obtained in high yields and
enantiopurity, which rendered this method a viable route to access 3-hydroxy-2,3,5-
trisubstituted THF motifs for use in total synthesis. We considered the possibility of using
one of these available THF diols for further elaboration to the target allylic alcohol III-3.

As far as the stereochemistry is concerned, of all the isomers, III-57 most closely

resembles the target allylic alcohol III-3 (Figure III-15).

 

 

 

 

OTMS O X OTMS»? X
HO ? 4 1 RO , ‘
OTMS OTMS
Ill-538 Ill-53a
acid 'acid
OH QH OH OH OH OH OH
. on ., OR ,- OH
HO 4 2OH 2 OH HO‘ 4 2 OH
Ill-54 Ill-55 Ill-56 Ill-57 Ill-58
(5-exo) (5-endo) (5-endo) (5-exo) (5-endo)
retention at Ct inversion at C1 retention at 01

Figure III-l4: Isomeric THF diols available from a common epoxy diol precursor

119

OH, OH OH

'7 01,. / HO 4 07.1 X
16” ‘3 U

14 3 2",
threo 081 erythro OH
Ill—3 Ill-56

Figure III-15: Stereochemical similarities and differences between the target THF unit

III-3 and an available precursor III-56

Trio] III-S6 has the same absolute configuration about the THF ring as that of
target triol III-3. The only difference lies in the threo (III-3) vs. erythro (III-56)
relationship between the side chain carbinol stereocenter (C17 in III-3 and C5 in III-56)
and the THF ring system. In order to use III-56 as a precursor to III-3 following three
transformations would be necessary: (1) inversion of the C5 stereocenter (ii) installation
of the aliphatic chain and (iii) elaboration of the pendant group (X) to the allylic alcohol
functionality.

In case of acetogenins containing 2.5-disubstituted THF rings flanked by

hydroxyl groups, inversion of such side chain carbinol stereocenters has been achieved in
two major ways:‘5 i) using Mitsunobu inversion of alcohols ii) via formation of a

terminal epoxide that involves 8N2 displacement at the stereocenter in question. It seemed

to us that inversion of the C5 stereocenter in III-56 using one of these protocols would

 

 

 

 

 

. QH
Mltsunobu : 4 0.,1, x
OH iar‘ltvggsion OR 2.: _l
HOWX selective Ill-59 0" 1 add 12'
_ . Ill-3
2., n .
OH protectlo S 2 2. manlpulate
Ill-56 N , 0., x X
displacement 2.. 4
at C5 III-60 OP

Figure III-16: A route to transform III-57 to the target allylic alcohol III-3

120

necessitate selective protection of the C2 hydroxyl group (Figure III-l6) due to similar
steric environments of the two hydroxyl groups. Overall, this approach did not appear
concise and straightforward. Also, if the oxygenated stereocenters in cyclization
precursor III-5 (Figure III-2) were derived from asymmetric transformations instead of
the naturally available chiral pool, a variety of stereoisomeric epoxydiols of type III-5
could be easily obtained merely by using enantio- and diastereomeric reagents and
reactants. Such a strategy would offer easy access to unnatural analogs of mucoxin.
Sharpless asymmetric dihydroxylationS6 and epoxidation reactions are extremely

reliablefl‘58 to establish oxygenated stereocenters in high enantio- and

diastereoselectivity. These methods are also highly versatile since both the antipodes of
chiral reagents employed are easily available. Thus, these protocols were ideally suited

for our purpose to synthesize epoxy diols of type III-5.

 

 

 

 

OR 1. SAD _ .
l ///\/ 1 steps 2. F13SiCl 93mg 1. \lillltftlg t'
‘ A OB ' oe lnalon
\th homologation W 2 3' 'R2 Wow“
III—61 Ill-62 4. oxidation Ill-63 OS‘Ra 2. DlBAL-H
98‘93 1- SAE 993 erg-0E12 9:1
WOH 4* WSW ————- Wsph
. . -, - 13
‘5 I . 2. PhSSPh 16 ; '0 16
OSIR3 BU P 0R3 14 Tb
111454 3 Ill-5 "ms H
1.protectdiol 0:“ O OH/
2. Pummerer ‘6 “.13
rearrangement 'OR 1
Ill-3

Figure III-17: Proposed synthesis of the left hand (C l3-C34) fragment of mucoxin

In our synthetic strategy (Figure III-17), we chose to incorporate the long alkyl

chain from the outset. Thus, readily available l-iodoheptadecane III-61 would be

121

homologated using suitably protected 3—butyn-l-ol and the resultant homopropargylic
alcohol would be transformed into the homoallylic alcohol III-62. Asymmetric
dihydroxylation of the trans olefin III-62 should afford diol III-63, which after suitable
manipulations should generate allylic alcohol III-64. Epoxy diol III-5 would then be
accessed via asymmetric epoxidation of III-64 followed by treatment with the Hata

reagent to install the thiophenyl group. Exposure of III-5 to Lewis acid should lead to
simultaneous deprotection / cyclization event (Chapter I)14 to afford the THF diol III-65

having all the stereogenic centers correctly established. Finally, Pummerer rearrangement
to convert the thiophenyl group in III-65 into an aldehyde functionality and subsequent
addition of vinyl magnesium bromide in a chelation controlled manner should provide
chiral allylic alcohol III-3. The transformations needed to elaborate the [3,y-dihydroxy
aldehydes similar to III-63 to the epoxy diol systems of type III-5 (Figure III-17), were
optimized during the course of our method development (Chapter 1). Therefore, our
immediate goal was to access aldehyde III-63 in a quick and efficient manner. Several
approaches toward this goal were tried.

The first approach involved introduction of the aldehyde functionality in a
masked form by alkylation of l-nonadecyne III-68 with bromoacetaldehye diethyl acetal
(Scheme Ill-1). III-68 was prepared in good yield via alkylation of l-heptyne (III-66)

with l-bromododecane followed by isomerization of the internal alkyne III-67 to the

terminal alkyne III-68 by way of an alkyne zipper reaction.59‘60 Homologation 0f111'68

with bromo (or iodo) acetaldehye diethyl acetalm‘62 to obtain desired alkyne III-69,

however proved low yielding under a variety of temperature and solvent conditions. The

fact that bromo (or iodo) acetaldehye diethyl acetal was fully consumed in the reaction —

122

X/\|/O\/

 

 

 

 

N 1- BUU 4 \ KNH(CH2)3NH2 A o\/ A
2 ’(Tl‘rBr \ NH2(CH2)3NH2 ‘5 F 3””
Ill-66 ill-67 1‘ n,es% Ill-68 THF/DMPU
:51 5913;? upto 22%
X=Br
Nal, acetone
reflux, 95%
X=|
j 1. LiAlH4
2. SAD 1L
0 --------------- - Ill-63 + x o o
15 § 0’] 3. SiR3Cl mL Li A?:V W68 +2 /‘
Ill-69 4. H+IH20

 

 

 

Scheme III-2: Alkyne zipper reaction strategy

as detected by GC analysis and D2 0 quenching experiments — suggested that [3-
elimination of the acetals by lithium acetylide III-70 (Scheme Ill-2) might be a side-
reaction resulting in lower yields.

We also attempted to prepare propargylic ester III-72 (Scheme III—3), which upon
treatment with LAH would directly provide the homoallylic alcohol III-62 through
simultaneous reduction of the alkyne and ester functionalities. Quenching of lithium

acetylide of III-68 with ethyl bromoacetate however led to decomposition.

nBuLi

BrACOZEt

1 nBuLi 0°C-rt
\ 1
m ’ mocha + III-68

Ill-68 2 el=3 0E12

 

 

 

 

Ill-72 12%., 83%

3 N2/\ 002E1F ;

3R :: Li +BF3-OE12 (R—=)'B
,20 0C : LAH WORZ

Nfco El H O ------------- T ‘6
,____2__,__2_, n——\-_— III-62; 92:11
~20 °C. -N2 87%- ill-71 COZE‘
92%

 

 

 

Scheme III-3: Propargylic ester strategy

123

Layton has developed a method for preparation of propargylic esters (III-71,
Scheme III-3) using trialkynylboranes.63 Treatment of a lithium acetylide with BF3-OEt2

at —20 °C generated the corresponding trialkynylborane which upon immediate exposure
to ethyl diazoacetate and subsequent hydrolysis afforded corresponding propargylic ester
(III-71) in high yields. This protocol was also unsuccessful in our hands. Reaction of
1-nonadecyne III-68 led to the propargylic ester III-72 in only 12% yield and the rest of
the starting alkyne was recovered unchanged. It must however be noted that, in our case,
treatment of the lithium acetylide of III-68 with BF3,-OEt2 resulted in a white precipitate,
which could not be solubilized even after addition of ethyl diazoacetate. Since such

precipitation has been reported in the original procedures,63 we think that the

organoborane species, due to the long hydrocarbon chain might be insoluble in the
reaction medium.

Next, a Wittig olefination approach for the direct preparation of the trans
homoallylic alcohol III-62 from a long chain aldehyde was explored. Schlosser has
developed a method for the synthesis of trans-alkenols using a modified Witti g reaction
of w-hydroxyalkyl-triphenylphosphonium bromides (Figure III-18).64'65 m-Hydroxyalkyl-

triphenylphosphonium bromides of varying chain lengths (III-73) after conversion to the

 

UB' RCHO
i o - 1. H01
9 6) 2PhLi-LiBr e (.9 '9 _ -30 0 L10 ...R Et20
5' PhaPWOH ———————+Br P113P n 01.1 _ _ (+) a
2PhL1-LlBr er Ph3P 01.1 2-KO‘BU
1"-73 Ill-74 n 700/ 80°
ill-75 , °' /°
wad
RWOH
Ill-76

E:Z upto > 99:1

Figure III-18: Schlosser’s B—oxido ylide route to trans alkenols

124

corresponding ylides III-74 were treated with an aldehyde at low temperatures (in order
to prevent decomposition of the corresponding oxaphosphetane to the olefin). Treatment
of the oxaphosphetane intermediate with PhLi-LiBr complex lead to the formation 0f [3-
oxido ylide III-75, which is allowed to equlibrate t0 the more stable trans isomer. Upon
reprotonation with HCI and breaking the oxaphosphetaneeLiBr complex with KO‘Bu, the
corresponding trans alkenols III-76 were isolated in good yields and high selectivity.

In order to explore the possibility of using Schlosser’s modified Wittig olefination

protocol in our synthesis, octadecanal III-78 was synthesized by BAIB / TEMPO
mediated oxidation of l—octadecanol (Scheme III-4).66 Wittig reaction of III-78 with
3-hydr0xypropyl triphenylphosphonium bromide III-8067 using nBuLi-LiBr complex for

ylide generation provided the trans alkenol III-62 in 58% yield. However, we were faced

with some difficulties. First, alkenol III-62 contained minor impurities (possibly the

 

 

 

 

 

BAIB, TEMPO ”“3 - .
OH ital/H BrMoH BrPhsPMOH
15 CH2012, rt. 89% 0 20:13:33; lrellux
III-77 Ill-78 Ill-79 . ’° Ill-80
1. nBuLi-LiBr (1 :1.5), 1. MeLi-LiBr (1 :1 .5),
THF / Et20 (5 13), rt THF I E120 (5 13), n
2. Ill-78, THF -30°C 30 min a 2. Ill-78, THF -30'C 30 min b
3. nBuLi-LiBr in THF/EIZO (5:3) rt 3. MeLi-LiBr in THF/Et20 (5:3) rt
4. HCI in E120 then men, 58% l 4. HCI in Et20 then KO‘Bu, 40%
1 1
OH2
\ OR
\H1/6\/\/ W 2
III-62; 82 = H Ill-62; R2 = H
BSTFA 1. Purification using
AgNO3 coated
OTMS silica gel
‘6 111.31 2. BSTFA
(94% pure by GC) OTMS
W
Ill-81
(>99% pure by GC)

Scheme Ill-4: Application of Schlosser’s method to synthesize trans alcohol III-62

125

cis isomer; as indicated by 1H NMR), which could not be separated via column
chromatography.‘ GC analysis of the TMS derivative III-81 also showed minor impurity
peaks. We did not want to proceed with isomeric mixtures at such an early stage in the
synthesis. Secondly, use of THF / EtzO (5/3) solvent mixture was reported to be critical
for achieving high trans selectivity. Thus, while using the commercially available nBuLi
in hexanes, it was necessary to remove hexanes and freshly prepare a stock solution of
nBuLi-LiBr complex in THF/ EtZO (5/3) prior to the reaction. This procedure proved
tedious and impractical especially for large-scale operations. When we switched to
commercially available MeLi in EtZO as the base, the desired alkenol III-62 was obtained

in lower yields (ca. 40%). Use of AgNO3 coated silica gel for chromatography is known
to facilitate separation of isomeric mixtures of unsaturated hydrocarbons.68 Purification

of the alkenol III-62 using this technique indeed separated the impurities to furnish III-
62 in >99% purity as indicated by GC analysis of the TMS derivative III-81. However,
the yields and efficiency of the purification technique could not be reproduced on large
scales needed to bring up multi gram quantities of material.

Next, we turned to the alkynylation reaction of a suitable primary iodide via an
5N2 displacement reaction to obtain the corresponding long chain homopropargylic
alcohol that can be reduced to trans homoallylic alcohol III-62. The requisite iodide
substrates III-84 and III-86 were prepared as shown in Scheme Ill—5. 1-heptadecanol

III-83 although commercially available, is expensive and was therefore prepared from
l~bromohexadecane (III-82) by carbon homologation.69 After considerable

experimentation, we found that the homologation worked reproducibly on large scales

—_¥

' Due to the presence of the long chain alkyl groups, purification by crystallization was also not
feasible.

126

1. NaH / BnBr

 

 

1. Mg / E120 PPh3, imid. THF rt
\i‘TBr \(‘1'0H ——-—‘ l H0/\/OH ’ - BrtO/\/l
‘5 2. (CH20)n ‘5 12 ,toluene '5 ‘ '
o 2. PPhs, lmld.
Ill-82 rt. 65 /o Ill-83 11.90% III-84 III-85 12' toluene, rt Ill-86

65%, (two steps)

. 1. nBuLi, THF 0 °C rt
1. nBuLl (2 eq.) OH W

2. Ill-39, THF/HMPA ‘5 Ill-68 2.111-41,THF/HMPA
III-87 25% Ill-88 80%
\H/VOBn_9<“—’ LAH +WOBF}
Ill-89 ‘5 |"_90

Scheme III-5: Iodide alkynylation route

 

 

 

(200 grams of III-82) only when paraformaldehyde was cracked to generate molecular
formaldehyde, which was then bubbled through an ethereal solution of l-hexadecyl
magnesium bromide. Iodination of III-83 using triphenyl phosphine and iodine furnished
l-iodo hexadecane in 90% yield. Similar iodination of mono benzyl protected ethylene
glycol III-85 provided the iodide III-86.

Treatment of the dianion of 3-butyn-l-ol III-87 with iodide III-84 produced the
homopropargylic alcohol III-88 only in low yields. Under a variety of different reaction
conditions which included changing the reactant stoichiometry, solvent prOportions and

temperature, the yields of III-88 went up to only ca. 25%.”73 l-octadecene — a [3

elimination product of the iodide III-84 was often obtained as a side product. On the
other hand, alkynylation of iodide III-86 with l-nonadecyne III-68 under similar
conditions (Scheme III-5), met with success and the benzyl protected homopropargylic
alcohol III-89 was isolated in good (80%) yields. Unfortunately, reduction of III-89 to

the corresponding trans olefin III-90 proved difficult, even after refluxing with LAH in

diglyme for several hours,74 alkyne III-89 was recovered unaffected. Realizing that LAH

127

reduction reactions of propargylic and homopropargylic alcohols containing free
hydroxyl groups are more facile due to pre formation of organoaluminates,75 we finally

resorted to a somewhat lengthier route to homoallylic alcohol III-62 (Scheme Ill-6).

nBuLi, THF

OH ‘ ' OTBS -30 °C to -10 °C
///\/ TBSCI, lmld. 4; ///\/ ;

DMF, n, 73% _ Ill-84, THF : HMPA 16 _
Ill-87 m 91 (3 : 1), 0 °C 80% Ill 92

TBAF,THF \fi/VOH LAH W0”
.__._______.. —= \

- ° 0° 16 ' l , 125 °C ‘5

10 C’ 90/ III-88 3%sz Ill-62

OTBS

\\

 

 

Scheme III-6: Synthesis of trans homoallylic alcohol III-62

Thus, reaction of the lithium acetylide of TBS protected 3-butyn—l-ol III-91 with
the iodide III-84 in THF'HMPA (3:1) at 0 °C afforded the TBS ether III-92 in consistent
yields of 80%.76’78 TBAF mediated deprotection of III-92 provided the homopropargylic
alcohol III-88 (90%). LAH reduction of the free alcohol III-88 delivered the homoallylic

alcohol III-62 in high yields (87%) after optimization of the work up procedure. Simply
quenching the LAH reaction with 1—2 N HCl, followed by extraction of the aqueous
layer,79 afforded alcohol III-62 only in 25%-53% yield depending upon the reaction
scale. The optimized work up involved first quenching the reaction by drop wise addition
of H20 and 15% NaOH, heating the resultant mixture at 50 °C for 45 min and filtration to
separate the white precipitate.80 The precipitate so obtained was further dissolved in 1.5
N HCl (concentration of HCI was critical to ensure maximum recovery of the product)
and extracted with EtOAc several times. Following this work up procedure, the alcohol

III-62 was obtained in greater than 85% yields, independent of the reaction scale. With

an optimized reaction sequence and sufficient amounts of III-62 in hand, we now focused

128

Ph \/\/O

Bu SnH 0.001 M Ph 0 Ph 0
Br 3 i ) *7 V6? 1.5 hydrogen \/\0 3

 

‘

 

_V

H
AIBN lPhH K; atom transfer H
reflux, 4h
“F93 Bu38nH Ill-94 Ph W0 "F95
-PhCH2' [3th H Bugan
0 Ill-96 Ill-97

Figure III-l9: Curran’s self-oxidizing protecting group

our attention on further functionalization of III-62 to chiral aldehyde III-63.
In 1992, Curran introduced a new class of ‘self oxidizing’ protecting groups.8'

The concept is outlined in Figure III-l9. o—Bromobenzyl ether of 3-phenyl-l-propanol
Ill-93, when treated with Bu3SnH / AIBN (at 0.001 M), bromine abstraction generated
the aryl radical species III-94. After a 1,5 hydrogen atom transfer III-94 is transformed
into the a-alkoxy radical III-95, which upon spontaneous homolytic fission produces
3~phenyl propanaldehyde III-96 (typical yields range from 55 to 60%). Maintaining a
low concentration of Bu3SnH is critical because trapping of III-94 or III-95 by hydrogen
transfer from Bu3SnH generates the reduced product (III-97). Thus, in the process of

reductive removal of the protecting group, the substrate undergoes oxidation to the

 

 

 

AD mix-a
NaH, O'Bf-PhCHQBf OCH Ph-4-Br MGSOQNHz
|||.52 A, W 2 A?
THF/DMF, reflux, 78% Ill-98 ‘BuOH : H20 (1 :1)
rt, 75%
9H TMSCI QTMS Buaan ems
‘ OCHZPh-Br-o ’ OCHzPh-d-Br -————x. 7
W EtaN, THF m AIBN, PhH W010
OH ""99 reflux 95% O Ill-100 reflux OTMS
' H2 III-63
Pd/C 9x
' OCHZPh-4-8r
W

(DY
Ill-101X=H,Y=TMS/X=TMS,Y=H

Scheme III-7: Attempted use of Curran’s self-oxidizing protecting groups in our system

129

corresponding aldehyde. This technique seemed useful so as to cut down on the number
of steps.

To test the feasibility of this tactic in our synthesis, differentially protected triol
III-100 was synthesized as shown in Scheme III-7. Protection of the homoallylic alcohol
III-62 as o-bromo—benzyl ether III-98 (78% yield), Sharpless asymmetric
dihydroxylation of III-98 (75% yield) and protection of the 1,2 diol III-99 as the bis-
TMS ether (95% yield) produced the required triol III-100. Preliminary attempts at
oxidative removal of the o-bromobenzyl group in III-100, following the reported
procedures resulted only in recovery of the starting material. The necessity to use high
dilutions for the oxidative deprotection reaction posed a practical limitation. In our case,
a large amount of solvent was necessary for a small scale reaction (80 mL of PhH for 50
mg of the substrate) in order to maintain 0.001 M concentration, which rendered the
process inconvenient especially on multigram scales. Therefore this approach was not
pursued further. Also, under hydrogenolysis conditions, one of the TMS groups in 111-
100 was cleaved prior to the removal of o—bromo-benzyl group (III-101).

During a separate project" it was shown that primary benzyl or p-methoxybenzyl
ethers in triol systems similar to III-100, were not amenable to selective cleavage in
presence of secondary bis—TMS ether groups. We therefore decided to consider using
more robust TES groups to block the 1,2 diol functionality (Scheme Ill-8). Accordingly,
PMB ether III-102 was synthesized (NaH / PMBCI) in 91% yield. Sharpless asymmetric
dihydroxylation reaction of the trans olefin III-102 at 0 °C was slow (82%, 4 d).

However when potassium osmate was added externally so as to increase the amount of

M

' Borhan, B.', Sivakumar, M. Unpublished results.

130

 

OH NaH,PMBCl ()ng
W _______+ wavy
Ill-62 TBAL THF. Ill-102
60 °C, 91%
TESCI .
' OPMB
EthlDMAP WV
THF, rt, quant.

AD mix—a
M eSOQNHZ

 

OH
‘BuOH : H20 (1 :1 ) 0H Ill-103

0 °C, 92%) (only
product observed)

Scheme III-8: Synthesis of the differentially protected triol III-104

osmium to 0.1 mol%, the reaction was completed in 17 h and furnished the diol III-103

in 92% yield.56 Treatment of diol III-103 with TESCI, Et3N / DMAP, produced the

differentially protected triol III-104 in quantitive yields.

Selective deprotection of the primary PMB ether was next examined. DDQ

mediated PMB cleavage using a mixture CHZCI2 and H20 in various ratios provided the
alcohol III-105 in up to 65% yields (Scheme III-9).82 Once again, a major side reaction

was TES cleavage in addition to the PMB removal, which probably occurred due to the

acidity of dichlorodicyano hydroquinone generated during the reaction. Accordingly, the
use of pH 7 buffer, 83 led to higher yields (78%) of alcohol III-105. The isolated yields

also depended upon the work up procedure — the optimum work up involved quenching

 

ores DDQ
' OH t
m CH20I2 2 H20
Ill-105 various ratios

0 °C 60%-65%

QTES

16

ores
III-104

 

DDQ
CH2C|2 I pH 7 QTES
8 Av ' 0“
phosphate buffer ‘5 (STES
(‘10 2 1) 0 °C, 78% m,105

Scheme III-9: Selective deprotection of the PMB group in III-104

131

the reaction with NaHCO3 and extraction with CHZCIZ.‘ With sufficient amount of the

bis-TES protected triol III-105 available, we proceeded with its further elaboration.

 

QTES QTES
. OH PhllOAclz v Ph§P=CHCOgEtA QTES /
16 i W010 THF reflux 91°/' . 002Et
OTES TEMPO OTES ' ' o 16 OTES
Ill-105 CHzClz. 96% Ill-63 Ill-106
ores
DIBAL—H A MO”

 

E120, 0 °C 89% ‘6 ones In.“
Scheme III-10: Synthesis of allylic alcohol III-64
Transformation of III-105 to allylic alcohol III-64 proceeded uneventfully

(Scheme III—10). Oxidation of III-105 by means of catalytic TEMPO and bis-acetoxyiodo
benzene (BAIB) as a stoichiometric oxidant,66 afforded aldehyde III-63 in excellent

yields. This oxidation proved more convenient and efficient with BAIB/TEMPO than
conventional Dess—Martin oxidation. Wittig olefination of the aldehyde III-63 with
(carbethoxymethylene)triphenyl phosphorane in refluxing THF generated the (1,3
unsaturated ester III-106 (91%) exclusively as the trans isomer. Finally, DIBAL-H
reduction of III-106 provided the allylic alcohol III-64 in 89% yield.

We now directed our efforts toward manipulation of III-64 to the epoxy sulfide

III-5. Earlier, we had optimized the Sharpless asymmetric epoxidation conditions for
allylic alcohol systems structurally related to III-64 (Chapter 1).”84 Using our optimized

conditions (Table III-1, entry 1), the epoxyalcohol III-107 was obtained in a maximum

yield of 30% (dr = 6.7 : 1) even after careful purification of the reagents and solvents

M

’ A non-aqueous workup involving filtration of precipitated salts afforded only 33% yield of

product, while the use of other solvents such as EtOAc or CHC13 in aqueous extractions resulted
In emulsions.

132

QTES Tartrate I Ti(O'Pr)4

 

ores

 

 

 

 

 

 

 

W0” a MC“
‘5 5m, 'BuOOH, 0112012 16 5mg
Ill-64 MS 4” Ill-107
entry (:gr/a;:./:;(Icl)liljzrt))4) dr yield (%)
l (D)-DET / Ti(O‘Pr)4 (5.0/ 3.6) 6.7: 1 29
2 (D)-DET/Ti(O‘Pr)4(1.2/1.0) 4.2: 1 35
3 (D)-DET / Ti(O‘Pr)4 (0.24/ 0.2) 2.5 : 1 68
4 (D)-DIPT / Ti(O‘Pr)4 (0.24/02) 8.3 : 1 67
5 (D)-DIP'T / Ti(O‘Pr)4 (0.6 / 0.5) 10 : 1 7o
6 (D)-DIPT/Ti(OiPr),,(1.2/ 1.0) 100: 1 73

 

 

 

 

 

Table III-l: Optimization of the Sharpless asymmetric epoxidation of III-64

and utilizing several different work up procedures. Upon decreasing the reagent
stoichiometry (entries 2 and 3), the yields went up but only at the cost of the
diastereoselectivity. Since the diastereomers were not amenable to separation by column
chromatography or crystallization techniques, we decided to maximize the diastereomeric
ratios. Epoxidation using catalytic (D)-DIP'T instead of (D)-DET (entry 4) significantly
improved the diasteroselectivity while keeping the yields high enough for material
throughput. With 50% catalyst (entry 5) the diastereoselectivity further increased and
gratifyingly, use of a complete equivalent of the catalyst (entry 6) afforded the desired

triol III-107 in excellent diastereomeric ratio (100:1) and good yields (73%). Next,
treatment of III-107 using (PhS)2 and Bu3P85 efficiently installed the thiophenyl pendant

group in one step (Scheme III-11) which provided us with large amounts of epoxysulfide

133

 

(PhS)2

QTES Q" Bu3P. TEA QTES 0,
W0” M ‘ SPh
16 i 0°C tort 16 i
ores m-1o7 94% OTES Ill-5

Scheme III-11: Use of the Hata reagent to install the thiophenyl pendant group

Ill-5 to further explore the proposed synthetic scheme.

At this point, it is appropriate to discuss the rationale behind our choice of
thiophenyl as the directing group. Our total synthesis, by design, called for an endo
selective epoxide opening of a suitable epoxydiol (III-108, Figure III-20) to access the
THF diol with appropriately positioned hydroxyl groups (III-109) and a functional group

handle (X) for further elaboration to bis-THF III-110.

QTES 0 OH OH
M 5-endo '6 0 2 X steps '5 0 2 O ‘1,
. . x ____. . .—__. .
16 ; 3 16 16
OH OH
Ill-108 Ill-109 Ill-110

Figure III-20: An endo selective epoxide opening of III-108 to generate 3-hydroxylated

trisubstituted THF III-109

From our earlier studies (Chapter 1), two such directing groups, namely, vinyl and
thiophenyl, had emerged that led to endo selective epoxide opening. Vinylic epoxide
Opening reactions involved inversion of configuration at the reactive epoxide carbon
whereas ring opening of epoxysulfides resulted in a net retention at the point of
cyclization. The major pathway followed in BF3-tOEt2 mediated simultaneous silyl
deprotection / epoxide opening reaction of epoxy sulfide systems such as III-111 (Figure
III-21) involved the generation of episulfonium intermediates (III-112). These reactive

intermediates spontaneously cyclized to produce five membered rings (referred to as THF

134

 

 

 

' 1
OH
OTMS

.09 BF3’OE12 R0 ‘- O SPh

no 6. ,- SPh ——————+ ___, H .. H

TMSO Ill-111 , OH

TMSO \ Ill-113 (> 99 I 1
III-112 P“ regio-and diastereo

—J selectivity)

(5-exo / 6-endo
retention at C2)

Figure III-21: Cyclization of an epoxy sulfide derived from 2—deoxy-D-ribose (Chapter 1)

via episulfonium ion formation

diols) in very high regio-and stereoselectivities (>99%). The THF diols so generated
retained (via a double inversion) the configuration at C2 (III-113). To sum up, during
cyclization of an epoxydiol such as III-108 (Figure III-20), configurations at C1 and C2
in the product III-109 are determined by (i) the geometry and stereochemistry of the
epoxide and (ii) the mode of epoxide opening (inversion vs. net retention at the reactive

carbon).

In our case, trans allylic alcohol III-64 was selected as the asymmetric
epoxidation precursor since in general, 3-E allylic alcohols have been shown to provide
the corresponding epoxy alcohols in higher enantiomeric purity than their 3-Z

counterparts.86 With this choice of the double bond (hence the epoxide) geometry, either

the 2R,3R (III-107, Figure III-22) or the 25,35 epoxy alcohol (III-114) could be accessed
by appropriate choice of the tartrate reagent. Furthermore, in order to generate a THF diol
having 2,3 cis relative configuration (III-65 or III-115), the endo selective epoxide
opening of either of the precursors III-107 or III-114 would have to proceed via net
retention at C2. As described earlier, out of the two endo directing groups, viz., vinyl and

thiophenyl (Chapter I) only the latter retains the stereochemistry of the carbon atom at the

135

OH

ores 6 01.2
i 9“ F? 0H steps We»),
W ——-—> g R
(D)-DIPT 16 H H ’OH

OTES Ill-107 Ill-65 retention at C2
W0”
16 i 3 OH
OTES :
Ill-64 (Um QTES o s OH 16 o 2 sen
W steps a s
1 1 _—_’_’
6 OTES s 0H
Ill-114 Ill-115 retention at C2

Figure III-22: Stereoisomeric THF diols originating from trans alcohol III-64

point of cyclization. Finally, of the two epoxides III-107 and III-114, only III-107 would

provide the requisite 2,3-cis-5—Irans relative configuration (III-65) across the THF ring.

A THF diol system stereochemically akin to III-65 but containing a vinyl
functional group (III-119, Figure III-23) could potentially be accessed from the cis
vinylic epoxide III-118 via preferential nucleophilic attack at C2. However, as mentioned
earlier, cis epoxy alcohols (III-117) may not be obtained in high diastereoselectivity
using the Sharpless asymmetric epoxidation reaction.86 Another complication associated
With intramolecular endo opening of cis vinylic epoxides in general, is that due to steric
interactions between the Ir system and the incoming nucleophile (III-118, Figure III-23),
the Jr-bond may not remain aligned parallel to the empty p orbital of the incipient

carbocation.87'89

136

OH OH

OTES OTES
: o -DIPT :

 

16 i 3 16 :
ores T680 0
Ill-116 Ill-117
OH ores OTES

' 01,2 \ 1 _.\OH T ..\0H

m 16 i

.9 R TESO
R OH 0+

   

III-119 inversion at C2

 

 

 

Figure III-23: C is-vinylic epoxide may exhibit reduced errata-selectivity during

intramolecular cyclization reaction

Thus, the two conformations A and B, in which the n-bond resides parallel to the p-
orbital at C2 would be higher in energy due to the proximity of the at system to the
incoming nucleophile. This steric barrier is reduced when the It system rotates away
(conformation C) which, however, causes loss of rr—overlap and hence the carbocation
stabilization at C2. This phenomenon is likely to diminish endo selectivity in case of cis
vinylic epoxide opening reactions. Therefore, we anticipated that this tactic would not be

applicable in our synthesis.

Taken together, our strategy of using the trans epoxysulfide III-5 as the
cyclization precursor was benefited by the fact that III-5 could be obtained in very high
diastereomeric ratios from allylic alcohol III-64 and that the possibility of any steric
interference to cyclization via episulfonuim formation (as discussed above for the vinyl

epoxide case) was minimized.

The stage was now set to investigate the in situ deprotection / cyclization reaction

of epoxy sulfide III-5. When III-5 was treated with BF3'OEt2 under previously optimized

conditions (Scheme III-12),14 two sets of products (III-65 and III-120, separable by

137

column chromatography) were isolated. The major product, III-65, after per—acetylation

to III-121 was shown to be an endo selective epoxide opening product having the desired

2,3-cis~5—rrans relative stereochemistry about the THF ring. The structure and relative

stereochemistry of III-121 was established by means of 2D COSY and 1D NOESY

ores

-’r. 2

s i 3
OTES

Ill-5

 

h BF3 0512 (6 eq) 9”
WSW? + mixture of isomeric THF diols
Et20 (o 07 M)
0°C 10 11.72%30H
Ill-65 (major) Ill-120 (minor)
(ca 2 6 1)
ACQO DMAP 1. HPLC seperation
CH2C|2 2. A020, DMAP
WSW + unidentified mixture oi
isomeric THF diols
OAc OAc
Ill-121 Ill-122 (ca, . 3)

Scheme III-12: BF3-OEt2 mediated cyclization of the epoxy sulfide III-5 using

previously optimized conditions

experiments. lH NMR of the minor product III-120 (ca 20%) revealed a mixture of

isomeric THF diols, which were separable into two fractions by HPLC. The minor

fraction (5%) was a single isomer whereas the major portion (15%) was again a mixture

of at least two isomeric THF diols (as judged by 1H NMR). COSY analysis of the per-

acetate derivative of the minor fraction suggested another endo epoxide opening product

(III-122). However, no conclusive information regarding the relative stereochemistry of

III-122 could be obtained using 1D NOESY experiments due to overlapping signals.

More rigorous stereochemical assignment of III-122 or structure analysis of the other

15% fraction was not pursued. No further improvement in the endo selectivity could be

accomplished by varying solvents, concentration or the stoichiometry of BF_,°OEt2 (Table

138

III-2). Overall, cyclization reaction of III-5 was clearly not as endo selective as that of

the original epoxysulfide systems (III-108, Figure III-20 and Chapter I).

 

 

 

 

 

 

 

w 0 °c ,0 r, W39“ Eomngfiuiiig diols
Ill-5 Ill-6;DH Ill-120
entry Solvent (concentration) BF3°OEt2 (eq.) III-6S : III-120
1 E90 (0.07 M) 3 2.5 : l
2 EtZO (0.04 M) 6 2.8 : 1
3 CHzCl2 (0.07 M) 3 trace : major
4 CHZCI2 (0.04 M) 6 trace : major

 

 

 

 

Table III-2: Cyclization of III-5 under various conditions

We next tried to understand the reduced regioselectivity in the cyclization of III-5
and rationally design experiments to improve the same. In order for our strategy (see
Figure III—2,1 and accompanying discussion) to be successful, epoxysulfide III-121 must
rearrange to the episulfonium intermediate (III-122) and the major product must arise
from intramolecular trapping of III-122. However, it is also possible that the pathway
involving direct opening of the epoxide at C3 is kinetically competitive with that

involving intermediacy of the episulfonium ion (Opening at C2).

 

 

A
r , ,A reso '
0 ES ,9 A+ O
TESO TESO ‘
Ill-121 Ill-122

Figure III-24: Payne like equilibration of epoxy sulfide III-121 under acidic conditions

139

 

This raises the possibility that an acid catalyzed Payne rearrangement-like equilibrium
(Figure III—24) may be Operative between the activated epoxide (III-121) and the

episulfonium ion (III-122).

_ \
OTt+,R3 1, en, /I

 

R,\Ci>\/sea TMSO“ n, as N OTMS C91 . ~ N
R2 F1: OTMS K2(303 F12 OTMS O
2 30%- 80%
Ill-123 [H.124 Ill-125

Figure III-25: Rayner’s conditions for intermolecular trapping of episulfonium ions

Rayner and co-workers in their studies involving intermolecular trapping of

episulfonium ions by nitrogen nucleophiles have suggested that a Payne like equilibration
may not be involved.90 They proposed that the starting epoxy sulfide (III-123, Figure

III-25) is completely converted to the reactive episulfonium ions (III-124), which is
subsequently trapped by the external nucleophile. Although the major products isolated in
their experiments (for example III-125) were via trapping of episulfonium ion
intermediates, the possibility of an equilibrium between III-124 and activated III-123
cannot be ruled out. Thus, the same result would be obtained if the trapping of
episulfonium III-124 with the nucleophile were much faster than of the activated
epoxide. In this scenario, the presence of this Payne-like equilibrium would be
inconsequential to the product distribution.

Should a Payne—like equilibrium exist in our system, the outcome of the
cyclization event would depend upon which of the two activated species, III-121 or
III-122 (Figure III-26) is trapped faster and that in turn, would be dictated by which of
the two hydroxyl groups CS-OH or C6-OH is more available for nucleophilic attack.

Thus, in this scenario, three different routes (3, b or c, Figure III-26) leading to three

140

isomeric products, III-126 (S-exo), III-127 (S-endo) and III-128 (6—end0) are available.
Since, in course of our previous studies (Chapter 1) products resulting from attack on the
less substituted carbon of the episulfonium ion were not observed, those pathways are not
shown in Figure III—26. On the other hand, if the starting epoxy sulfide is completely
converted into the reactive episulfonium intermediate III-122 prior to cyclization, only
pathways, b and c are accessible. In either scenario, the major product will be decided by

which of the two nucleophiles (CS—OH or C6-OH) is more available for cyclization.

OTES + ,A + TESO ‘9

M ————A M
e 5, J ; sen R 5,
TESO ‘
a

Ill-1 21 III-122

a A
OH on
o, o.
e 6 O"-32 SP“ R ’- sen n " sen
5 3
:5 3 \‘

 

 

HO‘ OH HO OH
Ill-126 Ill-127 Ill-128
(5-exo) (5-endo) (6-endo)

Figure III-26: Possible route for cyclization of epoxy sulfides under acidic conditions;

endo / exo notation is relative to epoxide.

Assuming that the hydroxyl groups must be freed from silyl blocking groups prior
to cyclization, their availability would be determined by the relative rates of silyl
deprotections. The facility of silyl group removal is governed mainly by their nature and
the steric environment, which precisely are the major structural differences between our
earlier 2-deoxy-D-ribose derived epoxy sulfide III-129 (Chapter 1) and the epoxy sulfide

III-S (Figure III-27).

141

1111'

the

kit

iiil

ores OTMS

o.
Wynn M
16 :5 3 TBDPSO : 2 sen
OTES TMSO
Ill—5 III-129

Figure III-27: Comparison of structures of epoxy sulfides Ill-5 and III-29

Due to the disposition of the bulky TBDPS group in close proximity, the C6-OH
in III-129 is probably more sterically hindered as compared to the C6-OH in III-5. On
the other hand, since the C6-OH in III-129 is protected as a TMS ether and that in III-5
as a TES ether, the former might be easier to cleave. Thus due to the two factors
seemingly working in opposite directions, the relative reactivities of the hydroxyl
nucleophiles are hard to predict. Nonetheless, under the exact same reaction conditions,
III-129 afforded the episulfonium intercepted cyclized product via participation of C5—
OH in much higher selectivity (>99:1) than III-5 (ca. 3:1). This might suggest that C6-
OH in III-5 is participating to a greater extent, derailing the reaction along unwanted

paths (a or c, Figure III-26).

ores 0 on o TMSO o
WSW TBAF M861 TMSC' Msph
16 :5 3 ——-’ 16 i5 3 16 i5 3
OTES OH Et3N, DM AP OTMS
Ill-5 Ill-130 Ill-131

 

BFg'OEtZ (6 eq.)
LEtZO' O 0010 rt
1

 

 

QH
? 0,
W813" + mixture of isomeric THF diols
s".
OH
Ill-65 III-120
(ca. 3 : 1)

 

 

 

Scheme III-l3: Cyclization of three different epoxy sulfides under the same conditions

142

The following experiments provided more evidence in that direction. Epoxy
sulfide III-130 containing free hydroxyl grOUps was prepared in order to decouple the
silyl removal and epoxide opening events (Scheme III-l3). Exposure of Ill-130 to
BF3°OEt2 under the same reaction conditions (Scheme III-12) again afforded a mixture of
III-65 and III-120 in about the same proportion. We also reprotected the hydroxyl groups
as TMS ethers (III-131) thinking that TMS groups might provide the right balance of
relative rates of silyl deprotection and episulfonium formation. However, treatment of
III-131 with BF3-iOEt2 also produced the same mixture of products. Thus, considering
these experiments, a possible explanation for the reduced endo selectivity in the
cyclization of III-5 might be that the C6-OH is sterically less hindered and hence more
available as a nucleophile (than C6-OH in III-129) thereby diverting the reaction along

the undesired pathways.

 

OH
T :
W A W h , , .
16 :5 3 CH CI 50°C ' ‘5 a. + isomeric THF leIS
OTES 2 2, OH
""5 75" Ill-65

unseperable mixture

Scheme III-14: Another attempt to improve the endo selectivity in the cyclization of

III-5

In a separate project,91 it was shown that epoxy sulfide III-129 upon heating with
pTSA in CHzCl2 at 50 °C also produced the endo selective cyclized product via
episulfonium formation in high selectivity (ca. 19 : 1). When III-5 was treated with
pTSA under those conditions (Scheme III-12), the desired THF diol III-65 was produced
apparently in higher yields (75% vs. earlier yields of 56%) but unfortunately, even after

purification the product contained inseparable isomeric impurities (ca. 20% as judged by

143

1H NMR). Since we did not want to proceed with isomeric mixtures at this point in the
total synthesis, we decided to go with the conditions that produced the desired product
III-65 in highest selectivity and purity (entry 2, Table III-2). The isolated yield (56%) of
III-65 under these conditions was acceptable for purposes of bringing up more material
for the total synthesis. Moreover, III-65 could be easily separated from other isomeric

THF products by flash column chromatography.

 

 

 

 

 

 

QH TBSOTf TBSQ TBSO.
’ o... ’ o.,_ mCPBA ' 0., sen
WSW 2,6 lutidine WSP“ CH C' o o W
'9 CH2C12, 0 °C ", 2 21 O '9 "
,, ores ores
Ill-65 H 91 /° Ill-132 quam- Ill-133
TBSQ OCOCF3 1. solid NaHC03 T880 0
TFAA : OI, CH3CN 01.2 H
- sen e ,6
2,6 lutidine ‘6 , -.3
01-12012 0 °C ’bTB 2. slow column chromatography was
’ "1.134 using wet (10% H20) 8102 “1-135
60%
TBSQ reso Cl CuO reso o
7 Or ’ . CuCl -2H 0 T o,
. . .3
bras reflux "OTBs acetone, relux ’bTBs
O
Ill-132 111-1 36 3° /° Ill-135

Scheme III-15: Preparation of the aldehyde III-135

Equipped with large amounts of THF diol III-65, we set out to investigate its
transformation to the key allylic alcohol III-3. Scheme III—15 outlines further
manipulation of the THF diol III-65. TBS protection of III-65 using more reactive
TBSOTf as the silylating agent (reaction using TBSCl was incomplete after 24 h)

proceeded smoothly in 91% yield to afford bis-TBS ether III-132, which was now set up
for a Pummerer rearrangement to install the aldehyde functionality.”93 The

rearrangement was carried out in two different ways. First, using conventional Pummerer

92 .93

rearrangement conditions, the phenyl sulfide III-132 was oxidized to the

144

 

\ci

 

 

corresponding sulfoxide III-133 by dropwise addition of a CHZCI2 solution of mCPBA
which proved critical to avoid over oxidation of III-132 to sulfone. The crude sulfoxide
III-133 was next treated with TFAA in presence of 2,6 lutidine to obtain the
a-trifluoroacetoxy phenyl sulfide III-134. Hydrolysis of the rearranged product III-134
to the aldehyde III-135 proved tricky. Treatment of III-134 with a variety of hydrolyzing
agentsgfl’95 including sat. aq. NaHCO3, aq. CuClz, aq. HgClz, wet SiOz, 5% HCl, and
NaZCO3 in MeOH either led to incomplete hydrolysis or decomposition of the material.
While the hydrolysis conditions were being explored another route to obtain the desired
aldehyde III-135 was examined. The sulfide III-132 was directly converted to or-chloro
phenyl sulfide III-136 by treatment with NCS in refluxing CCl4, which was then
hydrolyzed using cupric salts in acetone.%’97 This Pummerer like rearrangement however
provided the aldehyde III-135 in only 30% yield. We then refocused our attention to the
a-trifluoroacetoxy phenyl sulfide III-134 to further explore its hydrolysis. Ultimately,
we found that treatment of a CH3CN solution of III-134 with solid NaHCO3 for 18 h

followed by slow elution of the product on a wet silical gel (10% H20) column provided

the aldehyde III-135 in 60% yield.

3) Synthesis of a model allylic alcohol

We had planned to access the target allylic alcohol III-3 via a 1,2 chelation
controlled addition of vinyl magnesium bromide to the aldehyde III-135. With the
requisite aldehyde available, we were only a step away from III-3. In order to extensively

investigate the proposed regioselective intermolecular epoxide opening strategy (Figure

145

Ill-2), we needed sufficient amount of the allylic alcohol III-3 in hand. At this point,

instead of bringing up more material to acquire adequate quantities of III-3, we decided

 

OH 0 T880 0 1.mCPBA ‘ . T880 0 20
WSW resort Mean 2. TFAA, 2.61utidine WH
on ,3 —~—’ on ., T’ on -.3

’OH ’ores 3. NaHCO3, wet 8102 'ores
Ill-137 R = TBDPS Ill-138 Ill-139
1350 on
”M981 R0 0122 /
.......... >3
’ores
Ill-140

Scheme III-16: Synthesis of a model aldehyde III-139

to switch to a model allylic alcohol (III-140, Scheme III—l6), which could be obtained via
a shorter reaction sequence. THF diol III-137 (derived from 2-deoxy—Deribose) was
available form our earlier studies (Chapter I). Structurally (constitution and
stereochemistry), III-137 is akin to the real THF diol III-65 (Scheme Ill—12), the only
differences being the stereochemistry at C6 and an alkoxy methyl side chain, instead of
the long alkyl chain. Since these differences reside in the side chain remote to the
reacting end of the allylic alcohol, we thought that III-137 would serve as an appropriate
model THF diol. The aldehyde III-139 was synthesized from III-137 using the same

transformations as before (Scheme III-16).

146

OP OH

Mg H 9*” H 9e
0 o ' on HO o ‘ on no 0,.2 /
M —. sag/V “kg/V
Nu H
OP op OP

“I

1.2 Chelation desired diastereomer
Nu H H 9e H QP 0e OH
0 on H o : on no 0.._2 /
O \ OH —=— ,3
Miro" oe OP
1,3 Chelation undesired diastereomer

Figure III-28: 1,2 vs. 1,3 Chelation control in addition of vinyl magnesium bromide to

aldehyde III-139

We then turned to investigate the addition of vinyl magnesium bromide to
aldehyde III-139 to prepare model allylic alcohol III-140. 1,2 Chelation controlled

addition of organometallic reagents across a-tetrahydrofuranyl aldehydes is well
precedented in the acetogenin literature.98'100 In our case, in addition to a 1,2 chelation,
1,3 complexation of the aldehydic oxygen with the ring hydroxyl group was likely to
occur during organometallic addition reactions]01 As shown in Figure III-28, 1,2

Chelation (III-141) would lead to the desired diastereomer of the allylic alcohol III-140,
Whereas 1,3 complexation (III-142), due to attack on the opposite face of the aldehyde
would produce the unwanted diasteromer III-143, epimeric at the newly formed

stereocenter. Thus, to minimize any potential 1,3 chelation event, bulky TBS groups were

used to protect the hydroxyl groups in III-139.

147

 

T880 0 TBSO OH

 

 

 

 

 

 

 

R0 0'22 H ”MQBT ~ R0 01.2 /
,3 EtZO / conditioE .9
ores ores
Ill-139 Ill-140
conditions yield (%) dr
MgBrZOOEtQ, 0 °C, 2 h 57 7 : 1
~20 °C to —30 °C, 1 h 68 10: 1
—40 °C, 2 h 80 10 : l

 

 

 

 

 

Table III-3: Synthesis of model allylic alcohol III-140

After some experimentation (Table III-3), the desired allylic alcohol III-140 was
Obtained in high diastereoselectivity (10:1) and yields (80%). The absolute configuration
of III-140 at the newly formed stereocenter was established by Trost’s O-methyl

mandelate analysis102 (absolute configuration assignment of chelation controlled addition

products of the real aldehyde III-135, using Trost and Mosher ester analysis is discussed

in detail in Chapter IV)

b) Determination of the enantiomeric excess and the absolute configuration of diol
III-59

As mentioned earlier (Scheme III-12), the relative configuration of the THF diol
III-65 (which would eventually become the hydroxylated THF portion (C l3-C37) of
mucoxin, was established by 1D NOESY analysis. Since all the stereocenters in III-65
originated from asymmetric transformations, we decided to independently confirm the
stereochemical outcome of the asymmetric dihydroxylation. For this purpose, the diol

III-103 (Figure III-29) was chosen. III-103 was obtained from trans diol III-102 via a

148

89

am

tfi

d1.-

Sharpless asymmetric dihydroxylation reaction (Scheme Ill-8). According to the

Sharpless mnemonic device (Figure III-29) for predicting the enantioselectivity in the
asymmetric dihydroxylation reaction,56 the northeast (NE) and the southwest (SW)
quadrants are more open to the olefin substituents. The SW quadrant is considered an

attractive area for large aliphatic groups. If the olefin III-102 is positioned accordingly

(Figure III-29), AD-mix 01 should react from the bottom face, leading to the desired S,S

diol III-103. The

 

 

AD-mixB
NVV - NE
OPMB _ _/OPMB OH
H AD-leCI WH E WOPMB
__ H 16 is
16 HO OH OH
SVV SE III-103HOMe
OH
en
. _ o
AD-mix (1 Ha 9‘”) MFA OPMB DCC IDMAP
Ill-102 {\M
H” O(R)-MPA

"F141

Figure III-29: Mnemonic device for Sharpless asymmetric dihydroxylation reaction as

applied to trans olefin III-102

% ee of III-103 was determined after its derivatization to the bis-(R)—MPA ester III-141.

1H NMR of III-141 showed only one set of Ha, Hb protons indicating that diastereomeric
ratio of III-141 was >98:2.

In order to independently confirm the absolute configuration of diol III-103, we
decided to use exciton coupled circular dichroism (ECCD) spectroscopy. Use of ECCD
for determination of the absolute configuration of 1,2 and 1,3 diols is well

precedented.m3'104 For this purpose, the diol is first derivatized to install chromophoric

149

groups at the chiral centers in question. A chromophore, when exposed to circularly
polarized light, undergoes electronic excitation. When two such chromophores are close
in space, their electronic transition dipole moments interact through space. Consequently,
the excited states of the individual chromophores split each other resulting in two excited
states having different energy levels for the system as a whole. The CD spectrum of such
a coupled chromophoric system becomes bisignate or a ‘split’ CD. The split CD either
shows a positive signal at longer wavelength and a negative signal at shorter wavelength
(termed as a positive couplet), or vice versa (negative couplet). A positive CD couplet
results from chromophores arranged in positive helicity. A positive helical system, in
turn, is defined as one in which the transition dipole moments of the two interacting
chromophores are oriented in a clockwise manner going from the front to the back
chromophore (Figure Ill-30).
One of the most common chromophores used for the derivatization of diols is
p-dimethylamino benzoate group. This group has a large coefficient of absorption (8 =
28,200 (CH3CN); 11m, = 307 nm), which could lead to strong CD signals. Also, its

transition dipole (La, Figure III-30) is oriented parallel to the CO bond and since the

 

:29 we ECCD

M82N ~l” O O
R . . .
Laߢ921 La: transrtion dipole moment
0 H

Figure III-30: A positively helical system comprises of two interacting chromophores

twisted in a clockwise direction going from the front to the back chromophore

150

 

(111'

511:

C01

lit

twist of the adjacent transition diploes in turn directly correlates to the sign of the ECCD
spectrum, the absolute sense of twist between the vicinal C—O bonds (absolute
configuration of the diol) can be predicted from the sign of the ECCD spectrum. The

requisite di-benzoate derivative III-144 was synthesized from the diol III-103 as shown

in Scheme 111.17.105

0

 

 

 

 

1 . KOIBU I'l'103 O/LszHli'pNMez
Ill-142 PhH, reflux Ill-143 50% MezNP‘CsH‘tTr Ill-144
40% o
O O

DDQ 4 OJLCSprNMez BSTFA O/lLC6H4‘pNMez
(301120121120 (9:1) ,5 on 50 °C ,5 OTMS
o o, rt 50% Meng‘CeHitYO 111-145 MeZNP‘CsHfifO III-146

o 0

e
A

09 9 D
H 0 Hscmzmi‘ 0 O (croutons
no 110 ”O o H no H
U H
B C
O = M92N-C6H4-CO

Scheme III-17: Synthesis of dibenzoate derivatives of the diol III-103 for ECCD

analysis

Out of the three possible staggered conformations (A, B and C) of such a di-
benzoate system, B is ECCD inactive since the angle between the two transition diploes
is 180°. Conformation A bears two gauche interactions and therefore would be lower in
energy than C, which involves three such interactions. The transition dipoles in the
predominant conformation A are oriented in a clockwise direction going from the front to

the back chromophore and should lead to a positive ECCD signal. Thus, if the absolute

151

configuration of the original diol is S,S, its di-benzoate derivative is expected to produce
a positive ECCD signal. With III-144 in hand we now initiated the ECCD analysis.
Unfortunately, no distinct ECCD spectrum was observed for III-144 in various solvents
(CHzClz, MeCy and MeCN). We thought that the PMB group might also be behaving as a
chromophore causing additional dipole interactions with that of the two benzoate groups.
The PMB group, therefore was deprotected to generate the free alcohol III-145. The
ECCD sign in case of III-145 was found to be solvent dependent. In polar solvents such
as MeCN and CH2C12 : MeOH (l : 1) a positive spectrum was obtained while in less polar
CHZCIZ, the sign switched to negative. It is likely that the free hydroxyl group in III-145
developed intra / intermolecular hydrogen bonding with the p-dimethylamino benzoate
groups thereby affecting the stability and population of the conformations. Moreover, the
extent of such hydrogen bonding possibly is dependent on the polarity of solvents.
Ultimately the TMS ether III-146 provided consistent results. In a range of solvents, a
positive ECCD was observed. A representative spectrum of III-146 in MeCN is shown in
Figure III-31. Thus, the 5,5 configuration of diol III-103 was confirmed.

101
0“ .

280 320 340

 

 

 

mo! (:0

 

 

 

nm

 

 

 

Figure III-31: ECCD spectrum of III-146 in MeCN

2. Synthesis of vinylic epoxide III-4

With model allylic alcohol III-104 in hand, our next goal was the vinylic epoxide
III-4. Epoxide opening reactions, in general are facile under acid catalyzed conditions.

152

We anticipated that III-4, due to carbocation stabilizing vinyl group adjacent to the

42.55

epoxide moiety, could be activated under mildly acidic conditions and that the 01,6 —-

unsaturated ester functionality at the other end should be compatible with such mild
acidic medium. As described earlier (Figure III-2), we planned to employ Knochel’s

three component coupling protocol to build the carbon skeleton of the target epoxide III-

4.20 Retrosynthetically, III-4 was broken down into three fragments, alkynyl iodide III-6,
1,4-diiodobutane III-7 and (bromomethyl) acrylate III-8. Iodide III-6106 and acrylate

III-8107 are easily accessible, whereas 1,4-diiodobutane III-7 is commercially available.

Lithium acetylide of the TBS protected propargyl alcohol (III-148) was quenched
with I2 to efficiently obtain the alkynyl idode III-6 (67% overall yield Scheme III—18).
Synthesis of the (bromomethyl) acrylate III-8 on the other hand proved problematic.

Mitsunobu esterification of (bromomethyl) acrylic acid (III-149) with 3-buten~2-ol

(III-150) has been reported to provide acrylate III-8 in 70% yield.107 However, initially

we only obtained III-8 in about 30 -— 35% yield.

 

 

 

 

 

 

 

TBSCI nBuLi T880
HO \\ ————->TBSO/\\ \‘E—fi
lmid.,DMF l -78°C-rt
2i
Ill-147 ””7504, Ill-148 THF 89% Ill-6
O O
B/YLOHI +/OkH/ PPh3/DIAD BYLO
f / —> r
Et 0,69% M
Ill-149 Ill-150 2 Ill-8
O
\Q J H
DCC / DMAP 7\ TOW/”wick
EDC/DMAP O M
ANHCY BOP / 5131\1 “M 51
04%;, $0012 / DMAP, Py
111- 153 iii-152

Scheme III-18: preparation of the three component coupling partners, III-6 and III-8

153

28,108

Several Mitsunobu esterification conditions including different solvents,

reagent stoichiometry and order of reagent-addition were examined, but did not lead to
improved results. During this exploration, a common side product III-151 (formed by

N-alkylation of DIAD) was observed. Such N-alkylation of diazoesters is known to occur
when the acid component is less reactive due to steric bulk or weak nucleophilicity.109

We were finally, able to obtain a consistent yield (69%) of III-8 by drop wise addition of
an ethereal solution of III-150 and PPh3 to a solution of III-149 and DIAD in ether.
Meanwhile, several other esterification reactions involving DCC, EDC, BOP and SOCl2
were also attempted without any success. Interestingly, in the DCC coupling reactions,
the cyclized product III-153 was cleanly obtained probably via an intramolecular
displacement of all ylic bromide in the DCC—acid complex III-152.

(Bromomethyl) acrylic acid III-149 was prepared following a reported procedure
(Scheme III-19).l ‘0 Thus, diethyl malonate III-154 was transformed into the diol III-155

via treatment with formalin solution; III-155 upon heating with aq. HBr afforded III-149

O

37% (CH20)n EtOZC 002E148%.H8r
EtOZCVC02El _> in Br/YKOH

KHCO3, 84% OH OH 125°C
35% (3 crops) Ill-149

 

 

Ill-154 Ill-155

 

O O

NBS 1. NBS
YLOH X: Br/YLOH ~>< You

M OH /THF 2. Jones
"l'156 892344) reeflux "l'149 Odeatlon ""157

Br2, CH2012 B, 0 K81 / DMF
l1, 80°/o j/‘LOH LiBr, LlQCO3

Bf _ AgOTf I CH2CI2
"' '58 KOH lEtOH

 

 

 

Scheme III-19: Synthesis of bromomethylacrylic acid III-149

154

b1.

in acceptable yields. Since tedious crystallization was necessary to recover the product

prepared using this protocol, we also investigated alternate routes to III-149 (summarized

in Scheme III-19). However, both radical bromination (III-156 and III-157)'”"12 and

bromine addition / elimination (III-58)l 13’] 15 were unsuccessful.

The three component coupling reaction involved sequential, one pot coupling of
alkynyl iodide III-6 and (bromomethyl) acrylate III-8 with diiodide III-7 (Scheme
III-20). Treatment of III-7 with activated metallic zinc at 40 °C and subsequent exposure
to CuCNOLiCl complex generates putative bis-heterobimetallic species III-159. The
organocopper end of III-159 being more reactive, preferentially couples with the first
electrophile III-6 at low temperatures (-60 °C to —35 °C); the acrylate III-8, added
second then couples with the organozinc portion to provide the highly functionalized

intermediate III-160 in good yield (45%).”6‘117 It must be pointed out that use of

anhydrous pentane as a co-solvent was essential to obtain reasonable (40 to 50%) yields.
The yields obtained in our system, albeit lower than in Knochel’s systems (60 — 80%),

were acceptable since the entire carbon skeleton of the right hand fragment was installed

in a single step.

155

 

 

Zn, THF CuCN°2LiCl Ill-6 (0.7 eq)
l/W' *' t 'ZnM/CMCNflnI 4'
III 7 40 °C, 20 h THF / Pentane Ill-1 59 - 60 °C to -— 35 °C
' 0 °C, 30 min. 15 h

 

 

 

O
Ill-8 1.5 3 h
( 9Q) ; T1350 M TBAF

- 78 “C to rt, 45% / THF. - 10 “C, 85%

 

III-160
O
M0 H2, Lindlar's cat. 0
HO / M ' ‘ : HOMO
"F161 qurnollne cat, Ill-162 M

EtOAc, 11; 93 °/o

D-(—)—DIPT, Ti(O‘Pr)4

O
DCC O O
: H 0W0 _.____... thkOx/Hgn/ko
Ct1202 3131 MS, _-.25 c 0 III-163 M 9W“ "o" M
68 /o, 92 /o selectlvrty Ill-164
DMP
0142012, 11.89%

 

 

 

H o + o
0W0 PthCHgBr : W0
0 Ill-165 M NaHMDSWHF 0 111-4 M
'1OOC,70°/o
O
TBSO, _ | Br/YLO
Ill-6 Ill-8

 

 

 

Scheme III-20: Synthesis of vinylic epoxide III-4

Further elaboration of III-160 to the target vinylic epoxide III-4 was
straightforward (Scheme III-20). Sequential TBAF deprotection (85% yield) of TBS
ether III-160 and partial hydrogenation of the propargylic alcohol III-161 (93%)
produced the cis-allylic alcohol III-162 poised for the Sharpless asymmetric epoxidation
reaction. After some experimentation, the cis—epoxide III-163 was obtained in good
selectivity and yields. The selectivity of epoxidation reaction was determined after
derivatization of the epoxy alcohol III-163 to the corresponding MPA ester III-164. The

final transformations included Dess-Martin periodinane oxidation of III-163 to the

156

corresponding aldehyde III-165 (89%) and subsequent Wittig olefination of III-16S

(70%) to secure the target vinylic epoxide III-4.

3. Attemped intermolecular epoxide opening

With the requisite substrates, viz. III-140 and III-4 available, efforts were now
focused on the proposed regio-and stereoselective epoxide ring opening reaction. We first
decided to try Mioskowski’s optimized conditions55 for our epoxide opening reaction.
Initially, we chose a commercially available primary alcohol III-167 as the nucleophile
(Scheme III—21). The epoxide III-4 and the alcohol III-167 (1.1 eq.) coupled at ambient
temperature in presence of catalytic BF3°OEt2, to afford the ring opened product III-168
in 50% yields. The regiochemistry of III-168 was established by COSY analysis of its
acetate derivate III-169. We were greatly encouraged by this result because the sensitive
oufi unsaturated ester functionality seemed to tolerate the reaction conditions reasonably

well and excess amount of alcohol was not required.

0 \
1313-0512 (10 mol%) 0“ HA
/ . . o _ W0», 0
a: + NO 7 7
O M 7 \ O

H
CHQClz, rt; 500/0

 

 

(1.1eq.)
111-4 Ill-168
III-167
A020, DMAP OAc W
*7 \ o.,, o
CHzC|2.fl;quant. W (\ka
\ O

Ill-169

Scheme III-21: A trial intermolecular ring opening of the vinylic epoxide III-4 using

Mioskowski’s conditions

We then decided to move on to model alcohol III-140 hoping to further optimize

the reaction to increase the yields. Reaction of III-4 and III-140 under the same

157

conditions (Table III—4, entry 1) resulted in rapid decomposition, and the desired product
was isolated only in 12% yield, after careful chromatographic purification. We then

reduced the amount of catalyst and temperature as summarized below (entries 2-4, Table

 

 

 

 

 

 

ores OH TBSO / /
Ill 90 . O ‘2 / conditions 17 l
-4 + H“ . H R0 "’0 12‘0 9 .
bras OTBS OH 0 o/k/
111-140, R = TBDPS III-170
(1 - 3 eq.)
conditions result
BFfOEt2 (10 mol%), CHZCIZ, rt, 30 min III-170 (12%)
BF3-0Et2 (1 mol%), CH2C12,O °C, 1 h no reaction
BF3°OEt2 (4 mol%), CHzClz, 0 °C, 4 h no reaction
BF3°OEt2 (4 mol%), CHZCIZ, rt, 12 h III-170 (20%)
Cu(OTf)2 (10 mol%), CHZClz, rt decomposition

 

 

 

 

Table III-4: Preliminary attempts at optimization of the coupling of III-4 and III-140

III-4). To our disappointment, the yield increased only up to 20%. Also, the reaction
using Cu(OTf)2 (another catalyst that was shown to be as efficient as BFyOEt2 in
Mioskowski’s studies) lead only to decomposition. In all cases, unreacted alcohol III-140
was recovered.

At this point, it appeared to us that this reaction might need extensive
investigations that would involve screening of a variety of acid promoters, solvents and
temperature conditions. We therefore decided to further simplify the system to model
vinylic epoxides III-174 and III-175 and a model alcohol III-177, which could be
accessed quickly as shown in Scheme III-22. Commercially available ethyl 6-hydroxy

hexanoate III-171 via sequential PCC oxidation and E-selective Wittig olefination was

158

transformed to the 01,[3-unsaturted aldehyde III-172. III-172 upon Luche reductionllg‘l ‘9

followed by SAE furnished epoxy alcohol III-173. Finally Wittig olefination of the
aldehyde derived from III-173 with two different ylides provided the corresponding

vinylic epoxides III-174 and III-175. The allylic alcohol III-177 was obtained simply by

 

 

 

 

 

1. FCC / NaOAc 1. NaBH4, CeCla-7H20
O CHzclz, n, 72% 0 EtOH, 91%
/\O’u\/\/\/OH 2 Ph PCHCHO ' AOWCHO ;
_ - 3 _ 2. Ti O‘Pr ,D-DIPT
"' 17‘ PhH, reflux, 66% "' ‘72 téuooh Ms 4 A
72%
1. DMP, CHch2
0 rt, 64% o 121
/\ WOH + t /\ W
O "'0 2. PhapARBr" O "0
"H73 Ill-174, R = H
KHMDS, toluene :THF III-175, R = Me
(1 :1), -10°C, 86%
0 OH
0*“ ———9—~m (W
81
THF, 0 °C
111-1 76 79 /° 111-177

Scheme III-22: Synthesis of simplified model vinylic epoxides and an allylic alcohol

addition of vinyl magnesium bromide to cyclohexane carboxaldehyde III-176. With the
requisite substrates in hand, the optimization process was continued.

After some experimentation, we found that by slow addition of BF3°OEt2 (2
mol%) to a mixture of III-174 and III-177 (Scheme III-23) at ambient temperature the

yield of the desired product III-178 could be improved to 42%. However, the same

159

Cy

OH \\ <
O R / BFa‘OEIQ (2 mol%) 0 0 Fl
”0W + = W
0 CHZCIQ, rt, /\0 ,

 

 

 

111-174, R = H 111-177 III-178, (42%) R = H (311
111-175, R = Me (1 eq.) 111-179, (17%) R = Me
BF3'OE12 (2 mol%) 0 /
111-4 + 111-140 a- 111-170 /\o
CH2012, rt; 20% Ill-180, 35% o

 

 

 

Scheme III-23: Further optimization studies on the ring opening using model systems

procedure failed to improve the efficiency of the coupling of III-4 and III-140 beyond
20%. We suspected that in case of the terminal vinylic epoxides such as III-174 and
III-4, generation of undetected 1,4 addition products might be responsible for lower
yields. Thinking that the methyl substituted vinylic epoxide III-175 might diminish the
likelihood of the 1,4 addition pathway, it was treated with III-177 under the optimized
conditions. This, however led to even lower yield (17%) of the desired product III-179.
Furthermore, a side product III-180 formed via an intramolecular 1,2 hydride

0

migration12 was isolated in 35% yields. The studies described so far indicated that the

epoxide opening reaction might be acutely sensitive to the steric environment around the

reaction centers.

MC),
OH

0 Me / Rh[(CO)QCI]2 W
flow + > /\O ;

 

 

'(3 (20 mol%) OH
THF, rt, <23°/o
Ill-177 III
111-175 (5 eq.) -179
0 Me OH Rh[(CO)2Cl]2 Oallyl OH
AOW + W (20 mol%) \v0 4 i / + VOW
O (20 eq.) THF, rt, 60°/o O OH O Cally!
III-175 Ill-181 Ill-182

Scheme III-24: Application of Lautens’ conditions to model systems

160

We next turned to examine Lautens’ conditions (Scheme III-24).42 In their

studies, when terminal vinylic epoxides were used, a mixture of 1,4 and 1,2 addition
products was produced. We therefore chose III-175 as the electrophile. Also, in the
original report, 10 —- 3O equivalents of the alcohol were used with 5 mol% catalyst
loading. We modified those conditions to S equivalents of the nucleophile III-177, 20
mol% catalyst and the reaction was run at 3 M concentration. Under these conditions,
only 23% material was recovered which contained the desired product III-179 along with
unidentified side products. Interestingly when allyl alcohol was used, a l : 1 mixture of
regioisomeric products III-181 and III-182 was isolated. No further experimentation
using Lautens’ rhodium catalyst was continued.

Since the terminal vinylic epoxide III-174 proved superior to III-175 in terms of
yields and regioselectivity (Scheme III-23), investigations were continued using III-174.
Using the epoxide III-174 and the alcohol III-177, we now screened a range of Lewis
and protic acids (Scheme III-25) under different solvent and a wide window of
temperature (-78 °C to reflux). With the exception of Sn(OTf)2 and TMS(OTf)2 mediated
reactions where the desired product III-178 was obtained in 25%~40% yields, all other

reactions led either to recovery of the starting materials or decomposition.

ZITCIZITHF
Zn(OTf)2/THF
Sn(OTf)2 /CH2C12
TMSOTf/CH2012 V
Ill-174 + III-177 /\ = III-178
CsF/DMF
CSA/(3112012
Tron/0142012
AI(OAr)3 / 0112012

 

Scheme III-25: Screening of various acid catalysts for 8N2 opening of the model epoxide

I61

All the trials so far, led to us to think that the nucleophilicity of alcohols decreases
significantly with increase in steric bulk and hence under strongly activating conditions,
the vinylic epoxides followed intramolecular rearrangement (for example, III-180,
Scheme III-23) or other decomposition pathways.

One of the tactics used to increase nucleophilicity of such hindered alcohols is
their derivatization to the corresponding tin ethers. The enhanced nucleophilicity of tin
ethers as compared to the parent alcohols has been attributed to the more polar character
of Sn~O bond than H-O bond.121 In carbohydrate chemistry, hindered glycosyl accepters

are often derivatized as tin ehers which facilitates their O-glycosidation reactionm"24

This precedent prompted us to explore the use of tributyl tin ether derivative (III- 183,
Scheme III-26) of the model alcohol III-177 as a nucleophile. III-183 was conveniently

accessed by treatment of III-177 with bis-tributyl tin oxide in refluxing toluene
accompanied by azeotropic removal of H20.”123 Typically, tin ethers are used in

conjunction with lanthanide triflates. Three different triflates (Scheme III-26) were
examined for the coupling of III-183 and III-174 in refluxing toluene, all of which
resulted only in recovery of the starting materials. In our earlier experiments, BF3-0Et2
proved to be most effective catalyst. Unfortunately, in this case, all BF3°OEt2 mediated
coupling reactions (Scheme III-26) failed. From all our unsuccessful attempts at coupling
vinylic epoxides with alcohols as well as other reports, it became clear that activated
vinyl epoxides in absence of an effective nucleophile, are notorious for rapidly

undergoing internal rearrangement and other decomposition processes.

162

OH OSnBu3 Ill-174

(BU3Sn)20 catal st
/ / ._____.y no reaction
toluene toluene

 

 

reflux rt to reflux catalyst :Zn(OTf)2, Sn(OTf)2, Yb(OTf)3
Ill-177 III-183
L Ill-174, BF3’OEt2 _
4* no reaction
CHQCIZ
toluene

toluene, Bu4N+Br
Scheme III-26: Attempted epoxide opening reactions using a tributyl tin ether

While in search of alternative ways to activate an epoxide, which would avoid
other unwanted rearrangement pathways, we thought that the thiophenyl group (i.e., use
of an epoxy sulfide instead of a vinylic epoxide) might serve the purpose. In the course of
this and the earlier project (Chapter I) we had clearly established the effectiveness of a
thoiphenyl directing group in epoxide activation via an episulfonium ion formation and
its subsequent trapping by an internal hydroxyl group. Intermolecular trapping of

episulfonium ions (generated from epoxy sulfides) by nitrogen nucleophiles has been
reported by Rayner;90 however use of alcohols or other oxygen nucleophiles for this

purpose is not known. We thought that an epoxy sulfide activated via episulfonium ion is

less likely to self decompose than activated vinyl epoxides since the former does not

OH
0 1 L A 9“ 1 ? 1 1. Pummerer \1 \
RM . . ~ RMSPh :
2 SP" 0 RM.» = R o R
OH

m_ C 2. Wittig
184 912% P“ 91” olefination

Ill-185 III-186 Ill-187
I QH OH
. Pummerer R @122

2. RM 0
R M

1
III-188

 

 

Figure III-32: Design of an epoxy sulfide substrate for regioselective ring opening by

alcohols

163

111‘

involve highly reactive allyl cation type species which is prone to rearrangements and
other decomposition processes. Thus, an episulfonium ion if trapped rcgioselectively at
C2 (III-185, Figure III-32) would generate an intermediate phenyl sulfide (III-186),
which can be easily manipulated to the requisite RCM precursor III-187. Moreover, the
aldehyde intermediate en route to III-187 could be manipulated into a variety of other
useful functionalities (for example III-188) thus offering an efficient entry into
synthetically useful fragments.

To test our proposal, the required epoxy sulfide III-189 was quickly obtained

from available epoxy alcohol III-173 using the Hata reagent (Scheme III-27).85

 

 

O (Ph8)2 O
/\ W0“ /\ WSPh
0 "1c BugP, TEA 0 "15)
"F173 659% "L189
OH
BFg'OEtg 1
/ (20 mol%) W Cy
Ill-189 + +7 /\0 . 0
"l'177 gg/ZCIZ 0°C to 1'1 "'_190 OH
(2 eq.)

0 1
AC 0 ;
————2———-> flow/O Cy
DMAP 6Ac
Ill-191

Scheme III-27: Synthesis and acid catalyzed intermolecular coupling reaction of an

epoxy sulfide with an alcohol nucleophile

After preliminary optimization, we found that slow addition of BF;.,°OEt2 to a solution of
pre-mixed epoxy sulfide III-189 and the alcohol III-177 at 0 °C followed by warming
the reaction to ambient temperature provided a ring opened product in 75% yield. COSY
experiment of the acetate derivative III-191, however, showed that the undesired

regioisomer III-190 was produced via a [1,2] thiophenyl migration event. Although the

164

desired regioselectivity in Opening of the epoxy sulfide III-189 was not obtained, the
reaction was much cleaner and higher yielding than any of the attempted vinyl epoxide
openings suggesting that rearrangement / decomposition pathways were reduced in this
case.

A popular tactic employed for controlling the regioselectivity and increasing the
facility of intermolecular epoxide opening reaction is to tether the incoming nucleophile

to the epoxide prior to the desired bond formation via metal mediated chelate

127

complexesm’125 Miyashita126 and Saigo have independently contributed to this area

through the development of stereospecific epoxide substitution by use of

 

R'\ ,R'
O 1 R'3AI,CH2C12 o’Al\Rv OH
RMSPh —> w ""‘_'" RJY\SPh
'30 °C 10 '50 DC R 2113 R.
91 % to 97% ‘Ph
Ill-184 Ill-192 Ill-193
OH
(R'O)3Al R/k‘flsph
? OR'
Ill-194

Figure III-33: Regio-and stereoselective alkyl group transfer to epoxy sulfides

organoaluminum reagents. It was shown that organoaluminum reagents efficiently
transfered alkyl or alkynyl groups to 2,3 epoxy sulfides (III-184, Figure III-34) under
mild conditions with complete regio-and stereocontrol. Presumably, the trialkyl
aluminum initially acts as a Lewis acid to generate the episulfonum intermediate bearing
the ‘ate’ complex (III-192). An alkyl group is then transferred to C2 (the choice of
solvent was critical to the regioselectivity) to afford the substitution product (III-193)

with a net retention of configuration.

165

Inspired by these studies, we set out to investigate whether trialkoxy aluminum
species would transfer an alkoxy group in a similar manner, to afford the corresponding
C2 ring opened product (III-194). Aluminum aryloxides have been known in the

literature as effective Lewis acids for oxygen containing substrates. Their Lewis acidity is

OH

/
AIMe3 + W mIAKORh] JEL. no reaction
Ill-177 -w / -78 °C to rt
(3 eq.) R =

Scheme III-28: Attempted preparation and reaction of a trialkoxy aluminum with the

epoxy sulfide III-189

tuned by the steric and electronic nature of the aryloxy ligandsng“30 On the other hand,

use of alkoxy aluminums in an analogous manner has not been investigated. Aluminum
aryloxides can be easily prepared by reacting AlMe3 with an appropriate aromatic alcohol
and depending upon the nature and stoichiometry of the reagents, di~or tri arylxoxy

aluminums can be generated.129 Accordingly, we treated AlMe, with our alcohol

nucleophile III-177 (Scheme III—28) and the resultant solution was exposed to epoxy
sulfide III-189. However III-189 remained unchanged for a prolonged time even at room
temperature. A likely explanation for the lack of reactivity is that alkoxy aluminum
Species are not acidic enough to generate episulfonium intermediates.

Being aware of a report that used (C6F5O)3Al as a Lewis acid to promote epoxide

rearrangments,120 we next treated AlMe, with a 2 : 1 mixture of C6FSOH and III-177

(hoping to generate an aluminum species that would promote episulfonium generation as

well as transfer the desired alkoxy group). When the resultant solution was reacted with

166

Ill-189, the epoxy sulfide was completely consumed and two products III-195 and

III-196 were formed Scheme IIl~29). Unfortunately, the major product (76%) resulted

CN1
C) (3R
/ Ill-189 Wsph
AlMe3 + + CGFSOH -——————~ /\0 .
CH2012 OH
111-177 (29 ) -78 °C tort Ill-195R:
(1 eq.) ‘1' 051%,, (76%) ......
/
"ween=
(13%)
OH 0“ Ill-189
AlMea + / + Ph Ph ————* decomposition
CH2012
Ill-177 2 -78 Ctort
(1 eq.) ( eq)

Scheme III-29: Attempted alkoxy group transfer to the epoxy sulfide III-189

from transfer of pentafiuoro phenoxy group transferred to C2 while the minor product
(13%) contained the desired product. Lastly, when 2,6-diphenyl phenol was used in a
similar manner (in an attempt to prevent aryl group by increasing steric bulk), the
reaction resulted in decomposition.

After the unsuccessful attempts at the intermolecular epoxide opening with
desired regioselectivity, we considered yet other ways to access the target RCM precursor

III-2. Cyclic sulfates and sulfites derived from vicinal diols have served as effective

131-135

epoxide surrogates especially in intermolecular ring opening processes. Cyclic

sulfates are inherently reactive toward ring opening than their epoxide analogs possibly
due to the internal O-S-O angle strain and a partial double bond character of the ring O-S

bond (III-198, Figure III-35).131 Cyclic sulfites (such as III-199) on the other hand, can

be activated by Lewis acids via coordination with the lone pair on the sulfur atom.

167

I.) ‘ O O
‘K 0

R1 R2 R11 :R2984 R1 F12

III-197 III-198 Ill-199

Figure III-34: Cyclic sulfates and sulfites as epoxide surrogates

Accordingly, we decided to explore cyclic sulfate and sulfite analogs of the vinyl

epoxide III-174. The requisite diol precursor 111-200 was easily obtained by
stereoselective hydrolysis of III-174 (Scheme III-3O).136 Unfortunately, all attempts to
prepare the cyclic sulfate from 111-200 failedm'137 The reaction mediated by sulfuryl
chloride produced the vinylic epoxide III-174 presumable through the intermediacy of
chlorohydin III-201.131 The cyclic sufite III-202 which, was accessed from III-200 in
high yields,138 failed to combine with the alcohol III-177 under a variety of

acidicconditions. ‘39

 

0 THF : H20 15%HCIO4 0 0“
flow i flow

 

 

 

111-174 11. 30% “(-200 OH
o\ lo
\S’
so 01 0’ 0' CI
111-200——2—2—> ‘1. ( / —.» x / —» 111-174
3001 EtaN' DMAP 5” 5Q
2 111-201
Et N ' -
Ti-iF M Ill-174
87%

 

01 / 111-177
., Ho OT
/\0 . ’o __(__f.)3+ no reaction/decomposition

’ b-S’, won)3
111 202 o Zn(OTf)2
11011

Scheme III-30: Attempted preparation and ring opening of cyclic sulfates and sulfites

168

Due to the failure of acid promoted coupling of vinylic epoxides and equivalents thereof
with alcohols, we decided to explore the desired CO bond formation under basic or
neutral conditions as a last resort. Given the poor nucleophilicity of alcohols, we decided
to employ substrates such as III-203 (Figure III-35) containing a good leaving group at
anallylic position. One might anticipate that such allylic electrophiles would be reactive
enough toward nucleophiles under mildly basic or neutral conditions. In the carbohydrate

literature, O—glycosidation reactions of hindered secondary eletrophiles are often

x FtO‘M+ QR
/ -————- ‘ /
1“ a or HEW
OP ROSHR'a OP
ill-203

Figure III-35: 5N2 displacement of allylic electrophiles with alkoxides

facilitated by treatment of their triflate derivatives with stannylated glycosyl

122-124

acceptors. These couplings are usually carried out under near neutral conditions and

are compatible with benzoate or acetate protecting groups elsewhere in the substrates.

To continue efforts in this direction, the selectively protected diol III-206
containing a free allylic alcohol functionality was synthesized as shown in Scheme III-31.
Synthesis and isolation of triflate III-207 proved challenging. During the preparation,

subzero temperatures had to be maintained along with careful control of the reagent

‘40'142 As can be imagined, 111-207 was extremely

stoichiometry and the order of addition.
sensitive to aqueous work up. Even after meticulous non-aqueous work up

procedures,l‘”‘143 III-207 could not be completely freed of DMAP derived salts. Even

when the cleanest samples of 111-207 were treated with sodium or tin alkoxides of

III-177, no desired product was obtained. Similarly, tosylate III-199, tough relatively

169

more stable to isolation procedures, could not be purified from TsCl derived side

 

products.
PMBOH
O BF3’OE12 O OPMB NaH 0 OPMB
W ____-_. W ' M
A0 "'6 CH2012 A0 3 $88: /\0 i /
"-1 4 o "1.2 4 OH A _ OBn
' 7 60 /o 0 60% 111205
DDQ 0 0“ T120, DMAP 0 ON
011202 : H O : CH2012 .
(9 : 1), 920/? Ill-206 03" -40 .C to 0 °c ".407 OBn
"w FtO‘Na+
TsCl
JPY / DMAP R W +ROSflBU3
O OTs 0 OR
AAA/W MW
""209 08" Ill-208 58"

Scheme III-31: Attempted preparation and displacement reactions of allylic triflate and

tosylate

In conclusion, a synthetic scheme involving a convergent assembly of fully
functionalized left (Cl3—C37) and right (C1-C12) hand fragments of mucoxin via regio-
and selective intermolecular epoxide opening was designed (Figure 111-2). The advanced
coupling partners, viz., the allylic alcohol 111-140 and the vinylic epoxide III-4 were
synthesized as planned. However, a maximum yield of only 20% was obtained in the
attempted coupling reactions of III-140 and 111-4 using conventional acid catalyzed
conditions. The desired CO bond formation was also attempted under several other
acidic, basic and neutral conditions using model nucleophiles 111-177 and 111-183,
vinylic epoxides III-174, III-175 and vinylic epoxide equivalents III-189, III-202, III-
207 and III-209. None of these attempts met with success. Alcohols, inherently are
moderate nucleophilies and in our experience, their nucleophilicity depletes rapidly with

increase in their steric bulk. Under acid catalyzed reactions, the nucleophile is unable to

170

compete with the internal 1,2 hydride transfer and other rearrangement / decomposition
pathways of the activated vinylic epoxides, and is recovered unscathed. The ester
functionality in all the vinylic epoxides examined may also be responsible for
accelerating self-destruction of the epoxides under acidic conditions.

After the failure to access the prOposed RCM precursor III-2, the global synthetic
strategy was revised. The left hand (C12-C37) segment as the aldehyde III-135 was
conserved in the new designs, whereas, the right hand piece (C 1-Cl3) was functionalized
in several different ways. The new routes and culmination of the total synthesis of the

proposed structure of mucoxin is the subject of Chapter IV.

171

D. Experimental section
General Procedures:

All reactions were carried out in flame dried glassware under an atmosphere of

dry nitrogen or argon. 4 A molecular sieves were dried at 160 °C under vacuum prior to
use. Unless otherwise mentioned, solvents were purified as follows. THF and EtzO were

either distilled from sodium benzophenone ketyl or used as is from a solvent purification

system. CHZCIZ, toluene, CH3CN and Et3N were distilled from CaHz. DMF, diglyme, and

DMSO were stored over 4 A mol. sieves and distilled from CaHz. All other commercially

available reagents and solvents were used as received.
1H NMR spectra were measured at 300, 500 or 600 MHz on a Varian

Gemini-300, a Varian VXR-SOO or a Varian Inova-6OO instrument respectively. Chemical

shifts are reported relative to residual solvent (6 7.27, 2.50 and 4.80 ppm for CDCl3,
(CD3)ZSO and CD3OD respectively). 13C NMR spectra were measured at 125 MHz on a

Varian VXR-SOO instrument. Chemical shifts are reported relative to the central line of
CDC]3 (s 77.0 ppm). Infrared spectra were recorded using a Nicolet IR/42 spectrometer
FT-IR (thin film, NaCl cells). High resolution mass spectra were measured at the
University of South Carolina, Mass Spectrometry Laboratory. Optical rotations were
measured on a Perkin—Elmer polarimeter (model 341) using a 1 mL capacity quartz cell
with a 10 cm path length.

Analytical thin layer chromatography (TLC) was performed using Whatman glass

plates coated with a 0.25 mm thickness of silica gel containing PF254 indicator, and

172

compounds were visualized with UV light, potassium permangenate stain, p-
anisaldehyde stain, or phosphomolybdic acid in EtOH. Chromatographic purifications
were performed using Silicycle 60 A, 35-75 pm silica gel. All compounds purified by
chromatography were sufficiently pure for use in further experiments, unless indicated
otherwise. GC analysis was performed using HP (6890 series) GC system containing
Altech SIB-54, 30 m x 320 mm x 0.25 mm column. Analytical and semi—preparative

HPLC normal phase separations were performed using HP 1100 series HPLC system.

a. lite/32‘; mm
14 15
2. CHZO
Ill-82 rt, 59.1% Ill-83

A l—L three-necked round-bottom flask fitted with a reflux condenser and a 100
mL addition funnel was charged with magnesium turnings (24.1 g, 0.99 mol) and EtzO
(300 mL). To this mixture, 1,2 dibromoethane (5.5 mL, 63.9 mmol) was added over 30
min upon which EtzO started refluxing slowly. To the activated magnesium, l-
bromohexadecane III-82 (100 mL, 0.33 mol) was added via the addition funnel over 1h.
After completion of the addition, the reaction mixture was stirred for an additional 2 h.
The addition funnel was then replaced by a wide glass tube, which was connected to the
side-arm of a filtration flask via a rubber tubing. The filtration flask fitted with an inlet
for nitrogen was charged with paraformaldehyde (50 g) and heated to 180 °C -— 200 °C.
The formaldehyde generated by cracking paraformaldehyde in this manner was slowly
bubbled into the Grignard reagent by a current of dry nitrogen. After 1 h the bubbling

was stopped and the reaction was allowed to stir at ambient temperature for 2 h. The
reaction mixture was then diluted with H20 (200 mL), slowly poured into 300 g of

cracked ice, and 320 mL of 30% H2504 was added to it and stirred at ambient

173

 

temperature for 30 min. Layers were separated and the aqueous portion was washed with
EtzO (3x300 mL). The combined organic layers were washed with brine (300 mL), dried
over MgSO4, concentrated and the crude product was purified by flash column
chromatography [hexanes (1.5 L), 4 : 1 hexanes : EtOAc (3 L)] to yield l-heptadecanol
III-83 as a white solid (50 g, 59.1%). mp. 54-55 °C; Spectroscopic data for III-83
matched to that reported by Aldrich.
Partial data for III-83: 1H NMR (500 MHz, CDCl,) 0 3.66 (t, J = 6.6 Hz, 2 H), 1.6-1.53

(m, 2 H), 1.33-1.27 (m, 28 H), 0.90 (t, J = 6.6 Hz, 3 H)

 

PPhg, imld.
OH WI
‘5 12 / toluene 15
111-83 90% 111-84

To a solution of l—heptadecanol III-83 (68 g, 0.265 mmol) in dry toluene (2.3 L),
triphenyl phosphine (171 g, 0.652 mmol), and imidazole (45 g, 0.661 mmol) were added
at ambient temperature and stirred under N2 until a clear solution was obtained. To this
solution I2 (136 g, 0.535 mmol) was added and stirring was continued for 1 h at the same
temperature after which the reaction was quenched by adding aqueous saturated sodium
sulfite solution until the yellow color disappeared. The layers were then separated,
aqueous layer was washed with l : 4 EtOAc : hexanes (3x400 mL), and the combined
organic layers were dried over NaZSO4 and concentrated. Purification by flash column
chromatography (hexanes) afforded iodide Ill-84 as a white solid (87.3 g, 90%). Data for
III-84: 1H NMR (500 MHz, CDCl,) 0 3.18 (t, J = 7.07 Hz, 2 H), 1.82 (q, J = 7.06 Hz, 2

H), 1.40-1.22 (m, 28 H), 0.88 (t, J = 6.95 Hz, 3 H); l3C NMR (125 MHz, CDC13) 0 33.9,
32.2, 30.8, 30.0, 29.9, 29.8, 29.7, 29.6, 28.8, 22.9 (multiple carbons), 14.3, 7.2; 1R (thin

film) 2953, 2916, 2846, 1471, 1423, 1296, 1255, 1213, 1192, 1165, 725, 603 cm“;

174

HRMS (E1) calcd for C,7H3SI, 366.1784 m/z (M)+; observed, 366.1797 m/z; mp = 33-34

°C.

TBDMSCl
OH OTBS
Ill-87 7'32,- 111-91

To a solution of 3-butyn-l-ol 111-87 (27 mL, 29.16 g, 0.416 mmol) and imidazole
(61 g, 0.896 mmol) in DMF (100 mL) cooled to 0 °C, a solution of t-butyldimethylchloro
silane (64.5 g, 0.428 mmol) in DMF (125 mL) was added and stirred at the same
temperature for 40 min under N2. The reaction was then warmed to ambient temperature
and stirred for 3 h after which H20 (500 mL) was added. The aqueous layer was extracted
with 4:1 hexanes : EtOAc (4x400 mL), and the combined organic layers were dried
(NaZSO4) and concentrated. After flash column chromatography, the silyl ether III-91
was obtained as a colorless oil (62 g, 73 %). Data for III-91: 1H NMR (500 MHz, CDC13)
6 3.74 (t, J = 7.1 Hz, 2 H), 2.40 (dt, J = 7.2, 2.7 Hz, 2 H), 1.95 (d, J = 2.7 Hz, 1 H), 0.90
(s, 9 H), 0.07 (s, 6 H); 13C NMR (125 MHz, CDC13) 0 81.7, 69.5, 69.4 62.0, 26.1, 23.1,
18.5, -5.1; IR (thin film) 3330, 2954, 2860, 2753, 2711, 2123, 1839, 1590, 1471, 1388,
1255, 1106, 1006, 916, 837, 777, 643 cm"; HRMS (CI, CH4) calcd for ClonoOSi,

185.1362 m/z (M + H)*; observed, 185.1361 m/z.

1.nBuLl
-30 °C to -10 °C OTBS

OTBS
é/V ;

2. 111-84
111-91 THF , HMPA ‘6 111-92

(3/1), -78 °C to 0 °C
85%

\\

 

To a solution of the silyl ether III-91 (13.63 g, 74.08 mmol) in THF (113 mL)
cooled to -30 °C, nBuLi (7.8 mL of 9.97 M solution in hexanes, 77.8 mmol) was added
dropwise and the solution was warmed to -10 °C over 1 h. The lithium acetylide was

175

cooled to -78 °C after which a solution of iodide III-84 in 3 : l THF : HMPA (147 mL)
was added and stirred for 10 min at the same temperature. The reaction was then warmed
to 0 °C and after 1 h H20 (300 mL) was added. The aqueous layer was extracted with
3,0 (3x 400 mL). The combined organic layers were dried over MgSO4 and
concentrated to afford a crude oil, which was purified by flash column chromatography
(hexanes -> 19 : 1 hexanes : EtOAc) to yield the silyl protected homopropargylic alcohol
III-92 (26.7 g, 85%) as a yellow oil. Data for III-92: 1H NMR (500 MHz, CDC13) 0 3.69
(t, J = 7.06 Hz, 2 H), 2.36 (dt, J = 7.3, 2.4 Hz, 2 H), 2.12 (dt, J = 7.2, 2.4 Hz, 2 H), 1.45
(q, J = 7.1 Hz, 2 H), 1.39-1.23 (m, 30 H), 0.90 (s, 9 H), 0.88 (t, J = 7.01 Hz, 3 H), 0.07 (s,
6 H); 13C NMR (125 MHz, CDC13) 6 81.5, 76.8, 62.5, 31.9, 29.7, 29.6, 29.4, 29.2, 29.1,
289,259, 23.2, 22.7, 18.2, 183, 14.1, -5.3; IR (thin film) 2923, 2854, 1466, 1383, 1362,
1253, 1105, 1059, 1007, 916, 837, 777, 721 cm"; HRMS (C1, CH4) calcd for C27H54051,

421.3866 m/z (M - H)"; observed, 421.3874 m/z.

OTBS TBAF ITHF OH
é -—————-—» //
16 111-92 '20 C' 90 /° 16 111-113

To a solution of the silyl protected homopropargylic alcohol III-92 (35.9 g, 0.085
mol) in THF (100 mL), TBAF (130 mL of 1M solution in THF, 0.13 mol) was added at
—20 °C under N2. After stirring for 30 min at the same temperature, H20 (200 mL) was
added. The layers were separated, aqueous layer was extracted with E50 (3x 400 mL).
The combined organic layers were dried over MgSO4 and concentrated. The crude
product was purified by flash column chromatography (4 : 1 hexanes : EtOAc) to furnish
the homopropargylic alcohol III-88 as a white solid. Data for III-88: lH NMR (500

MHZ, CDCI3) 6 3.67 (t, J = 6.2 Hz, 2 H), 2.43 (dt, J = 6.2, 2.4 Hz, 2 H), 2.15 (dt, J = 7.2,

176

2.4 Hz, 2 H), 1.76 (s(br), 1 H), 1.48 (q, J = 7.1 Hz, 2 H), 1.37-1.25 (m, 30 H), 0.88 (t, J =
6.6 Hz, 3 H); 13C NMR (125 MHz, CDCl,) 0 82.9, 76.2, 61.4, 31.9, 29.8, 29.7, 29.6, 29.4,
29.2, 29.0, 28.9, 23.2, 22.7, 18.8, 14,1; IR (thin film) 2953, 2914, 2848, 1470, 1049,
1018, 874, 752 cm“; HRMS (CI, CH4) calcd for CmeO, 307.3001 m/z (M — H)“,
observed, 307.3003 m/z; mp = 61—62 °C.

OH LAH ldiglyme
125 °C, 87%
‘5 Ill-88 Ill-62

 

A 1 L two-necked round-bottom flask fitted with a stir bar and a reflux condenser
was charged with LAH (7.7 g, 0.203 mol). A solution of the homo propargylic alcohol
111-88 (35 g, 0.113 mol) in diglyme (350 mL) was carefully added dropwise at 0 °C to
the reaction mixture. While stirring vigorously, the mixture was heated to 125 °C. After
17 h, the reaction was cooled to room temperature upon which 7.7 mL of H20 was added
dropwise. 7.7 mL of 15% NaOH was then added followed by 22 mL of H20. The
resultant mixture was heated at 50 °C for 45 min and filtered after cooling to ambient
temperature. The filtrate was diluted with EtOAc (500 mL) and washed with 1.5 N HCl
(5x100 mL) to remove diglyme from the organic layer. The organic layer was dried
(NaZSO4) and concentrated. Chromatographic purification of the crude product (19 :l
hexanes : EtOAc -) 2.3:1 hexanes : EtOAc) yielded the E- homo allylic alcohol III-62
(30.5 g, 87%) as a white solid. Data for III-62: 1H NMR (500 MHz, CDC13) 0 5.58-5.52
(m, 1 H), 5.40-5.34 (m, 1 H), 3.62 (t, J = 6.3 Hz, 2 H), 2.26 (dt, J = 12.5, 6.08, 2 H), 2.00
(dt J = 14.3, 7.3 Hz, 2 H), 1.48 (8 (br), 1 H), 1.36-1.25 (m, 30 H), 0.88 (t, J = 6.8 Hz, 3

H); 13C NMR (125 MHz, CDCl,) 5 134.4, 125.7, 62.1, 36.1, 32.7, 31.9, 29.7, 29.6, 29.5,

29.4, 29.2, 22.7, 14.1; IR (thin film) 3448, 3136, 2914, 2848, 1637, 1470, 1047, 1020,

177

926, 890, 715 cm"; HRMS (CI, CH4) calcd for C2,H420, 309.3157 m/z (M - H)*;

observed, 309.3142 m/z; mp = 55-56 °C.

 

1. NaH
\ OH 0 °C to rt \ OPMB
\(ng _ \(ng
III-62 2. PMBCl, TBAl "“02

THF, 60 °C, 91 °/o

A 1 L round—bottom flask was fitted with a reflux condenser was charged with a
stir bar and NaH (14 g of 60 wt% dispersion in oil, 0.36 mol). A solution of the homo
allylic alcohol III-62 (37 g, 0.12 mol) in THF (400 mL) was added dropwise at 0 °C. The
mixture was warmed to room temperature and stirred for an additional 1 h.
4-Methoxybenzyl chloride (25 g, 0.16 mmol) and TBAI (16.5 g, 0.045 mol) were added
and the reaction mixture was heated to 60 °C for 18 h. The reaction was cooled to
ambient temperature and carefully quenched by adding saturated NH4C1 solution. The
layers were separated, and the aqueous layer was extracted with 3,0 (3x300 mL).
Combined organic layers were dried (MgSO4) and concentrated to furnish a crude solid
which upon purification by flash column chromatography (49 : 1 hexanes : EtOAc)
afforded the PMB protected homo allylic alcohol III-102 (47 g, 91%) as a white solid.
Data for III-102 1H NMR (500 MHz, CDC13) 6 7.27 (d, J = 8.7 Hz, 2 H), 6.88 (d, J = 8.7
Hz, 2 H), 5.54—5.48 (m, 1 H), 5.45-5.39 (m, 1 H), 4.46 (s, 2H), 3.81 (s, 3 H), 3.47 (t, J =
7.0 Hz, 2 H), 2.31 (dt, J = 13.5, 6.6 Hz, 2 H), 2.0 (dt, J = 13.9, 6.9 Hz, 2 H), 1.37-1.28
(m, 30 H), 0.9 (t, J = 6.9, 2 H); 13C NMR (125 MHz, CDC13) 0 159.2, 132.7, 130.7, 129.4,
129.3, 126.2, 113.8, 113.7, 72.5, 70.1, 55.3, 33.1, 32.7, 32.2, 29.8, 29.7, 29.6, 29.5, 29.4,
29.2, 14.2; IR (thin film) 2954, 2918, 2848, 1969, 1896, 1614, 1522, 1462, 1361, 1246,
1176, 1097, 1030, 964, 822 cm"; HRMS (E1) calcd for C29H5002, 430.3811 m/z (M)*;

observed, 430.3799 m/z; mp = 38-39 °C.

178

H OMe

”I

 

 

AD mix-u QH (FD-MFA Q 0
\ 0PMB ; ’ 013111113 +7 T OPMB
W W DCC, DMAP W
MeSOQNHz 16 OH 6 o 0
01-12012, n

KQOSOQ(OH)4 830/0 I

""102 ‘BuOH2H20 (1:1) ""103 MeO :‘ ran
0 °C, > 98% ee 92% H

111-141

A 2 L two-necked round flask fitted with a mechanical stirrer was charged with

AD mixflor (97.8 g). tBuOH (330 mL) and H20 (330 mL) were added followed by

methanesulfonamide (6.6 g) and KzOsO4-2HZO (144 mg). This mixture was stirred until

a clear solution was obtained which was cooled to 0 °C upon which the olefin III-102 (30
g, 0.07 mol) was added in one portion. The reaction was vigorously stirred at 0 °C for 20
h after which time sodium sulfite (100 g) was added at the same temperature. The
mixture was then warmed to room temperature and stirred for 45 min, then diluted with
EtOAc (500 mL) and washed with H20 (200 mL). The aqueous layer was extracted with
EtOAc (3x300 mL), combined organic layers were dried (Na2804) and concentrated to
yield a crude solid which was purified by flash column chromatography (9 : 1 hexanes :
EtOAc 9 2 : 3 hexanes : EtOAc) to yield the diol III-103 (32.5 g, 92%, > 98% ee as
determined after derivatization to bis-(R)-methoxyphenylacetate). To a solution of

111-103 (50 mg, 0.11 mmol), (R)-MPA (54 mg, 0.32 mmol) and DCC (67 mg, 0.32

mmol) in CH2C12 (2 mL) was added DMAP (2 mg, 0.02 mmol) at room temperature.
After 10 h, the reaction was quenched by saturated NaHCO3 solution (1 mL). The

aqueous layer was extracted with CH2C12 (5x2 mL), combined organic layers were dried,
concentrated and the solvent was evaporated to afford bis-(R)-methoxyphenylacetate

(111-141). 1H NMR of the crude material indicated the presence of a single diastereomer.

179

Data for 111-103: [(11020 —2.0 (c 1.0, CHC13), ‘11 NMR (500 MHz, CDCl,) 6 7.24

(d, J = 8.4 HZ, 2 H), 6.88 (d, J = 8.4 HZ, 2 H), 4.45 (S, 2 H), 3.80 (S, 3 H), 3.72-3.62 (m, 3
H), 3.42—3.39 (m, l H), 1.89-1.74 (m, 2 H) 1.44-1.50 (m 3 H), 1.33-1.21 (m, 31 H), 0.88
(t, J = 6.8 HZ, 3 H); 13‘C NMR (125 MHZ, CDC13) 0 159.4, 129.8, 1294,1139, 74.3, 73.7,
73.1, 68.3, 55.3, 33.6, 33.2, 32.0, 29.7, 29.6, 29.5, 29.4, 25.8, 22.7, 14.1; 1R (thin film)
3354, 2916, 2848, 1612, 1514, 1467, 1369, 1248, 1178, 1114, 1035, 814 cm"; HRMS
(E1) calcd for C,,H,,o,, 464.3866 m/z (M)“; observed, 464.3875 m/z; mp = 75-77 °C.
Data for III-141: 1H NMR (500 MHz, CDC13) 0 7.45-7.42 (m, 4 H), 7.39—7.31 (m,
6 H), 7.18 (d, J = 8.6 Hz, 2 H), 6.85 (d, J = 8.6 Hz, 2 H), 5.11 ((11, J = 2.4, 6.6 HZ, 1 H),
4.874.84 (m, 1 H), 4.70 (S, 1 H), 4.67 (S, 1 H), 4.20 (dd, J = 11.5, 19.2 HZ, 2 H), 3.79 (S,
3 H), 3.39 (S, 3 H), 3.35 (S, 3 H), 3.17—3.13 (m, 1 H), 3.08-3.03 (m, 1 H), 1.43-0.93 (m,
34 H), 0.88 (t, J = 6.9 HZ, 3 H); ”C NMR (125 MHZ, CDC13) 0 170.4, 170.2, 159.4,
136.7, 136.6, 130.5, 129.5, 129.0, 128.9, 128.8, 127.6, 127.5, 127.4, 113.9, 82.7, 82.3,
75.0, 72.7, 71.9, 65.8, 57.5, 57.4, 55.5, 32.2, 30.8, 30.4, 30.0, 29.9, 29.8, 29.7, 29.6, 29.5,

29.4, 25.2, 22.9, 14.3.

 

9H TESCI, Et3N QTES
‘ OPMB ¢ ' OPMB
1s 5,, SMAP 111111: 15 5,58
111-103 ' qua” ' 111-104

To a solution of the diol 111-103 (29 g, 0.062 mol) in THF (600 mL) triethyl
amine (202 mL) was added followed by triethylsilyl chloride (63 mL, 0.374 mol) and
DMAP (2.9 g, 0.024 mol) at ambient temperature. The reaction was stirred under N2 for 3
h after which time was quenched by adding saturated NaHCO3 solution (400 mL). The
aqueous layer was extracted with 1 : 5 EtOAC : hexanes (3x500 mL) to afford crude oil

which was purified by column chromatography (35 : 1 hexanes : EtOAc) providing the

180

fully protected triol III-104 as a colorless oil (43 g, quant.). Data for III-104: [01],)20

—17.5 (c 3.0, CHCl3) 1H NMR (500 MHZ, CDCl3) 0 7.26 (d, J = 8.6 Hz, 2 H), 6.87 (d, J =
8.6 Hz, 2 H), 4.43 (s, 2 H), 3.81 (s, 3 H), 3.78-3.74 (m, l H), 3.58-3.51 (m, 3 H), 2.04-
2.00 (m, l H), 1.63—1.43 (m, 3 H), 1.35-1.19 (m, 30 H), 1.00-0.88 (m, 21 H), 0.63—0.51
(m, 12 H); 13C NMR (125 MHZ, CDCl,) 0 159.0, 131.0, 129.1.113.6, 75.3, 72.3, 72.1,
67.5, 55.3, 31.9, 30.6, 30.2, 29.9, 29.7, 29.6, 29.4, 26.7, 22.7, 14.1, 7.0, 6.9, 6.6, 6.4, 5.8,
5.2, 5.1; IR (thin film) 3324, 2924, 2856, 2071, 2003, 1876, 1614, 1587, 1513, 1461,
1414, 1379, 1301, 1247, 1172, 1099, 1014, 825, 738 cm"; HRMS (EI) calcd for

C41H8004Siz, 692.5595 m/z (M-H)‘; observed, 692.5567 m/z.

 

9153 000 OTES
OPMB 4, 3 OH
16 5TES 51:?le1zoPhgsphate 16 5E8
u er :
111-104 0 °C, 78% 111-105

To a 0 °C solution of PMB ether III-104 (15 g, 0.02 mol) in 460 mL of CHZCl2 :
phosphate buffer (10 :1), DDQ (5.7 g, 0.03 mol) was added in one portion. After stirring
the reaction under N2 at the same temperature for 90 min, saturated NaHCO3 solution
(200 mL) was added. The mixture was warmed to ambient temperature and carefully
extracted with CHzCl2 (3x200 mL) so as to avoid emulsions. The combined organic
layers were dried (NaZSO4), concentrated and the crude oil was purified by flash column

chromatography (3% EtOAc in hexanes) to afford 12.6 g (78%) of the primary alcohol

III-105.

Data for III-105: [01],)20 —24.3 (c 2.34, CHC13) lH NMR (500 MHz, CDCl3) 0 3.78-3.58
(m, 4 H), 2.88 (t, J = 5.7 Hz, 1 H), 1.95—1.90 (m, 1 H), 1.67-1.57 (m, 2 H), 1.40-1.46 (m,
1 H), 1.37-1.17 (m, 30 H), 0.99-0.90 (m, 18 H), 0.87 (t, J = 7.0 Hz, 3 H), 063-052 (m,

181

12 H); 13C NMR (125 MHz, CDCl3) 0 75.9, 75.1, 61.1, 34.7, 32.1, 30.6, 30.0, 29.9, 29.8,
29.5, 26.8, 22.9, 14.2, 7.0, 5.3, 5.2; IR (thin film) 3471, 2928, 2851, 1468, 1411, 1374,

1242, 1080, 1023, 723 cm"; HRMS (CI, CH4) calcd for C33H7203Siz, 571.4942 m/z (M-

H)+; observed, 571.4927 m/z.

 

QTES OH Phl(OAc)2 QTES
. 5 W010
16 OTES TEMPO /CH2C12 16 éTES
111-105 '1' 96 /° Ill-63

To a solution of alcohol 111-105 (15.5 g, 0.03 mol) in CHzCl2 (50 mL) at room

temperature, bisacetoxyiodo benzene (9.61 g, 0.03 mol) was added. After addition of

TEMPO (437 mg, 3.0 mmol) the clear orange solution was stirred at rt for 2 h. The
reaction was then diluted with CHzCl2 (150 mL) and treated with saturated sodium sulfite
solution until it became colorless. Upon separation of the layers aqueous layer was
extracted with CHzCl2 (3x200 mL). The combined organic layers were dried over
NaZSO4, concentrated and purified by column chromatography (2% EtOAc in hexanes) to
furnish aldehyde III-63 as a colorless oil (14.8 g, 96%).

Data for III-63: 1611,,20 -21.6 (c 1.95, 010,) '11 NMR (500 MHz, CDC13) a 9.67 (1, J =
2.22 Hz, 1 H), 4.20-4.17 (m, 1 H), 3.62—3.59 (m, 1 H), 2.65 (ddd, J: 1.8, 4.0, 15.9 Hz, 1
H), 2.43 (ddd, J = 2.9, 8.2, 15.7 Hz, 1 H), 1.66-1.60 (m, 1 H), 1,47-1.31 (m, l H), 1.30-
1.12 (m, 30 H), 0.92-0.88 (m, 18 H), 0.86 (t, J = 7.1 Hz, 3 H), 0.62-0.48 (m, 12 H); 13C
NMR (125 MHz, CDC13) 0 201.9, 75.1, 70.8, 46.1, 32.1, 30.6, 30.0, 29.9, 29.8, 29.5,

26.7, 22.9, 14.2, 7.0, 5.3, 5.1; IR (thin film) 2930, 2978, 2855, 2716, 1732, 1640, 1414,

182

1381, 1327, 1240, 1103, 1007, 976, 833, 743, 673 cm"; HRMS (Cl, CH4) calcd for

C33H7OO3Si2, 569.4785 m/z (M-H)+; observed, 569.4775 m/z.

 

QTES Ph3PCHC02Et QTES
Wcuo 2 Moog:
‘6 OTES THF, reflux, 91% ‘6 OTES
Ill-63 111-106

A solution of aldehyde III-63 (15.1 g, 0.02 mol) and
(carbethoxymethylene)triphenylphosphorane (13.9, 0.05 mol) in THF (245 mL) was
heated to reflux for 16 h. After cooling the solution to rt, the solvent was evaporated and
the crude product was purified by column chromatography (EtOAc : hexanes = l : 99) to
afford or,[3-unsaturated trans ester III-106 as a yellow oil (15.1 g, 91%).

Data for III-106: [611020 -301 (c 1.99, C110,) '11 NMR (500 MHz, CDC13) s 7.37-6.98
(m, l H), 5.85 (d, J = 15.5 Hz, 1 H), 4.19 (q, J: 7.1 Hz, 2 H), 3.71-3.61(m, l H), 3.60-
3.58 (m, 1 H), 2.57-2.52 (m, 1 H), 2.23-2.17 (m, l H), 1.67-1.20 (m, 35 H), 1.02—0.92 (m,
18 H), 0.90 (t, J = 7.0 Hz, 3 H), 065—055 (m, 12 H); 13C NMR (125 MHz, CDC13) 6
166.6, 147.9, 123.0, 75.6, 75.0, 60.2, 60.1, 34.2, 34.1, 32.1, 30.3, 30.0, 29.9, 29.6, 26.8,
22.9, 14.5, 14.4, 14.3, 7.1, 7.0, 5.4, 5.2; IR (thin film) 2926, 2878, 2855, 1729, 1657,
1464, 1414, 1379, 1368, 1318, 1264, 1238, 1167, 1100, 1047, 1005, 984, 849, 743,673

cm"; HRMS (CI, CH4) calcd for C37H76OzSiz, 639.5204 m/z (M-H)+; observed, 639.5213

m/z.
QTES DlBAL-H QTES
‘6 OTES Et20. 0°C ‘6 OTES
Ill-106 89% 111-64

To a cold (0 °C) solution of ester III-106 (15.2 g, 23.8 mmol) in diethyl ether

(245 mL), DIBAL-H (75 mmol, 50 mL of 1.5 M solution in toluene) was added under

183

N2. After stirring for 30 min. at the same temperature, saturated potassium-sodium
tartrate solution (240 mL) was added and the mixture was brought to rt. EtzO (250 mL),

H20 (50 mL) and glycerol (12 mL) were added and the resultant heterogeneous mixture

was stirred overnight. The two layers were then separated and the aqueous layer was

extracted with diethyl ether (2x200 mL). The combined organic layers were dried
(MgSO4), concentrated and the crude product after chromatographic purification (5%
EtOAc in hexanes) afforded allylic alcohol III-64 as a colorless 011 (14.2 g, 89%).

Data for III-64: [0111320 -27.1 (c 2.22, CHC13) 'H NMR (500 MHz, CDC13) 6 5.74-5.64
(m, 2 H), 4.10-4.07 (m, 2 H), 3.62-3.55 (m, 2 H), 2.43-2.39 (m, l H), 2.07-2.00 (m, l H),
1.64-1.24 (m, 33 H), 0.99-0.93 (m, 18 H), 0.88 (t, J = 7.0 Hz, 3 H), 0.63-0.51 (m, 12 H);
13C NMR (125 MHz, CDC13) 0 131.6, 130.9, 75.7, 64.1, 33.9, 32.1, 30.4, 29.9, 29.8,
29.6, 26.8, 22.9, 14.3, 7.1, 7.0, 5.4; IR (thin film) 3333, 2928, 2978, 2853, 1460, 1414,

1379, 1329, 1238, 1098, 1007, 972, 909, 743, 673 cm‘l; HRMS (CI, CH4) calcd for

C35H74O3Siz, 597.5098 m/z (M-H)+; observed, 597.5090 m/z.

 

ores D-(—)-DIPT 91159
W051 —; WOH
16 i Ti(O‘Pr)4, ‘BuOOH 16 a "'0
ores
CHZCIZ, 411 mol. OTES
Ill-64 sieves 01 > 33:1, 73% "1.107

A two necked round bottom flask charged with 4 A mol. sieves (1.17 g) and

01202147 mL) was cooled to -20 °C. To this, Ti(OiPr)4 (3.04 mL, 10.0 mmol) and a
CHzCl2 soltution of D-(-—)-DET (2.91 g, 12.4 mmol in 41 mL CHZClz) were added in that
order and stirred at the same temperature under N2 for 30 min. After cooling the complex

to ~30 °C, tBuOOH (13 mL of 3.1 M solution in toluene, 40 mmol) was added dropwise

184

and the mixture was stirred for another 45 min. A solution of allylic alcohol (6.21 g, 10.4

mmol) in CHZCIZ (31 mL) was added via a syringe pump over 45 min. The reaction was

warmed to —20 °C, stirred for 2 h and then quenched by adding saturated Nast4 and

NaZSO3 solutions (6.8 mL each). EtzO (25 mL) was added and the resultant yellow
mixture was vigorously stirred at rt for 4 h. The yellow gelatinous mass was further
diluted with EtzO (200 mL), celite was added and the mixture was filtered through a pad
of celite. The filter cake was washed with EtzO (ca. 600 mL) until it turned dry and

granular. The filtrate was concentrated and epoxy alcohol III-107 was isolated in 73%
yield after purification by column chromatography (7% EtOAc in hexanes).

Data for III-107: [0111320 —16.7 (c 1.06, CHC13) 'H NMR (500 MHz, CDC13) 6 4.05-4.03
(m, l H), 3.93 (dt, J = 4.3, 8.4 Hz, 1 H), 3.76-3.72 (m, 2 H), 3.23 (dt, J = 2.3, 7.2 Hz, 1
H), 3.04 (m, 1 H), 2.09 (5 (br), 1 H), 1.98-1.76 (m, 4 H), 1.63-1.37 (m, 30 H), 1.15-1.05
(m, 18 11), 1.03 (1, J = 7.0 112, 3 11), 077-067 (m, 12 11); 13C NMR (125 MHz, CDC13) o
75.6, 73.6, 62.2, 58.8, 54.9, 33.7, 32.2, 30.4, 30.0, 29.9, 29.8, 29.6, 26.8, 22.9, 14.4, 7.2,
7.1, 5.3, 5.2; IR (thin film) 3438, 2932, 2878, 2855, 1462, 1414, 1379, 1329, 1238, 1098,

1009, 976, 903, 874,743, 673 cm"; HRMS (CI, CH4) calcd for 0,117,051,, 613.5047

m/Z (M-H)+; observed, 613.5052 m/z.

 

OTES OTES
1‘ P118 , BU P :
WOH ( )2 3 ; W813“
‘5 ones 0 TEA,0°C tort 16 (ms ’0
94%
Ill-107 Ill-5

To a solution of dipehyldisulfide (6.4 g, 29.3 mmol) in triethylamine (20 mL),

was added tibutylphosphine (7.1, mL, 29.3 mmol) at ambient temperature under N2. This

185

solution was cooled to 0 °C and into it was cannulated a pre-cooled solution of epoxy
alcohol 111-107 (5.96 g, 9.67 mmol) in Et3N. The reaction was warmed to ambient
temperature over 6 h. The reaction mixture was quenched with water (50 mL) and the
aqueous solution extracted with EtOAc (3x150 mL). The combined EtOAc extracts were
dried over NaZSO4 and concentrated under reduced pressure. The crude product was

purified by flash column chromatography (hexanes -) 2% EtOAc in hexanes) to afford
III-5 as a colorless oil (6.43 g, 94%).

Data for 111-105: 16'ti —26.2 (c 1.62, C110,) '11 NMR (500 MHz, 030,) o 7.41 (d, J =

1.3 112,2 11), 7.40-7.18 (m, 3 11), 3.75 (dt, J = 4.2, 8.2 112, 1 11), 3.57—3.55 (m, 1 11), 3.13

(dd, J = 5.1, 13.9 112, 1 11), 2.97-2.87 (m, 3 11), 1.69-1.31 (m, 4 11), 1.30115 (m, 30 11),
0.98-0.91 (m, 18 11), 0.88 (1, J = 7.1 112, 3 11), 0.62-0.53 (m, 12 11); 13C NMR (125 M112,
030,) 6 135.9, 130.0, 129.2, 126.7, 75.4, 73.5, 58.2, 57.3, 36.5, 33.9, 32.2, 30.4, 30.1,

30.0, 29.9, 29.8, 29.6, 26.8, 23.0, 14.4, 7.2, 5.3; IR (thin film) 3077, 3061, 2853, 1806,

1586, 1482, 1439, 1416, 1379, 1327, 1302, 1238, 1184, 1092, 1007, 970, 943, 916, 747,

699 cm"; HRMS (131) calcd for 0,11,80,50,, 706.5210 m/z (Mr; observed, 706.5223

m/z.
QTES SPh erg-oer2 9“ O
W -———-—’ WSW) + mixture otregio-lstereoisomeric
OTES E120, 76% 15 ' THF diOlS
'OH
Ill-5 III-65 111-1 20
(2.8 : 1)
QAC

 

CHQCIQ; quant. :OAC
Ill-1 21

 

186

To a solution of 111-5 (4.0 g, 5.66 mmol) in 150 mL E120 at 0 °C was added

BF3°OE12 (4.3 mL, 33.8 mmol) drop wise. After complete addition, the mixture was
slowly allowed to attain room temperature over 5 h. The reaction mixture was quenched

with NaHCO, solution (50 mL) and extracted with EtOAc (3x100 mL). The combined

extracts were dried over MgSO, and concentrated under reduced pressure to afford a

mixture of regio- and stereoisomeric products. Flash column chromatography provided
III-65 (1.52 g, 56%) as a white solid along with an inseparable mixture of isomers (543

mg). III-65 (50 mg, 0.11 mmol) was subjected to the acetylation conditions by treatment
with acetic anhydride (43 mg, 0.42 mmol) and DMAP (52 mg, 0.42 mmol) in CHZCl2 (1
mL) at room temperature to furnish III-121 as a colorless oil (61 mg, 99%).

Data for III-5: [01]D20 -—36.2 (c 0.32, CHCI3) 'H NMR (500 MHz, CDCl3) 6 7.41-7.20 (m,
5 H), 4.45-4.44 (m, 1 H), 4.10 (dt, J = 6.2, 9.5 Hz, 1 H), 4.00 (ddd, J = 3.1, 5.4, 8.8 Hz, 1
H), 3.81-3.34 (m, 1 H), 3.27 (dd, 5.3, 13.0 Hz, 1 H), 3.14 (dd, J = 9.1, 13.3 Hz, 1 H),
2.17-2.12 (s(br), l H), 2.04-1.80 (m, 2 H), 1.58-1.25 (m, 32 H), 0.88 (t, J = 7.0 Hz, 3 H);
13C NMR (125 MHz, CDCl,) 6 195.1, 135.6, 129.9, 129.3, 126.8, 100.9, 91.5, 81.5, 74.2,

73.0, 37.7, 33.7, 32.7, 32.2, 29.9, 29.8, 29.6, 25.8, 22.9, 14.3; IR (thin film) 3440, 3400,

2918, 2841, 1585, 1464, 1414, 1325, 1173, 1092, 1026, 964, 949, 879, 810, 729, 683
cm“; HRMS (CI, CH4) calcd for C29H5003S, 477.3402 m/z (M-H)+; observed, 477.3398
m/z.

Partial data for III-121: lH NMR (500 MHz, CDCl3) 6 7.39-7.18 (m, 5 H), 5.34 (m, 1 H),
4.83 (dt, J = 4.9, 8.4 Hz, 1 H), 4.21 (m, l H), 4.12 (m, l H), 3.19 (dd, J = 5.7, 13.5, 1 H),

3.06 (dd, J = 8.4, 13.4, 1 H), 2.09-1.86 (m, 2 H), 2.05 (s, 3 H) 2.00 (s, 3 H), l.58-l.49 (m,

187

2 11), 1.27—1.21 (m, 30 11), 0.87 (1, J = 6.6 112, 3 11); 13C NMR (125 MHz,CDC13)6

171.1, 170.0, 135.8, 130.3, 129.2, 126.8, 80.2, 78.7, 75.1, 74.7, 35.6, 32.9, 32.2, 31.0,

29.9, 29.8, 29.7, 29.6, 25.6, 22.9, 21.4, 21.2, 14.4;

OH
7 o. TBSOTf Ems
Wsph ________._ Wsph
16
.‘OH 2,6 lutidine ‘5
CH2C12, 8713/0 OTBS
Ill-65 Ill-132

2,6-Lutidine (1.3 mL, 11.2 mmol) was added to a 0 °C solution of diol III-65

(1.72 g, 3.59 mmol) in 18 mL CHZCIZ. A solution of TBS-OTf (1.9 mL, 8.25 mmol) in 10

mL CHZCI2 was then added and the reaction mixture stirred at 0 °C for 30 min. When
TLC indicated completion of the reaction, water (50 mL) was added and the aqueous

solution was extracted with CH2C12 (3x100 mL). The combined extracts were dried over

NaZSO4 and concentrated under reduced pressure to afford the crude product as a

colorless oil. After purification by flash column chromatography, 2.26 g of 111-132 was

obtained (87 %).

Data for 111-132: (611920 —71.0 (c 0.57, C110,) 111 NMR (500 M112, CDCl,) 6 7.37-7.14
(m, 5 11), 4.40 (m, 1 11), 4.21 (dt, J = 5.3, 10.4 112, 1 11), 3.96 (dt, J = 3.1, 6.8 112, 1 11),
3.59 (m, 1 11), 3.15 (m, 2 11), 1.88-1.79 (m, 2 11), 1.38-1.22 (m, 32 11), 0.91 (s, 9 11), 0.89
(1, J = 6.6 112,3 11), 0.87 (s, 9 11), 0.12 (s, 3 11), 0.11 (s, 3 11) 0.07 (s, 3 11) 0.05 (s, 3 11);
l3C NMR (125 MHz, 030,) s 137.0, 129.1, 128.7, 125.8, 81.7, 80.7, 74.6, 73.0, 37.3,

32.7, 32.2, 30.1, 29.9, 29.8, 29.6, 26.2, 26.0, 22.9, 18.5, 18.3, 14.4, —4.0, —4.2, —4.3, —4.6;

1R (thin film) 2926, 2856, 1585, 1470, 1439, 1389, 1362, 1254, 1194, 1078, 1057, 1007,

188

960, 835, 775, 737, 690 cm"; HRMS (ES) calcd for 0,11,80,50,, 707.5289 772/:

(M+H)+; observed, 707.5269 m/z.

 

 

QTBS QTBS 1. TFAA, 2,6 lUi OTBS O
.- 0, mCPBA 7 02., 3,0 ‘ 01-1202 : 0,
W“ W? = ,6 " H
~. CH Cl ,0 °C P“ .
2
111-132 111-133 60% (3 steps) 111-135

To a 0 °C solution of sulfide III-132 (1 g, 1.4 mmol) in 16 mL CHZCIZ, was
added a solution of mCPBA (350 mg, 1.4 mmol) in CHZCI2 (16 mL). After 30 min at 0
°C, the reaction mixture was quenched with sat. NaHCO3 solution (30 mL) and the
aqueous mixture was extracted with CHZClz (3x50 mL). The combined extracts were
dried over NaZSO4 and concentrated under reduced pressure to afford sulfoxide III-133

as a mixture of stereoisomers. The crude product so obtained was dissolved in CHzCl2
(11.3 mL), cooled to 0 °C and 2,6-lutidine (0.56 mL) was added. TFAA (0.69 mL, 4.95
mmol) in CHZCI2 (11.3 mL) was then added and the mixture stirred at 0 °C for l h. The

reaction mixture was quenched with sat. NaHCO, solution and extracted with CHZCIZ.

The combined extracts were dried over NaZSO4 and concentrated under reduced pressure
to afford a clear oil. This material was taken up in 1:1 acetonitrile - water (50 mL) and
solid NaHCO3 (2.5 g) was added. The reaction mixture was stirred at ambient
temperature for 16 h, upon which the solution was diluted with 25 mL water. The

aqueous solution was extracted with CH2C12 (3x25 mL) and the combined extracts were

dried over NaZSO4 and concentrated under reduced pressure. Slow chromatography on

189

wet silica gel containing 10% water (3% EtOAc in hexanes) afforded 515 mg of aldehyde

111-135 in 61% yield over the three steps.

Data for III-135: [04D20 —33.0 (c 0.91, CHCI3) 1H NMR (500 MHz, CDC13) 6 9.58 (d, J =
2.2 Hz, 1 H), 4.73-4.72 (m, l H), 4.46-4.44 (m, 1 H), 4.28—4.16 (m, 1 H) 3.68-3.66 (m, 1
H), 1.87-1.84 (m, 2 H), 1.40-1.22 (m, 32 H), 0.93—0.84 (m, 21 H), 0.08 (s, 3 H), 0.07 (s, 3
H), 0.06 (s, 3 H), 0.05 (s, 3 H); 13C NMR (125 MHZ, CDC13) 6 203.1, 87.2, 82.8, 76.0,
74.4, 38.0, 33.2, 32.2, 30.0, 29.9, 29.8, 29.6, 26.2, 26.0, 25.9, 25.8, 22.9, 18.4, 18.1, 14.3,
—4.1, ~43, —4.5, -—5.1; IR (thin film) 2928, 2855, 1738, 1475, 1464, 1441, 1389, 1362,

1256, 1186, 1076, 1065, 998, 939, 837, 808, 777, 737, 691, 664 cm"; HRMS (CI, CH4)

calcd for C35H7204Si2, 611.4891 m/z, (M-HY; observed, 611.4898 m/z.

OH TBSOTf T880

 

TBDPSO 0., TBDPSO 0..
M9“ 2.6 lutidine WSW
'3 CHQClQ, 0 c’C ",
OH 89° /° OTBS
111-137 111-138

TBS protection of diol III-137 was performed using the same procedure
described above for 111-132. Thus, 1.98 g (3.90 mmol) of III-37 afforded 2.51 g (91%) of

111-138.
Partial data for 111-138: 1H NMR (500 MHz, CDC13) a 7.68-7.64 (m, 4 H), 7.47-7.13 (m,

11 H), 4.66-4.60 (m, l H), 4.45—4.43 (m, 1 H), 4.05-4.00 (m, l H), 3.96 (dt, .1 = 2.9, 6.8
Hz, 1 H), 3.54 (dd, J = 4.9, 10.5 Hz, 1 H), 3.38 (dd, J = 7.8, 10.3 Hz, 1 H), 3.17 (d, J =

6.8, 1 H), 2.19-2.11 (m, 1 H), 1.83 (ddd, J: 1.2, 5.1, 12.7 Hz, 1 H), 1.03 (s, 9 H), 0.94 (s,

9 H), 0.77 (s, 9 H), 0.15 (s, 3 H), 0.14 (s, 3 H), -0.04 (s, 3 H), -0.08 (s, 3 H); 13c NMR

190

(125 MHZ, CDCl3) 5 137.0, 135.8, 133.6, 129.9, 129.1, 128.7, 127.9, 125.9, 81.7, 79.5,

73.3, 73.2, 65.8, 49.7, 34.5, 32.5, 27.0, 19.4, 18.2, 7.0, -4.6, -6.9.

OTBS OTBS 1. TFAA, 2,5 lUl OTBS o

O! mCPBA O, + O - I
WSW" 1. Slat CHQC 2 ¢ WH
OR OR ., Ph

 

 

-. CH Cl ,0 °C . OR «,
Was 2 2 T880 2. N3HCOg, CH3CN ’OTBs
H20
61% (3 steps)
III-138 111-139

R = TBDPS

Pummerer rearrangement of sulfide III-138 (2.51 g, 3.47 mmol) to secure
aldehyde 111-139 (1.36 g, 61%) was carried out following the representative procedure
described above (for 111-135). Flash column chromatography was performed using 3%
EtOAc in hexanes.

Partial data for 111—139: [(1)020 —37.9 (c 0.89, CHCI3) 'H NMR (500 MHz, CDCI3) a 9.62
(d. J = 2.2 Hz, 1 H), 7.67-7.65 (m, 4 H), 7.45-7.37 (m, 6 H), 4.84-4.80 (m, 1 H), 4.77-
4.75 (m, 1 H), 4.16-4.10 (m, 1 H), 4.09-4.07 (m, 1 H), 3.60 (dd, J = 4.6, 10.4 Hz, 1 H),
3.43 (dd, J = 7.3, 10.4, 1 H), 2.17-2.19 (m, 1 H), 1.89-1.85 (m, 1 H), 1.05 (s, 9 H), 0.86

(s, 9 H), 0.81 (s, 9 H), 0.07 (s, 3 H), 0.04 (s, 3 H), -0.02 (s, 3 H), -0.07 (s, 3 H); 13C NMR
(125 MHZ, CDCl3) b 203.1, 135.8, 135.7, 133.5, 133.4, 130.0, 129.9, 127.9, 127.8, 87.2,

81.7, 76.3, 72.9, 65.9, 35.2, 27.0, 26.0, 25.8, 19.4, 18.2, 18.1, —4.5, -4.6, -5.0; IR (thin

film) 2957, 2889, 1959, 1909, 1821, 1736, 1471, 1255, 1151, 1068, 998, 887, 806, 777,

 

702 cm".
TBSO o TBSO OH
TBDPSO 0.,2 H flMgB, TBDPSO 07.2 /
3 j 3
i E120, -40 °C, 2 h '9
OTBS 80%, 10 :1 OTBS
III-139 Ill-14o

191

To a solution of aldehyde III-139 (205 mg, 0.32 mmol) in diethyl ether (4.2 mL),
vinylmagnesium bromide (0.8 mL) was added at —40°C and stirred under N2 for 2 h. The
reaction was the quenched by addition of saturated NH4C1 solution (5 mL), layers were

separated and the aqueous layer was extracted with EtzO (3x50 mL). The combined

organic layers were dried (MgSO4), concentrated and the crude product was purified by
flash column chromatography to furnish the allylic alcohol III-140 (170 mg. 80%; dr 2

10:1 by 1H NMR) as a colorless oil.

Partial data for III-140: 1H NMR (300 MHZ, CDC13) 8 7.68-7.65 (m, 4 H), 7.47-7.37 (m,

6 H), 6.00-5.89 (m, 1 H), 5.43 (d, J = 17.0 Hz, 1 H), 5.20 (d, J = 10.7 Hz, 1 H), 4.66 (m,
1 H), 4.42-4.40 (m, 2 H), 4.08-4.07 (m, 1 H), 3.75-3.72 (m, 1 H), 3.56 (dd, J = 4.9, 15.2
HZ, 1 H), 3.41 (dd, J = 7.7, 9.9 Hz, 1 H), 3.29 (s(br), 1 H), 2.25-2.16 (m, l H), 1.90-1.83

(m, 1 H), 1.04 (s, 9 H), 0.94 (s, 9 H), 0.82 (s, 9 H), 0.14 (s, 6 H), 0.02 (s, 3 H), —0.06 (s, 3

 

H).
i
o C H - NM
OH MegN—Q-COCI W 4 p 92
Wows ; ,5 OPMB
OH DMAP, CHzle, 50% Meng-CSH4 jfO
o
"“03 111-144

To a solution of diol III-103 (54 mg, 0.11 mmol) in CHZCl2 (1 mL) pa
dimethylaminobenzoyl chloride (107 mg, 0.93 mmol) and DMAP (100 mg, 0.82 mmol)

were added and stirred for 15 h. The reaction was then quenched with H20 (3 mL) and
the aqueous layer was extracted with CH2C12 (2x5 mL). The combined organic layers

were dried (NaZSO4), concentrated under reduced pressure and the crude material was

192

purified by flash column chromatography (30% EtOAc in hexanes) to afford bis-ester
111-144 (37 mg, 50%).

Partial data for III-144: 1H NMR (300 MHz, CDC13) 0 8.02-7.91 (m, 4 H), 7.18 (d, J =
8.6 Hz, 2 H), 6.85 (d, J = 8.6, 2 H), 6.62-6.71 (m, 4 H), 5.58-5.56 (m, 1 H), 5.37—5.35 (m,
1 H), 4.04 (s, 2 H), 3.78 (s, 3 H), 3.58-3.47 (m, 2 H), 3.04 (s, 6 H), 3.02 (s, 6 H), 2.02-

1.98 (m, 2 H), 1.78-1.61 (m, 2 H), 1.56-1.20 (m, 30 H), 0.88 (t, J = 6.9 Hz, 3 H).

 

O i
OACBprNMeZ DDQ __ O C5H4—pNMe2
OPMB CH2C'23H20 (9'1) OH
5 UCJtSG% w

Mesz‘CeHa, YO Mesz'CsH4 Y0
O 0

111-144 111-145
DDQ (14 mg, 0.06 mmol) was added to a solution of PMB ether 111-144 (37 mg,
0.05 mmol) in 9 : 1 CHZCI2 : H20 (1.1 mL) at 0 °C. After 30 min, the reaction mixture

was carefully poured into saturated NaHCO3 solution (2 mL). Extraction of the aqueous

layer with CHZCI2 (3x5 mL) followed by evaporation of the solvent and purification

using column chromatography furnished alcohol III-14S in 50% yield (16 mg).

Partial data for 111-145: 1H NMR (500 MHz, CDC13) 0 7.96-7.94 (m, 4 H), 6.67-6.65 (m,
4 H), 5.41 (dt, J = 3.3, 10.6 Hz, 1 H), 5.36-5.32 (m, 1 H), 3.65 (5 (br), 1 H), 3.56-3.52 (m,
1 H), 3.05 (s, 6 H), 3.04 (s, 6 H), 3.01-2.90 (m, 1 H), 1.97-1.69 (m, 3 H), 1.55-1.20 (m,
31 H), 0.86 (t, J = 7.1 Hz, 3 H); 13C NMR (125 MHz, CDC13) 6 168.1, 166.7, 153.8,
153.6, 131.9, 131.7, 116.3, 117.2, 74.4, 71.4, 58.4, 48.9, 40.3, 40.2, 34.1, 32.1, 31.4, 32.1,

30.0, 29.9, 29.8, 29.7, 29.6, 29.5, 29.4, 25.4, 22.9, 14.3.

193

O O
O C5H4-DNM82 OiC5H4‘pNM82
BSTFA
15 OH ' "“"’ OTMS
O 50 00 15 O
Mesz—C5H4 Y Meng-C5H4
a r
0
111-145 Ill-146

Alcohol III-145 (15 mg, 0.02 mmol) was dissolved in bis-(trimethylsilyl)trif1uoro
acetamide (0.2 mL) and the solution was heated to 50 °C for 30 min. After cooling to
room temperature, the volatiles were removed under reduced pressure and the TMS
derivative 111-146 was used for ECCD analysis without further purification.

Partial data for III-146: 1H NMR (500 MHz, CDC13) 6 8.00-7.94 (m, 4 H), 6.69-6.65 (m,
4 H), 5.47-5.44 (m, 1 H), 5.35-5.32 (m, 1 H), 3.68-3.65 (m, 2 H), 3.05 (s, 6 H), 3.04 (s, 6
H), 2.01-1.94 (s, 2 H), 1.73-1.69 (s, 2 H), 1.37-1.21 (m, 30 H), 0.89 (t, J = 7.1 Hz, 3 H),
0.08 (s, 3 H), 0,07 (s, 6 H); 13C NMR (125 MHz, CDC13) a 166.7, 166.5, 153.5, 131.7,

117.5, 117.4, 110.9, 74.2, 71.4, 59.2, 40.3, 34.5, 32.1, 31.3, 30.0. 29.9, 29.8, 29.7, 29.6,

29.5, 25.5, 22.9, 14.3, 1.2, —0.4.

TBSCI

HO \ -—————-—> TBSO \
/\ lmid., DMF /\
rt., 75%
111-147 111-148

To a solution of propargyl alcohol 111-147 (11.2 g, 0.20 mol) in DMF (70 mL), 1-

butyldimethylchlorosilane (36.2 g, 0.24 mmol) and imidazole (34 g, 0.50 mol) was
added. After stirring for 15 h, the reaction was poured in H20 (200 mL) and extracted

with pentane (3x300 mL). Purification by vacuum distillation afforded the TBS protected
Propargyl alcohol Ill-148 as a colorless oil (25.5 g, 75%). Spectroscopic properties of

111-148 match those reportedmb

194

 

Partial data for III-148: 1H NMR (500 MHZ, CDC13) 6 4.30 (t, J = 2.2 Hz, 2 H), 2.38 (t, J

= 2.2 Hz, 1 H), 0.9 (s, 9 H), 0.2 (S, 6 H).

 

nBuLi
T880\ 1.880%,
12, -78°C-rt
111-14a THF» 89% 111-6

To a —78 °C solution of III-148 (11 g, 64.7 mmol) in THF (60 mL), n-BuLi (32

mL of 2.02 M solution in hexanes, 64.7 mmol) was added. After stirring for 1 h, a

solution of 12 (18.9g, 74.4 mmol) in THF (35 mL) was added and the reaction was

warmed to rt. After 15 min, the reaction was diluted with EtzO (200 mL) and washed

with saturated sodium thiosulfate solution (3x150 mL). The organic layer was dried,
concentrated and purified by column chromatography (2% EtOAc in hexanes) to afford
17.0 g of alkynyl iodide III-6 as a brown liquid (89% yield). Spectroscopic properties of

III-6 match those reported. ‘06

Partial data for III-6: 1H NMR (500 MHz, C130,) 6 4.47 (s, 2 H), 0.92 (s, 9 H), 0.13 (s,

6 H).
O O
B /\n/kOH OH PPh3 / DIAD B filJLO
——-—t——’ f
r + M Et 0 69°/ N
2 . °
Ill-149 111-150 III-8

To a solution of bromomethylacrylic acid III-149 (7 g, 42.4 mmol) and DIAD

(8.34 mL, 42.4 mmol) in ether (64 mL), was added a solution of alcohol 29 (5.52 mL,
63.6 mol) and Ph3P (11.1 g, 42.4 mol) in ether (64 mL), dropwise at 0 °C. The reaction
was stirred at the same temperature for 30 min, and then at room temperature for 16 h.

The mixture was filtered and washed with EtzO (100 mL). After concentration of the

195

filtrate, the crude product was purified by flash column chromatography (3% EtOAc in

hexanes) to furnish acrylate III-8 as a yellow liquid (6.5 g, 69%). Spectral properties of
III-8 matched those reported.107
Partial data for III-8: 1H NMR (500 MHz, CDC13) a 6.31 (app 8, 1 H), 5.95 (app 3, 1 H),

5.88—5.82 (m, 1 H), 5.46 (quin, J = 6.4 Hz, 1 H), 5.26 (d, J = 16.7 Hz, 1 H), 5.18 (d, J =

10.1 Hz,1 H), 4.19 (s, 2 H), 1.38 (d, J: 6.7 Hz, 3 H).

 

 

 

 

 

 

 

 

 

 

CuCN-2LiC1 111-6 0.7 e
/\/\/‘ Zn, THF fl 4 ‘ZnA/VCLKCNMN ( Q) a:
l O
40 °C, 20 h THF / Pentane -60'C to -35 C
0°C, 30 min. 15h
Ill-7 111-159
0
O TBSO
Ill-8 (1.5 eq) 3h \____—_:__| BrAH‘o
7 / O M
-78 C to rt, 45% / m_8
III-160

 

 

 

1,2 dibromoethane (27 11L, 5 mol%) in THF (1.5 mL) was added to zinc dust (398
mg, 6.08 mmol) and the suspension was refluxed for 30 min. Upon cooling the mixture to
rt, TMSCl (23 11L, 3 mol%) and 1,4 diiodobutane III-7 (0.2 mL, 1.52 mmol) in THF (5
mL) were added and the mixture was heated at 40°C for 20 h after which CG analysis
indicated complete consumption of the diiodide. The suspension was then allowed to
settle at room temperature. The supernatant liquid was transferred to a pre—cooled (-
60°C) solution of CuCN (136 mg, 1.52 mmol) and LiCl (129 mg, 3.04 mmol) in 3 : 1
THF : pentane (2 mL). The resulting mixture was stirred at 0°C for 1h after which
alkynyl iodide 111-6 (315 mg, 1.06 mmol) in 1 :1 THF : pentane (1.5 mL) was added at -
60°C, stirred at -35°C for 20 h and again cooled to —78 °C. Allyl bromide 111-8 (500 mg,

2.28 mmol) was added at —78 °C and the reaction was warmed to room temperature.

196

After 4 h saturated NH4C1 solution (15 mL) was added to quench the reaction. The
aqueous layer was extracted with EtzO (3x50 mL), organic layers were combined, dried

(Na2804), concentrated and the product was purified by column chromatography (1% —
4% EtOAc in hexanes) to afford III-160 as a yellow liquid (174 mg, 45% yield).

Partial data for III-160: 1H NMR (300 MHz, CDC13) 6 6.16 (app 3, 1 H), 5.95-5.83 (m, 1
H), 5.52 (app s, 1 H), 5.44-5.40 (m, 1 H), 5.23 (d,J= 17.2, 1 H), 5.15 (d, J: 11.2, 1 H),
4.30428 (m, 2 H), 2.31 (t, J = 6.6 Hz, 2 H), 2.21 (t, J = 6.6 Hz, 2 H), 1.58-1.42 (m, 6 H),

1.36 (d, J = 6.6 Hz, 3 H), 0.97 (s, 9 H), 0.13 (s, 6 H).

O O

O TBAF O
TBSO // M HO 4 M

THF, 40°C, 85%
111-160 Ill-161

 

To a solution of III-160 (650 mg, 1.78 mmol), in THF (10 mL) cooled to -10°C,
TBAF (3.6 m1. of 1.0 M solution in THF, 3.6 mmol) was added and stirred for 45 min

after which the reaction was poured into water (15 mL) and extracted with EtzO (3x15
mL). Combined organic layers were dried over NaZSO4, concentrated and crude product
was purified by column chromatography (5% EtOAc in hexanes) to furnish propargylic

alcohol III-161 as a colorless liquid (378 mg, 85% yield).
Partial data for 111-161: 1H NMR (300 MHz, CDC13) 6 6.16 (app 3, 1 H), 5.89-5.84 (m, 1
H), 5.54 (app 3, 1 H), 5.48-5.41 (m, 1 H), 5.25 (d,J= 17.5 Hz, 1 H), 5.16 (d, J = 10.4 Hz,
1 H), 4.34-4.26 (m, 2 H), 2.33 (t, J := 6.2 Hz, 2 H), 2.23 (t, J = 6.6 Hz, 2 H), 1.76-1.36 (m,

9 H).

197

O O

HOMO H2, Llndlar 5 cat. HOMO
M t M

quinoline cat,
Ill-161 EtO Ac, rt 93% 111-162

 

A mixture of propargylic alcohol III-61 (600 mg, 2.4 mmol), Lindlar’s catalyst

(110 mg) and quinoline (0.3 mL) in ethyl acetate (20 mL) was vigorously stirred under

H2 (1 atm) for 2 h. The reaction mixture was then filtered over a pad of celite and the
residue was washed with ethyl acetate (40 mL). The filtrate was washed with 5% CuSO4

(2x5 mL). The organic portion was dried (Na2804), concentrated and the crude product

was purified by flash column chromatography to afford the cis allylic alcohol III-162 as a

colorless liquid (562 mg, 93%).
Partial data for III-162: 1H NMR (300 MHz, CDC13) 6 6.16 (app 5, 1 H), 5.94-5.83 (m, l

H), 5.62—5.53 (m, 3 H), 5.44—5.40 (m, 1 H), 5.30 (d, J = 16.0, 1 H), 5.20 (d, J = 10.4 Hz, 1
H), 4.20 (d, J = 6.3 Hz, 1 H), 2.31 (t, J = 7.3 Hz, 2 H), 2.12-2.05 (m, 2 H), 1.48-1.27 (m,

9 H).

O D-(—)-DlPT, Ti(O‘Pr)4
HOMO $ HOMO
N 0142012 4 A° MS, -25°c 0 M
Ill-162 68 /°' 92 /° 69 111-163

 

To a flame dried round bottom flask charged with pre-activated 4 A MS (100 mg)
and CH2C12 (3 mL), Ti(OiPr)4 (230 mg, 2.5 mmol) was added and the mixture was cooled
to -30°C. To this, a solution of D-(-)-DIPT (225 mg, 0.95 mmol) in CH2C12 (3 mL) was

added and the mixture was stirred for 30 min before t-BuOOH (0.9 mL of 4.01 M

solution in toluene, 3.52 mmol) was added to it. After stirring for another 30 min at the

same temperature, a solution of allylic alcohol 111-162 (400 mg, 1.59 mmol) in CHZCIZ (6

198

mL) was added dropwise and the reaction was stirred at -25°C for 18 h. Saturated
6132504 solution (0.8 mL) and saturated NaZSO3 (1.6 mL) were added and the reaction

was diluted with ether (12 mL). The mixture was stirred vigorously for 3 h, stored at 0 °C
overnight and then filtered through a celite pad. The filtrate was washed with anhydrous
ether (500 mL), concentrated and the crude product was purified by column
chromatography (5% EtOAc in hexanes) to furnish the epoxy alcohol III-163 as faint
pink liquid (290 mg 68% yield, 92% ee). The % ee of 111-163 was determined after

derivatization to the corresponding (S-MPA ester.
Partial data for III-163: 'H NMR (300 MHz, C130,) 6 6.24 (app 3, 1 H), 5.94-582 (m, 1

H). 5.52 (app 3, l H), 5.43-5.39 (m, 1 H), 5.25 (d, J = 17.5 Hz, 1 H), 5.14 (d, 10.9 Hz, 1
H), 3.84 (dd, J = 3.9, 12.1 Hz, 1 H), 3.67 (dd, J = 6.7, 12.1 Hz, 1 H), 3.18-3.13 (m, 1 H),

3.05-3.00 (m, 1 H), 2.31 (t, J: 6.7 Hz, 2 H), 2.05 (s(br), 1 H), 1.55-1.23 (m, 11 H).

O 0

HOME) DMP OHCMO
'9 M 7 9 M

CHQCIZ, rt, 89%
111-163 111-165

 

To a suspension of DMF reagent (3.2 g, 7.55 mmol) in CHZCl2 (20 mL), a
solution of the epoxy alcohol 111-163 (1.19 g, 4.44 mmol) in CH2C12 (10 mL) was added
at rt and the reaction was stirred for 2 h. After diluting with EtZO (26 mL), the mixture

was poured in a saturated solution of NaHCO3 (26 mL) containing NazSzo3 (9.5 g) and
stirred vigorously for 5 min. The layers were separated and the aqueous layer was washed
with CHZCl2 (2x30 mL). The combined organic layers were dried, concentrated and the
crude product was purified by flash column chromatography (2% EtOAc in hexanes) to

afford epoxy aldehyde 111-165 (1.05 g, 89%)-

199

Partial data for 111-165: lH NMR (300 MHz, CDC13) a 9.31 (d, J = 2.7, 1 H), 6.14 (app 3.
1 H), 5.93-5.82 (m, 1 H), 5.51 (app 5, 1 H), 5.24 (d, J = 17.2 Hz, 1 H), 5.13 (d, J = 10.5
Hz, 1 H), 3.35-3.32 (m, 1 H), 3.28-3.24 (m, 1 H), 2.30 (t, J = 6.7 Hz, 1 H), 1.8-1.34 (m, 8

H), 1.23 (t, J = 6.7 Hz, 3 H).

 

o + o
OHCMO Ph3PCH3Br ; W0
0 M NaHMDS, THF 0 M
111-165 -10 °C, 70% 111-4

To a suspension of methyl triphenylphosphonium bromide (710 mg, 1.99 mmol)
in THF (10 mL), NaHMDS (1.58 mL of 1.0 M solution, 1.58 mmol) was added at 0°C
and stirred at rt for 30 min. The resultant ylide was cooled back to —10°C. To this cooled

mixture, a solution of aldehyde III-16S (350 mg, 1.32 mmol) in THF (3 mL) was added
dropwise. After 10 min at —10°C, the reaction was quenched with sat. NH4C1 (20 mL)
diluted with ether (50 mL). The organic layer was washed with H20 (10 mL), brine (10

mL), dried and concentrated. The crude product was purified by column chromatography

to afford vinylic epoxide 4 as a colorless liquid (244 mg, 70% yield).

Partial data for 111-4: ‘H NMR (300 MHz, CDC13) a 6.10 (app 5, 1 H), 5.85-5.79 (m, 1
H), 5.69-5.62 (m, 1 H), 5.45 (app s, 1 H), 5.43 (d, t = 17.7 Hz, 1 H), 5.35-5.33 (m, 1 H),
5.28 (d, J = 10.6 Hz, 1 H), 5.20 (d, J = 17.6 Hz, 1 H), 5.08 (d, J = 10.6 Hz, 1 H), 3.33 (m,
1 H), 3.00 (m, 1 H), 2.52 (d, J = 7.7 Hz, 3 H); 13C NMR (125 MHz, CDC13) a 166.5,

141.3, 138.0, 132.9, 124.5, 120.4, 115.7, 71.3, 58.8, 57.3, 31.9, 29.1, 28.5, 27.8, 26.2,

20.1.

200

 

0
We BF3» OEt2 (10 mol%) W0 0” EA
.3 \‘I N : 7 ‘3,
O M+ 7 \ O

OH
q.) 01202, rt, 50%

 

(1.16
Ill-4 Ill-1 67 111-1 68
\
WOWW A020, DMAP J
7 <~ .
\ O CHQClQ, 1'1, quant.
III-159

A mixture of vinylic epoxide III-4 (22 mg, 83.0 umol) and alcohol III-167 (15

mg, 92.0 umol) was dissolved in CHZCI2 (0.18 mL). To this, BF3-OEt2 (10 11L of 0.83 M

solution in CHZCIZ, 8.3 umol) was added at once at room temperature. After 15 min, the

epoxide was completely consumed and several other spots appeared as judged by TLC.

The reaction was then diluted with CH2C12 (10 mL) and quenched with H20 (2 mL). The

aqueous layer was extracted with CHZCl2 (2x10 mL), combined organic layers were
dried, concentrated. Careful chromatographic purification (5% —- 7% EtOAc in hexanes)
afforded the adduct III-168 (17 mg, 50%). Structure of 111-168 was confirmed by 1H
homo decoupling experiments of the acetate derivative III-169.

Partial data for III-168: 1H NMR (500 MHz, CDC13) 6 6.15 (app S, 1 H), 5.95-5.81 (m, 2
H), 5.78-5.60 (m, 1 H), 5.57 (app 8, 1 H), 5.45-5.39 (m, l H), 5.30-5.24 (m, 3 H), 5.18—
5.13 (m, 1 H), 5.03-4.93 (m, 2 H), 3.59-3.39 (m, 3 H), 3.28-3.23 (m, 1 H), 2.77 (S (br), 1
H), 2.31 (t, J = 7.1 Hz, 2 H), 2.09-2.04 (t, 6.6 Hz, 2 H), 1.57-1.27 (m, 15 H).

Partial data for 111-169: 1H NMR (500 MHz, CDC13) 6 6.15 (app s, 1 H), 5.95-5.88 (m, 2
H), 5.83-5.87 (m, 1 H), 5.42 (app s, 1 H), 5.39 (m, 1 H), 5.21-5.25 (m, 3 H), 5.18 (m, 1
H), 4.95-5.10 (m, 3 H), 3.63-3.69 (m, 2 H), 3.45-3.55 (m, 1 H), 3.31-3.27 (m, 1 H), 2.28

(t, J = 6.7 Hz, 1 H), 1.98-2.04 (m, 5 H), 1.81-1.20 (m, 23 H).

201

OTBS OH BF 3.032 TBSO

/ /
ROW (4 mol%) 17 1
111-4 + H“ , H M HO "’0 1~°—"'o 9 ,
bras CHQClzt rt OTBS OH k;

0 0
111-140, R = TBDPS 111-170 (20%)

+ Ill-140 (55%)

Intermolecular ring opening of vinylic epoxide III-4 (20 mg, 76 umol) with
alcohol III-140 (51 mg, 77 umol) using 8133-0Et2 (3.5 111 of 0.9 M solution in CHzClz,
3.2 umol) was effected by the procedure described above. Adduct III-170 was obtained
in 20 % yield (14 mg) along with recovered alcohol III-140 (28 mg, 55%).

Partial data for 111-170: ‘H NMR (500 MHz, CDCI3) 3 7.62-7.78 (m, 5 H), 7.32-7.43 (m,

5 H), 6.15 (app S, 1 H), 5.98-5.62 (m, 3 H), 5.45 (app 8, 1 H), 5.39-5.42 (m, 1 H), 5.15—
5.30 (m, 6 H), 4.69-4.57 (m, 1 H), 4.20-4.37 (m, 2 H), 4.01-4.13 (m, 1 H), 3.78-3.60 (m,
2 H), 3.59-3.42 (m, 2 H), 3.38-3.29 (m, 1 H), 2.89 (s(br), 1 H), 2.23 (t, J = 6.7, Hz, 2 H),
2.02—2.18 (m, 1 H), 1.98-1.80 (m, ’l H), 1.43—.120 (m, 11 H), 1.04 (s, 9 H), 0.9 (s, 9 H),

0.8 (s, 9 H), 0.1 (s, 3 H), 0.08 (s, 3 H), 0.06 (s, 3 H), —0.02 (s, 3 H).

0 PCC / NaOAc 0

AOWOH o i AOJWCHO
CHQCIZ, rt, 72 a

111-171 111-21 0

 

To a suspension of FCC (22.9 g, 0.11 mol) and sodium acetate (2.4 g, 0.03 mol)

in CHzCl2 (100 mL), was added a solution of ethyl 6-hydroxyhexanoate III-171 (9.85 g,
0.06 mol) in CHzCl2 (24 mL) at room temperature. After 2 h, the reaction was diluted

with EtZO (150 mL) and filtered through a celite pad. The filtrate was concentrated under

reduced pressure and the crude material was purified by column chromatography (5%

EtOAc in hexanes) to afford aldehyde 111-210 (6.83 g, 72%).144

202

Partial data for III-210: 1H NMR (500 MHz, CDC13) 6 9.76 (t, J = 1.7 Hz, 1 H), 4.11 (q,
J = 7.1 Hz, 2 H), 2.48—2.43 (m, 2 H), 2.36-2.08 (m, 2 H), 1.69-1.63 (m, 4 H), 1.24 (t, J =
7.3 Hz, 3 H); 13C NMR (125 MHz, CDC13) 6 202.4, 173.5, 60.6, 43.7, 34.2, 24.5, 21.7,

14.5.

/\o CHO 4* /\o / CHO

PhH, reflux, 66%
"L210 "L172

 

Aldehyde 111-210 (6.8 g, 43.0 mmol) in benzene (30 mL) was added to a
suspension of (triphenylphosphoranylidene)acetaldehyde (13.1 g, 43.0 mmol) in benzene
(30 mL). The mixture was heated to reflux for 15 h and then cooled to room temperature.
Volatiles were evaporated and the crude product was purified by flash column
chromatography (15% EtOAc in hexanes) to obtain 5.2 g of a,B—unsaturated aldehyde
III-172 (66%) as a colorless liquid.

Partial data for 111-172: 1H NMR (500 MHz, 030,) a 9.51 (d, J = 7.7 Hz, 1 H), 6.84 (dt,

J = 6.7, 8.8 Hz, 1 H), 6.18-6.09 (m, 1 H), 4.13 (q, J = 7.15 Hz, 2 H), 2.41-2.31 (m, 4 H),

1.74-1.53 (m, 4 H), 1.24 (t, J = 7.1 Hz, 3 H).

O
M NaBHtCeC'S‘Hzo W21
/\o / CHO 4: /\o /

EtOH, 91%
111-172 111-211

 

To a solution of aldehyde III-172 (5.2 g, 28.3 mmol), cerium trichloride (10.6 g,

28.3 mmol) and sodium borohydride (1.07 g, 28.3 mmol) were added at room

temperature. After completion of the reaction (30 min), H20 (3.5 mL) was added and the
volatiles were removed under reduced pressure. The residue was taken up in EtzO (300

mL) and H20 (150 mL). Layers were separated, the aqueous layer was extracted with

203

EtZO (3x150 mL), the combined organic layers were dried (MgSO4) and concentrated.

After purification by column chromatography (30% EtOAc in hexanes), allylic alcohol

III-211 was produced in 91% yield (4.78 g).
Partial data for 111-211: 1H NMR (500 MHz, CDC13) 6 5.70-5.66 (m, 2 H), 4.17-4.10 (m,

4 H), 2.31 (t, J = 7.2 Hz, 2 H), 2.11-2.05 (m, 2 H), 1.70-1.60 (m, 2 H), 1.58-1.38 (m, 2

H), 1.27 (t, J = 7.14 Hz, 3 H).

 

o Ti(O‘Pr)4, D-DIPT o
/\ W011 ‘= ”OW/OH
O ttattoorl, MS 4 A ’0
111-211 72% 111-173

To a flame dried round bottom flask charged with pre-activated 4 A MS (1.54 g)

and CHzCl2 (46 mL), Ti(OiPr)4 (1.5 mL, 5.14 mmol) was added and the mixture was

cooled to —30°C. To this, a solution of D-(-)-DIPT (1.3 mL, 6.17 mmol) in CHzCl2 (46
mL) was added and the mixture was stirred for 30 min before t-BuOOH (13.9 mL of 4.01
M solution in toluene, 56.0 mmol) was added to it. After stirring for another 30 min at the
same temperature, a solution of allylic alcohol III-211 (4.78 g, 25.7 mmol) in CH2C12 (18
mL) was added dropwise and the reaction was stirred at ~25°C for 18 h. Saturated
NaZSO4 solution (5.4 mL) and saturated Na2S03 (30.8 mL) were added and the reaction
was diluted with ether (150 mL). The mixture was stirred vigorously for 3 h and then
filtered through a celite pad. The filtrate was washed with anhydrous ether (1 L),
concentrated and the crude product was purified by column chromatography (5% EtOAc
in hexanes) to furnish the epoxy alcohol III-173 as colorless liquid (3.74 72% yield,

>99% ee). The % ee of 111-173 was determined after derivatization to the corresponding

(S)-MPA ester.

204

Partial data for III-173: 1H NMR (500 MHZ, CDC13) 6 4.12 (q, J = 7.14 Hz, 2 H), 3.88

(dd, J = 2.2, 12.6 Hz, 1 H), 3.61 (dd, J = 4.1, 12.6 Hz, 1 H), 2.97-2.90 (m, 2 H), 2.31 (t, J

= 7.4 Hz, 2 H), 1.72-1.42 (m, 6 H), 1.25 (t, J = 7.14 Hz, 3 H).

JOK/\/\/\/ DMP, CHZCIZ j)\/\/\/\
/\o OH -— /\o ,0 CHO
0 rt, 64%
111-173 111-212

 

To a suspension of DMP reagent (10.2 g, 24.0 mmol) in CHZCI2 (40 mL), a
solution of the epoxy alcohol III-173 (2.42 g, 12.0 mmol) in CHzClz (20 mL) was added
at rt and the reaction was stirred for 2 h. After diluting with EtzO (50 mL), the mixture

was poured in a saturated solution of NaHCO3 (80 mL) containing Na28203 (20 g) and
stirred vigorously for 5 min. The layers were separated and the aqueous layer was washed
with CHzClz (2x100 mL). The combined organic layers were dried, concentrated and the
crude product was purified by flash column chromatography (2% EtOAc in hexanes) to
afford epoxy aldehyde III-212 (1.53 g, 64%).

Partial data for 111-212: 1H NMR (500 MHz, CDC13) 6 9.01 (d, J = 6.04 Hz, 1 H), 4.13
(q, J = 7.1 Hz, 2 H), 3.26-3.22 (m, l H), 3.16—3.13 (m, 1 H), 2.33 (t, J = 2.5 Hz, 2 H),

1.78-1.52 (m, 6 H), 1.26 (t, J = 7.1 Hz, 3 H).

 

O O R
+ A - W
KHMDS, toluene Z THF Ill-174 R = H
"F212 (1 I 1), -10 °C, 86% 111-175: R : Me

To a slurry of ethyltriphenylphosphonium bromide (1.97 g, 5.3 mmol) in 4 : 1
toluene : THF (10.6 mL) at —20 °C, KHMDS (9.16 mL of 0.5 M solution in toluene, 4.58
mmol) was added and the orange mixture was warmed to room temperature. After 1 h,

the yilde was cooled back to ~20 °C and a solution of aldehyde 111-212 (530 mg, 2.65

205

mmol) in THF (5.3 mL) was added. The reaction was continued at -—10 °C for 1 h after
which EtOH (0.19 mL) was added and the solids were filtered off through a celite pad.
The crude material was purified by column chromatography (10% EtOAc in hexanes) to
furnish vinylic epoxide 111-175 in 86% yield (483 mg) as colorless oil.

Vinyl epoxide 111-174 was prepared by the same procedure using
methyltriphenyl-phosphonium bromide. Thus 446 mg (85%) of 111-174 was obtained

from 530 mg of III-212.

Partial data for 111-175: 1H NMR (500 MHz, CDC13) 6 5.79-5.73 (m, 1 H), 5.08-5.04 (m,
1 H), 4.15 (q, J = 7.3 Hz, 2 H), 3.38-3.34 (m, 1 H), 2.84-2.80 (m, 1 H), 2.31 (t, J = 7.1
Hz, 2 H), 1.79 (dd, J = 1.7, 7.1 Hz, 3 H), 1.75-1.48 (m, 6 H), 1.24 (t, J = 7.3 Hz, 3 H).
Partial data for 111-174: 1H NMR (500 MHz, 030,) o 5.62-5.40 (m, 2 H), 5.26-5.23 (m,
1 H), 4.11 (q, J = 7.1 Hz, 2 H), 3.10—3.07 (m, 1 H), 2.84-2.80 (m, 1 H), 2.31 (t, J = 7.1

HZ, 2 H), 1.72—1.45 (m, 6 H), 1.25 (t, J = 7.2 Hz, 3 H).

O OH

O)LH flM Br O/K/
THF, 0 °C
0
111-176 79 /° 111-1 77

Cyclohexylcarboxaldehyde 111-176 (1.12 g, 10 mmol) in T HP (10 mL) was added

to a solution of vinylmagnesium bromide (12 mL of 1.0 M solution, 12 mmol) in THF

(10 mL) at 0 °C. After 3 h, the reaction was quenched by addition of saturated NH4C1
solution (10 mL). The layers were separated and aqueous layer was extracted with EtZO

(3x15 mL). The combined organic layers were dried over MgSO4, concentrated and the

crude material was purified by flash column chromatography (20% EtOAc in hexanes) to

206

afford alcohol III-177 as a colorless oil (1.1 g, 79%). Spectral data for 111-177 matched
that of the reported.‘45
Partial data for III-177: 1H NMR (500 MHz, CDC13) 6 5.92-5.81 (m, 1 H), 5.24-5. 12 (m,

2 H), 3.87-3.45 (m, 1 H), 1.87-0.99 (m, 11 H).

o (PhS)2 o
/\ Wort ...—.... A W801
O "'0 1311313, TEA O "'(3
Ill-173 65 /° 111-189

To a solution of dipehyldisulfide (3.28 g, 15.0 mmol) in triethylamine (9 mL),
was added tributylphosphine (3.5 mL, 15.0 mmol) at ambient temperature under N2. This
solution was cooled to 0 °C and to it was canulated a pre-cooled solution of epoxy
alcohol 111-173 (1.0 g, 5.0 mmol) in Et3N. After stirring at ambient temperature for 6 h,

the reaction mixture was quenched with water (50 mL) and the aqueous solution

extracted with EtOAc (3x150 mL). The combined EtOAc extracts were dried over
NazSO4 and concentrated under reduced pressure. The crude product was purified by

flash column chromatography (5% EtOAc in hexanes) to afford epoxy sulfide III-189 as

a colorless oil (955 mg, 65%).
Partial data for 111-189: ‘H NMR (500 MHz, CDCI3) 8 7.44-7.19 (m, 5 H), 4.11 (q, J =

7.1 Hz. 2 H), 3.213 12 (m, 1 H), 2.96-2.86 (m, 2 H), 2.68-2.61 (m, 1 H), 2.27 (t, J = 7.3

Hz, 2 H), 1.67-1.24 (m, 6 H), 1.26 (t, J = 7.2 Hz, 3 H).

 

\ CV
OH H
O / BF3'OE12 (2 mol%) 0 O
/\ W + l l t /\ W
O ”O CHQCIQ, r1, 4270 0 (SH
Ill-174 Ill-177 111-178

207

Coupling of vinylic epoxide III-174 (50 mg, 0.25 mmol) and alcohol 111-177 (35

mg, 0.25 mmol) was performed using the same representative procedure as above but by
dropwise addition of BFyOEt2 (6 1.1L of 0.85 M solution in CHZCIZ, 5 umol) at room
temperature. Adduct III-178 was obtained in 42% yield as a mixture of diastereomers (35
mg).

Partial data for 111-178: 1H NMR (500 MHz, CDC13) 6 5.84-5.74 (m, 1 H), 5.64-5.72 (m,
2 H), 5.52-5.62 (m, 1 H), 5.32-5.28 (m, 1 H), 5.24-5.18 (m, 3 H), 5.16-5.04 (m, 4 H),
4.08-4.14 (m, 4 H), 3.74—3.68 (m, 3 H), 3.64-3.59 (m, 1 H), 3.55 (t, J = 7.0 Hz, 1 H), 3.46

(t, J = 6.9 Hz, 1 H), 2.38-2.21 (m, 4 H), 2.02 (s(br), 2 H), 1.82-0.90 (m, 40 H); 13C NMR
(125 MHZ, CDC13) 6 174.0, 173.9, 138.2, 137.7, 135.4, 135.3, 120.0, 118.4, 118.3, 117.1,

84.6, 82.4, 82.1, 80.5, 73.6, 72.4, 60.4, 60.3, 42.5, 34.5, 34.4, 32.0, 31.9, 29.3, 29.2, 29. 1,

29.0, 26.8, 26.7, 26.4, 26.3, 26.2, 26. 1, 25.7, 25.5, 25.2, 14.4.

O OH BF3’OE12
/\ WSW) OM (20 mol%)
0 "'6 + A

CHQClg 0°C 10 rt

111-189 111-177 75%

\
o SPh 1
(3H
111-190

A solution of epoxy sulfide III-189 (30 mg, 0.1 mmol) and alcohol III-177 (29
mg, 0.2 mmol) in CHZCl2 (1 mL) was cooled to —10 ° C. BF3°OEt2 (24 11L of 0.8 M, 0.02
mmol) was added dropwise and the reaction was stirred at —10 °C to 0 °C for 10 h.

Saturated NaHCO3 solution (0.5 mL) was then added dropwise and the reaction was

diluted with CHZCl2 (10 mL) and H20 (5 mL). After separation of layers, the aqueous

208

layer was extracted with CHZCI2 (2x10 mL), combined organic layers were dried,
concentrated to afford a crude oil. Purification by flash column chromatography (5% —
7% EtOAc in hexanes) furnished ring opened product III-190 (32 mg, 75%).
Partial data for III-190: 1H NMR (500 MHz, CDC13) 6 7.43-7.20 (m, 5 H), 5.64-5.60 (m,
1 H), 5.21-5.06 (m, 3 H), 4.12 (q, J = 7.3, 2 H), 3.70-3.66 (m, l H), 3.56-3.49 (m, 1 H),

3.44-3.39 (m, 1 H), 3.36-3.29 (m, 1 H), 2.26 (dt J=1.6, 7.5 Hz, 2 H), 1.91 (d, J = 5.5 Hz,

3 H), 1.78-1.22 (m, 17 H).

 

OH
/ Ill-189 , WSW
AIM93 + + CGFSOH /\0 .
CHQCIQ OH
111-177 -78 °C to rt
(1 eq) (2 60») 111-195 H = C5F5.
(76%) ....
O /
flow/SPh 111-196 R = dv
we) (13%)
111-189

 

 

 

To a solution of pentafluorophenol (110 mg, 0.60 mmol) and alcohol III-177 (42
mg, 0.30 mmol) in CH2C12 (1 mL) was added trimethyl aluminum (0.15 mL of 2 M
solution in toluene, 0.30 mmol) dropwise at room temperature. After 1 h, the brown
solution was cooled to —78 °C and epoxy sulfide 111-189 in CH2C12 (0.6 mL) was added.
The reaction was then warmed to room temperature over 90 min, after which saturated

NaHCO3 solution (2 mL) was added dropwise. The mixture was diluted with CHzCl2 (5
mL) and H20 (3 mL) and the layers were separated. The aqueous layer was extracted

with CHZCl2 (3x5 mL), the combined organic layers were dried, concentrated to obtain a

crude oil. Upon purification by column chromatography (5% EtOAc in hexanes), two
ring opened products 111-195 (37 mg, 76%) and 111-196 (6 mg, 13%) were isolated as

colorless liquids.

209

Partial data for III-195: 1H NMR (500 MHZ, CDC13) 6 7.30-7.21 (m, 5 H), 4.21—4.20 (m,

1 H), 4.14 (q, J = 7.1 Hz, 2 H), 3.97-3.96 (m, 1 H), 3.39 (dd, J = 7.3, 14.1 Hz, 1 H), 3.13

(dd, J = 4.9, 13.9 Hz, 1 H), 2.32—2.29 (m, 2 H), 2.03 (s(br), 1 H), 1.71—1.29 (m, 6 H), 1.25
(t, J = 7.3 Hz, 3 H); 13C NMR (125 MHZ, CDC13) a 173.9, 137.2, 135.1, 132.5, 130.5,

129.5, 129.3, 128.0, 127.2, 86.0, 71.6, 71.4, 60.5, 55.8, 34.9, 34.3, 33.8, 33.1, 29.9, 25.7,

25.5, 24.9, 14.4.
Partial data for III-196: 1H NMR (500 MHZ, CDC13) 6 7.42-7.25 (m, 5 H), 5.68-5.50 (m,

2 H), 5.22-5.12 (m, 2 H), 4.25 (q, J = 7.1 HZ, 2 H), 3.90-3.81 (m, 1 H), 3.80-3.65 (m, 1

H), 3.60-3.54 (m, 1 H), 3.42-3.48 (m, '1 H), 3.38-3.25 (m, 3 H), 3.18-2.98 (m, 1 H), 2.21-

2.35 (m, 2 H), 1.78—1.20 (m, 20 H).

210

E. References

10.

11.

12.

13.

14.

15.

16.

17.

18.

Cave, A.; Figadere, B.; Laurens, A.; Cortes, D. in Progress in the Chemistry of
Organic Natural Products; Herz, W., Kirby, G. W., E., M. R., Steglish, W.,
Tamm, C., Eds; Springer-Verlag: New York, 1997; Vol. 70, pp 81—287.

Zeng, L.; Ye, Q.;Ober1ies, H.; Shi, G.; Gu, Z.—M.; He, K.; McLaughlin, J. L. Nat.
Prod. Rep. 1996, I3, 275.

Gu, Z.-M.; Zhao, G.-X.; Oberlies, N. H.; Zeng, L.; McLaughlin, J. L. Rec. Adv.
Phytochem. 1995, 29, 249.

Fang, X. P.; Rieser, M. J.; Gu, Z. M.; Zhao, G. X.; McLaughlin, J. L. Phytochem.
Anal. 1993, 4, 27.

Rupprecht, J. K.; Hui, Y. H.; McLaughlin, J. L. J. Nat. Prod. 1990, 53, 237.

Alali, F. Q.; Liu, X.-X.; McLaughlin, J. L. J. Nat. Prod. 1999, 62, 504.

Schaus, S. E.; Branalt, J.; Jacobsen, E. N. J. Org. Chem. 1998, 63, 4876.

Evans, P. A.; Cui, J.; Gharpure, S. J.; Polosukhin, A.; Zhang, H.-R. J. Am. Chem.
Soc. 2003, 125, 14702.

Takahashi, S.; Nakata, T. J. Org. Chem. 2002, 67, 5739.

Yang, W. Q.; Kitahara, T. Tetrahedron 2000, 56, 1451.

Shi, G.; Kozlowski, J. F.; Schwedler, J. T.; Wood, K. V.; MacDougal, J. M.;
McLaughlin, J. L. J. Org. Chem. 1996, 61, 7988.

Marshall, J. A.; Hinkle, K. W.; Hagedorn, C. E. Isr. J. Chem. 1997, 37, 97.

Nonfon, M.; Lieb, E; Moeschler, H.; Wendisch, D. Phytochemistry 1990,29,
1951.

Narayan, R. S.; Sivakumar, M.; Bouhlel, E.; Borhan, B. Org. Lett. 2001, 3, 2489.
Koert, U. Synthesis 1995, 115.
Hoppe, R.; Scharf, H. D. Synthesis 1995, 1447.

Casiraghi, G.; Zanardi, F.; Battistini, L.; Rassu, G.; Appendino, G. Chemtracts
1998, 1 I , 803.

Louie, J.; Bielawski, C. W.; Grubbs, R. H. J. Am. Chem. Soc. 2001, 123, 11312.

211

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

Knochel, P.; Yeh, M. C. P.; Berk, S. C.; Talbert, J. J. Org. Chem. 1988, 53, 2390.
Achyutharao, S.; Knochel, P. J. Org. Chem. 1991, 56,4591.

Alvarez, E.; Candenas, M. L.; Perez, R.; Ravelo, J. L.; Martin, J. D. Chem. Rev.
1995, 95, 1953.

Marmsater, F. P.; West, F. G. Chem. Eur. J. 2002, 8, 4347.
Koert, U. Angew. Chem, Int. Ed. 1995, 34, 298.
Trost, B. M.; Tenaglia, A. Tetrahedron Lett. 1988, 29, 2931.

Sinou, D.; Frappa, 1.; Lhoste, P.; Porwanski, S.; Kryczka, B. Tetrahedron Lett.
1995, 36, 1251.

Godleski, S. A. in Comprehensive Organic Synthesis; lst ed.; Trost, B. M.,
Fleming, 1., Eds; Pergamon Press: Oxford, 1991; Vol. 4, pp 585-662.

Foumier-Nguefack, C.; Lhoste, P.; Sinou, D. Tetrahedron 1997, 53, 4353.

Mitsunobu, O. in Comprehensive Organic Synthesis; lst ed.; Trost, B. M.,
Fleming, 1., Eds; Pergamon Press: Oxford, 1991; Vol. 6, pp 88-93.

Trost, B. M.; Fleming, 1., Eds. Comprehensive Organic Synthesis; lst ed.;
Pergamon Press: Oxford, 1991; Vol. 6.

Birkinshaw, T. N. in Comprehensive Organic Functional Group Transformations;
Katritzky, A. R., Meth-Cohn, 0., Rees, C. W., Roberts, S. M., Eds; Pergamon
Press: Oxford, 1990; Vol. 1, pp 204—220.

Wood, H. B.; Ganem, B. J. Am. Chem. Soc. 1990, 112, 8907.

Toda, J.; Sano, T.; Tsuda, Y.; Itatani, Y. Chem. Pharm. Bull. 1982,30, 1322.

Schultz, A. G.; Lucci, R. D.; Fu, W. Y.; Berger, M. H.; Erhardt, J.; Hagmann, W.
K. J. Am. Chem. Soc. 1978, 100, 2150.

Voyle, M.; Kyler, K. S.; Arseniyadis, S.; Dunlap, N. K.; Watt, D. S. J. Org.
Chem. 1983, 48, 470.

Schwartz, R. E.; Helms, G. L.; Bolessa, E. A.; Wilson, K. E.; Giacobbe, R. A.;
Tkacz, J. S.; Bills, G. F.; Liesch, J. M.; Zink, D. L.; Curotto, J. E.; Pramanik, B.;
Onishi, J. C. Tetrahedron 1994, 50, 1675.

Ley, S. V.; Sternfeld, F. Tetrahedron 1989, 45, 3463.

212

37.

38.

39.

41.

42.

43.

45 .

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

Peredamiranda, R.; Garcia, M.; Delgado, G. Phytochemistry 1990, 29, 2971.
Posner, G. H.; Rogers, D. Z. J. Am. Chem. Soc. 1977, 99, 8214.

Otera, J .; Yoshinaga, Y.; Hirakawa, K. Tetrahedron Lett. 1985, 26, 3219.
lranpoor, N.; Baltork, I. M. Tetrahedron Lett. 1990, 31 , 735.

Caron, M.; Sharpless, K. B. J. Org. Chem. 1985, 50, 1557.

Fagnou, K.; Lautens, M. Org. Lett. 2000, 2, 2319.

Trost, B. M.; VanVranken, D. L. Chem. Rev. 1996, 96, 395.

Righi, G.; Pescatore, G.; Bonadies, F.; Bonini, C. Tetrahedron 2001, 57, 5649.

Keinan, E.; Sahai, M.; Roth, 2.; Nudelman, A.; Herzig, J. J. Org. Chem. 1985, 50,
3558.

Trost, B. M.; McEachern, E. J.; Toste, F. D. J. Am. Chem. Soc. 1998, 120, 12702.
Kawata, S.; Hirama, M. Tetrahedron Lett. 1998, 39, 8707.

Friesen, R. W.; Danishefsky, S. J. J. Am. Chem. Soc. 1989, 111, 6656.

Halcomb, R. L.; Danishefsky, S. J. J. Am. Chem. Soc. 1989, 11], 6661.

Gordon, D. M.; Danishefsky, S. J. Carbohydr. Res. 1990, 206, 361.

Randolph, J. T.; Danishefsky, S. J. J. Am. Chem. Soc. 1993, 115, 8473.

Danishefsky, S. J.; McClure, K. F.; Randolph, J. T.; Ruggeri, R. B. Science 1993,
260, 1307.

Liu, K. K. C.; Danishefsky, S. J. J. Org. Chem. 1994, 59, 1892.
Liu, K. K. C.; Danishefsky, S. J. J. Org. Chem. 1994, 59, 1895.

Prestat, G.; Baylon, C.; Heck, M. P.; Mioskowski, C. Tetrahedron Lett. 2000, 41 ,
3829.

Kolb, H. C.; Vannieuwenhze, M. S.; Sharpless, K. B. Chem. Rev. 1994, 94, 2483.

Katsuki, T.; Sharpless, K. B. J. Am. Chem. Soc. 1980, 102, 5974.

213

58.

59.

60.

61.

62.

63.

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

Gao, Y.; Hanson, R. M.; Klunder, J. M.; Ko, S. Y.; Masamune, H.; Sharpless. K.
B. J. Am. Chem. Soc. 1987, 109, 5765.

Brown, C. A.; Yamashita, A. J. Am. Chem. Soc. 1975, 97, 891.
Kimmel, T.; Becker, D. J. Org. Chem. 1984, 49, 2494.

Speicher, A.; Eicher, T. Synthesis 1995, 998.

Tsuboi, S.; Masuda, T.; Takeda, A. Chem. Lett. 1983, 1829.

Hooz, J.; Layton, R. B. Can. J. Chem. 1972, 50, 1105.

Schlosser, M.; Christmann, K. F. Angew. Chem, Int. Ed. 1966, 5, 126.
Schlosser, M.; Tuong, H. B.; Schaub, B. Tetrahedron Lett. 1985, 26, 311.

DeMico, A.; Margarita, R.; Parlanti, L.; Vescovi, A.; Piancatelli, G. J. Org.
Chem. 1997, 62, 6974.

Couturier, M.; Dory, Y. L.; Rouillard, F.; Deslongchamps, P. Tetrahedron 1998,
54, 1529.

Heath, R. R.; Tumlinson, J. H.; Doolittle, R. E.; Duncan, J. H. J. Chromatogr. Sci.
1977, I5, 10.

Gilman, H.; Catlin, W. E. Org. Synth., CV1, 188.

Mori, K.; Tomioka, H. Liebigs Ann. Chem. 1992, 1011.
Mori, K.; Ogita, H. Liebigs Ann. Chem. 1994, 1065.
Freeman, F.; Kim, D. Synthesis 1989, 698.

Bengtsson, M.; Liljefors, T. Synthesis 1988, 250.
Magoon, E. F.; Slaugh, L. H. Tetrahedron 1967, 23, 4509.
Doolittle, R. E. Synthesis 1984, 730.

Chattopadhyay, S.; Mamdapur, V. R.; Chadha, M. S. Synth. Commun. 1990, 20,
1299.

Sankaranarayanan, S.; Chattopadhyay, S. Tetrahedron-Asymmetry 1998, 9, 2627,

Fisher, 1. G.; Tyman, J. H. P. Synth. Commun. 1998, 28, 1323.

214

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

96.

Baurle, S.; Peters, U.; Friedrich, T.; Koert, U. Eur. J. Org. Chem. 2000, 2207.

Paquette, L. A. in Encyclopedia of Reagents for Organic Synthesis; Paquette, L.
A., Ed.; Wiley: Chichester ; New York, 1995; Vol. 5, pp 3009-3014.

Curran, D. B; Yu, H. S. Synthesis 1992, 123.

Horita, K.; Yoshioka, T.; Tanaka, T.; Oikawa, Y.; Yonemitsu, O. Tetrahedron
1986, 42, 3021.

Evans, D. A.; Kaldor, S. W.; Jones, T. K.; Clardy, J.; Stout, T. J. J. Am. Chem.
Soc. 1990, 112, 7001.

Reed, L. A.; lto, Y.; Masamune, S.; Sharpless, K. B. J. Am. Chem. Soc. 1982,
104, 6468.

Sekine, M.; Kume, A.; Hata, T. J. Chem. Soc., Chem. Commun. 1981, 969.

Johnson, R. A.; Sharpless, K. B. in Catalytic Asymmetric Synthesis; 2nd ed.;
Ojima, 1., Ed; Wiley-VCH: New York, 2000, pp 231-280.

Nicolaou, K. C.; Duggan, M. E.; Hwang, C. K.; Somers, P. K. J. Chem. Soc.,
Chem. Commun. 1985, 1359.

Nicolaou, K. C.; Prasad, C. V. C.; Somers, P. K.; Hwang, C. K. J. Am. Chem. Soc.
1989, II], 5330.

Nicolaou, K. C.; Prasad, C. V. C.; Somers, P. K.; Hwang, C. K. J. Am. Chem. Soc.
1989, III, 5335.

Gill, D. M.; Pegg, N. A.; Rayner, C. M. J. Chem. Soc., Perkin Trans. I 1993,
1371.

Sivakumar, M.; Borhan, B. Tetrahedron Lett. 2003, 44, 5547.

De Lucchi, 0.; Miotti, U.; Modena, G. Org. React. (N. Y.) 1991, 40, 157.
Padwa, A.; Gunn, D. E., J r.; Osterhout, M. H. Synthesis 1997, 1353.
Sugihara, H.; Tanikaga, R.; Kaji, A. Synthesis 1978, 881.

Dilworth, B. M.; McKervey, M. A. Tetrahedron 1986, 42, 3731.

Bakuzis, P.; Bakuzis, M. L. F.; Fortes, C. C.; Santos, R. J. Org. Chem. 1976,41,
2769.

215

97.

98.

99.

100.

101.

102.

103.

104.

105.

106.

107.

108.

109.

110.

111.

112.

113.

114.

115.

lmbroisi, D. D.; Simpkins, N. S. J. Chem. Soc., Perkin Trans. 1 1991, 1815.
Koert, U.; Wagner, H.; Pidun, U. Chemise/1e Berichte 1994, 127, 1447.

Berninger, J.; Koert, U.; Eisenberghohl, C.; Knochel, P. Chem. Ber. 1995, 128,
1021.

Amdt, S.; Emde, U.; Baurle, S.; Friedrich, T.; Grubert, L.; Koert, U. Chem. Eur.
J. 2001, 7, 993.

Reetz, M. T. Angew. Chem, Int. Ed. 1984, 23, 556.

Trost, B. M.; Belletire, J. L.; Godleski, S.; McDougal, P. G.; Balkovec, J. M.;
Baldwin, J. J.; Christy, M. E.; Ponticello, G. S.; Varga, S. L.; Springer, J. P. J.
Org. Chem. 1986, 51, 2370.

Harada, N.; Nakanishi, K. Circular Dichroic Spectroscopy : Exciton Coupling in
Organic Stereochemistry; University Science Books: Mill Valley, CA, 1983.

Berova, N.; Nakanishi, K.; Woody, R. Circular Dichroism .° Principles and
Applications; 2nd ed.; Wiley-VCH: New York, 2000.

Harada, N.; Chen, S. L.; Nakanishi, K. J. Am. Chem. Soc. 1975, 97, 5345.

Trost, B. M.; Tanoury, G. J.; Lautens, M.; Chan, C.; Macpherson, D. T. J. Am.
Chem. Soc. 1994, 116, 4255.

Furstner, A.; Dierkes, T. Org. Lett. 2000, 2, 2463.
Mitsunobu, 0. Synthesis 1981, 1.

Hughes, D. L.; Reamer, R. A.; Bergan, J. J.; Grabowski, E. J. J. J. Am. Chem. Soc.
1988, 110, 6487.

Singh, R.; Whitesides, G. M. J. Am. Chem. Soc. 1990, 112, 1190.

Inokuchi, T.; Asanuma, G.; Torii, S. J. Org. Chem. 1982, 47, 4622.

Jorapur, V. S.; Duffley, R. P.; Razdan, R. K. Synth. Commun. 1984, 14,655.
Ando, M.; Wada, T.; lsogai, K. J. Org. Chem. 1991, 56,6235.

Oki, M.; Shionoiri, K.; Otake, K.; Ono, M.; Toyota, S. Chem. Lett. 1991, 597.

Nakamura, K.; Kimura, T.; Kanno, H.; Takahashi, E. J. Antibiot. 1995, 48, 1134,

216

116.

117.

118.

119.

120.

121.

122.

123.

l 24.

125.

126.

127.

128.

129.

130.

131.

132.

133.

l 34.

135.

Knochel, P.; Jeong, N.; Rozema, M. J.; Yeh, M. C. P. J. Am. Chem. Soc. 1989,
111, 6474.

Knochel, P.; Achyutharao, S. J. Am. Chem. Soc. 1990, 112, 6146.
Luche, J. L. J. Am. Chem. Soc. 1978, 100, 2226.

Luche, J. L.; Rodriguezhahn, L.; Crabbe, P. J. Chem. Soc., Chem. Commun. 1978,
601.

Kita, Y.; Furukawa, A.; Futamura, J.; Ueda, K.; Sawama, Y.; Hamamoto, H.;
Fujioka, H. J. Org. Chem. 2001, 66, 8779.

David, S.; Hanessian, S. Tetrahedron 1985, 41, 643.
Blunden, S. J.; Cusack, P. A.; Smith, P. J. J. Organomet. Chem. 1987, 325, 141.

Danishefsky, S. J.; Gervay, J.; Peterson, J. M.; McDonald, F. E.; Koseki, K.;
Griffith, D. A.; Oriyama, T.; Marsden, S. P. J. Am. Chem. Soc. 1995, 117, 1940.

Hodosi, G.; Kovac, P. Carbohydr. Res. 1998, 308, 63.

Sasaki, M.; Miyazawa, M.; Tanino, K.; Miyashita, M. Tetrahedron Lett. 1999, 40,
9267.

Sasaki, M.; Tanino, K.; Miyashita, M. J. Org. Chem. 2001, 66, 5388.

Liu, C. Q.; Hashimoto, Y.; Kudo, K.; Saigo, K. Bull. Chem. Soc. Jpn. 1996, 69,
2095.

Maruoka, K.; Oishi, M.; Shiohara, K.; Yamamoto, H. Tetrahedron 1994, 50,
8983.

Saito, S.; Yamamoto, H. Chem. Commun. 1997, 1585.

Marx, A.; Yamamoto, H. Synlett 1999, 584.

Byun, H. 8.; He, L. L.; Bittman, R. Tetrahedron 2000, 56, 7051.

Gao, Y.; Sharpless, K. B. J. Am. Chem. Soc. 1988, 110, 7538.

Berridge, M. S.; Franceschini, M. P.; Rosenfeld, E.; Tewson, T. J. J. Org. Chem.
1990, 55, 1211.

Tewson, T. J. J. Org. Chem. 1983,48, 3507.

He, L. L.; Byun, H. S.; Bittman, R. J. Org. Chem. 1998, 63, 5696.

217

136.

137.

138.

139.

140.

141.

142.

143.

144.

145.

Moghaddam, M. F.; Motoba, K.; Borhan, B.; Pinot, F.; Hammock, B. D. Bioehim.
Biophys. Acta 1996, 1290, 327.

Myers, A. G.; Goldberg, S. D. Angew. Chem, Int. Ed. 2000, 39, 2732.

Bianco, A.; Brufani, M.; Manna, F.; Melchioni, C. Carbohydr. Res. 2001, 332,
23.

Sanders, W. J .; Kiessling, L. L. Tetrahedron lett. 1994, 35, 7335.
Capaccio, C. A. 1.; Varela, O. Tetrahedron-Asymmetry 2000, 1], 4945.
Corey, E. J .; Helal, C. J. Tetrahedron Lett. 1996, 37, 5675.

Galinis, D. L.; Fuller, R. W.; McKee, T. C.; Cardellina, J. H.; Gulakowski, R. J.;
McMahon, J. B.; Boyd, M. R. J. Med. Chem. 1996, 39, 4507.

Shinkai, S.; Tsuji, H.; Hara, Y.; Manabe, 0. Bull. Chem. Soc. Jpn. 1981, 54, 631.
Trost, B. M.; Verhoeven, T. R. J. Am. Chem. Soc. 1980, 102, 4743.

Bergmeier, S. C.; Stanchina, D. M. J. Org. Chem. 1997, 62, 4449.

218

CHAPTER IV

TOTAL SYNTHESIS OF THE PROPOSED STRUCTURE OF MUCOXIN

A. Revised strategies for the coupling of left- (C 13-C37) and right-hand (Cl-C12)

fragments of mucoxin

As discussed in Chapter III, our original synthetic approach to mucoxin called for
a late stage coupling of the fully functionalized allylic alcohol IV-l and vinylic epoxide
IV-2 (Figure IV-l) via a regioselective C-O bond formation. Since the proposed
intermolecular allylic alkylation strategy was not successful, we turned to explore
alternative routes to couple the two halves. 1n redesigning the synthesis, we decided to
rely on C-C bond forming reactions because we felt that, a broader range of

methodologies could be explored for intermolecular C-C bond formation as compared to
C-0 bond formation.1 Nonetheless, in order to keep the synthesis concise and convergent,

we wanted to conserve the global Strategy to couple the two fragments in their

functionally elaborated forms.

l 3.37 l ,3?
TBSQ OH 0 35 -‘ TBSQ OH 35
717 0,, / e o 7 7 0., 0., O
14.? /10 O 14..’1‘l\10\\ O
ores ores
lV-1 IV-2 IV-3

Figure IV-l: Original regio- and stereoselective intermolecular epoxide opening strategy
The hydroxylated THF portion (Cl2-C37) of mucoxin was available from earlier

studies (Chapter 111) in the form of an aldehyde (IV-4; X = H, Figure 1V-2). Since

carbonyl group is a versatile functionality and has been used extensively in C-C bond

219

forming reactions both as an electrophile as well as a nucleophile,2 we decided to

conserve aldehyde IV-4 as the left hand coupling partner in our revised synthetic plan. A
general design of the right hand segment IV-5 is shown in Figure IV-2. Accordingly, the
plan required accessing a fragment containing a terminal acrylate (IV-5), an appropriate
reactive group (M) at the other end and a suitable functionality along the linker that can

be elaborated to a 2,5 di-Substituted THF ring.

 

TBSQ o 351W... 37
’7 O... Asuitable functionality
16 13 x + M‘ito installtheCB-C12TH1—7 2 ,0

14 l'OTBS O
lV-4 IV-5

 

 

 

Figure IV-2: General representation of the revised strategy

An obvious C-C bond forming reaction involving a carbonyl reacting partner is

addition of an organometallic reagent to the carbonyl group. Organomagnesium

5

(Grignard)3'4 and organolithium reagents are probably the most commonly used species

for this purpose. Although carbonyl addition reactions of magnesium and lithium

organics are highly facile and reliable, these organometallics, owing to the highly polar
nature of the metal—carbon bond, exhibit low chemoselectivity in their reactions}5 Thus
in addition to carbonyl functionalities, they also react with several moieties including
epoxides, nitriles, halides and in some cases even silyl and benzyl protecting groups.6
Their reactivity can be attenuated by techniques such as transmetallation to the
corresponding copper7 or titanium8 species. Nevertheless, since such transition metal

reagents were derived from the corresponding organomagnesiums or organolithiums,

highly functionalized organometallics are not accessible. Clearly, in our case, the right

220

hand piece IV-S could not be derivatized as a Grignard or organolithium species due to
the sensitive ester group.
Functionalized organozinc reagents bearing electrophilic carbonyl groups and

their equivalents are stable and can be generated from the corresponding alkyl halides.

9.10

Although organozinc compounds have been known for several decades, they have

found only limited utility in organic synthesis possibly due to their lack of inherent
reactivity. However, the discovery that organozincs can be efficiently transmetallated to a

11.12

variety of more reactive transition metal salts, opened avenues for new applications.

During the past few years, mostly through the work of Knochel, these reagents have

emerged as effective alternatives to the conventional main group organometallic
reagentslms Organozincs can be prepared under mild conditions (that not require pre-
formation of the corresponding organomagnesium or lithium species) by direct insertion
of elemental zinc into carbon-halogen bonds, or via zinc—iodine or boron-zinc
exchange.”‘16'l7 Due to the availability of such methods of preparation and their inherent

low reactivity, several organozinc reagents containing reactive functional groups like
esters, ketones, nitriles, amides, nitro groups and epoxides have been prepared.

Organozincs so generated can be reacted with various electrophiles with or without
transition metal catalysts depending upon the reactivity of the latter.15 Thus, organozinc

mediated coupling reactions offer an attractive strategy to combine fragments bearing

sensitive functional groups.

221

1. Evaluation of coupling strategies involving organozinc additions

Being aware of the scope and recent discoveries on organozinc reagents, our first
plan was to couple tri-substituted THF aldehyde IV-4 with an organozinc species derived
from a suitably functionalized right hand fragment of type IV-S (M 2 Zn, Figure lV-2).
To quickly test the feasibility of this approach, our immediate target was to access the
functionalized primary iodide IV-6 (Figure lV-3) designed as a model system. Also, a
model tetrahydrofuranyl aldehyde IV-8 that closely mimicked the real aldehyde IV-4 was
available from our earlier Studies (Chapter 111). Chelation controlled addition of the

organozinc obtained from iodide IV-6 to aldehyde IV-8 would afford the corresponding

18,19

coupled product (IV-9). A subsequent stereoselective epoxidation / cyclization of the

bis-homoallylic alcohol IV-920 should install the 2,5 di-Substituted THF ring to complete

assembly of the bis-THF core unit.

0 Zn or E122n O

/‘\Q/u\/\/\=/\/l % flow an

IV-6

 

IV-7 (Y = l or a second equiv.
of the alkyl group)

 

 

 

 

TBSO OH O . .

lV-8 O epoxrdatlon
-—- W,“ _____.
OR ,_ cyclization

T880 0

TBSO OH HO O... H

’ores

. O
OH 13038 IV-8; Fl = TBDPS
lV-tO

Figure IV-3: Design of the new synthetic strategy

222

The requisite primary iodide IV-6 was readily obtained from the commercially

available ethyl 6-hydroxyhexanoate IV-ll following a three-step sequence (Scheme

1. KHMDS O

2. TMSCI PPh3, imid.
PhaRMOH Bl" —= A0WOH
4, |v.12 12. toluene
3. AcOH : THF : Hgo ”'13 50%
(6 :3 :1),0 °C, 79%

 

 

 

O o 0
WI PCC I NaOAc
/\O —- ”(D/W0 ¢ AOJKAA/OH
CH202, rt
lV-6 M12 72% lV-11

 

 

 

 

Scheme IV-l: Synthesis of the model iodide

lV-l). PCC oxidation of IV-ll (72%) delivered the aldehyde IV-12. Witti g olefination of

IV-12 with 3-hydroxy-pr0pyltriphenylphosphonium bromide was carried out by in situ

2122 After treatment of the

TMS protection of the ylide prior to addition of the aldehyde.
reaction mixture with aqueous acid in the same pot, cis homoallylic alcohol IV-13 was

obtained in >95% diastereoselectivity. Finally, PPh3 / l2 mediated iodination of IV-13

produced the desired iodide 1V-6.23

Organozincs are known to undergo nucleophilic addition to aliphatic aldehydes in
the presence of Lewis acids or transition metal activatorsz‘l’27 First, iodide IV-6 was
treated with activated metallic zinc to generate the organozinc iodide intermediate IV-l4,
which was then reacted with aldehyde IV-8 that had been pre-complexed with BF3°OEt2

(Scheme IV-2). Although IV-8 was usually recovered unchanged, iodide IV-6 was
always completely consumed (as indicated by GC and TLC analysis). Based on this as

well as our previous experience with alkylzinc reagents (Chapter 111), we think that the

223

desired alkylzinc iodide (IV-14) was generated but probably was not reactive enough to

add to the activated aldehyde.

 

of t d 2
lV-8

 

40 °C, 6 h
IV-G lV-‘14 BF3'OE12
13122n (5 eq.) lV-8 “330 O” 0
4, —’__—————><—. R0 0,. __ o/\
Cul, 50 °C TlCl4, ~78 °C .
17 h "ores [v-15

Scheme IV-2: Attempted organozinc additions to aldehyde IV-8

Since dialkylzinc reagents are known to be more reactive than alkylzinc halides.
we next attempted to generate the dialkyl zinc species from IV-6. Thus, IV-6 was treated
with EtZZn and catalytic CuI to obtain the corresponding dialkyl zinc via zinc-halogen

exchange.”29 However, when aldehyde IV-8 was added to the dialkyl zinc reagent in the

presence of TiCl4, no desired secondary alcohol (IV-15) was obtained. In all of our

attempts, a part of the starting material (iodide IV-6) was always recovered unchanged
indicating that the exchange process remained incomplete.
In addition, a number of operational difficulties were encountered. First, the

process30 calls for the use of neat EtZZn which is extremely flammable when exposed to

atmosphere. Therefore all the operations needed to be carried out in a dry box. Secondly,
this protocol typically uses excess EtZZn to drive the equilibrium towards the product

side. The excess reagent then has to be carefully and completely removed under vacuum
before treatment with the aldehyde, in order to avoid competing ethyl group transfer. We
felt that such a procedure would be cumbersome and unsafe especially on multi-gram

scales. Hence the zinc-halogen exchange route was abandoned.

224

In absence of Lewis acids or transition metal catalysts, aldehydes are not reactive
towards alkyl zincs. On the other hand, the coupling of acid chlorides with organozinc
reagents is much more efficient, and requires no further activation. Therefore we

redirected our attention toward using the corresponding acid chloride as the electrophile
(IV-17, Scheme lV-3). IV-l7 was prepared by NaClO2 / NaHZPO4 mediated oxidation3 1
of IV-8 to the corresponding acid (IV-18, 80%) and its subsequent treatment with
(COCl)2 / DMF.32 Gratifyingly, the primary iodide (IV-6) derived alkyl Zinc, after
transmetallation to the corresponding organocopper species, reacted with acid chloride

IV-17 to afford ketone IV-18 in 60% yield (Scheme IV-3). We anticipated that

stereoselective carbonyl reduction of IV-18 would generate the desired threo 01-

tetrahydrofuranyl secondary alcohol (IV-19).

reso o NaH p0 7630 o T880 0
R0 0". 2 4 _ 0.. (COC1)2, DMF _ o.._
H v ' OH ~ 0'
NaClOz, ‘BuOH op hexaneS. rt on

 

 

 

 

bores 30% bras quant bTBS
lV-8; R = TBDPS IV-16 IV-17
1. Zn, 40 °C, 6 h “380 O O O
“(,5 ¢ W0“
2. CuCN-21.10 OR ,
3. lV-17 toms
0 °C, 60% lV-18

stereoselective reduction
of the ketone

TBSO OH O
04, __ O/\

OH

OTBS lV-1 9

Scheme IV-3: Synthesis of ketone IV-18 via organozinc addition to acid chloride IV-l7

Along Similar lines, we also tried to access epoxy ketone IV-23 (Scheme IV-4) by
addition of the epoxy iodide (IV-20) derived organozinc reagent to acid chloride lV- 17.

If successful, this strategy would bypass the proposed stereoselective epoxidation /

225

cyclization sequence (Figure IV-3) to install the second (C8-C12) THF ring of mucoxin,
thereby making the synthesis more convergent. Stereoselective ketone reduction and in
situ cyclization of IV-24 would directly afford bis-THF unit IV-10 (Figure IV-3).
However, in the attempted coupling of IV-20 with IV-17, the crude product did not Show
any diagnostic 1H N MR peaks corresponding to the desired product IV-23 (for example,
the epoxy methines or a-keto methylene protons). Instead, unusual upfield signals
belonging to a cyclopropyl ring were observed. Although epoxides are known to be
compatible with organozincs and the corresponding organocopper reagents, we surmised
that juxtaposition of the two functionalities in IV-21, might trigger an internal

rearrangement to produce cyclopropyl alkoxide IV—22.

 

 

 

O 1. mCPBA 0 1. Zn, 40 °C, 6 h
/\ WOH *7 /\ WI *
O 2. PPh;,, 12 O O 2. CuCN-21.10
imid., toluene 3. lV-17
M13 50% IV-20
o 1860 o o
o
/\O/U\/\/\_/\/ an . W0“
oyU on ., o
’0188
M21 0 IV-23
L 7 47 /\o
0an
lV-22

Scheme IV-4: Attempted addition of epoxy iodide IV-20 to acid chloride IV-17 via the

the organozinc reagent

With ketone IV-18 (Scheme IV-3) in hand, we focused our attention on its
stereoselective reduction. a-Oxygenated ketones, by appropriate choice of the hydride
source and nature of the oxygen substituent, can be reduced to the corresponding erythro

or threo a-oxy alcohols. Using metal hydrides such as LiA1H4, NaBH4 and ZnBH4,

erythro products can be obtained, provided that an a—oxygen is available for chelation (as

226

in or-hydroxy ketones, a-keto lactones and a-keto tetrahydrofurans, etc.). These reactions
occur via a chelation controlled transition state.”35 On the other hand, bulky, non-

chelating metal hydrides such as L- and K-selectride afford the corresponding threo

products through a Felkin-Anh transition State, irrespective of the nature of the ct-oxygen
substituent in the parent ketone.”38 Hydride reduction of ketone IV-18 following these

two routes is shown in Figure IV-4. In our case, the desired threo isomer (IV-19) would

be obtained via a Felkin-Anh transition state IV-26.

  

 

M OTBS OTBS
—-—. Y E ‘— O/\
"H" Y H OH ”
ores OTBS ores
lV-24
Chelation controlled T.S. IV 25
OTBS OTBS TBSO OH O
TBSO HO O H - OR 0., _ o’\
RO —* H E
H Y OR 3’
“H“ OTBS OTBS

Felkln-Anh T.S.

 

O
_ WA
Y_'|.1 4 o

 

 

Figure IV-4: Chelation controlled vs. Felkin-Anh transition state for reduction of ketone

IV-18

In preliminary studies, we found that NaBH4 reduction of ketone IV-18 produced

alcohol IV-27 quantitatively, but in poor diastereoselectivity (dr = 3 : 2, Scheme IV-S).
Also, the diastereomers were not separable by column chromatography. On the other
hand, exposure of IV-18 to the selectrides (Scheme IV-5) resulted in complete
consumption of the starting material, but no desired product could be isolated. It appeared
that the hydride transfer step was occurring and the intermediate borinate was being

produced. However oxidation of the borinate species to free the alcohol product appeared

227

T880 0 O

 

o __ o/\
on
OTBS lV-18

NaBH4, EtOH _ K-selectride

0 °C, rt, quant L-selectride

dr = 3 : 2
TBSO OH O TBSO OH O

O/,_ __ O/\ O’.‘ _,__,. OK
on .3 OR .,’
T
OTBS M27 0 BS 1V-19

Scheme IV-5: Attempted hydride reduction reactions of ketone IV-18

to be the problematic Step. Overall, the stereoselective ketone reduction approach proved
synthetically unviable.

Concurrent with the organozinc addition approach, another strategy involving a
HWE olefination reaction39 to couple the right and the left hand portions of mucoxin was

also examined. For this purpose, aldehyde IV-8 was further functionalized to generate
B-keto phosphonate IV-29 (Scheme IV-6). Addition of the anion of diethyl

methylphosphonate furnished B-hydroxyphosphonate IV-28, which was oxidized to

B-ketophosphonate 1V-29 in 83% yield with the Dess—Martin periodinane.40 The

 

 

 

 

   

 

 

 

TBSO OH O T880 0 O
M PO OE 1.nBuLi 0... Ploenz DMP 04. Ploala
e t ,
( )2 2.1V-8,-78°C 0'“ CHzCI2.rt OR
970/0 OTBS 83°/o OTBS
lV-28 lV-29
T880 0 O O O
KHMDS -. .~ O/\ HWOA
.. O _
IV-30 0“ "01133 0 IV 30
lV-31

Scheme IV-6: Model studies on HWE olefination approach

228

aldehyde partner IV-30 was similarly prepared by oxidation of the corresponding epoxy
alcohol (not shown), which was available from earlier studies (Chapter 11]).
Unfortunately, the intended HWE olefination to acquire 01,13-unsaturated ketone IV-31
was unsuccessful. When NaH was used as the base, the starting materials were recovered

unchanged. Furthermore, the use of KHMDS as the base afforded an intractable mixture,
containing none of the desired enone — discerned from the 1H NMR spectrum of the cure

product.

2. Conventional organometallic addition using chelation control to couple the two

halves of mucoxin

In view of the failed coupling strategies described above, we decided to move
away from our original plan of combining the two halves of mucoxin in fully
functionalized forms. In search of more a straightforward, workable route while still
keeping the synthesis concise, we came up with the following design (Figure IV-5).
Chelation controlled addition of an organomagnesium or lithium reagent of general
structure IV-32 to aldehyde IV-8 should produce the bis-homoallylic alcohol IV-33.
Further, one pot stereoselective epoxidation / cyclization‘ of IV-33 would furnish
bis-THF containing compound IV-34. Finally, the primary iodide derived from IV-34
would be coupled to the (it-bromomethyl lactone IV-35 via formation of the
corresponding organozinc. From our earlier experience (Chapter III) and literature

reports,“ we anticipated that alkylzinc iodides would couple efficiently with

bromomethyl acrylate type substrates such as IV-35.

" Several possible methods for stereoselective epoxidation of IV-33 are discussed later in
the same section.

229

T880 0 TBSO OH

 

 

 

 

 

 

 

 

HO O Chelation R0 0” op
H + M W0 ' “'- 3
, controlled .,
OTBS addition Ores
lV-8 lV-32 lV-33
stereoselective TBS? OH 1. 1310190th0 (1")
epoxidation / ' O.,' 0., Op 2. deprotect P
cyclization OR "-OTBS 3. iodination O
lV-34 4. couple lV-35 55,/wt%)
' lV-35

 

Ores

lV-36
Figure lV-S: Revised stepwise strategy to assemble fragments IV-8, IV-32 and IV-35

As part of the revised synthetic strategy, our immediate goal was to optimize the
chelation controlled addition of an appropriate organometallic reagent (IV-32) to
aldehyde IV-8. We began by synthesizing suitable homoallylic halide precursors that
would allow access to the corresponding organometallic reagent. Commercially available
1,5-pentanediol after mono protection and PCC oxidation afforded aldehyde IV-40 in

66% overall yield (Scheme IV-8). Z—selective Wittig olefination of IV-40 by way of

 

 

 

 

 

1 NaH PCC
2. TlPSCl or _ . = CHzclg, rt
lV-37 TBDPSCl 13,33: 2 = 155,33 82% M40
THF, BOO/o '
1. KHMDS
2. TMSCl
* _ 4 /\/==\/\/\
P03P/\/\OH Br v HO OR
4. M40 1. MsCl, EtaN
W41 3. AcOH :THF : H20 IV-42 CH2C12, 0 °c
(6 13 I 11.0 °C. 79% 2. Nal, acetone
reflux, 77%
PPh3, imid.
Br/VWOTBDPS #08 CH 0 * lW/‘ortps
1'
M43 . 4' ,2 2 M44; R :TBDPS
0 c, 35 /° lv-45; R = TIPS

Scheme IV-7: Synthesis of the requisite homoallylic halides

230

in situ TMS protection of the ylide derived from 3—hydroxypropyltriphenylphosphoni um

bromide as described earlier (Scheme IV-l),2 1'42 delivered homoallylic alcohol IV-42.

Bromide IV-43 was prepared by bromination43 of the alcohol using PPh3 / CBr4, while

iodides IV-44 and IV-4S were accessed via displacement of the corresponding mesylates

by NaI.44

The chelation controlled addition of Grignard reagents“48 derived from halides
IV-43, IV-44 and IV-45 to aldehyde IV-8 required some optimization. These studies are
summarized in Table IV-l.

Initially, the formation of the Grignard reagent proved to be tricky. When iodide

IV-45 (entry 1) was treated with activated Mg in refluxing diethyl ether3’4 and aldehyde

IV-8 subsequently added (entries 1 and 2),49 no addition products were obtained. Under
these conditions, dienes IV-46a and IV-46b were obtained as major products along with
the reduced product IV-46c. We surmised that the enhanced reactivity of the allylic
iodide might be responsible for the relative facility of the competing fi-elimination and

homo-coupling pathways.

\ __ _
WOT-”38 TIPSOWOTIPS WOTWS
1V-46a lV-46b lV-46c

Accordingly, when bromide lV-43 was subjected to the same reaction conditions (entry
3), the desired product was obtained in 30% yield along with a significant amount of

recovered aldehyde. Notably, when the alkylmagnesium iodide (entry 4) was generated at

low temperature via lithium—halogen exchange (‘BuLi, -—90 °C) followed by

231

transmetallation (MgBrZOOEtZ, —78 °C to 0 °C)50‘51, no elimination or homo coupled

 

 

 

 

 

 

 

products
x’\/::\/\/\orq1 [v.3 TBSO OH
IV-43‘ x = Br R = TBDPS ~ - ROWOTBDPS
M44} x = l, H, 1=T1313Ps °°ndmons
lV-45; X = l, R, = TIPS OTBS [v.47
Additives for
entry halide conditions pre- . result
complexation
of IV-8
1 was Mg, EtzO, reflux an’Z 1“ 1V-8 +1v-46a-ca
THF
2 IV-45 Mg, EtZO, reflux no additive IV-8 + IV-46a-c“1
‘ . . 1v-47 (30%) + IV-8
3 IV-43 Mg, EtzO, reflux no additive (60%)
4 [v.45 ‘BuLi, MgBr2°OEt2b, Eco Mites-013,b mm + IV-8
5 1V-44 'BuLi, MgBrz-OE12°,THF MgBrz-OEtzc IV-47 (67%, dr = 4: 1)
6 IV-44 ‘BuLi, MgBrZOOEtZC, EtZO MgBr20OEt2c IV-47 (74%, dr = 9: 1)
L

 

 

 

 

 

 

 

Table IV-l: Optimization of the chelation controlled addition
("1 all the products were detected by GC-MS; b commercially available; c freshly prepared)
were observed. However, aldehyde IV-8 remained unreacted, while the reduced product
IV-46c was detected suggesting not only that metallation had occurred, but also that we
were successful in suppressing the unwanted side reactions of the Grignard reagent.

MgBrz°OEt2 is known to be highly moisture sensitive,52 and it is likely that when

232

commercially available solid MgBrz-OEt2 contained enough moisture to quench the

metallated species (entry 4). Indeed, when freshly preparedsz’53 MgBr2°OEt2 was used

(entry 5), the yield of the desired adduct jumped to 67% (dr 2: 4 : 1). The yield and

diastereoselectivity were further improved (entry 6) when EtzO was used as a solvent

instead of THF.

Before further investigations began, homoallylic alcohol IV-SO bearing a terminal
PMB ether (instead of silyl ether IV-47) was synthesized (Scheme IV-8) in order to
facilitate protecting group manipulations. Iodide IV-48 was obtained following a Similar
sequence as before (Scheme IV-7). By carefully controlling the temperature and amount

of MgBr2-OEt2, adduct lV-50 was obtained in 88% yield as a single diastereomer after

chromatographic purification.

 

 

tBuLi, -100 °C IV-8, MgBr2-0E12
lN—WOPMB 47 BVMQWOPMB :
was then MgBrzOEtZ M49 -40 C. 88 4. (it > 9 .1
-90 °C to 0 °C
TBSO OH
R0 0... _ OPMB
3
"'ores lV-50

Scheme IV-8: Synthesis of bis-homoallylic alcohol IV-50

As the first step toward investigations on the proposed stereoselective
epoxidation-cyclization sequence to install the (Ii-substituted THF ring (Figure IV-5), bis-
homoallylic alcohol IV-50 was subjected to mCPBA mediated epoxidation (Scheme 1V-
9). As expected, no diastereoselectivity was observed and after treatment of the reaction
mixture with glacial AcOH in the same pot, the bis-THF unit IV-Sl was obtained as an

inseparable mixture of diastereomers (ca. 1: 1). Since IV-51 was easily synthesized, we

233

reso OH reso OH

 

 

 

mCPBA, CH2012 O
HO 0,. __ 3 OPMB g 0., OPMB
, then AcOH op; .,
OTBS 85% bras
1V-50; R = TBDPS lV-51
TBSO OTBS OPMB reso ores
TBSOTf ’ O.,_ O DDQ i o,“ o OH
2,6 lutidine OR -, 0112012 ; H20 OH
CH2012. 89% OTBS (9 : 1), quant OTBS
lV-52 IV-53
ppha' imid. TBS? OTBS Zn, then TBSO ores O

 

 

W CuCN-2L1Cl V _ W

i I [<7 ' O

12, toluene OR ._ 0 OR .‘ 4

84% Ores [Br/#0 Ores l
lV-54 M lV-55

Scheme IV-9: Feasibility studies of the new strategy described in Figure IV-5

decided to test the viability of further transformations in our proposed synthetic plan
(Figure IV-5, Scheme IV-9). TBS protection of IV-51 to produce tris—TBS ether IV-52
(89%) and subsequent PMB deprotection of lV-52 to reveal the primary alcohol IV-53
proceeded smoothly. Iodination of IV-53 secured the target iodide IV-54 in 84% yield.
However, our preliminary attempts toward organozinc mediated coupling of the iodide
with the bromomethyl acrylate were unsuccessful.

At this point, the two issues that needed to be addressed were stereoselective
epoxidation of bis—homoallylic alcohol IV-50 and the final coupling of iodide IV-54 with
the bromomethyl acrylate. Several methods for the stereoselective epoxidation reaction
were considered. The most commonly used tactic for the conversion of stereodefined bis-

homoallylic alcohols to the corresponding THF units is a one pot, hydroxyl directed
VO(acac)2 / 'BuOOH mediated epoxidation / cyclization reaction.20 Transition metal

catalyzed, tert-butyl peroxide mediated epoxidation of olefins was first reported by

234

Indictor and Brill.54 Among various transition metal catalyzed epoxidations, vanadium-

catalyzed hydroxyl directed epoxidation of alkenols has been used most commonly in

‘BuOOH ------------- - ‘BuOH
HOV><‘O/\<é
is” o 8:00 Bu
Ho l\O ‘0
C Q’Yloy

(tyBU
ROI, 1\

slow 13,2510)
8

Figure IV-6: Sharpless’ mechanism for vanadium catalyzed epoxidation of allylic

alcohols

organic synthesis. The first mechanistic proposal for VO(acac)2 catalyzed epoxidation of

allylic alcohols was put forth by Sharpless and co-workers (Figure IV-6).55 After initial

oxidation and ligand exchange at the metal center (A), the peroxide is activated by
bidentate coordination to vanadium (B). The subsequent rate-determining step (C)
involves oxygen transfer to the olefin.

This mechanism has been extended to construct working transition state models to

explain the observed diastereoselectivities in epoxidation of various allylic, homoallylic,
bis- and tris-homoallylic alcohols.20 In particular, such a transition state model for
secondary bis-homoallylic alcohols containing trisubstituted olefins was originally
proposed by Kishi.56 A representative example from Kishi’s studies is Shown in Figure

IV-7. During epoxidation / cyclization of trisubstituted alkenol IV-56, THF IV-58 was

235

produced as the major diastereomer via intermediacy of epoxide (IV-57). To explain the

facial selectivity of the olefin epoxidation, two transition states A and B were invoked.

Irrespective of the nature of R and R’, A is the lower energy TS since the iPr group

R Et H
R 1. VO(acac)2/ R. R \ R a \/\v
R. tBuOOH A R' Pr H r Ito
i o" | ' WW. 0 El O’V Rik}
P’ OH 2_ AcOH OH A B transannular
lV-56 lV-58 interactions

1 a 1 ll 11
Rm major epoxide minor epoxide
OH

lpp‘“ R,R'=H;AIB=921
R=Me,R'=H;A:B=6:1
R=H,R'=Me;A:B>20:1

 

lV-57

 

Figure IV-7: Kishi’s transition state analysis to explain the diastereoselectivity observed

in directed epoxidation of bis-homoallylic alcohols

occupies an ‘outside’ position whereas B, due to the proximity of the iPr and Er groups

suffers from transannular interactions. In the absence of any substitution at the or carbon
(R, R’ = H) a 9:1 selectivity in favor of IV-57 was obtained. When R = Me and R’ = H,
the selectivity was lowered due to additional 1,3 diaxial interactions (of R and Er) in A.
Finally, when the configuration at the a-carbon is switched (R = H and R’ = Me), B is
highly disfavored due to the 1,3 diaxial interactions (of R’ and E1) in addition to the
preexisting transannular interactions.

Applying a similar model to our secondary bis-homoallylic alcohol (IV-50), two
transition states, A and B (Figure IV-8) can be drawn. Due to the cis-1,2 substitution
pattern of the olefin, transition state A suffers from steric interactions between X and the

incoming electrophilic oxygen. B, though devoid of such steric compression, experiences

an allylic AL3 strain57 between X and the axial hydrogen. From this analysis, the relative

236

preference for the two transition states was not readily apparent. Moreover, a brief
literature search revealed that high diastereoselectivities for VO(acac)2 promoted directed

epoxidations of secondary bis- homoallylic alcohols have been observed only in the case

 

 

 

 

 

TBSO OH OH
HO O... __ OPMB __ Ok/E/x
, 3 "7 reso
’OTes IV-50 O :90 O... ‘1.
H )( Ores
x OPMB
.. a. cacao. «W
H \ H 2N 7*
V x '11 ‘BUOOH
O’ \K
A 3

Figure IV-8: Application of Kishi’s T.S. models to bis-homoallylic alcohol IV-50

of trisubstituted olefins.

Also, in our hands, preliminary trials to epoxidize IV-SO using VO(acac)2 /

tBuOOH were not successful. Under several different conditions (ranging from ambient

temperature to 80 °C), no epoxide product was ever observed. This indicated that olefin
IV-50 was inherently unreactive towards epoxidation under these conditions. Even if this
type of epoxidation were successful, the Strategy suffers from an inherent deficiency. The
stereoselectivity of the epoxidation would be derived from the substrate (existing carbinol
stereocenter) rather than from the reagents. Thus, diastereomeric THFS that would result
from cyclization of the opposite epoxide stereoisomer would be difficult. Since we were
aiming to establish a versatile synthesis of mucoxin, that would allow access to unnatural

stereoisomers, the directed metal catalyzed strategy was not pursued further.

237

Among other protocols for the asymmetric epoxidation of unfunctionalized
olefins, are methods developed by Shi and Jacobsen / Katsuki.58 Recently, Shi and co—

workers have developed a new chiral ketone catalyst (IV-59, Figure IV—9) for

 

 

 

.4? 0 cat. IV-59 o
\ cat. lV-59 . Q j J
oxone 87% O OXOHG. 61% o
910/0 e’e 97°/o ee 0
lV-60 lV-61 lV-54 (+)-IV-65
cat. lV-59 .oQ O

 

/ \ / t A
/ O
Ph/W oxone. 82% Ph/\l d [“803
91 °/ '
[v-52 ° ee IV-63

0‘" :- ores
)fo lV-59

 

 

 

 

Figure IV-9: Representative examples of Shi asymmetric epoxidation of cis olefins

asymmetric epoxidations of cis- and terminal olefins.’ Although the corresponding
oxiranes were obtained in high enantiOpurities and complete diastereospecificity, a major
limitation of this method is only conjugated olefins or olefins bearing an adjacent acetal
functionality (for example, IV-60, IV-62 and IV-64) are optimal substrates. In case of
alkyl substituted olefins (only one example reported) ca. 65% ee was obtained.(’1
Moreover, no further data on the diastereoselectivity of such unconjugated olefins is
available.

In 1990, Jacobsen62 and Katsuki63 independently reported asymmetric
epoxidation of unfunctionalized olefins using Mn—salen complexes as chiral catalysts.
Although 1,2 di-substituted cis-olefins produced the corresponding epoxides in high

enantiopurities, as in Shi epoxidations, the optimum results were obtained only for

 

’ The earlier ketone catalysts proved to be highly enantioselective only for trans and trisubstituted

olefins.”6O

238

conjugated and acetal containing olefins.58 In addition, during epoxidation, the

diastereomeric purity of the starting olefin was lost. For example, cis-[fl—methylstyrene

(IV-66, R = Me, Figure IV-IO) produced a mixture of the corresponding cis- and trans

 

 

r— _
R
Ph Ph R
I
Mn” 0
O lV-67
Ph __R + an __.
lV-66 l
R Ph
' o .... ._ ,.
Ph Mn” ’0‘ R
— .._l
lV-68

Figure IV-10: Proposed radical intermediate during oxygen transfer step in Jacobsen
epoxidation
epoxides. It is believed that a radical intermediate is involved during the oxygen transfer,
which undergoes bond rotation to favor formation of the trans epoxide (IV-68).63 Taken

together, none of the abovevmentioned epoxidation protocols appeared feasible for use in
our system.

In this context, were also aware of Sharpless’ method for the stereospecific
conversion of 1,2-diols to epoxides (Figure IV—II).64 Thus, vicinal diol IV-69 is first
converted to the corresponding ortho acetate (IV-70), which when treated with an acyl or
TMS halide, leads to formation of regioisomeric acetoxy halides (IV-72 and IV-73) via

the intermediacy of acetoxonium ion IV-71. Upon basic hydrolysis, the halohydrin esters

239

OMe
OH MeC(OMe)3 >< AcX or ego

 

 

 

R 92 o o
1 OH cat. PPTS REE MeasiX REE
IV-69 ”'70 lV-71
OAc X
K CO ?
R1/<?’R2 2 3 R‘/'\;-/R2 + 911sz
MeOH X OAc
X = Cl, Br, I
lV-74 lV-72 lV-73

Figure IV-ll: Sharpless’ protocol for stereospecific conversion of vicinal diols into

epoxides

undergo intramolecular halide displacement to generate the corresponding epoxide
(IV-74). After each step, the corresponding intermediate is isolated simply by
evaporation of the volatiles and the epoxides are obtained in 82% — 97% yield with
complete retention of configuration at both the vicinal carbinol centers. Formation of
both acetoxy halohydrin IV-72 and IV-73 involves inversion at one of the original diol
stereocenters, which undergoes another inversion during epoxide ring formation. Thus
the regioseletivity in acetoxy halohydrin formation is immaterial and the final epoxide is
obtained with complete stereochemical fidelity.

We felt that this type of stereoselective epoxide formation was suitable in our
synthesis for several reasons. First, asymmetric dihydroxylation unlike asymmetric
epoxidation reactions, does not require any specific structural elements in the parent
olefin and thus is a much more general way to oxidatively functionalize olefins.
Secondly, the dihydroxylation process has been optimized for a variety of olefins with

different substitution patterns to obtain the corresponding diols in high enantioselectivity

65.66

and yields. In particular, for cis 1,2-di-substituted olefins (which is the substitution

240

pattern of our substrate IV-50), with appropriate choice of ligands, upto 90% ee has been
obtained. Also, all four stereoisomeric epoxides are accessible from the appropriate diol
precursors, which in turn are easily available simply by permutations of the olefin
geometry and both antipodes of the chiral ligand. This would allow stereochemical
diversity in our synthetic scheme to efficiently access the unnatural isomers of the natural
product.

Since our ultimate goal was to construct the bis-THF unit IV-Sl (Scheme IV-9)

starting from bis-homoallylic alcohol IV-SO (whether or not via an epoxide intermediate),

1. MeC(OMe)3

 

OA
“0 0“ cat. PPTS 0&0 o\/kc
R 2. acid “0 n
1. deprotect P
lV-76 IV-77 lV-78 2' cycm
SAD
1. MeC(OMe)3

HO OH

H _\ 1. protection (P) w cat. PPTS m
¢ p = P
U— R 2. SAD R 2. AcX R

3. K2003 I MeOH
lV-75 lV-79 IV-80

 

 

Figure IV-12: Proposed one pot cyclization of triols (IV-76) to the corresponding cyclic

hydroxy ethers (IV-78)

based on Sharpless’ proposed mechanism for diol to epoxide conversion?4 we put forth

the following proposal (Figure IV- 12).

An acetoxonium ion containing an appropriately positioned hydroxyl group (IV-
77) may be intramolecularly trapped by the hydroxyl nucleophile. This event should
result in formation of the corresponding cyclic ether unit defined by the general
representation IV-78. We anticipated that in a triol system such as IV-76, the 1,2 ortho

acetate would be preferentially generated leaving the isolated hydroxyl free for

241

nucleophilic attack. Also, the acid used for generation of the acetoxonium intermediate
preferably should not contain a good nucleophile unlike in Sharpless’ protocol (Figure
IV—l 1), which might compete to trap the cation. Certainly, one might predict that even in
presence of an external nucleophile, intramolecular trapping of the acetoxonium ion

would be faster. In the ring closure of medium sized (5-7) cyclic ethers, an exo-let mode
is generally favored over an endo-tet according to Baldwin’s rules.67 Accordingly we

anticipated that cyclization of hydroxy olefin IV-50 (Scheme IV—9), following this route,
should lead to the required 2,5 (Ii—substituted THF ring (IV-51) bearing an adjacent
secondary carbinol on the side chain. If successful, this strategy (from now on referred to
as 1,2,n triol cyclization) would offer a quick and efficient entry to Cyclic ethers of type
IV-78 (Figure IV-12) starting from alkenols such as IV-75 in two steps, namely,
Sharpless asymmetric dihydroxylation and a one—pot triol cyclization. Clearly, the
alternative route involving pre-formation an epoxide (IV-80, Figure IV-12) would be
lengthy and less efficient since it would call for additional protection / deprotection steps.
In addition this sequence would obviate the need for a hydroxyl (or other) functionality in
the substrate to direct the epoxidation, allowing us to generate stereoisomeric analogs of
mucoxin as described in Chapter III.

Our immediate goal now, was to test the feasibility of the proposal. We decided to
use a simplified model triol IV-82 (Scheme IV-IO) for this purpose. IV-82 would also
serve to test the compatibility of the PMB protecting group (which was present in the real
substrate IV-50) with the cyclization conditions. The triol was obtained by

dihydroxylation of the bis-homoallylic alcohol (IV-81), which in turn was prepared using

242

Grignard addition of the available iodide (IV-48) to cyclohexane carboxaldehyde. The

stage was now set to attempt the proposed triol cyclization reaction.

 

 

'BuLi, -100 °C OH AD mix-a
MgBr2°OEt2 __ OPMB MeSOQNHg
IM‘WOPMB _; >
lV-48 Cy-CHO “1.31 ‘BuOH :H20
-40 °C, 70% (1 :1 ), 80%
OH 1.MeC(OMe)3 / PPTS, rt OAc

 

Wows CH2Cl2 i Won
4
HO OH 2. BF3‘OEt2(10mOl°/o)

-30 °C to 0 °C, (3H2c3I2
lV-82 IV-83; Ft = OPMB (40%)
IV-84; R = H (40%)

Scheme IV-10: One pot cyclization of a model triol IV-82

From the outset, BF3°OEt2 was chosen as the acid promoter as it is an effective
oxygen-coordinating Lewis acid in epoxide activations. After treatment of IV-82 with
trimethyl orthoacetate and catalytic PPTS, rapid consumption of the starting material was

accompanied by appearance of two new spots on TLC at higher Rf values. Volatiles were
then evaporated and the crude product was exposed to BF3°OEt2 (10 mol%, —30 °C) in

CHzClz. Upon warming to 0 °C, the reaction was quenched and the purification of the

crude material afforded two products IV-83 and IV-84 (each as a mixture of
diastereomers) in 80% overall yield. No other regioisomeric cyclic products were
detected. Although PMB deprotection under the reaction conditions could not be
prevented, we were pleased to obtain the desired cyclized products. Later, we also found

that isolation of the ortho acetate intermediate was not necessary and similar yields of
IV-83 and IV-84 were obtained by addition BF3°OEt2 in the same pot. Thus, the triol
cyclization, as proposed, was efficiently accomplished in a two-step one-pot procedure.

Further studies to improve functional group compatibility of the reaction and to expand

243

its scope to access a variety of heterocycles have been undertaken by another graduate

student in our laboratories.

B. Completion of the total synthesis of the proposed structure of mucoxin

Encouraged by the model studies, our next goal was to test the applicability of the
triol cyclization with a bis-homoallylic alcohol such as IV-SO. Since IV-50 was also a
model system derived from a model aldehyde IV-8 (Scheme IV—8), we first decided to
synthesize the real trisubstituted THF containing bis-homoallylic alcohol IV-85 (Figure

IV-I3), which would be used for completion of the total synthesis.

TBSO OH TBS? O
? ‘ O,
OI'- __ 08” :3 a H + INA/OB”
15 4 15 4
bras OTBS
lV-85 IV-86 lV-87

Figure IV-13: Assembly of the real aldehyde (IV-86) and partially functionalized right

hand piece IV-87

The aldehyde precursor (IV-86) was available from earlier studies (Chapter III).
Chelation controlled addition of the Grignard reagent derived from iodide IV-87 to
aldehyde IV-86, should furnish the requisite substrate IV-85. We decided to use iodide

IV-87 — a slightly modified version of the previous iodide (IV-48, Scheme IV—9), for two
reasons. First, since the PMB protecting was found to be unstable to the BF3°OEt2
mediated triol cyclization reaction (Scheme IV-IO), it was replaced by a more robust
benzyl group‘s"68 Second, in view of our unsuccessful attempts to couple iodide IV-54

(Scheme IV-9) with (bromomethyl) acrylate, we decided to explore alternative ways

244

(vide infia) to install the terminal butenolide ring. This required the use of a nine carbon

iodide (IV-87) rather than the earlier eight carbon unit lV-48.

Our efforts began by synthesis of IV-87 (Scheme IV-l 1). Commercially available
1,6 hexanediol (IV-88) was transformed into aldehyde IV-90 via mono benzylation

(78%) followed by PCC oxidation (82%). Cis—selective Witti g olefination of IV-90 with
3-hydroxypropyltriphenylphosphonium bromide via in situ TMS protection of the yilde42

generated the homoallylic alcohol IV-91 (83%, > 10:1 diastereoselectivity).

Displacement of the mesylate obtained from IV-9l by NaI afforded the requisite iodide

in 77% yield.44 Chelation controlled addition involved first, generation of Grignard

 

 

 

 

H
NaH, BnBr PCC W
HO\/\/\/\OH HOWOBO O OBn
TBA1,THF CH Cl ,rt
lV-88 50 °C, 78% IV-89 82°21 2 IV-90
1. KHMDS
+ 2. TMSCI 08”
Ph PMOH Br’ = HOW
3 4.1v-9o 1. MsCl, 51311
3. AcOH :THF : H20 IV-91 CHQCIQ, 0 °C
(6 : 3 : 1), 0 °C, 83% 2. Nal, acetone
reflux, 77%
T88? 0H lV-86 ‘BuLi, -100 °C 08
‘ 0,. _ OBn 4 4r l/v—W n
is -. “ 31932255: MgBrz'OEtg IV-87
r 2 ' -
OTBS 850/0 E120
lV-85

Scheme IV-ll: Synthesis of the real bis-homoallylic alcohol (IV-85)

reagent from lV-87 by low temperature lithium-halogen exchange / transmetallation

sequence, followed by treatment with MgBrz-OEt2 pre—complexed aldehyde IV-86 at —40

°C. The adduct (IV-85) was obtained in 85% yield as a single diastereomer (> 20:1

selectivity based on 1H NMR of the crude product) after chromatographic purification.

245

 

 

With the desired bis-homoallylic alcohol IV-85 in hand, we now set out to
examine the triol cyclization reaction. This required first accessing the corresponding
triol using Sharpless asymmetric dihydroxylation reaction. According to the empirical

65.69

mnemonic device to predict enantioselecivity in the dihydroxylation reaction, the

southwest (SW) and the northeast (NE) quadrants are more open to accommodate the
olefinic substituents (Figure IV—l4). The SW quadrant is considered an attractive area for
soft, large and / or flat groups. Thus is it preferentially occupied by aryl and large alkyl
groups in that order. Moreover oxygen-containing groups have a lesser tendency to

occupy this position.70 This mnemonic is most reliable in case of monosubstituted and

trans 1,2 (Ii-substituted olefins. When an olefin is oriented according to the constraints,

AD-mix (3

 

 

NW 1 NE HO OH
Rs”) (”HM
RL H RL 58 \RM
1). f H
SW t SE HO OH
AD-mix a

Figure IV-14: Empirical mnemonic device for the asymmetric dihydroxylation reaction

AD—mix~a reacts from the bottom face.

While positioning olefin IV-85 according to the mnemonic, we reasoned that the
unbranched alkyl portion might preferentially occupy the SW corner. The highly
oxygenated THF ring containing substituent would then be placed in the SE area (Figure
IV-IS). IV-85, so oriented, would generate corresponding triol IV-94 when treated with

AD-mix-a. Subsequent triol cyclization involving inversion of configuration at the point

246

—- = W
W ~ OBn We}: 0

IV-85 lV-85

 

NW fl NE
_ TBSQ fl OH
C/xb BOW W08.
4 15 i 4
OBn HO OH OH
SW SE

 

Ill

bras

 

 

lV-94
TBSQ OAc
. - 0,, 01.. OBn cyclization
AD—mlx a 15 5 T
1v-93 bras
IV-95

Figure IV-lS: Application of the asymmetric dihydroxylation mnemonic to olefin IV-85

 

 

agérgxmi TBSQ OH OH 1. MeC(OMe)3 / PPTS (1 0 mol%)
2 2 7 0, CH Cl , rt 5 min
K20804°2H20 -. OH 2 BF - E 2 1%
o co. 88% ‘n 1350...? 5 "‘° ’
dr = 5 I 1 ’
lV-94

TBSQ OAc

   
 

+ TBS cleaved bis-THF products

bras

 

 

 

 

resort
(91%) 2,6 lutidine (8%)
CH2012
1. MeC(OMe)3, PPTS
F~ t rt
”94 B 3052, ¢
2. K2003 MeOH, rt
3. TBSOTf, 2,61utidine, 0 °C OTBS

lV-97
(90% over three steps)

Scheme IV-12: Application of triol cyclization method to the real system

of cyclization, should lead to bis-THF unit (IV-95) bearing correct configuration at all the

stereocenters.

247

Accordingly, asymmetric dihydroxylation of IV-85 with AD-mix-a afforded the
triol (IV-94, Scheme lV-12, only the major isomer shown) in 88% combined yield and
ca. 5:] diastereoselectivity. The diastereomers were easily separable by flash column
chromatography and the major isomer was isolated in 73% yield. Cis 1,2-di-substituted
olefins are known to be the most difficult class of substrates for the asymmetric

dihyroxylation. Use of new chiral ligands, viz., DHQD-IND and DHQ—IND has
significantly improved the enantioselectivites in certain substrates.71 In our case, since

the major diol (IV-94) was isolated in enantiopure form and in good yields no further
attempts were made to improve the diastereoselectivity by variation of the chiral ligands).
Armed with sufficient amounts of the trio] IV-94 we next investigated its

cyclization reaction (Scheme IV—12). Using our original conditions (Scheme IV-IO), i.e.,
10 mol% BF3-OEt2, -30 °C to 0 °C, IV—9S was obtained as the major (55%) product
along with 20% of a mixture of TBS deprotected bis-THF products. After some
experimentation, it was found that rapid addition of 25 mol% of BFyOEt2 at ambient

temperature and immediate (IO-15 min) quenching of the reaction maximized the yield of
the desired bis—THF (IV-95) up to 91%. Furthermore, reprotection on the small amount
of cyclized product that had lost one of the silyl groups afforded IV-9S in >95% yield.
Thus, under these optimized conditions, the triol cyclization of IV-94 proceeded almost
quantitatively to afford IV-9S as a single regio- and stereoisomer. Furthermore, we found
that triol IV-94 could be converted to fully protected bis-THF unit IV-97 following a
three-step sequence, viz., cyclization, acetate hydrolysis and TBS protection (Scheme IV-

12) in excellent yield without purifying any of the intermediates. Differentially protected

248

bis-THF IV-97 was suitable for further elaboration along the proposed synthetic scheme
(Figure IV—S).
The mnemonic for asymmetric dihydroxylation is not completely reliable to

predict facial selectivity of complex unknown olefins, particularly with a cis-1,2

65 .69

substitution pattern. Therefore, before proceeding further, we decided to

independently establish the absolute configuration of the vicinal diol generated via the

asymmetric dihydroxylation reaction of IV-85 (Scheme IV-12).

 

 

0,. __ OBn
- 12 4

lV-95
MeOH

 

 

l'bres lV-85

 

 

Scheme IV-l3: Chiral alcohols (IV-85 and IV-98) used in Mosher’s ester analysis

We planned to use Mosher’s ester analysis for this purpose.” The three free

hydroxyl groups in IV-94, being sterically similar would be hard to differentiate while
forming the Mosher’s monOester derivative. To simplify the derivatization and analysis
process, we decided to use cyclized product IV-98 (Scheme IV-13). lV-98 was prepared
by base hydrolysis of acetate IV-95. Mosher’s ester analysis of IV-98 would establish the
absolute configuration at C8 and indirectly that of C9 since both C8 and C9 carbinols
originated via dihyroxylation of cis olefin IV-85. Also, a similar Mosher’s ester analysis
of IV-85 (Scheme IV—13) would ascertain the configuration at C12. Finally, NOESY
experiments would to confirm the relative stereochemistry across the C9-C12.THF ring.
As per the plan, both, (S)— and (R)- a-methoxy-a-trifluoromethylphenylacetate

(MTPA) ester derivatives of IV-85 were synthesized (IV-99 and IV-100, Scheme lV-14).

249

The DCC / DMAP mediated coupling was most efficient when freshly prepared MTPA

chlorides were used.32

MeQ

 

 

 

 

Ph '- CF21
lV—85 I
MeO ph (000:) MGO‘Ph DCC / DMAP T859 0 O
,.‘ OH 2 _: -‘ Cl AV ' O”. 12 __ OBn
FSC 0 DMF F3C O CHZCI2, rt 15 ‘ 4
hexanes, rt 85% '1
(S-MPTA OTBS 1v-99
Ph
MeO ‘- CF 3
lV-85 I
P11 QMe (coon) Pn OMe occ / DMAP T839 0 O
_, OH 2 4' F C _- CI ; ' O 12 __ OBn
FSC 0 DMF 3 O CHQCIQ, rt 15 _ 4
hexanes, rt 87% '1
(FD‘MPTA OTBS lV-1OO

Scheme IV-l4: Synthesis of Mosher’s esters of IV-85

Table IV-2 shows esters lV-99 and IV-100 drawn (only relevant structural features

shown for clarity) in conformations proposed by Mosher‘ that explain the correlation

between observed 1H NMR chemical shifts and the absolute configuration of the parent

alcohol IV-85 at C12.72

 

lV-99, (S)-MTPA lV-100, (R)-MTPA
T880
0 = i 0,, ‘5‘ R = 1"" MOB“
W X = TBS
OTBS

—_

' Based on ORD and CD studies,73 it has been proposed that the electronegative CF3 group
eclipses the carbonyl group in the CD active conformation. No detailed explanation of this
conformational bias is provided.

250

 

 

 

 

 

 

 

 

 

 

proton chemical shift (6) in IV-99 chemical shift (6) in IV-100
H8 5.30 5.34
H9 5.22 5.30
H12 5.46 5.40
H 14 4.37 4.33
1116 4.32 4.25
“17 3.65 3.59

 

 

 

 

Table IV-2: Mosher’s ester analysis of IV-99 and IV-100

Also, listed in Table IV-2 are chemical shifts of protons relevant in determination of
the configuration. In IV-99, the olefin containing side chain is juxtaposed with the phenyl
group of the MTPA ester. Therefore, those protons fall within the shielding cone of the

phenyl group and are expected to shift upfield compared to the same protons in the other

diastereomer (IV-100). As can be seen in Table IV-2, H8 (5.30 6) and H9 (5.22 6) in

IV-99 are more upfield than H8 (5.34 6) and H9 (5.30 6) in IV-100. Similarly, the

trisubstituted THF ring in the (R)-MTPA derivative (IV-100), is shielded by the phenyl
group and all the oxygenated methines (H12, H14, H16 and H17) in that portion of the
molecule are shifted upfield compared to the corresponding protons in IV-99 (Table IV—
2). From this analysis, the stereocenter at C12 was established to be (S) which is also the
expected configuration based on a chelation controlled transition state.

Next, we attempted to determine the configuration of bis-THF IV-98 at C8
carbinol using the same technique. IV-98 was derivatized as S (IV-101) and R (IV-102)

251

MTPA esters, again via DCC / DMAP mediated coupling with appropriate acetyl

chlorides (Table IV-3). In IV-lOl, the alkyl side chain is shielded by the phenyl group of
the MTPA ester and hence is expected to show upfield 1H chemical shifts compared to
the same protons in IV-102. On the other hand, the bis-THF portion in IV-102, being in

the phenyl-shielding cone, would exhibit relatively upfield—shifted 1H signals than those

protons in IV-101. In both the derivatives, 1H NMRs signals of the short, five carbon

alkyl side chain overlapped with that of the T HF ring methylenes (C10, C11 and C15) as
well as the long, 17 carbon side chain on the other side. Therefore, the short alkyl chain
portion was not used for the analysis. As indicated in Table IV-3, all the oxygenated
methines (proton numbering corresponds to the carbon numbering in 1V-98) belonging to

the bis-THF portion in IV-102 are shifted upfield relative to those in IV-101 as

expected.‘
TBSO OH/'8
‘7? 6 01., 01., 8 OBn
15 13 5
14",
OTBS
IV-98
MTPA-Cl MTPA-Cl
from (S) aV W (H) acid
shielded
H3 0
RM... ,U\(Ci=3
H3 0 0 "‘OMe
R\U“ /U\(CF3 Ph
0 "'Ph .
OMe shielded
lV-101 lV-102
TBSQ
17'- 5 0,, 0,,
MM 0 s Wtf‘
14"OTBS

 

 

' Only H9 did not fit in the trend, possibly because it resided outside shielding cone of the phenyl
group.
252

 

 

 

 

 

 

 

 

 

 

proton chemical shift (6) in IV-101 chemical shift (6) in IV-102
H8 5.14 5.12
H9 404 4.08
H12 4.31 4.27
H,3 3.66 3.61
Hl4 4.22 4.16
Hl6 4.26 4.24
H17 3.71 3.70

 

 

 

 

Table IV-3: Mosher’s ester analysis of IV-101 and IV-102

Thus, the configuration of IV-98 at C8 was determined to be (5). Also, since asymmetric
dihydroxylation of a cis-olefin can in principle, produce only IR, 23 or IS, 2R diols, the
original configuration at C9 (in IV-94, Scheme IV-12) is expected to be (R). Since the
cyclization of IV-94 to produce bis-THF IV-95 (Scheme IV-12) involves inversion of
configuration at the point of cyclization, the configuration of C9 in IV-95 must be (S).

In order to further confirm our stereochemical assignment of IV-98, bis-THF IV-104
(Table IV-4), epimeric at C8 and C9 was similarly analyzed. Triol IV-103 was obtained
via asymmetric dihydroxylation of IV-85 (Scheme IV-l2) using AD-mix-B, which upon
cyclization and acetate deprotection furnished the bis-THF (IV-104, Table IV-4). The
corresponding (S) (IV-105) and (R) (IV-106) MTPA esters were accessed as before. As
expected the bis-THF portion of IV-l05 showed upfield 1H N MR shifts relative to that of

IV-106 (Table IV-4), which verified the (R) configuration at C8 (and hence again (R) at

253

C9 as discussed before) in IV-104. Furthermore, 1D NOESY experiments clearly showed
a strong nOe correlation between H9 and H12 (Figure IV-16) indicating a cis geometry
across the THF ring, whereas no nOe correlations were observed across the (Ii—substituted

THF ring in IV-lOl.

 

Figure IV-16: nOe correlations in IV-101 and lV-105 containing trans and cis di-

substituted THF rings respectively

The Mosher’s ester analysis taken together with the nOe correlations confirmed

that bis-THF lV-98 (produced as the major diastereomer), possessed the requisite relative

stereochemistry in C8-C12 portion. The minor diastereomer IV- 104 on the other hand,

 

 

TBSQ OH QH 1. MeC(OM e)3, PPTS Tsso QH/' R

W081} 2. BFB‘OE‘ZI '1 W080

' fi- 15 1
15 4 5
'2 OH 3. K CO , MeOH 14';
ores {93°}: ores
lV-103 lV-104
(FD-MTPA-Cl (S)-MTPA-Ci

H3 0
/U\(CF3
O 0 "'OMe
R P

h

Ushielded

lV-106

 

TBSQ

17’ 6 0,, 0
MW 0 E W“

14 3,
OTBS

254

 

 

 

 

 

 

 

 

 

proton chemical shift (6) in IV-l05 chemical shift (6) in IV-l06
H8 5.07 5.04
H9 4.02 4.07
H12 4.26 4.13
H 13 3.60 3.52
H 14 4.02 3.96
H16 4.30 4.28
H17 3.72 3.75

 

 

 

 

Table IV-4: Mosher’s ester analysis of IV-105 and IV-106

contained the undesired cis—di-substituted THF ring. Equipped with sufficient amounts of

the fully protected version (IV-97, Scheme IV—12) of the desired diastereomer we then

proceeded toward the final stages of the synthesis.

One of the tactics used to install the terminal butenolide in acetogenins, is
outlined in Scheme IV—15 (A).38‘74 a—Phenylthio lactone (IV-107) is alkylated to produce
a-di-substituted derivative IV-108. The thiophenyl group is then oxidized to the
corresponding sulfoxide, which upon heating undergoes syn-elimination to furnish the

corresponding internal a,{3—unsaturated lactone (IV-109).

255

 

O O O
PhS base 1.oxid tion R
A 0 “*7" R O a ’ \ 0
RN Phs 2.elimination

 

 

 

lV-107 lV-108 lV-109
1. LDA (2 eq.) 1. LDA (2 eq.)
0 2. o 2
Phs /__\ ' L_\ Phs
B o : PhSAn’OH 2 O
3. PTSA, PhH o 3. PTSA, PhH
88% 88%
lV-111 lV-11O lV-112

Scheme IV-lS: Synthesis of a—SPh lactones IV-lll and IV-112

We decided to adopt this strategy to introduce the terminal lactone in mucoxin.
Known a-SPh lactones IV-lll and IV-11275 were efficiently accessed from
commercially available phenylthioacetic acid IV-110. Di—anion of IV-l 10 when treated
with (S)- and (R)-propylene oxides generated the corresponding y—hydroxy acids (not
shown) which spontaneously cyclized upon exposure to catalytic PT SA in benzene. Thus,
both diastereomers IV-lll and IV-112 — referred to as S-y-methyl and R-y-methyl
lactones respectively, were conveniently synthesized. This was particularly advantageous
since we had randomly targeted an enantiomer of the bis-THF core (C8-Cl7) of
mucoxin.‘ By reacting the iodide derived from IV-97, (Scheme IV-12) with S-y-methyl
lactone lV-lll either natural mucoxin or its diastereomer would be produced. On the
other hand, combination of the iodide with IV-l 12 would furnish either the enantiomer or

C36 epimer of natural mucoxin. In either case, by comparison of the Optical rotation of

the synthetic samples with that of the natural product,” the absolute stereochemistry of

 

' The absolute stereochemistry of that part of mucoxin is unknown, However, the y-methyl

. . 7
stereocenter has been assrgned 5' configuration. 6

” Although the (10 of mucoxin has not been reported, we hoped to obtain an authentic sample.

256

mucoxin should be established. Accordingly, having the lactones (IV-111 and IV-112) in

hand, we now turned to orthogonally protected bis-THF IV-97 for further manipulations.

Iodide IV-114 was obtained in a straightforward manner from IV-97 by

sequential debenzylation (H2, Pd/C, 92%)77 and iodination (PPh3/IZ, 60%) of the resultant

primary alcohol (Scheme lV- l6).23

 

 

 

 

 

TBSQ OTBS H2, Pd /C TBSQ OTBS PPh3 imid.
1 O". O/,_ 08” _> t 01,. O/._ '
‘5 , 5 EtOAc :‘PrOH ‘5 . 5 '2, toluene
bras (1 : 1), rt, 92% bras 60%
lV-97 lV-113
TBSQ OTBS
' 0., 0.. I
15 5
bras
IV-114
LDA, lV-114 TBS? OTBS O 1. mCPBA, CHQCIQ, 0 °C
lV-111 *7 =
THF : HMPA 2. toluene, reflux
(4 : 1), 83% 83%

  

 
 

 

bores
lV-‘l 16

lV-117

Scheme IV-16: Completion of the total synthesis of proposed structure of mucoxin

(IV-117)

After some experimentation,38'74 alkylation of a—SPh lactone IV-lll with iodide

IV-ll4 was effected in 83% yield to secure intermediate IV-llS which contained the

complete carbon skeleton on mucoxin. The stage was now set for the B—elimination and

final deprotection reactions. IV-llS when submitted to mCPBA oxidation, afforded the

corresponding sulfoxide in quantitative yield. The crude sulfoxide upon heating

(refluxing toluene) underwent syn-fi-elimination to provide internal (1,6—unsaturated

257

lactone IV-116. Finally, global deprotection of IV-116 using HF-Py74 occurred

uneventfully to furnish target molecule IV-117, which was isolated in high purity after
HPLC purification. Also, coupling of iodide IV-114 with lactone IV-112 in an analogous
manner (Scheme IV-l7) provided IV-118, which was exactly identical to IV-117 in all

respects except the absolute configuration at C36.

0 1. LDA, lV-114, rt
“‘3 O 2. mCPBA, 0 °C
3. toluene, reflux

4. HF-Py, THF,rt

 

Scheme IV-l7: Synthesis of C36 epimer of IV-117

C. Comparison of spectroscopic data and conclusions

The structures (constitution) of IV-117 and IV-118 as shown (Schemes lV—l6 and

lV-l7), were confirmed by COSY experiments.

   

cis
lV-117, mucoxin (synthetic) IV-1 19, mucoxin (proposed structure)

Figure IV-17: Mucoxin: synthetic and originally proposed structures

lH and UC NMR spectra of IV-117 and IV-ll8 were found to be exactly identical
indicating that stereochemistry at C36 was inconsequential as far as NMR spectra were
concerned. However, both 1H and 13C spectra of IV-ll7 differed from the corresponding
published spectra of natural mucoxin having the proposed structure IV-119 (Figure

lV-l7). Partial 1H NMR spectra of the synthetic and natural samples are shown in Figure

258

1V-18. Since the major differences in the spectra reside in the hydroxyl-flanked bis-THF
(CS—C17) region, only that portion in each spectrum is shown (proton numbering

corresponds to the carbon numbering shown in the drawings above the spectra).

Table IV-S shows comparison of 1H chemical shifts of bis—THF portions of
IV-117 vs. natural mucoxin. The following differences and similarities in the spectra can
be noted. Oxymethines that show largest differences in chemical shifts are H17, and H14,
which are part of the trisubstituted THF ring. Other oxymethines, though slightly
different in chemical shifts (A6 = ca. 0.01 to 0.09), have the same splitting pattern. H13 in
the natural spectrum appears as a triplet with J = 3 Hz. Mclaughlin has used this splitting

pattern and the J value of H13 along with preliminary molecular modeling to propose the
relative stereochemistry of C12, C13 and C14 triad (Figure lV-l7) of mucoxin.“76 In

view of this, it becomes important to note that H13 in the synthetic spectrum appears as a
triplet as well with the exact same COUpling constant. Finally, chemical shifts of the THF

ring methylenes (H10, H11, and H15) also differ significantly. Moreover, of all three

methylenes, 5 value of H15, which again is part of the trisubstituted THF ring, deviates

the most.

 

’ The basis for this stereochemical assignment is discussed in more detail later.

259

HO OH

6 0 O
xWr
14
O

H
natural (IV-1 19)

H13 H111 H15

H12» H14

H16 H9

OH \

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

synthetic (IV—1 1 7)
H13
OH H10,H11,H15
a He. H17
HR H16 OH
H19 0“
H14
H10

H11 H15

1. J .. in

5,0 4.5 4.0 3.5 3.0 2.5 2.0 ppm

Figure IV-18: Comparison of partial 1H NMR spectra of the natural mucoxin and IV-117

260

 

 

 

 

 

 

 

 

 

 

 

 

proton IV-ll7 natural mucoxin A6 (IV-117—natural)
H8 3.41 3.42 -0.01
H9 3.96 3.95 +0.01
H10 1.84, 2.02 1.91, 2.05 -0.07, —0.03
Hll 2.02, 2.13 1.91, 2.05 0.11, 0.08
H12 4.35 4.31 0.04
H13 3.80 3.71 0.09
Hl4 4.44 4.32 0.12
H15 1.84, 2.02 1.91, 2.35 -0.07, -0.33
H16 4.09 4.04 0.05
H17 3.41 3.53 -0.12

 

 

 

 

 

Table IV-5: Comparison of 1H NMR chemical shifts of bis-THF portions (CS-C17) of

natural mucoxin vs. IV—117

Since neither a natural sample of mucoxin nor any other characterization data
besides the published spectra were available, we began further investigations using the
existing information. As a part of our efforts to locate the source of the discrepancies,
IV-122 - a diastereomer of IV-117 (epimeric at C8 and C9) was synthesized (Scheme
lV—18). Bis-THF intermediate IV-104, which was available via cyclization of triol
IV-103 (Table lV—4) was converted to the corresponding iodide (IV-121). Coupling of
iodide IV-121 with lactone IV-lll following a similar reaction sequence as before

(Scheme IV-18) afforded IV-122.

261

 

—E——#fi_u aun- n

 

 

 

TBSQ QH TBSQ ores
17, 60.. O . OBn 1.TBSOTt 2 17, 50., O , '
W 2 H2 Pa/c W
mores 3. 1333/3 / 12 ”bras
lV-104 ( °) rv-121
O
o 1.LDA,lV-121,rt H9 9H o36
Phs O 2. mCPBA,0°C -17 o,,_ ,2 o g 6 \
_= 15
3. toluene, reflux MbH
4. HF-Py, THF, rt
W411 (55% overall) Iv-122

Scheme IV-18: Synthesis of (8,9—epi) IV-117

Comparison of 1H NMRs of IV-122 and the natural sample indicated that
chemical shifts of all the oxymethines in the bis-THF (CS-C 17) portion differed (Figure
lV-19 and Table lV-6). The diagnostic Hl3 signal (t, J = 3 Hz) in the natural spectrum,
which was used to propose the relative configuration of C12, C13 and C14 stereocenters
(vide supra), is a dd (J = 1.5, 3.4 Hz) in IV-122. A6 for H9 in case of IV-122 is much

greater than that in IV-ll7 (Tables lV—S and 1V—6), which suggests that stereochemistry

of the di-substituted THF ring in IV-117 matches more closely to that in the natural

product. Also, the THF methylenes (H 10’ H 1 1 and H15) in IV-122 differ widely from those

in natural mucoxin. Taken together, 1H NMR of IV-117 matches more closely with the

natural spectrum than that of IV-122.

262

HO OH

OH
natural (IV-119)
“13 H1 is H 15

l.112‘Hl4

Pub H9

OH \ ,
= \

 

 

 

 

 

 

 

 

 

T r I l r 1 1 1 r 1 r 1 r I “—r 1 I i r m
5 4 3 2
Ho OH
1W;
14%
OH
synthetic (IV-122)
H10, 11111-115
UH
H13 OH
H12
H14
H9 H16
1 He H11 H15
H17
5.0 4.5 4.0 3.5 3.0 2.5 2.0 ppm

Figure IV-l9: Comparison of partial 1H N MR spectra of natural mucoxin and IV-122

263

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

proton IV-122 natural mucoxin A6 (IV-122-natural)

H8 3.44 3.42 0.02

H9 4.16 3.95 0.21
HIO 2.04, 2.04 1.91, 2.05 0.13, 001
HH 2.04, 2.22 1.91, 2.05 0.13, 0.17
Hl2 4.37 4.31 0.06
H13 3.95 3.71 0.24
H14 4.46 4.32 0.14
H15 1.92, 2.04 1.91, 2.35 0.01, 031
Hl6 4.08 4.04 0.04
Hl7 3.39 3.53 —0.14

 

 

Table IV-6: Comparison of 1H chemical shifts of bis-THF portions (C8-C17) of natural

mucoxin vs. IV-122

At this point, we decided to re-examine Mclaughlin’s reasoning for structure
elucidation of mucoxin. In the process we heped to delineate any ambiguities in their
proposed structure and possible sources of discrepancies between the synthetic and the
natural spectra.

Based on the reported COSY and HRMS analysis of the natural sample of

mucoxin, the proposed structure (constitution) appears to be correct. Figure IV-20 shows

the HRMS (El) fragmentation pattern of the tris-TMS derivative of mucoxin (IV-120).76

264

 

Figure IV-20: HRMS fragmentation pattern of the tris—TMS derivative of mucoxin. (* =

observed peak)

Also, as mentioned earlier, using COSY experiments the structure (constitution)
of our synthetic sample (IV-117) was clearly established. Therefore we felt that the
differences in the synthetic vs. natural spectra are most likely due to stereochemical
mismatches. In case of natural mucoxin, the relative configuration across both the THF
rings (C9—C12 and C13—C16, IV-119, Figure IV-17) was suggested to be trans based on
the lack of NOESY correlations. 1D NOESY correlations for synthetic compounds
IV-117 and IV-122 are shown in Figure IV-21. No nOe correlation peaks across either of
the THF rings (Hm—H13 or le—Hg) were observed in IV-117 (only relevant partial
structure shown). On the other hand, in case of IV-122, a strong nOe correlation was

observed across the (ii-substituted THF ring (Hg—H12) while no nOe signals were seen

between H13 and H16. This clearly suggests 2,5—cis relationship across the

no no no
nOe nOe nOe nOe

.1 H9<? )g 1) .a' H4?

 

lV-‘l 17 lV-1 22

Figure IV-21: nOe correlations in the two synthetic diastereomers

265

(Ii-substituted THF and 2,5«trans relation across C13-C16 THF ring in lV-122.
Moreover, since we have independently established the absolute configurations at C9,
C12 and C16 using Mosher’s ester analysis and ECCD techniques, it can be
unambiguously stated that both the THFS in IV-117 are trans. nOe correlations have been
routinely used to predict 2,5 relative configuration of (Ii-substituted THFS in acetogenins.
Such stereochemical assignments have proved reliable as confirmed by total synthesis or
X-ray analysis of these natural products.78 Thus, based on the above analysis we reasoned
that 2,5 stereochemical assignment across the two THF rings in natural mucoxin might be

correct. Also, strong nOe correlations between H13 and H14 in IV-117 and IV-122 (Figure

lV-21) confirmed cis-relation between them.

 

threo
HO H H OH HO H
o . 0 ”a . 0 H
‘5, H 12 8 012-013 bond rotation¢ ”I, H ,2 O ”I,
cis ”OH 14 0-..; OH
lV-119-A IV-119-B

Figure IV-22: Intramolecular hydrogen bonding in mucoxin as proposed by McLaughlin

We next turned to evaluate McLaughlin’s assignment of the relative configuration
of C12, C13, C14 triad (Figure IV-22, only relevant structural features shown). In 1H
NMR of natural mucoxin, I-I13 appeared as a pseudotriplet with 3 Hz coupling constant.

Based on this and molecular models, McLaughlin proposed that the C12-C13 bond
rotation might be restricted possibly due to intramolecular hydrogen bonding between

C14 hydroxyl and C9-C 12 THF ring oxygen (IV-119-B, Figure IV-22). In such a rigid

conformation, for H13 to maintain 3 Hz coupling constant with H12 and H 141 the two THF

266

 

 

rings must be threo to each other and C14~OH must be cis to the C13 side chain as shown

in IV-ll9-A (Figure IV-22).

Ho OH HO OH
. 5 0’" ‘2 0’“ a W
“at ”be
IV-123 lV-124
(cis-threo) (cis-erythro)
HQ OH Ho QH
14 14
OH OH
lV-125 lV-126
(trans-threo) (trans-eryhtro)

Figure IV-23: Truncated stereoisomeric bis-THF analogs of prOposed structure of

mucoxin

This stereochemical assignment seems tenuous since the only experimental
evidence presented is the coupling constant of H13 with H12 and H14. To gain further

insight into the conformations of the natural product, we decided to carry out basic
molecular modeling studies. Using molecular mechanics (MMFF94 force field),
conformational searches and energy minimizations of four diastereomeric bis—THF units

IV-123 — IV-126 (Figure lV-23) were carried out.‘ IV—123 is McLaughlin’s proposed
stereoisomer containing H13 and HM cis to each other, and H12 and H13 in a threo
relationship. IV-123 is referred to as the cis-threo isomer and all the remaining structures
are named similarly to reflect the relative configurations at the C12, C13, C14 triad

(Figure IV-23). IV-124 — IV-126 are hypothetical diastereomers containing different

relative stereochemistries at the C12, C13, C14 triad. However, threo relation of the side

 

' Spartan V 5.1.3, Wavefunction, Inc., 18401 Von Karman Avenue, Suite 370, lrvive CA 92612,
U.S.A.

267

chain hydroxyls with the THFS and 2,5 trans configuration across both the THFS in IV-

124 - IV-126 is conserved. Also, in all four diastereomers, the original long hydrocarbon

 

 

 

 

 

 

 

 

 

 

 

 

 

HQ OH
: 6 0a. 12 0,” 8
1:229
14",
OH
lV-123
(cis-threo)
lowest energy conformation
conformation Relative OH—Oa 813_14 (degrees) b @1243 (degrees) c
energy Distance (A)
(Kcal/mol)
1 0.00 1.78 -39.7 64.3
2 0.07 1.77 -39.7 64.5
3 0.37 1.77 -39.3 64.7
4 0.45 1.77 40.2 64.2
5 0.47 1.78 -40.4 63.9
6 0.73 1.77 —40.2 64.2
7 0.75 1.77 -40.4 63.9
8 0.78 1.7 8 -40.2 63.9
9 1.06 1.77 -40.2 63.9
10 1.55 1.79 -38.3 65.7

 

 

 

 

 

 

a distance between cr¢on and C9-C12 THF oxygen. b HB-HM dihedral angle. ° 1112-1113
dihedral angle.
_ . I, l ' ? _ -_
, f I,
/ ‘. ‘~;\

overlapped conformations
(E = 0.00 - 1.55 Kcal/mol)

Figure IV-24: Low energy conformations of cis-threo isomer IV-123

268

chains are truncated to ethyl groups for ease of energy minimizations.

Figure IV-24 shows conformations of IV-123 within 2 Kcal/mol of the lowest
energy conformation. In all the conformations, the OH —-0 distance between C14—OH
and C9-C12 THF oxygen is ca. 1.78 A, which is within hydrogen bonding range. The

average le-Hl3 and H13-H14 dihedral angles are 64.6° and 396° respectively with

variation of ca. 2° in each case. The overlapped conformations indicate that both the THF
rings are superimposible and only the side chains have different rotations. Using these

average dihedral angles, the corresponding theoretical coupling constants between H12-

H13 and Hrs-H1 4 can be estimated based on Karplus equation.79

 

10 T j. 7.— T I

 

 

 

 

 

Coupling Constant

 

 

 

 

 

O 30 60 90 120 150 180
Vicinal Dihedral Angle

Figure IV-25: Karplus equation plot for vicinal oxygenated systems

269

Besides dihedral angle, vicinal 1H—IH coupling constants also depend upon the
nature of substituents. Figure IV-25 shows variation of coupling constant (J) with

dihedral angle according to Karplus equation in case of vicinal oxygenated systems.”80

From the graph, 01243 = 65° corresponds to J 2 ca. 1.9 Hz and 0,}14 = 40° correspond to

J = 4.2 Hz. Thus, the estimated coupling constants are close to the experimentally

observed value of 3 Hz.
Similar analysis of cis-erythro diastereomer IV-124 provided interesting results

(Figure IV-26). In this case, the low energy (< 2 Kcal/mol) conformations can be divided
into two sets —- one having 01243 = 65°(A, Figure IV—26) and the other with 912-13 = 172°
(B).' Again from the Karplus equation plot (Figure IV-25) le-Hl3 coupling constant in
conformations A is estimated to be 9.0 Hz while that in conformations B would be ca. 1.9
Hz. As can be seen form the table (Figure IV-26), 60% of the p0pulation constitutes

conformations B. Notably, the OH—O distance between C14-OH and C9-C12 THF

oxygen in B is ca. 1.84 A whereas that in A is 2.38 A. Thus, the conformations with
812.13 = 65° are more within hydrogen bonding distance than those with 012.13 = 172°. If

such an intramolecular hydrogen bonding exists, conformations B are likely to be

favored.
Furthermore, when the 1H NMR spectrum of the synthetic material (IV-117) was
measured in methanol, H13 appeared as a doublet of a doublet (J = 3.3 and 7.3 Hz). While

the coupling constant between H13 and Hm would be expected to be more or less

independent of the solvent, that between H12 and H13 could vary with the nature of the

 

' Average values listed in both cases.

270

 

A (E = 0.00 Kcal/mol)

4 50H
IV-124

(cis-erythro)

B (E = 0.63 Kcal/mol)

 

ass-4°

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Conformation Relative ()H—Oa 013_14(degrees)b 01243 (degrees)°
energy Distance (A)
(Kcal/mol)

1 0.00 2.38 -33.2 -l71
2 0.48 2.37 -32.9 - 171
3 0.51 2.44 -34.2 - 17 3
4 0.63 1.84 40.1 -65.4
5 0.82 2.43 —34.1 - 173
6 0.93 1.84 -39.9 -65.4
7 1.03 1.84 -40.2 -65.6
8 1.19 1.85 -39.4 -64.4
9 1.33 1.84 -40.0 -65.6
10 1.53 1.85 -39.5 -64.5

 

a distance between C14—OH and C9-C12 THF oxygen. b H13-H14 dihedral angle. c le'H13

dihedral angle.

Figure IV-26: Low energy conformations of cis-erythro isomer IV-124

solvent due to the possibility of hydrogen bonding. Thus, it is conceivable that the

intramolecular hydrogen bond between the C14«OH and the THF oxygen (Figure IV-22)

is broken in CD3OD, thereby allowing free rotation about the C12—C13 bond. In this case,

the molecule is capable of attaining a conformation with a large value for 0,243 (~152°)

271

 

which corresponds to J = 7.3 Hz. In our conformational analysis, only IV-124 was found

to possess low energy conformations (set A) that would fit these observations.

Conversely in IV-123, 012.13 is close to 65° (J = 1.9 Hz) in all of the low energy

conformations, which would not fit the observed coupling constant in CD3OD.
From the above conformational analysis of IV-123 and IV-124, it can be stated
that the 3 Hz coupling constant may be maintained between H12 and H13, irrespective of

the threo/erythro relation between the two THFS. Thus, the possibility that the two rings
may be erythro cannot be ruled out just based on the observed coupling constants.
Certainly, such modeling studies are not reliable to decisively assign configurations of
unknown stereocenters without experimental evidence, but the analysis definitely
suggests a viable alternative to the originally prOposed structure of mucoxin.

The C13-C14 trans diastereomers IV-125 and IV-126 were analyzed along

similar lines. In case of the trans-threo isomer IV-125, low energy conformations can
again be divided into two sets based on 01243 values (Figure IV-27). 012,13 = 65°
corresponds to J = 1.9 Hz while 0,243 = 45° corresponds to J = 4.0 Hz according to the
Karplus plot (Figure IV-25). However, the 013“ values in the two sets of conformations
are —151° and —163° which would give coupling constants of about 6.8 Hz and 7.8 Hz
respectively. Thus, because the estimated trans H13-H14 coupling constants greatly

deviate from the observed value (3 Hz), trans-threo isomer IV-125 is not considered as a

valid alternative stereoisomer.

272

E = 0.00 Kcal/moi

 

HO

OH

14
OH

lV-1 25

(trans-threo)

1

457°

 

E = 0.15 Kcal/mol

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Conformation Relative energy 0H_0a 913-1 4 (degrees) b 912-13 (degrees) C
(Kcal/ “101) Distance (A)

l 0.00 4.61 -151 64.5
2 0.03 4.54 -153 63.9
3 0.15 2.18 —163 -45.7
4 0.22 4.52 ~155 63.2
5 0.30 4.65 - 148 65.6
6 0.32 4.65 -147 64.6
7 0.33 4.61 ~15] 64.2
8 0.48 2.19 -162 —45.7
9 0.53 4.53 -154 63.3
10 0.55 4.53 -155 63.3
1 l 0.60 4.66 -146 64.9
12 0.64 2.22 -162.5 46.7
13 0.79 2.18 -162.8 -45.8
14 0.85 4.53 ~154.5 63.4
15 0.88 2.18 -l62.5 -45.6
16 0.91 4.73 “142.4 64.6
17 0.97 2.22 - 162.4 -46.7

 

 

 

 

 

a distance between C14-OH and C9-C12 THF oxygen. b H13-H14 dihedral angle. c Hrz'Hr3
dihedral angle.

Figure IV-27: Low energy conformations of trans-threo isomer IV-125

273

 

In the case of trans-erythro isomer IV-126, all the low energy conformations

were found to have large dihedral angles (Figure IV—28). The estimated coupling
constants, viz., 9 Hz (912-13 2 178°) and 7.8 Hz (913-14 = 162°) do not match the observed

value of 3 Hz. Thus, IV-126 is also not considered a viable option.

HQ QH

1

4 OH
lV-126
(trans-eryhtro)

-162°

lowest energy conformer

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Conformation Relative energy Dbl—0° 913-1 4 (degrees) b 01243 (degrees) C
(Kcal/mol) Distance (A)
l 0.00 2.20 ~162 -l78
2 0.34 2.20 -162 —178
3 0.37 2.23 -161 -178
4 0.69 2.23 -162 —178
5 0.71 2.23 -161 -178
6 0.78 2.20 -162 ~178
7 1.03 2.23 -161 - 178
8 1.07 2.20 —l62 -178
9 1.34 2.23 -162 -178
10 1.51 2.24 -l61 -178
11 1.83 2.23 -161 -178

 

a distance between C14—OH and C9-C12 THF oxygen. b H13-H14 dihedral angle. C H, 2-H 13
dihedral angle.

Figure IV-28: Low energy conformations of trans-erythro isomer IV-126

274

 

Interestingly, during the total synthesis of another nonclassical actogenin —
jimenezingl (Figure IV-29), it was found that in the original proposed structure (IV-127

containing l9-or-H) the relative stereochemistry between the two rings was incorrectly
assigned. Diastereomer IV-126 (containing 19-B-H) was found to match the reported

spectra of natural jimenezin rather than IV-125.

H0, 0

H OH 10

Jimenezin, lV-127: 19 a-H
lV-128: 19 B-H

Figure IV-29: Jimenezin: proposed structure (IV-125) vs. real structure (IV-126)

Finally, based on X-ray crystal structures of previously known related
acetogenins-’8, McLaughlin has suggested that both the hydroxyl groups (C8 and C17)
flanking the bis-THF unit must be threo to the ring system (IV-119, Figure IV-17) in

mucoxin.

From all the above analysis, we propose IV-129 and its C8-C17 enantiomer

IV-l30 (Figure IV-30) as valid alternatives to the originally proposed structure (IV-119

Figure IV-l7) of natural mucoxin. Both the THFS in IV-129 and IV-130 are 2,5 trans and
therefore are not expected to show any NOESY correlations across the rings. Also, H13
and H14 being syn oriented should have the observed 3 Hz coupling constant. Finally, as
discussed above, we believe that the threo vs. erythro relationship between the two rings
is inconsequential for maintaining the 3 Hz coupling constant between H12 and H13. Thus,
both structures IV-129 and IV-130 are consistent with the experimental spectrosc0pic

data for mucoxin.

275

 

IV-129 lV-130
Figure IV-30: Possible alternative structure of mucoxin

In conclusion, total synthesis of the proposed structure of mucoxin (IV-117) has
been accomplished

in 32 steps (26 steps

 

along the longest

O
lV-117, Mucoxin (proposed structure)

hnearsequence)

Figures IV-31 and IV—32 depict the final synthetic scheme. The key
transformations are described below. The trisubstituted hydroxy THF portion (C12-C34)

was synthesized in the form of aldehyde IV-86 (Figure IV-3 1). The THF core of IV-86

276

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

OH 1. TBSCI, imid. OH 1. LAH. diglyme
//\’ T ¢ 3
/ 2- nBuLi \H" \H/V 2. NaH, PMBCI
1s ‘6 3. AD-mix-a
III-87 3. TBFA ITHF "F83 (73% three steps)
(53% three steps)
0H 1. TESCI, 513111 ores
’ OPMB DMAP ’ Ph3P=CHCOQEt
W F WCHO THF reflux 910/:
2. DDQ, CH2012 OTES ' '
IV-131 phosphate buffer lV-132
3. Phl(OAc)2, TEMPO
(75% three steps)
OTES 1. DIBAL-H OTES BF3'OE12 (6 eq.)
Woo Et 5 WM?“ :5
1e .4 2 2- D’ID'PT) 15 i 14 E120, 72%
OTES 3.1311282, BU3P OTES
(61% three steps)
IV-133 lV-134
QH TBSQ TBSQ 0
7,6 0". 183011 r 6 0.. 1, mCPBA r15 0..
Ween ' . A, 15 ,3 sen : 15 ,3 H
1..., 2.6 lutidine 1.2, 2. TFAA, 2,61utidine 14
OH CH2C12, 91°/o OTBS NaHC03, 600/0 OTBS
lV-135 IV-136 lV-86
HO\/W\0H m HOWOBn PCC OHCMOBH
TBAl, THF CH CI .11
lV-88 60 °C, 78% 1V-89 820/20 2 lV-90
1. KHMDS
2. TMSCI OBn
+ ’ .. /\/=’\/\/\/
Pn3P’\/\oHer 4 H0
4. lV-90 1. MsCl, E13N
3. AcOH :THF : H20 IV-91 CH2C12,0 °C
(6 :3 : 1), 0 °C, 83% 2. Nal, acetone
reflux, 77%
may 0H Iv-86 tBuLi, -100 °C
' 6 On. 12 __ OBn __#_ ; l/\/=\/\/\/OBH
15 13 4 MgBr ~OEt ,
“ho-1.88 Et20'2_40 OE: MgBr? OEtZ 'V‘87
85% 5'20
lV-85

Figure IV-31: Synthesis of hydroxy THF (C 12-C34) portion and its union with iodide

IV-87 via chelation controlled addition

was assembled using the method developed earlier (Chapters 1 and 111) via endo selective

cyclization of epoxy sulfide IV-134. The cyclization involved net retention of

277

configuration at C13, which provided the requisite C13-C14 cis relative configuration.
Pummerer rearrangement of the cyclized product (IV-136) afforded aldehyde IV-86.
IV-86 was then combined with iodide IV-87 by way of chelation controlled addition of
the corresponding Grignard derivative to furnish adduct IV-85 in high yield and

diastereoselectivity.

 

 

 

1880 OH AD-mix-cx T889 Hofi OH 1. EggéOMeh
: 0’“ 08“ I 0'" 12 8 OBn r
88% rs 95 4 2 BF ~OEt
dr : 5 ’ 1 -. OH . 3 2
lV—94
TBSQ OTBS
1. K2C03, MGOH : ' O". O... l
2. H2, Pd / c ‘5 .. 5
3. PPhg, imid., 12 0188
(55 /0 three steps) w.114

 

 

0 1. LDA, 1V-114, rt

°“3 0 2. mCPBA, 0 °C
3. toluene, reflux
4. HF-Py, THF,rt

lV-1 11 (55% overall) lV-1 17

 

 

Figure IV-32: Completion of the total synthesis

Sharpless asymmetric dihydroxylation of IV-85 (Figure IV-32) afforded the
requisite triol IV-94 in high yield (88%), albeit in modest diastereomeric ratio. However,
the 5 : 1 ratio was acceptable to us since the desired isomer IV-94 was easily separable
form the minor diastereomer by flash column chromatography and could be isolated in
good (73%) yield.

In order to install the di-substituted THF (C8-C12) a novel triol cyclization
method was developed. Triol IV-94 was converted to bis-THF IV-95 in a single
transformation involving generation and intramolecular trapping of acetoxonium ion of

the vicinal diol functionality. This triol cyclization proved highly efficient (98% yield)

278

and the desired bis-THF was obtained in a completely regio- and stereoselective manner.
Further investigations to expand the scope of this methodology are being pursued by
another graduate student in our laboratory. The two step protocol involving sequential
asymmetric dihydroxylation and triol cyclization to transform bis-homoallylic alcohol
such as IV-85 to bis—THF IV-95 is more efficient and versatile (in terms of yields and
potentially accessible stereoisomers) than traditional vanadium catalyzed directed

epoxidation / cyclization route. Finally, the terminal butenolide was introduced using

known or-SPh lactone IV-lll.
1H and 13 C NMR Spectra of synthetic product IV-ll7 differed from that of those

of natural mucoxin. Based on the reported COSY and HRMS analysis of the natural
sample and our own COSY experiments on synthetic IV-117, we believe that
constitutionally, the structures of synthetic and natural mucoxin are identical. Therefore,
it was reasoned that the discrepancies in the spectra are most likely due to stereochemical
mismatches. In our total synthesis, all the stereocenters were established using either

highly reliable asymmetric reactions, viz., Sharpless asymmetric epoxidation and

Mosher's
ster analysis

   

1D NOESY 1D NOESY cis

lV-1 1 7, mucoxin (synthetic) lV-1 19, mucoxin (proposed structure)
Figure IV-33: Summary of structure proof of synthetic material (IV-117)

dihydroxylation or well precedented transformations such as chelation controlled

organometallic addition and intramolecular epoxide opening whose stereochemical

279

outcomes are definitively predictable. Moreover as shown in Figure lV—33,
configurations at all stereogenic centers have been independently confirmed using
Mosher’s ester analysis, nOe correlations and ECCD techniques.

Therefore, we believe that the relative stereochemistry of IV-117 exactly matches
the proposed structure and that the differences in the synthetic vs. natural spectra are due
to incorrect stereochemical assignment in the original prOposed structure (IV-119, Figure
IV-27). After closer examination of McLaughlin’s reasoning in allocating the relative

configuration, we feel that assignment of the threo relationship between the two THF
rings, which was based on coupling constant between H12 and H13, is unconvincing.

Accordingly, our modeling studies (MM2) indicate that threo or erythro relationship
between the rings may be inconsequential to explain the observed coupling constant.
Thus, we prOpose two alternative structures (IV-129 and IV-l30, Figure IV-30)
containing erythro THFS, which also fit the reported spectroscopic data for natural

mucoxin.

D. Experimental section

General Procedures

All reactions were carried out in flame-dried glassware under an atmosphere of

dry nitrogen or argon. 4 A molecular sieves were dried at 160 °C under vacuum prior to
use. Unless otherwise mentioned, solvents were purified as follows. THF and Etzo were

either distilled from sodium benzophenone ketyl or used as is from a solvent purification

280

system. CHZCIZ, toluene, CH3CN and EN were distilled from CaHz. DMF, diglyme, and

DMSO were stored over 4 A mol. sieves and distilled from CaHZ. All other commercially
available reagents and solvents were used as received.

1H NMR spectra were measured at 300, 500 or 600 MHz on a Varian
Gemini-300, a Varian VXR-SOO or a Varian [nova-600 instrument respectively. Chemical

shifts are reported relative to residual solvent (5 7.27, 2.50 and 4.80 ppm for CDCl3,

(CD3)ZSO and CD3OD respectively). 13C NMR spectra were measured at 125 MHz on a
Varian VXR-SOO instrument. Chemical shifts are reported relative to the central line of
CDCl3 (a 77.0 ppm). Infrared spectra were recorded using a Nicolet IR/42 spectrometer

FT-IR (thin film, NaCl cells). High-resolution mass spectra were measured at the
University of South Carolina, Mass Spectrometry Laboratory using micromass VG-70 5
mass spectrometer. Optical rotations were measured on a Perkin—Elmer polarimeter
(model 341) using a 1 mL capacity quartz cell with a 10 cm path length.

Analytical thin layer chromatography (TLC) was performed using Whatman glass
plates coated with a 0.25 mm thickness of silica gel containing PF254 indicator, and
compounds were visualized with UV light, potassium permanganate stain, p-
anisaldehyde stain, or phosphomolybdic acid in EtOH. Chromatographic purifications
were performed using Silicycle 60 A, 35—75 pm silica gel. All compounds purified by
chromatography were sufficiently pure for use in further experiments, unless indicated
otherwise. GC analysis was performed using HP (6890 series) GC system containing
Altech 813-54, 30 m x 320 mm x 0.25 mm column. Analytical and semi-preparative

HPLC normal phase separations were performed using HP 1100 series HPLC system.

281

O O

CH 01 , rt
lV-11 72°}; 2 lV-12

 

To a suspension of PCC (22.9 g, 0.11 mol) and sodium acetate (2.4 g, 0.03 mol)

in CHZCl2 (100 mL), was added a solution of ethyl 6-hydroxyhexanoate IV-ll (9.85 g,
0.06 mol) in CHZClz (24 mL) at room temperature. After 2 h, the reaction was diluted

with E120 (150 mL) and filtered through a Celite pad. The filtrate was concentrated under

reduced pressure and the crude material was purified by column chromatography (5%

EtOAc in hexanes) to afford aldehyde IV-12 (6.83 g, 72%). Spectroscopic data for IV-12
was found to be identical to that reported previously.55

Partial data for mm: ‘H NMR (500 MHz, CDC13) a 9.76 (t, J = 1.7 Hz, 1 H), 4.11 (q, J
= 7.1 Hz, 1 H), 2.48—2.43 (m, 2 H), 2.36—2.08 (m, 2 H), 1.69-1.63 (m, 4 H), 1.24 (t, J =

7.3 Hz, 3 H); 13'c NMR (125 MHz, CDC13)6 202.4, 173.5, 60.6, 43.7, 34.2, 24.5, 21.7,

 

14.5.
1. KHMDS O
i 2' TMSCI W0”
Ph3P/\/\OH Br" = /\0 —
4. IV-12
3. ACOH I THF 2 H20 |V-13

(6 :3 : 1), 0 °C, 79%

KHMDS (57 mL of 0.5 M solution in toluene, 28.5 mmol) was added to a —20 °C
slurry of 3-hydroxypropyltriphenylphosphonium bromide (5.72 g, 14.25 mmol) in THF
(30 mL). The mixture was brought to room temperature and stirred for 1 h. After cooling
back to 0 °C, TMSCI (1.34 mL, 10.5 mmol) was added and stirring was continued at the
same temperature for 15 min. The reaction was then cooled to —78 °C upon which a THF
solution of aldehyde IV-12 (1.5 g, 9.5 mmol in 20 mL) was added. The reaction was

282

warmed to ——10 °C over 1 h and then treated with AcOH : H20 : THF (6 : 3: 1, 100 mL).
After being stirred at room temperature for 15 h, the reaction mixture was neutralized by

saturated NaHCO3. The aqueous layer was extracted with EtOAc (3x 200 mL), combined

organic layers were dried over NaZSO4, concentrated and purified by column

chromatography (20%-10% EtOAc in hexanes) to secure the homoallylic alcohol IV-l3

(1.5 g, 79% Z: E> 10: 1).

Partial data for IV-13: 1H NMR (500 MHz, CDC13) o 5.50—5.44 (m, 1 H), 5.39-5.27 (m, 1
H), 4.08 (q, J = 7.0 Hz, 2 H), 3.59 (t, J = 6.7 Hz, 2 H), 3.40 (s(br), l H), 2.31-2.23 (m, 4
H), 2.08-2.00 (m, 2 H), 1.65—1.55 (m, 2 H), 1.40—1.33 (m, 2 H), 1.21 (t, J = 7.2 Hz, 3 H);

13c NMR (125 MHz, CDC13) a 174.05, 132.5, 125.9, 62.4, 60.5, 34.4, 31.0, 29.3, 27.1,

 

24.7, 14.4.
0 . . 0
PPh , mid.
l2, toluene
IV-13 50% Me

To a solution of alcohol IV-13 (580 mg, 2.90 mmol) in toluene (20 mL), triphenyl
phosphine (1.91 g, 7.28 mmol), imidazole (500 mg, 7.34 mmol) and iodine (1.47 g, 5.79
mmol) were added at room temperature. After 30 min, saturated sodium sulfite solution
was added to the yellowish brown mixture until it turned colorless. Layers were
separated, the aqueous layer was extracted with EtOAc (3x20 mL), and the combined
organic layers were dried (NaZSO4) and concentrated. Upon purification by column

chromatography (2% EtOAc in hexanes), iodide IV-6 was isolated in 50% yield (450

mg).

283

Partial data for IV-6: 1H NMR (500 MHz, CDC13) a 5.52-5.48 (m, 1 H), 5.35-5.31 (m, 1
H), 4.11 (q, J = 7.1 Hz, 2 H), 3.12 (r, J = 7.3 Hz, 2 H), 2.64259 (m, 2 H), 2.29 (t, J = 7.5
Hz, 2 H), 2.06-2.02 (m, 2 H), 1.66-1.60 (m, 2 H), 1.43-1.37 (m, 2 H), 1.25 (t, J = 6.4 Hz,

3 H); ”C NMR (125 MH2,CDC13) 5 173.8, 132.2, 128.5, 60.4, 34.4, 31.7, 29.2, 27.3,

 

 

24.8, 14.5, 5.5.
T880 0 NaH p0 T880 0 T880 0
R0 0,, 2 4 0., (COCI)2, DMF 0.,_
‘ H > - OH > Cl
., NaClOZ, tBuOH OR -. hexanes, rt OR
01133 30% ’oTes quant ores
lV-8; R = TBDPS IV-16 lV-17

An aqueous solution of sodium chlorite (105 mg, 1.0 mmol) was added to a
solution of aldehyde IV-8 (200 mg, 0.32 mmol) in tBuOH (2.5 mL) followed by 0.75 mL
of 2—methyl-1-butene in THF (2 M, 1.5 mmol). Monobasic sodium phosphate (95 mg, 0.5
mmol) was added in one portion upon which the solution turned yellow. After stirring for
17 h, volatiles were evaporated and the residue was taken up in CHZCIZ. The salts were
removed by filtration and crude acid IV-l6 was used without further purification. To a
solution of IV-16 (197 mg, 0.30 mmol) in hexanes (5 mL), oxalyl chloride (132 uL, 1.5
mmol) and DMF (26 uL, 0.30 mL) were added and the mixture was stirred at room
temperature for 1 h. Supernatant liquid was separated from the solids and concentrated
under reduced pressure. The crude acid chloride IV-17 was dried under high vacuum
(0.05 mm) and used without purification.

Partial data for IV-16: 1H NMR (500 MHz, CDC13) d 7.68-7.64 (m, 4 H), 7.47-7.36 (m, 6
H), 4.81—4.76 (m, 1 H), 4.67-4.62 (m, 1 H), 4.37 (d, J: 3.3 Hz, 1 H), 4.18-4.11 (m, l H),

3.57 (dd, J = 4.7, 10.7 Hz, 1 H), 3.40 (dd, J = 7.1, 10.5 Hz, 1 H), 2.37-2.22 (m, 1 H),

284

1.94-1.88 (m, 1 H), 1.05 (s, 9 H), 0.89 (s, 9 H), 0.82 (s, 9 H), 0.10 (s, 3 H), 0.08 (s, 3 H), -

0.01 (3,3 H), -0.08 (s, 3 H).

Partial data for mm: 1H NMR (500 MHz, CDC13) d 7.68-7.63 (m, 4 H), 7.45-727 (m 6
H), 4.89-4.75 (m, 2 H), 4.74 (d, J = 4.4 Hz, 1 H), 4.08-4.05 (m, 1 H), 3.58 (dd, J = 4.5,
10.4 Hz, 1 H), 3.40 (dd, J = 7.6, 10.5 Hz, 1 H), 2.19—2.14 (m, 1 H), 1.95-1.91 (m, l H),
1.04 (s, 9 H), 0.90 (s, 9 H), 0.82 (s, 9 H), 0.13 (s, 3 H), 0.12 (s, 3 H). -001 (s, 3 H), 009
(s, 3 H); 13c: NMR (125 MHz, CDC13) is 171.9, 135.8, 135.7, 133.4, 130.0, 127.9, 90.0.

82.6, 75.1, 72.4, 65.7, 34.2, 27.0, 26.0, 25.8, 19.4, 18.2, 18.1, 1.3, —4.3, —4.6, 41.7, -5.0.

 

1. Zn, 40 °C, 6 h T830 0 0
ADM—WI ‘5 Oh, ._ O/\
2. CuCN-2L1Cl
3. IV-17 OR toms
lV-6 0 °C. 60% IV-18

A flask charged with Zn powder (55 mg, 0.84 mmol) was flame dried and flushed
with Ar. THF (0.8 mL) and 1,2 dibromoethane (2.6 ptL, 0.03 mmol) were added and the
mixture was heated to 65 °C for 30 min. After cooling to room temperature, TMSCI, (3.1
uL, 0.02 mmol) was introduced and the mixture was heated back up to 40 °C for 15 min.
Again after cooling to room temperature, a solution of iodide IV-6 (130 mg, 0.42 mmol)
in THF (0.5 mL) was added and the mixture was further heated to 40 °C for 6 h. The
suspension of organozinc reagent so generated was allowed to settle at room temperature.
In the mean time, a mixture of CuCN (38 mg, 0.42 mmol) and LiCl (36 mg, 0.84 mmol)
was dissolved in THF (0.5 mL). After cooling the solution to —60 °C, the organozinc
reagent was canulated into the CuCN°2LiCl complex. This mixture was warmed to 0 °C,
stirred for 45 min and cooled back to ——25 °C. Acid chloride IV-17 (217 mg, 0.32 mmol)

was added as a THF solution (0.5 mL) and the reaction was stirred overnight at 0 °C.

285

 

Saturated NH 4Cl solution (1.5 mL) and Etzo (5 mL) were added, layers were separated
and the aqueous layer was extracted with Etzo (3x20 mL). The combined organic layers

were dried (NaZSO4), concentrated and crude material was purified by column

chromatography (5% EtOAc in hexanes) to afford the ketone IV-18 (158 mg, 60%) as a

colorless liquid.

Partial data for IV-18: 1H NMR (500 MHz, 0001,) d 7.67-7.65 (m, 4 H), 7.43-737 (m, 6
H), 5.37-5.35 (m, 2 H), 4.78—4.74 (m, 1 H), 4.66-4.64 (m, 1 H), 4.25 (d, J: 3.8 Hz, 1 H),
4.12 (q, J = 7.1 Hz, 2 H), 3.61 (dd, J = 4.6, 10.4 Hz, 1 H), 3.45 (dd, J = 7.1, 10.4 Hz, 1

H), 2.78-2.71 (m, 1 H), 2.59-2.52 (m, 1 H), 2.31—2.28 (m, 4 H), 2.10-2.07 (m, 3 H), 1.85-
1.82 (m, l H), 1.67-1.61 (m, 2 H), 1.41-1.35 (m, 2 H), 1.25 (t, J: 7.1 Hz, 3 H), 1.05 (s, 9
H), 0.85 (s, 9 H), 0.81 (s, 9 H), 0.06 (s, 3 H), 0.01 (s, 3 H), -0.05 (s, 3 H), -0.07 (s, 3 H);
13c: NMR (125 MHz, CDC13) a 211.4, 173.9, 135.8, 133.6, 133.5, 1334.302, 129.9,
129.0, 127.9, 88.7, 81.1, 75.9, 73.1, 65.9, 60.4, 40.5, 35.2, 34.5, 29.4, 27.0, 26.0, 25.9,

25.8, 24.9, 20.9, 19.4, 18.2, 18.1, 14.4, -4.6, -4.7, -4.9.

 

TBSO OH 0
W0 J NaBH" 30” 4 W02
d 0°C,3rt,2 quant OR ,_
0188 r 0183
lV-18 lV-27
To an ethanol solution of ketone IV-18 (21 mg, 0.03 mmol in 2 mL), sodium

borohydride (5 mg, 0.13 mmol) was added in one portion at room temperature. After 2 h,

the reaction was quenched by H20 (1 mL) and extracted with EtOAc (3x5 mL).

Combined organic layers were dried over NaZSO4, concentrated and the crude material

286

 

 

was purified by column chromatography (20% EtOAc in hexanes) to afford alcohol IV-

27 as an inseparable mixture of diastereomers.

Partial data for IV-27: 1H NMR (500 MHz, 0001,) s 7.66-7.63 (m, 4 H), 7.44—7.36 (m, 6
H), 5.43-5.35 (m, 2 H), 4.61-4.44 (m, 3 H), 4.14-3.82 (m, 3 H), 3.67-3.54 (m, 2 H), 3.42-
3.38 (m, 1 H), 2.31-2.07 (m, 6 H), 1.85-1.80 (m, 1 H), 1.67-1.58 (m, 3 H), 1.47—1.43 (m,
l H), 1.41-1.37 (m, 3 H), 1.05 (s, 9 H), 0.85 (s, 9 H), 0.81 (s, 9 H), 0.06 (s, 3 H), 0.01 (s,
3 H), —0.05 (s, 3 H), -0.07 (s, 3 H); 13C NMR (125 MHz, CDC13) a 174.0, 163.1, 135.8,
135.7, 133.5, 130.5, 130.4, 130.3, 130.2, 130.0, 129.9, 129.8, 127.9, 84.5, 84.1, 79.3,
78.9, 75.6, 75.5, 74.7, 73.4, 73.1, 70.7, 65.9, 65.8, 60.4, 35.2, 34.5, 34.4, 34.2, 33.8, 33.6,
32.4, 29.9, 29.5, 29.4, 29.3, 29.0, 27.1, 27.0, 26.1, 26.0, 25.8, 24.9, 24.7, 23.8, 19.4, 18.2,

18.1, 14.5, -4.05, «4.21, -4.52, 4.62, -4.65, —4.79.

‘BuLi. -100 °C lV-8, MgBrQ-OEIQ

 

 

lM—WOPMB —; _:

then MgBrg-OEtZ -40 °C. 88% Or > 9 :1

”'48 -90 °C to 0 °C
1880 OH T830 0
o,,_ __ OPMB CH0
3
on ,
OR .0188 bTBS
IV-50 IV—8, R = TBDPS

 

 

 

Preparation of 1.0 M solution of MgBr2¢OEt2 in diethyl ether:53
A two necked round bottom flask fitted with a reflux condenser was charged with

Mg turnings (875 mg, 36 mmol) and a stir bar. After flame drying the flask under N2,

Etzo (30 mL) was added. 1,2 dibromoethane (2.6 mL, 30 mmol) was then added drop

wise with gentle stirring upon which the solvent started refluxing slowly. When the

addition was complete and refluxing ceased, the mixture was stirred for additional 1 h to

287

ensure completion of the MgBr2°OEt2 formation. The solution so prepared was used

immediately.

'BuLi (4.0 mL Of 1.3 M solution in pentane, 5.16 mmol) was added drOp wise to

pre-cooled (-100 °C) E120 (9mL). To this, a solution Of iodide IV-48 (1.78 g, 4.97 mmol)
in EtZO (14 mL) was added over 10 min.‘ After stirring for 5 min at —100 °C to -90 °C,

MgBrz'OEt2 in EtZO (5.2 mL of 1.0 M solution (freshly prepared as described above),
was added and the mixture was warmed 0 °C over 1 h. Meanwhile, a solution of aldehyde
IV-8 (660 mg, 1.03 mmol) in Etzo (9 mL) was cooled to —40 °C. MgBr2°OEt2 in Etzo

(3.9 mL of 1.0 M solution, 3.9 mmol) was added and stirred for 10 min. To this pre—
complexed aldehyde, solution of the above mentioned Grignard reagent was cannulated

at —40 °C and stirred overnight at the same temperature. The reaction was then quenched

by slow addition of saturated NH4C1 solution (10 mL) and H20 (20 mL). The aqueous
layer was extracted with EtZO (3x100 mL). Combined organic layers were dried

(NaZSO4), concentrated under reduced pressure to afford a yellow Oil. Purification by
column chromatography (2% EtOAc in hexanes) furnished the adduct IV-50 (808 mg,
88%) as a single diastereomer.

Partial data for mm: 1H NMR (500 MHz, 01301,) 6 7.67-7.64 (m, 4 H), 7.44-7.35 (m, 6
H), 7.26-7.24 (m, 2 H), 6.89—6.86 (m, 2 H), 5.40-5.35 (m, 2 H), 4.61-4.57 (m, 1 H), 4.44-
4.42 (m, 2 H), 4.43 (s, 2 H), 4.06-4.04 (m, 1 H), 3.86-3.82 (m, 1 H), 3.80 (s, 3 H), 3.66 (t,

J = 4.0 Hz, 1 H), 3.57 (dd, J = 4.9, 10.4 Hz, 1 H), 3.45 (t, J = 6.6 Hz, 2 H), 2.27-2.07 (m,

 

' Iodide IV-48 was prepared using the same procedure as for iodide IV-87, which is described
later in this section.

288

3 H), 1.86—1.82 (m, l H), 1.65—1.42 (m, 3 H), 1.21 (t, J =2 7.0 Hz, 2 H), 0.90 (s, 9 H), 0.70
(s, 9 H), 0.60 (s, 9 H), -0.09 (s, 3 H), -O.11 (s, 3 H), -O.12 (s, 3 H), —0.23 (s, 3 H); 13C
NMR (125 MHz, CDCI3) 0 158.8, 135.3, 133.0, 130.0, 129.7, 129.5, 129.3, 129.0, 127.5,
120.1, 113.5, 100.1, 84.0, 78.8, 75.2, 72.6, 72.3, 70.2, 69.8, 65.4, 55.0, 34.7, 33.3, 29.2,
27.3, 26.8, 26.5, 26.1, 25.6, 25.5, 23.3, 18.9, 17.7, 17.6, -4.7, —5.0, -5.1, -5.3; IR (thin

film) 2953, 2930, 2858, 1514, 1429, 1361, 1250, 1151, 1113, 1076, 1007, 835,777, 702

-I

 

cm .
1830 OH TBSO OH
mCPBA, CH2012 ;' o 0
ROWOPMB = Wows
, then ACOH OR ~.
0183 85% 0168
lV-50 IV-51

mCPBA (172 mg, 1.00 mmol) in CHZCI2 (9 mL) was added to a CHZCI2 solution
of hydroxy alkene IV-50 (430 mg, 0.48 mmol in 9 mL) and the reaction was stirred at
room temperature for 30 min. 10 mL glacial acetic was then added and after 10 h, the

reaction was quenched by saturated NaHCO3 solution (15 mL). Upon separation of the

layers, the aqueous layer was extracted with CHZCI2 (3x15 mL), combined organic layers

were dried, concentrated and the crude material was purified by column chromatography
(20% EtOAc in hexanes) to afford bis-THF IV-51 as an inseparable mixture of
diastereomers (ca. 1 : 1 ratio).

Partial data for 1v-51: 1H NMR (500 MHz, 0001,) 6 7.50-7.48 (m, 4 H), 7.29—7.21 (m, 6

H), 7.11 (d, J = 8.5 Hz, 2 H), 6.72 (d, J = 8.7 Hz, 2 H), 4.564.53 (m, 1 H), 4.27 (S, 2 H),
4.18-3.96 (m, 3 H), 3.66-3.62 (m, 1 H), 3.64 (s, 3 H), 3.60-3.45 (m, 1 H), 3.40-3.18 (m, 5
H), 2.00-1.30 (m, 12 H), 0.90 (s, 9 H), 0.70 (s, 9 H), 0.60 (s, 9 H), -0.09 (s, 3 H), -0. 11 (s,

3 H), -012 (s, 3 H), —0.23 (s, 3 H); 130 NMR (125 MHz, 0001,) a 158.8, 135.3, 133.1,

289

133.0, 130.5, 129.5, 129.4, 129.0, 127.4, 120.2, 113.5, 85.3, 82.1, 79.2, 78.9, 74.0, 73.7,
73.4, 72.5, 72.3, 69.9, 65.2, 55.0, 53.9, 47.3, 33.7, 33.5, 29.6, 29.5, 28.0, 27.9, 27.7, 26.5.
25.6, 25.4, 22.2, 18.8, 17.8, 17.6, -4.3, -5.0, -5.1, —5.2; IR (thin film) 3583, 3470, 2932,
2859, 2256, 2968, 1887, 1818, 1718, 1605, 1514, 1429, 1361, 1250, 1151, 1072, 939,

910, 808, 734, 702 cm".

 

TBSO OH 1050 ores OPMB
W T880" W
- OPMB 4v '
on ., 2,6-lutidine on -,
’ores CHZClz, 89% ores
IV-51 lV-52

To a 0 °C solution of alcohol IV-Sl (315 mg, 0.35 mmol) in CHZCl2 (10 mL),
2,6-lutidine (0.28 mL, 2.43 mmol) and TBSOTf (0.24 mL, 1.04 mmol) were added in that

order. After 30 min at the same temperature, saturated NaHCO3 solution (5 mL) was
added and the layers were separated. The aqueous layer was extracted with CH2C12 (3x15

mL), combined organic layers were dried over NaZSO4, and concentrated under reduced

pressure to afford a crude Oil. Upon purification of the oil by column chromatography
(1% EtOAc in hexanes), tris-TBS ether IV-52 was obtained in 89% yield (323 mg).

Partial data for IV-SZ: 1H NMR (500 MHz, CDC13) a 7.65-7.63 (m, 4 H), 7.44—7.35 (m, 6
H), 7.26-7.25 (m, 2 H), 6.88—6.86 (m, 2 H), 4.68-4.63 (m, 1 H), 4.42 (d, J = 2.4 Hz, 2 H),
4.32-4.30 (m, 1 H), 4.14-4.09 (m, l H), 4.00—3.92 (m, 2 H), 3.80 (s, 3 H), 3.76-3.72 (m, 1
H), 3.63 (dt, J = 2.9, 8.8 Hz, 1 H), 3.53-3.50 (m, 1 H), 3.45-3.34 (m, 3 H), 2.17-1.24 (m,
12 H), 0.88 (s, 9 H), 0.86 (s, 9 H), 0.80 (s, 9 H), 0.78 (s, 9 H), 0.10-0.05 (m, 18 H); 13C
NMR (125 MHz, CDCl3) 0 159.2, 135.8, 133.7, 133.5, 131.1, 129.9, 129.8, 129.4, 129.3,
127.9, 127.8, 113.9, 86.8, 85.3, 81.1, 81.0, 79.8, 79.6, 79.3, 79.1, 74.3, 74.1, 73.9, 73.8.

73.3, 72.8, 72.7, 72.6, 70.6, 70.5, 65.9, 65.8, 55.5, 34.4, 34.3, 31.5, 31.4, 30.6, 30.3, 30.2,

290

29.9, 28.5, 28.0, 27.0, 26.9, 26.8, 26.3, 26.2, 26.1, 26.0, 25.9, 25.8, 23.0, 19.3, 19.2, 18.4,

18.3, 18.2, 18.1, 18.0, -2.7, -3.7, -3.8, —4.0, -4.1, -4.3, -4.4, 4.5, -4.7, -4.8, -4.9.

 

Teso ores OPMB TBSQ ores
* 0,, 0 DDQ : ’ 0,, 0
. 0 OH
Ores (9 I 11.01181"t ’ores
lV-52 lV-53

DDQ (94 mg, 0.41 mmol) was added to a solution of PMB ether IV-52 (330 mg.

0.32 mmol) in 10% wet chloroform (7.1 mL) and the mixture was stirred for 30 min at 0

°C. The reaction was then poured into saturated NaHCO3 solution (5 mL), layers were
separated and the aqueous layer was extracted with CHZCIZ (3x10 mL). Combined

organic layers were dried (Na2304), concentrated and the crude product was purified by

column chromatography (5% EtOAc in hexanes) to furnish alcohol IV-53 as a colorless

Oil (287 mg, quant.).
Partial data for IV-53: 1H NMR (500 MHz, 0001,) 6 7.66-7.63 (m, 4 H), 7.41—7.34 (m, 6

H), 4.67-4.60 (m, 1 H), 4.33-4.23 (m, 1 H), 4.13-4.01 (m, 1 H), 3.98—3.88 (m, 1 H), 3.74-
3.72 (m, 2 H), 3.68-3.62 (m, 1 H), 3.62 (t, J = 6.0 Hz, 2 H), 3.53-3.46 (m, 1 H), 3.37-3.34
(m, 1 H), 2.19-1.18 (m, 12 H), 1.01-0.09 (m, 36 H), 0.05-0.08 (m, 18 H); 13C NMR (125
MHz, CDCI3) 0 135.8, 133.7, 133.6, 133.5, 132.2, 129.9, 129.8, 127.9, 127.8, 114.5,
86.8, 85.3, 81.1, 81.0, 79.8, 79.6, 79.3, 79.2, 74.3, 74.0, 73.9, 66.1, 65.9, 65.8, 63.2, 63.1,
55.8, 34.4, 34.3, 33.3, 33.2, 31.8, 31.5, 31.4, 29.9, 28.6, 28.1, 27.0, 26.4, 26.3, 26.2, 26.1,
26.0, 25.9, 25.8, 25.7, 22.9, 22.4, 19.3, 18.4, 18.3, 18.2, 18.1, 18.0, 15.5, 14.3, -3.8, -3.9, -
4.0, -4.1, -4.2, -4.3, -4.5, -4.7, -4.8, -4.9; IR (thin film) 3441, 3073, 2955, 2893, 2853,
1911, 1887, 1822, 1701, 1601, 1512, 1471, 1429, 1362, 1257, 1113, 1074, 1005, 939,

885, 775,702 cm".

291

TBSQ ores Pphs imid. TBSQ ores

 

o, 0 OH 45 o... o 1
l toluene
0H ., 2' OH .,
0188 34% 0183
ME; IV-54

To a solution of alcohol IV-53 (260 mg, 0.29 mmol) in toluene (10 mL), triphenyl
phosphine (192 mg, 0.73 mmol), imidazole (52 mg, 0.73 mmol) and iodine (160 mg, 0.57
mmol) were added at room temperature. After 30 min, saturated sodium sulfite solution
was added to the yellowish brown mixture until it turned colorless. Layers were

separated, the aqueous layer was extracted with EtOAc (3x10 mL), combined organic
layers were dried (NaQSO4) and concentrated. Upon purification by column

chromatography (2% EtOAc in hexanes), iodide lV-54 was isolated in 84% yield (253
mg).

Partial data for IV-54: 1H NMR (500 MHz, 0001,) 8 7.66-7.64 (m, 4 H), 7.43-7.36 (m, 6
H), 4.70-4.64 (m, 1 H), 4.33-4.31 (m, l H), 4.144.10 (m, 1 H), 4.03-3.92 (m, 2 H), 3.76-
3.75 (m, 1 H), 3.65 (dt, J = 2.9, 7.5 Hz, 1 H), 3.54-3.50 (m, l H), 3.39-3.35 (m, l H),
3.21-3.16 (m, 2 H), 2.19-1.77 (m, 6 H), 1.63-1.53 (m, 3 H), 1.44-1.26 (m, 3 H), 1.01 (5,9
H), 0.89 (s, 9 H), 0.81 (s, 9 H), 0.10-0.04 (m, 18 H); 13C NMR (125 MHz, CDCl,) 6
135.8, 133.6, 133.5, 133.4, 129.9, 129.8, 128.0, 127.9, 86.8, 85.3, 81.0, 79.8, 79.6, 79.3,
79.2, 74.3, 73.9, 73.8, 73.5, 73.3, 72.8, 65.8, 65.7, 34.4, 34.3, 34.2, 34.1, 30.5, 30.3, 28.5,
28.0, 27.6, 27.6, 27.0, 26.9, 26.3, 26.2, 26.1, 26.0, 25.9, 25.8, 19.4, 19.3, 18.4, 18.3, 18.2,
18.1, 18.0, 7.5, 7.3, -3.8, -3.9, -4.0, -4.1, -4.3, -4.5, -4.6, -4.7, -4.8; IR (thin film) 2955,

2930, 2856, 1471, 1429, 1361, 1253, 1113, 1074, 1005, 939,835, 775, 702 cm].

292

 

 

tBuLi, -100 °C OH
MgBr2-OEt2 __ OPMB
1/\/=”W\0PMB +—
Cy-CHO
IV-48 _40 0C, 70% lV-81

Alcohol IV-81 was prepared using the same representative procedure as described
above for IV-SO. Thus, 1.14 g (3.04 mmol) of iodide IV-48 afforded 766 mg (70%) of

alcohol IV-81.
Partial data for IV-81: 1H NMR (500 MHz, CDCl,) 0 7.24 (d, J = 8.8 Hz, 2 H), 6.85 (d, J

= 8.6 Hz, 2 H), 5.41—5.34 (m, 2 H), 4.40 (s, 2 H), 3.78 (s, 3 H), 3.44332 (m, 3 H), 2.12-

1.98 (m, 4 H), 1.78-0.99 (m, 17 H).

 

OH AD mix-or OH
WOPMB MeSOZNHZ WOPMB
‘BuOH : H20 ”0 0”
1V-31 (1 31 ). 80% lV-82

AD-mix-Ot (700 mg) was dissolved in 1 : l 'BuOH : H20 (5 mL). To this clear,

orange solution, methane sulfonamide (47.5 mg, 0.50 mmol) and potassium osmate (1
mg) were added and stirred until all the solids dissolved. The solution was then cooled to
0 °C upon which olefin IV-81 (180 mg, 0.50 mmol) was added in one portion. The
reaction was vigorously stirred for 16 h after which solid sodium sulfite (750 mg) was

then added at the same temperature. The mixture was warmed to room temperature and
stirring was continued for 45 min. EtOAc (20 mL) and H20 (5 mL) were added and the

layers were separated. The aqueous layer was extracted with EtOAc (4x20 mL),

combined organic layers were dried over N32504, concentrated and the crude product

293

was purified by column chromatography (EtOAc). Triol I-82 was isolated in 80% yield

(158 mg) as a colorless oil.
Partial data for IV-82: 1H NMR (500 MHz, CDCl,) 0 7.25 (d, J = 8.5 Hz, 2 H), 6.86 (d, J

= 8.6 Hz, 2 H), 4.93-4.83 (m, 1 H), 4.41 (s, 2 H), 3.92—3.84 (m, 1 H), 3.80 (s, 3 H), 3.57-

3.52 (m, 1 H), 2.06 (d, J: 12.1 Hz, 1 H), 1.92-0.87 (m, 20 H).

OH 1.MeC(OMe)3 / PPTS, rt OAc

A 4
HO OH 2. BF3'OE12 (10 mol%)

-30 °c to 0 °C, CH 01
lV-82 2 2 lV-83; R = PMB (40%)
lV-84; R = H (40%)

 

PPTS (0.5 mg, 1.98 umol) was added to a solution Of triol IV-82 (80 mg, 0.20
mmol) and trimethylorthoacetate (33 0L, 0.22 mmol) in CHZCI2 (1.5 mL) at 0 °C. After 1

h, the volatiles were removed under reduced pressure and the residue was taken up in

CHzCl2 (1 mL). Upon cooling this solution to —30 °C, BF3°OEt2 (2.7 uL, 0.02 mmol) was

added and the reaction was warmed to 0 °C over 30 min. Saturated NaHCO3 solution (2
mL) was slowly added, the layers were separated and the aqueous layer was extracted
with CHZCI2 (3x20 mL). Combined organic layers were dried (NaZSO4), concentrated
under reduced pressure and the crude product was purified by flash column

chromatography (2% EtOAc in hexanes) to furnish cyclized products IV-83 (35 mg,

40%) and IV-84 (25 mg, 40%).

Partial data for IV-83: 1H NMR (500 MHz, 0001,) 6 7.22 (d, J = 8.4 Hz, 2 H), 6.85 (d, J
= 8.6 Hz, 2 H), 4.90-4.83 (m, 1 H), 4.39 (s, 2 H), 3.91-3.83 (m, 1 H), 3.78 (s, 3 H), 3.57-
3.50 (m, 1 H), 3.40 (t, J = 6.4, 2 H), 2.05 (s, 3 H), 1.92-1.80 (m, 3 H), 1.70-1.52 (m, 10

H), 1.40—1.32 (m, 3 H), 1.28-1.15 (m, 4 H), 0.08-0.04 (m, 1 H); 13C NMR (125 MHz,

294

CDCl,) 0 171.1, 79.9, 79.8, 75.6, 75.4, 68.5, 68.4, 68.3, 31.9, 31.3, 28.1, 26.1, 25.2, 22.7,

21.4, 21.3,14.2,14.1,

 

NaH, BnBr
HOWOH HOWOBn
TBAI, THF
lV-88 60 °C, 73% lV-89

To a slurry Of NaH (7 g, 0.18 mol) in THF (300 mL), 1,6 hexanediol IV-88 (20 g,
0.17 mol) was added at 0 °C and stirred for 1 h while warming to rt. Benzyl bromide (20

mL, 0.17 mmol) was the added drOp wise followed by TBAI (2.6 g). The reaction was

heated to 60 °C for 15 h. After cooling to room temperature H20 (150 mL) was carefully
added. The layers were separated, aqueous layer was extracted with EtQO (3x300 mL)

and the combined organic layers after drying (MgSO4) were concentrated. Monobenzyl
ether IV-89 was obtained as clear oil (35.4 g, 78%) after chromatographic purification
(30% EtOAc in hexanes). This material was spectroscopically identical to a previously

reported compound.82

Data for IV-89: 1H NMR (500 MHz, 0001,) 6 7.35-7.25 (m, 5 H), 4.50 (s, 2 H), 3.56 (t,
J = 6.6 Hz, 2 H), 3.47 (t, J = 6.6 Hz, 2 H), 1.63 (quint, J = 6.6 Hz, 2 H), 1.54 (quint, J =
7.0 Hz, 2 H), 1.38-1.32 (m, 4 H); 130 NMR (125 MHz, 0001,) 6 138.8, 128.6, 127.9,
127.8, 73.1, 70.6, 62.8, 32.9, 29.9, 26.2, 25.9; [R (thin film) 3393, 3063, 2933, 2859,
1951, 1874, 1810, 1603, 1454, 1363, 1309, 1251, 1205, 1099, 1028, 909,735,675 cm";

HRMS (El) calcd for C13H2002, 208.1458 m/z (M)+; observed, 208.1463 m/z.

 

295

To a slurry of PCC (31.6 g, 0.15 mol) in CH2C12 (300 mL), a solution of alcohol
IV-89 (20.4 g, 98.1 mmol) in CHZCIZ (100 mL) was added at room temperature under N2

with vigorous stirring. After 2 h, anhydrous Etzo (400 mL) was added and the reaction

mixture was filtered through a celite pad. The filtrate was concentrated and the crude
brown oily material was purified by flash column chromatography (10% EtOAc in
hexanes) to afford aldehyde IV-90 as a clear liquid (16.6 g, 82%).

Data for IV-90: 1H NMR (500 MHz, 0001,) 6 9.75 (t, J = 2.2 Hz, 1 H), 7.39—7.20 (m, 5
H), 4.49 (s, 2 H), 3.47 (t, J = 6.4, 2 H), 2.40-2.48 (m, 2 H), 1.72-1.61 (m, 4 H), 1.46-1.38
(m, 2 H); ”C NMR (125 MHz, CDCl,) 0 202.8, 138.8, 128.6, 127.8, 127.7, 73.1, 70.2,
44.0, 29.7, 26.0, 22.1; IR (thin film) 3031, 2936, 2860, 2720, 1954, 1875, 1724, 1453,

1409, 1363, 1391, 1101, 1026, 906, 737, 703 cm‘l; HRMS (E1) calcd for 0,,H,,O,,

206.1307 m/z (M)+; observed, 206.1309 m/z.

 

1. KHMDS
2. TMSCI OBn
4' - A /\/:_W\/
PhsP/V\OH Br - HO
4. IV-QO
3. AcOH :HZO :THF IV-91

(6:3 :1).0°C,83%

KHMDS (175 mL of 0.5 M solution in toluene, 87.5 mmol) was added to 8 ~20
°C slurry of 3-hydroxypr0pyltriphenylphosphonium bromide (17.6 g, 43.7 mmol) in THF
(55 mL). The mixture was brought to room temperature and stirred for 1 h. After cooling
back to 0 °C, TMSCI (5.8 mL, 43.7 mmol) was added and stirring was continued at the
same temperature for 15 min. The reaction was then cooled to —78 °C upon which a THF

solution of aldehyde lV-90 (5 g, 24.3 mmol in 40 mL) was. The reaction was warmed to

—10 °C over 1 h and then treated with AcOH : H20 : THF (6 : 3: l, 250 mL). After 15 h

stirring at room temperature the reaction mixture was neutralized by saturated NaHCO3.

296

The aqueous layer was extracted with EtOAc (3x 400 mL), combined organic layers were
dried over NaZSO4, concentrated and purified by column chromatography (10% EtOAc in
hexanes) to secure the homoallylic alcohol IV-91 (9 g, 83%). The sample contained < 5%
of the Z isomer; however the exact ratio could not be determined due to overlapping

signals in the 1H NMR Spectrum.

Data for IV-91: 1H NMR (500 MHz, CDCl,) 0 7.37-7.25 (m, 5 H), 5.55-5.51 (m, 1 H),
5.39-5.34 (m, l H), 4.50 (s, 2 H), 3.60 (t, J = 6.9 Hz, 2 H), 3.47 (t, J = 6.9 Hz, 2 H), 2.32-
2.28 (m, 2 H), 2.08-2.04 (m, 3 H), 1.66-1.60 (m, 2 H), 1.42-1.37 (m, 4 H); l3C NMR (125
MHz, CDCl,) 0 175.5, 138.8, 133.0, 128.6, 127.9, 127.7, 125.5, 73.1, 70.6, 62.4, 30.9,
29.8, 29.7, 27.5, 26.0, 20.9; IR (thin film) 3386, 3028, 2861, 2063, 1950, 1872, 1809,

1714, 1654, 1605, 1453, 1366, 1250, 1204, 1050, 872, 735, 695 cm“; HRMS (El) calcd

for C16H2402, 248.1776 m/z (M)+; observed, 248.1769 m/z.

 

1. MsCl, Et3N
— W
H oWWOB" ‘
2. Nal, acetone
”'91 reflux, 77% “"3"

A solution of alcohol IV-91 (9.1 g, 36.7 mmol) in CHZCI2 (140 mL) was cooled

to 0 °C. To this, mesyl chloride (8.55 mL, 0.11 mol) and triethyl amine (17 mL) were
added and stirring was continued at the same temperature for 30 min. The reaction was
quenched with H20 (100 mL). The aqueous layer was extracted with CHZCIZ (3x200 mL)
and combined organic layers were concentrated. The residue and sodium iodide (25 g,

0.17 mol) were taken up in acetone (150 mL) and refluxed for 2 h. Upon cooling to room

temperature, the reaction was treated with saturated sodium sulfite until it became

297

colorless. The aqueous layer was extracted with EtOAc (3x200 mL). Chromatographic
purification (3% EtOAc in hexanes) of the crude product obtained by concentration of the
organic portion afforded iodide IV-87 (10.1 g, 77%).

Data for IV-87: 1H NMR (500 MHz, CDC1,) o 7.36-7.28 (m, 5 H), 5.55-5.52 (m, 1 H),
5.35-5.32 (m, l H), 4.52 (s, 2 H), 3.48 (t, J = 6.6 Hz, 2 H), 3.14 (t, J = 7.3 Hz, 2 H), 2.66-
2.61 (m, 2 H), 2.06-2.04 (m, 2 H), 1.66-1.63 (m, 2 H), 1.42-1.39 (m, 4 H); l3C NMR (125
MHz, CDCl,) 0 138.9, 132.7, 128.6, 128.2, 127.9, 127.7, 73.1, 70.6, 31.8, 29.9, 29.6,

27.7, 26.1, 5.7; IR (thin film) 3028, 3009, 2932, 2855, 2791, 1920, 1850, 1790, 1495,

1454, 1361, 1242, 1169, 1105, 1028, 734, 698 cm"; HRMS (El) calcd for CI6H23IO,

358.0794 m/z (M)+; Observed, 358.0800 m/z.

1. LDA (2 eq.)

 

2 o O
' L\ PhS
PhS/YOH *‘ {i
o 3. PTSA, PhH
88%
lV-110 lV-11‘l

Lactone IV-lll was prepared as a mixture of diastereomers (3 : 2) according to

the reported procedure and the spectral data of our sample matched the reported data.75

 

 

. TBSO OH
tBuLi, -100 °C IV-86 ? o. OBn
i/\/=\/\/\/OB” 4 ”A 15 '1 — 4
MgBrZ-OEtg M931201512 '-.
E, o Etgo, -40 °C OTBS
lV-87 2 85% IV-85
raso
-' o.,_ CHO
%
bras
lV-86

 

 

 

t-BuLi (7.6 mL of 1.3 M solution in pentane, 9.94 mmol) was added drop wise to

pre-cooled (-100 °C) Etzo (18 mL). To this, a solution of iodide IV-87 (1.78 g, 4.97

298

 

mmol) in Etzo (14 mL) was added over 10 min. After stirring for 5 min at —-100 °C to

-90 °C, MgBrZOOEtz in Etzo (12.4 mL of 1.0 M solution (freshly prepared as described
on page 288), was added and the mixture was warmed 0 °C over 1 h. Meanwhile, a

solution of aldehyde IV-86 (1 g, 1.63 mmol) in EtZO (18 mL) was cooled to —40 °C.

MgBrZOOEt2 in EtZO (5.0 mL of 1.0 M solution, 5.0 mmol) was added and stirred for 10
min. To this pre-complexed aldehyde, solution of the above mentioned Grignard reagent
was cannulated at —40 °C and the reaction mixture was stirred overnight at the same

temperature. The reaction was then quenched by slow addition Of saturated NH4C1
solution (20 mL) and H20 (50 mL). The aqueous layer was extracted with EtZO (3x100

mL). Combined organic layers were dried (NaZSO4), concentrated under reduced pressure
to afford a yellow oil. Purification by column chromatography (2% EtOAc in hexanes)
furnished the adduct 1-85 (1.17 g, 85%) as a single diastereomer.

Data for 1-85: [611,20 —17.1 (c 0.97, 0H01,) 1H NMR (500 MHz, 0001,) 6 7.33—7.25 (m,
5 H), 5.39-5.33 (m, 2 H), 4.52 (s, 2 H), 4.50-4.39 (m, 1 H), 4.25-4.21 (m, l H), 3.8 (dt, J
= 4.3, 8.8 Hz, 1 H), 3.68-3.62 (m, 2 H), 3.46 (t, J = 6.6 Hz, 2 H), 2.96 (8 (br), 1 H), 2.29-
2.05 (m, 4 H), 1.88-1.86 (m, 2 H), 1.65-1.47 (m, 4 H), 1.38-1.23 (m, 36 H), 0.91 (s, 9 H),
0.90 (s, 9 H), 0.88 (t, J = 7.0 Hz, 3 H), 0.11 (s, 3 H), 0.10 (s, 3 H), 0.08 (s, 3 H), 0.06 (s, 3
H); 130 NMR (125 MHz, 0001,) 6 139.0, 130.3, 129.7, 128.5, 127.8, 127.6, 85.0, 80.5,
75.0, 74.6, 73.1, 70.7, 38.2, 33.4, 32.9, 30.1, 29.9, 29.8, 29.6, 27.4, 26.2, 26.1, 26.0, 25.8.
23.7, 22.9, 18.4, 18.1, 14.3, -4.0, -4.1, -4.3, -4.8; IR (thin film) 3596, 3521, 3031, 3004,

2928, 2856, 1464, 1406, 1389, 1362, 1256, 1190, 1076, 1005, 955, 939, 835, 808,775,

299

733, 696, 662 cm"; HRMS (El) calcd for C51H9605S12, 844.6796 m/z (M)+; observed,

 

844.6789 m/z.
AD-miX-Ot
TBSQ OH MeSOQNHg TBSQ OH OH
.- 01“ __ OBn : 7 OI“ 12 Dan
15 4 K20504'2H20 15 69H 4
bores 0 °C' 88% horas
lV-94 I IV'103 :
lV-85 (5 3 1) IV-94
+
TesQ OH OH
' O 12 9 OBn
15 -. OH 4
’0138
IV-103

AD-mix-Ot (1.26 g) was dissolved in 1 : l tBuOH : H20 (13 mL). To this clear,

orange solution, methane sulfonamide (86 mg, 0.9 mmol) and potassium osmate (19 mg)
were added and stirred until all the solids dissolved. The solution was then cooled to 0 °C
upon which Olefin IV-85 (760 mg, 0.90 mmol) was added in one portion. The reaction
was vigorously stirred for 16 h after which solid sodium sulfite (1.35 g) was then added

at the same temperature. The mixture was warmed to room temperature and stirring was
continued for 45 min. EtOAc (50 mL) and H20 (20 mL) were added and the layers were
separated. The aqueous layer was extracted with EtOAc (4x50 mL), combined organic
layers were dried over NazSO4, concentrated and the crude product was purified by

column chromatography (8% EtOAc in hexanes). The desired diastereomer I-94 was

isolated in 73% (577 mg) yield as a colorless Oil.
Data for L94: [0111,20 —16.2 (c 0.87, CHCI3) 1H NMR (500 MHz, CDC13) o 7.33-7.24 (m,

5 H). 4.49 (s, 2 H), 4.42-4.41 (m, l H), 4.22-4.20 (m, l H), 3.88-3.87 (m, 1 H), 3.71-3.69
(m, l H), 3.64-3.58 (m 3 H), 3.47 (t, J = 6.6 Hz, 2 H), 3.23 (3 (br), 1 H), 3.03 (8 (br), 1

H), 1.88-1.86 (m 2 H), 1.68-1.22 (m, 44 H), 0.90 (s, 9 H), 0.89 (s, 9 H), 0.88 (t, J = 7.0
300

Hz, 3 H), 0.10 (s, 3 H), 0.09 (s, 3 H), 0.07 (s, 3 H) 0.06 (s, 3 H); 130 NMR (125 MHz,
0001,) 6 138.9, 128.5, 127.8, 127.7, 85.0, 80.6, 74.9, 74.8, 74.5, 74.4, 73.1, 71.3, 70.6,
38.3, 32.9, 32.1, 31.9, 30.1, 29.9, 29.8, 29.6, 27.7, 26.5, 26. 1, 25.9, 25.8, 22.9, 18.4, 18.1,
14.3, 4.0, 44.2, -43, -49; IR (thin film) 3596, 3521, 3031, 3004,2928, 2856, 1464, 1406,
1389, 1362, 1256, 1190, 1076, 1005, 955, 939, 835, 808,775,733, 696, 662 cm"; HRMS
(ES) calcd for 0,,H,,O,Si,, 879.6929 m/z (M+H)+; observed, 879.6931 m/z.

TBSQ OH OH 1. MeC(OMe)3 / PPTS TBSO OAc
’ . OBn CH2012. '1 ?

       
   

-., 0H 2. BF3-OE12, rt -.,
OTBS 91 % OTBS

lV-94 lV-95
PPTS (6 mg, 0.02 mmol) was added to a solution of triol IV-94 (200 mg, 0.22
mmol) and trimethylortho acetate (36 (LL, 0.23 mmol) in CHZCI2 (3 mL) at rt. After
complete consumption of the trio] (ca. 5 min, as judged by TLC), a solution of BF3°OEt2
(8 (1L, 0.06 mmol) in CHZCI2 (1 mL) was rapidly added to the reaction. After 10 min, the

reaction was slowly poured into saturated NaHCO, solution (5 mL) and the aqueous layer

was extracted with CHzCl2 (3x20 mL). Combined organic layers were dried (NaZSO4),

concentrated under reduced pressure and the crude product was purified by flash column
chromatography (2% EtOAc in hexanes) to furnish bis-THF acetate IV-95 (187 mg,
91%) as a clear oil.

Data for 1-95: [(11020 -293 (c 0.47, 0H01,) 'H NMR (500 MHz, 0001,) 6 7.35—7.25 (m,

5 H), 4.91-4.87 (m, 1 H), 4.48 (s, 2 H), 4.29-4.18 (m, 3 H), 4.05-4.01 (m, l H), 3.74-3.71
(m, 1 H), 3.63 (dd, J = 3.6, 7.7 Hz, 1 H), 3.44 (t, J = 6.4 Hz, 2 H), 2.08-2.00 (m, l H),

2.04 (s, 3 H), 1.96-1.80 (m, 4 H), 1.66-1.51 (m, 6 H), 1.48-1.18 (m, 35 H), 0.88 (s, 9 H),

301

0.87 (s, 9 H), 0.86 (t, J = 7.0 Hz, 3 H), 0.11 (s, 3 H), 0.09 (s, 3 H), 0.07 (s, 6 H); 13C
NMR (125 MHZ, CDC13) 6 171.1 138.9, 128.5, 127.8, 127.6, 85.5, 81.0, 79.3, 79.2, 75.8,

75.4, 73.7, 73.1, 70.6, 36.8, 32. 1, 32.0, 31.0, 30.1, 29.9, 29.6, 28.6, 27.9, 26.4, 26.2, 25.9,

25.6, 22.9, 21.4, 18.4, 18.1, ~39, —4.0, -4.5, —4.8; IR (thin film) 2926, 2854, 1739, 1463,

1354, 1244, 1100, 1056, 940, 833, 775 cm’l; HRMS (ES) calcd for C53H9807Siz,

903.6929 m/z (M+H)+; observed, 903.6913 m/z.

   
 

 

TBSQ OAc TBSQ OH
: OBn K2003 t T o". 0% 8 OBn
., 5 MeOH, 95% ‘5 ,,l 5
0038 ores
lV-95 lV-98

Acetate IV-95 (440 mg, 0.49 mmol) was dissolved in MeOH (7 mL). Solid

KZCO3 was added to this solution and the heterogeneous mixture was stirred vigorously
at room temperature for 17 h. The reaction was then diluted with CHZCI2 (20 mL) and
washed with NaHCO3 (5 mL) and H20 (10 mL). The aqueous layers were mixed and

extracted with CHZCl2 (3x15 mL). The combine organic layer was dried using NaZSO4,

the solvent evaporated and the crude product was purified by flash column
chromatography (5% EtOAc in hexanes). Bis-THF IV-98 was obtained in 95% yield

(401 mg) as a clear oil.

Data for 1-98: 161,,20 —26.1 (c 0.92, 0H01,) lH NMR (500 MHz, 01301,) 6 7.33-7.26 (m,
5 H), 4.49 (s, 2 H), 4.27-4.23 (m, 2 H), 4.16 (app q, J = 7.2 Hz, 1 H), 3.77 (app q, J = 6.9
Hz, 1 H), 3.72-3.70 (m, 1 H), 3.65 (dd, J = 3.2, 7.5 Hz, 1 H), 3.47 (t, J = 6.6 Hz, 3 H),
3.41-3.39 (m, 1 H), 2.57 (d, J = 3.1 Hz, 1 H), 1.97-1.22 (m, 43 H) 0.89 (s, 9 H), 0.88 (8,9

H), 0.90 (t, J = 7.0 Hz, 3 H), 0.09 (s, 3 H), 0.08 (s, 3 H), 0.06 (s, 6 H); 13C NMR (125

302

 

MHz, CDC13) 6 139.0, 128.5, 127.8, 127.6, 87.5, 82.5, 81.1, 79.0, 745,741, 73.7, 73.1,

70.6, 37.0, 33.6, 32.1, 30.1, 29.9, 29.8, 29.6, 29.0, 28.4, 26.5, 26.2, 26.1, 25.9, 25.7, 22.9,
18.4, 18.1, 14.3, —3.9, —40, —4.5, ~48; IR (thin film) 3581, 3476, 2926, 2854, 1805,

1755, 1463, 1361, 1253, 1100, 1057, 939, 835, 775, 697 cm’l; HRMS (ES) calcd for

0,,H9606512, 861.6824 m/z (M+H)+; observed, 861.6819 m/z.

 

TBSQ OH T Taso ores
t On. O’a,9 8 Dan TBSO f t ‘ O’o. 0". QB”
15 . 5 2,6 lutidine ‘5 -. 5
bras CH20'2. 0 °C bras
97%
lV-98 lV-97

To a 0 °C solution of alcohol IV-98 (172 mg, 0.20 mmol) in CHzCl2 (5 mL), 2,6
lutidine (0.15 mL, 1.2 mmol) and TBSOTf (0.14 mL, 0.6 mmol) were added in that order.

After 30 min at the same temperature, saturated NaHCO3 solution (2 mL) was added and
the layers were separated. The aqueous layer was extracted with CHZCIZ (3x15 mL),

combined organic layers were dried over NaZSO4, and concentrated under reduced

pressure to afford a crude oil. Upon purification of the oil by column chromatography

( 1% EtOAc in hexanes), tris-TBS ether IV-97 was obtained in 97% yield (189 mg).

Data for 1-97: [61,,20 —30.5 (c 0.83, 0H01,) 1H NMR (500 MHz, 0D01,) 6 7.34-7.26 (m,
5 H), 4.50 (s, 2 H), 4.31427 (m, 2 H), 4194.15 (111, 1 H), 3.97-3.93 (m, 1 H), 3.78-3.76
(m, 1 H), 3.75—3.71 (dd, J = 3.5, 7.7 Hz, 1 H), 3.46 (t, J = 6.6 Hz, 3 H), 2.00—1.16(m,4
H), 1.57-1.2 (m, 42 H), 0.90 (t, J = 7.0 Hz, 3 H), 0.89 (s, 9 H), 0.88 (s, 9 H), 0.87 (8,9 H)
0.08 (s, 3 H), 0.06 (s, 9 H), 0.05 (s, 3 H), 0.04 (s, 3 H); 130 NMR (125 MHz, 0D01,) 6
139.0, 128.5, 127.8, 127.6, 85.9, 81.7, 80.5, 79.2, 74.4, 74.0, 73.7, 73.1, 70.7, 36.3, 32.1,

31.9, 31.5, 30.1, 30.0, 29.9, 29.6, 28.7, 26.7, 26.6, 26.4, 26.2, 26.1, 25.9, 22.9, 18.4, 18.3,

303

 

18.1, 14.3, —3.9, —4.1, —4.4, —4.8; IR (thin film) 2904, 2855, 1990, 1871, 1463, 1366,

1254, 1098 cm"; HRMS (ES) calcd for C57H11006813, 975.7689 m/z (M+H)+; observed,

 

975.7697 m/z.
rsso ores H2, pd / c raso ores
’ 0,, 0,, 080 g 7 0,, 0., OH
‘5 _ 5 EtOAc :lPrOH ‘5 , 5
bras (1 :1), rt. 92% bras
lV-97 lV-113

Benzyl ether IV-97 (390 mg, 0.40 mmol) was dissolved in 1 : 1 EtOAc : iPrOH

(20 mL). To this solution, 10% Pd-C (111 mg) was added and the mixture was stirred
vigorously under H2 (1 atm). The hydrogenolysis was complete in 1 h after which the

reaction was filtered through a celite pad. The filtrated was concentrated and the crude
product was purified by flash column chromatography (5% EtOAc in hexanes) to furnish
alcohol IV-ll3 in 92% yield (326 mg) as a colorless oil.

Data for 1.113: [(11020 —29.4 (c 0.83, 0H01,) 1H NMR (500 MHz, 0D01,) 6 4.30-4.26
(m, 2 H), 4.17—4.13 (m, 1 H), 3.97-3.93 (m, 1 H), 3.76-3.72 (m, 2 H), 3.65-3.63 (m, l H),
3.62 (t, J = 6.6 Hz, 2 H), 2.00-1.64 (m, 4 H) 1.59-1.12 (m, 44 H), 0.89 (t, J = 7.0 Hz, 3
H), 0.88 (s, 9 H), 0.87 (s, 9 H), 0.86 (s, 9 H), 0.07 (s, 3 H), 0.05 (s, 3 H), 0.04 (s, 6 H),
0.04 (s, 6 H); 13C NMR (125 MHz, CDCl,) 0 85.9, 81.6, 80.5, 79.2, 74.3, 74.0, 73.6,

63.2, 36.3, 33.0, 32.1, 31.9, 31.5, 29.9, 29.8, 29.7, 29.6, 28.7, 26.5, 26.4, 26.2, 26.1, 26.0,
25.9, 22.9, 18.4, 18.3, 18.1, 14.3, 1.2, —3.9, —4.0, ~41, —4.4, —4.8; IR (thin film) 3385,

2926, 2855, 1600, 1463, 1360, 1255, 1079, 835, 774 cm"; HRMS (ES) calcd for

C50H1m06313, 885.7219 m/z (M+H)+; observed, 885.7217 m/z.

304

 

TBSQ ores TBSQ OTBS

 

0., 0., OH PPh3, imid. 0,, O", 1
‘5 .. 5 I2, toluene ‘5 ._ 5
’OTBs rt, 60% 0783
IV-1 13 1V-1 14

Alcohol IV-133 (304 mg, 0.34 mmol), triphenylphosphine (223 mg, 0.85 mmol)
and imidazole (61 mg, 0.90 mmol) were dissolved in toluene (12 mL). Upon addition of
iodine (231 mg, 0.91 mmol) the clear, colorless solution turned yellowish brown and
turbid. After 1 h vigorous stirring at room temperature, saturated sodium sulfite solution
was added to the reaction until the yellowish brown color disappeared. The layers were
separated and the aqueous layer was washed with EtOAc (3x15). After evaporation of the
solvent form combined and dried (Na2S04) organic layers a gummy material was
obtained. Purification of the crude material by column chromatography (3% EtOAc in
hexanes) afforded iodide IV-l 14 (203 mg, 60%) as a colorless oil.

Data for 1.114: [0.1020 -277 (c 1.09, 0H01,) 1H NMR (500 MHz, 0D01,) 6 4.32—4.29
(m, 2 H), 4.19-4.15 (m, 1 H), 3.98-3.94 (m, 1 H), 3.78-3.72 (m, 2 H), 3.65 (dd, J = 3.5,
7.7 Hz, 1 H), 3.18 (t, J = 6.6 Hz, 2 H), 2.00-1.23 (m, 46 H), 0.90 (t, J = 7.0 Hz, 3 H), 0.89
(s, 9 H), 0.88 (s, 9 H). 0.87 (s, 9 H), 0.08 (s, 3 H), 0.07, (s, 3 H), 0.06 (s, 6 H), 0.05 (s, 6
H); l3C NMR (125 MHz, CDCl,) 0 85.9, 81.6, 80.6, 79.3, 77.5, 77.2, 77.0, 74.2, 74.0,
73.6, 36.3, 33.8, 32.1, 31.7, 31.5, 31.0, 30.1, 29.9, 29.8, 29.6, 28.7, 26.5, 26.4, 26.2, 26.1,
26.0, 25.2, 22.9, 18.4, 18.3, 18.1, 143, 7.3, —3.9, —4.0, —4.1, —4.3, —4.4, —4.8; IR (thin

film) 2925, 2854, 1597, 1462, 1359, 1253, 1076, 835, 775 cm"; HRMS (ES) calcd for

C50H1030515i3, 995.6236 m/z (M+H)+; observed, 995.6259 m/z.

305

    

 

o
Phs LDA, 1v-114 TBS? ores O
o 2. O
THF : HMPA 4

0 °C to rt '2
(4 : 1), 83% OTBS
1v-111 IV-115

A solution of diisopropylamine (5.8 piL, 0.06 mmol) in THF (0.5 mL) was cooled to —78
°C and n-BuLi (24 uL of 2.5 M solution, 0.06 mmol) was added to it. After 15 min,
lactone IV-lll (12.6 mg, 0.06 mmol) in THF (0.4 mL) was added and stirring was
continued for 30 min during which time the solution was warmed to 0 °C. Iodide lV-llS

(30 mg, 0.03 mmol) was then added as a solution in 1 : 1 THF : HMPA (0.5 mL). The
reaction was allowed to attain room temperature. After 15 h, H20 (1 mL) and EtOAc (5
mL) were added and the layers were separated. The aqueous layer was extracted with
EtOAc (3x5 mL), combined organic layers were dried (NaZSO4), concentrated and the
crude product was purified by column chromatography (1% —- 3% EtOAc in hexanes) to
afford sulfide IV-l 15 as a mixture of diastereomers (27 mg, 83%).

Data for I-115: ]H NMR (500 MHz, CDCl,) 0 7.56-7.52 (m, 2 H), 7.43-7.33 (m, 3 H),
4.53-4.46 (m, 1 H), 4.32-4.28 (m, 2 H), 4.19-4.15 (m, 1 H), 3.97-3.93 (m, l H), 3.78-3.72
(m, 2 H), 3.65 (dd, J = 3.5, 7.7 Hz, 1 H), 2.52 (dd, J = 3.5, 7.7, 1 H), 2.01-1.91 (m, 2 H),
1.90-1.21 (m, 50 H), 0.90 (t, J = 7.0 Hz, 3 H), 0.89 (s, 9 H), 0.88 (s, 9 H), 0.87 (s, 9 H),
0.08 (s, 3 H), 0.07 (s, 3 H), 0.06 (s, 12 H); 13C NMR (125 MHz, CDC13) 0 177.3, 137.3,
137.1, 130.8, 130.1, 129.9, 129.2, 129.1, 85.9, 81.6, 80.6, 79.3, 74.3, 74.0, 73.7, 73.6,
73.4, 56.4, 40.3, 36.7, 36.3, 32.1, 31.8, 31.5, 30.2, 30.1, 29.9, 29.6, 28.7, 26.5, 25.4, 26.2,

26.2, 26.0, 25.0, 22.9, 21.7, 18.4, 18.1, 14.3, —3.9, —4.0, —41, —4.3, —4.4, -4.8; IR (thin

film) 2926, 2854, .1770, 1464, 1385, 1360, 1255, 1184, 1068, 968, 939, 835, 806, 775,

306

705, 692 cm"; HRMS (ES) calcd for 0,,H,,,,o.,ssr,, 1075.7671 m/z (M+H)+; observed,

1075.7690 m/z.

TBSQ ores o 1. mCPBA TBSQ ores O
T CH2C12,0°C :

  
 

 

 

  

.1 2. toluene, reflux a
OTBS 83% (two steps) OTBS

lV-115 IV-116

   

To an ice cold solution of IV-llS (30 mg, 0.03 mmol) in CHZCl2 (1 mL), ca. 75%
mCPBA (6.8 mg, 0.03 mmol) in CHZCl2 (1 mL) was added drop wise. After 20 min,
saturated NaHCO, solution (1 mL) was carefully added and the layers were separated.

The aqueous layer was extracted with CHZCl2 (3x5 mL). The combined organic layers

were dried and concentrated to afford the corresponding sulfoxide. The crude sulfoxide
was taken up in toluene (2 mL) and heated to reflux for 4h. After cooling the solution to
room temperature, the solvent was evaporated under reduced pressure and the crude
material was purified by column chromatography (5% EtOAc in hexanes) to afford
IV-116 (24 mg, 83%).

Data for 1-116: (61,,” —17.9 (c 0.42, 0H01,) 1H NMR (500 MHz, 0D01,) 6 6.99 (d, J =

1.6 Hz, 1 H), 5.00—4.99 (m, 1 H), 4.31428 (m, 2 H), 4.19-4.15 (m, 1 H), 3.98-3.94 (m, 1
H), 3.78-3.74 (m, 2 H), 3.65 (dd, J = 3.5, 7.7 Hz, 1 H), 2.27 (t, J = 7.3 Hz, 2 H), 2.02-
1.12 (m, 49 H), 0.92088 (m, 30 H), 0.08-0.05 (m, 18 H); 130 NMR (125 MHz, 0D01,) 6
174.0, 149.0, 134.5, 85.9, 81.6, 806,793, 74.3, 74.0, 73.7, 36.3, 32.1, 31.7, 315,301,
29.9, 29.8, 29.7, 29.6, 28.7, 27.6, 265,264, 26.2, 26.1, 26.0, 25.4, 22.9, 19.4, 18.3, 18.1,

14.3, 1.2, —3.9, —4.0, -4.1, —4.4, —48; IR (thin film) 2954, 2927, 2854, 1761, 1463, 1361,

307

1319, 1257, 1081, 1026, 835, 802, 775 cm’l; HRMS (ES) calcd for C55H1()807Si31

965.7481 m/z (M+H)+; observed, 965.7480 m/z.

  
 

 

reso ores O
r I HF‘PY

 

THF, rt
OTBS 80%

lV-116 lV-117

   

To a solution of IV-116 (9 mg, 9.31 umol) in THF (0.5 mL) taken in a

polyethylene vial, HF°pyridine (32 11L) was added at room temperature. After stirring for
12 h, the reaction was neutralized by saturate NaHCO, solution. H20 (1 mL) and EtOAc
(5 mL) were added and the layers were separated. The organic layer was washed with

saturated CuSO4 (2x2 mL) and the combined aqueous layers were extracted with EtOAc

(3x5 mL). The organic layers were mixed, dried over NaZSO4, and the solvent was
evaporated to afford a waxy material. Sequential purification by column chromatography
(EtOAc, 10% MeOH in EtOAc) and HPLC (10% iPrOH in EtzO) triol lV-117 as a
colorless wax (4.6 mg, 80%).

Data for 1.117; [61,20 +3.2 (0 0.40, 0H01,) 1H NMR (500 MHz, 0D01,) 6 7.00 (d, J =

1.5 Hz, 1 H), 5.02-4.98 (m, 1 H), 4.46-4.43 (m, 1 H), 4.35 (dt, J = 3.4, 6.8 Hz, 1 H), 4.11—
4.08 (m, 1 H), 3.97—3.94 (m, 1 H), 3.81 (t, J = 3.1 Hz, 1 H), 3.75 (d, J = 5.4 Hz, 1 H),
3.43—3.40 (m, 2 H), 2.32 (d, J = 4.4 Hz, 1 H), 2.28 (dt, J = 1.5, 7.8 Hz, 2 H), 2.19 (d, J =

5.4 Hz, 1 H), 2.15-2.11 (m, l H), 2.06-1.98 (m, 3 H), 1.84 (ddd, J = 4.4, 9.5 Hz, 13.4 Hz,

1 H), 1.75-1.69 (m, 1 H), 1.57-1.26 (m, 43 H), 0.89 (t, J = 6.9, 3 H); 130 NMR (125

MHZ, CDC13) 6 174.0, 149.2, 134.4, 84.1, 83.3, 81.6, 79.3, 77.6, 74.7, 74.3, 73.6, 48.9,

39.0, 34.0, 33.6, 32.1, 29.9, 29.8, 29.7, 29.6, 29.4, 29.3, 28.0, 27.6, 25.8, 25.6, 25.3, 22.9

308

(multiple carbons), 19.4, 14.3; IR (thin film) 3404, 2917, 2850, 1749, 1590, 1465, 1319,

1072, 1045, 995, 873, 798, 719 cm"; HRMS (ES) calcd for C37H66O7, 623.4887 m/z

(M+H)+; observed, 623.4879 m/z.

 

o
PhS LDA, lV-114 T889 0 o ores o
0 e
0 cc to r1 i’ores
(4 I 1), 82°70

Sulfide IV-129 was prepared following the same procedure as for IV-115 using
lactone IV-112 (17.6 mg, 0.08 mmol) and iodide IV-114 (42 mg, 0.04 mmol). Other
reagents and solvents were used in appropriate proportions. IV-129 was obtained in 82%
yield (37 mg).

Data for 1-129: 1H NMR (500 MHz, CDCl,) 0 7.56—7.51 (m, 2 H), 7.40-7.33 (m, 3 H),
4.50—4.46 (m, 1 H), 4.30—4.26 (m, 2 H), 4.18-4.13 (m, 1 H), 3.96-3.92 (m, 1 H), 3.78—3.72
(m, 2 H), 3.65 (dd, J = 3.5, 7.7 Hz, 1 H), 2.52 (dd, J = 3.5, 7.7 Hz, 1 H), 2.01-1.91 (m, 2
H), 1.90-1.21 (m, 50 H), 0.90 (t, J = 7.0 Hz, 3 H), 0.89 (s, 9 H), 0.88 (s, 9 H), 0.87 (s, 9
H), 0.08 (s, 3 H), 0.07 (s, 3 H), 0.06 (s, 12 H); 130 NMR (125 MHz, 0D01,) 6 171.2,
137.1, 130.8, 130.2, 129.9, 129.2, 129.1, 85.9, 81.6, 80.5, 79.3, 74.3, 74.0, 73.7, 73.4,
56.4, 42.7, 40.4, 36.7, 36.3, 32.1, 31.8, 31.5, 30.2, 29.9, 29.8, 29.6, 28.7, 26.6, 26.4, 26.3,
26.2, 26.1, 26.0, 25.9, 24.9, 22.9, 21.7, 20.9, 18.3, 18.1, 14.3, —3.9, —4.0, —4. 1, —4.3, —4.4,

—4.8; IR (thin film) 2927, 2854, 1770, 1463, 1385, 1359, 1255, 1184, 1070, 939, 835,
775, 692 cm"; HRMS (ES) calcd for C62HH4O78813, 1075,7671 m/z (M+H)+; observed,

1075,7632 m/z.

309

Teso ores O 1. mCPBA

W CHZC'Z'OOC
15 4 O ;
PhS ._

'a -. 2. toluene, reflux -,
OTBS ’ 83% (two steps) OTBS

IV-129 1V -1 30

 

 

Oxidation of IV-129 to the corresponding sulfoxide and subsequent elimination
was carried out by the same procedure as described for IV-ll6. Thus, 42 mg of IV-129

afforded 33 mg (83%) of IV-130.
Data for I-130: [01],,20 —30.6 (c 0.68, CHC13) l11 NMR (500 MHz, CDCl,) 0 6.99 (d, J =

1.6 Hz, 1 H), 5.00—4.99 (m, 1 H), 4.31-4.28 (m, 2 H), 4.19—4.15 (m, 1 H), 3.98-3.94 (m, 1
H), 3.78—3.74 (m, 2 H), 3.65 (dd, J = 3.5, 7.7 Hz, 1 H), 2.27 (t, J = 7.3 Hz, 2 H), 2.02—
1.12 (m, 49 H), 0.92-088 (m, 30 H), 0.08-0.05 (m, 18 H); 130 NMR (125 MHz, 0D01,) 6
174.0, 149.0, 134.5, 85.9, 81.6, 80.6, 79.3, 74.3, 74.0, 73.7, 36.3, 32.1, 31.7, 31.5, 30.1,
29.9, 29.8, 29.7, 29.6, 28.7, 27.6, 26.5, 26.4, 262,261,260, 25.4, 22.9, 19.4, 18.3, 18. 1,

14.3, 1.2, —3.9, —4.0, —4.1, —4.4, -4.8', [R (thin film), 2927, 2859, 1761, 1463, 1359, 1319,

1257, 1080, 939, 835, 806,775 cm"; HRMS (ES) calcd for 0,,H,0,o,s1,, 965.7481 m/z,

(M+H)+', observed, 965.7473 m/z.

 

   

HF-Py
THF, rt
OTBS " 80%
1v-130 1v-118

Triol IV-118 was obtained by TBS ether removal of IV-130 using the same
procedure as for IV-117. Thus, 11 mg (0.01 mmol) of IV-130 furnished 5.6 mg of IV-
118 (80% yield).

Data for I—118: [61,20 —23.8 (c 0.50, 0H01,) 1H NMR (500 MHz, 0D01,) 6 7.00 (d, J =

1.5 Hz, 1 H), 5.02-4.98 (m, 1 H), 4.44-4.42 (m, 1 H), 4.34 (dt, J = 3.4, 6.8 Hz, 1 H), 4.11-

310

4.08 (m, l H), 3.97—3.93 (m, 1 H), 3.80 (t, J = 3.1 Hz, 1 H), 3.76 (d, J = 5.4 Hz, 1 H),
3.43-3.40 (m, 2 H), 2.42 (3 (br), 1 H), 2.34 (5 (br), 1 H), 2.28 ((11, J = 1.5, 7.8 Hz, 2 H),
2.15-2.11 (m, 1 H), 2.06-1.96 (m, 3 H), 1.83 (ddd, J = 4.4, 9.5, 13.4 Hz, 1 H), 1.75-1.69

(m, 1 H), 1.57—1.26 (m, 43 H), 0.88 (t, J = 6.9, 3 H); 130 NMR (125 MHz, 0D01,) 6

174.0, 149.2, 134.4, 84.1, 83.4, 81.6, 79.3, 77.6, 74.7, 74.3, 73.6, 48.9, 39.0, 34.0, 33.6,
32.1, 29.9, 29.8, 29.7, 29.6, 29.4, 29.3, 28.0, 27.6, 25.8, 25.6, 25.3, 22.9, 19.4, 143; IR
(thin film) 3374, 2919, 2848, 1756, 1465, 1439, 1319, 1201, 1076, 952, 873, 800, 721

cm"; HRMS (ES) calcd for C37H6607, 623.4887 m/z (M+H)+; observed, 623.4888 m/z.

1. MeC(OMe)3 / PPTS

TBSO OAc
CHQCIQ, rt '7 '

 

 

OBn _fi. O... r O 9 6 OBn
., 2. BF3-OE12, rt ‘5 ., 5
’ores 91% ’ores
lV-103 lV-131

Cyclization of triol IV-103 to bis-THF IV-l3l was carried using the same
procedure as for IV-95. Thus, 80 mg (0.09 mmol) of IV-103 produced 75 mg of IV-13l

(91%).

Partial data for 1.131; 1H NMR (500 MHz, 0D01,) 6 7.39—7.23 (m, 5 H), 487.481 (111, 1
H), 4.46 (s, 2 H), 4.28421 (m, 2 H), 4.03-3.91 (m, 2 H), 3.72-3.71 (m, 1 H), 3.63 (dd, J
= 3.3, 8.0 Hz, 1 H), 3.42 (t, J = 6.3 Hz, 2 H), 2.02 (s, 3 H), 1.90—1.22 (m, 46 H), 0.88 (t, J
= 7.0 Hz, 3 H), 0.85 (s, 9 H), 0.84 (s, 9 H), 0.04 (s, 6 H), 0.02 (s, 6 H); 130 NMR (125
MHz, 0D01,) 6 171.1, 138.9, 135.0, 129.8, 128.5, 127.9, 127.8, 127.7, 863,809,797.

79.6, 75.6, 74.3, 73.8, 73.1, 70.6, 36.7, 32.1, 31.8, 31.0, 30.1, 29.9, 29.8, 29.6, 27.9, 27.8,

26.8, 20.4, 26.3, 26.2, 25.9, 25.6, 22.9, 21.4, 18.4, 18. 1, 14.3, -3.8, -4.0, —4.5, -48.

311

 

“339 9A“ t K co MeOH TBSQ OTBS
? 0’" 1 O 9 6 OB" . 2 3’ > T on. 1 O 9 £3 OBn
‘5 . s 2. TBSOTf 15 . ,
"oras 2,61utidine, 01-12012 20,88
0 °C, 92%
1v-131 1v-132

Tri-TBS ether IV-132 was prepared by basic hydrolysis and subsequent TBS
protection of acetate IV-l31 following the same procedure as described for acetate

IV-95. 75 mg (92%) of IV-132 was obtained from 75 mg (0.08 mmol) of IV-13l.

Partial data for I-132: 1H NMR (500 MHz, CDCl,) 6 7.37-7.22 (m, 5 H), 4.49 (s, 2 H),
4.34-4.27 (m, 2 H), 4.02—3.97 (m, 1 H), 3.94-3.90 (m, 1 H), 3.82-3.78 (m, 1 H), 3.72-3.70
(m, 1 H), 3.65 (dd, J = 3.1, 8.1 Hz, 1 H), 3.46 (t, J = 6.6 Hz, 2 H), 1.92-1.84 (m, 4 H),
1.82-1.12 (m, 42 H), 0.90-0.87 (m, 30 H), 0.08-0.02 (m, 18 H); 13C NMR (125 MHz,
CDC13) 6 139.0, 128.5, 127.8, 127.6, 86.9, 81.7, 80.5, 79.6, 74.2, 73.9, 73.8, 73.1, 70.7,

36.3, 32.1, 31.9, 31.4, 30.2, 30.0, 29.9, 29.8, 29.7, 29.6, 28.2, 26.7, 26.5, 26.4, 26.2, 26.1,

26.0, 25.9, 25.8, 22.9, 18.4, 18.1, 14.3, —3.8, -4.0, -4.1, -4.3, -4.4, -4.9.

 

 

o
Phs O 1. LDA, 1v-121, rt rose @183 om5
{a 2.mCPBA,O°C _ ‘17 o,“ '2 O 9 g \
' 15
3. toluene, reflux wines
lV-1 11 (68% three steps)
IV-133
TBSQ QTBS
: 6 0a, 0 g 1
15 5
14",
OTBS
lV-121

 

 

 

Alkylation of lactone IV-lll with iodide IV-121, oxidation of the resultant
sulfide and elimination of the sulfoxide were effected as in case of IV-114. Thus, 40 mg
of IV-121 afforded 26 mg of IV-133 (68% overall yield).

Partial data for 1-133: 1H NMR (500 MHz, CDCl,) 6 6.98 (d, J = 1.5 Hz, 1 H), 5.02-4.88
(m, 1 H), 4.35-4.28 (m, 2 H), 4.04-3.98 (m, 1 H), 3.96-3.88 (m, 1 H), 3.81-3.75 (m, 1 H),

312

3.71-3.69 (m, 1 H), 3.63 (dd, J = 3.3, 7.5 Hz, 1 H), 2.26 (t, J = 7.2 Hz, 2 H), 1.91-1.72
(m, 3 H), 1.59-1.18 (m, 46 H), 0.91-0.87 (m, 30 H), 0.08-0.05 (m, 18 H); 13'C NMR (125
MHZ, CDC13) 6 174.0, 149.0, 134.5, 86.9, 81.6, 80.5, 79.6, 74.1, 73.9, 73.7, 36.3, 32.1,

31.4, 30.2, 29.9, 29.6, 28.2, 27.7, 26.4, 26.3, 26.2, 26.0, 25.9, 25.8, 25.4, 22.9, 19.5, 19.4,

18.4, 18.3, 18.1, 14.4, 14.3, —3.8, -3.9, -40, —4.1, -4.4, -4.9.

       
   

 

TBSQ ores 0 HQ 9H 0
O '- e
.. THF, rt 15 w
0733 80% ’DH
IV-1 33 IV-1 22

HF-pyridine mediated TBS cleavage of IV-133 (10 mg, 0.01 mmol) to afford triol

IV-122 (5 mg, 80%) was performed as described before (for IV-116).

Data for 1.122: [61020 —22.0 (c 0.30, 0H01,) 1H NMR (500 MHz, 0D01,) 6 7.00 (d. J =
1.5 Hz, 1 H), 5.02498 (m, 1 H), 4.47-4.46 (m, 1 H), 4.37 (dt, J = 1.5, 7.6 Hz, 1 H), 4.18-
4.14 (m, 1 H), 4.09407 (m, 1 H), 3.95 (dd, J = 1.5, 3.4 Hz, 1 H), 3.46-3.42 (m, 1 H),
3.41—3.37 (m, 1 H), 3.19 (d, J = 8.8 Hz, 1 H), 2.27 (dt, J = 1.5, 7.8 Hz, 2 H), 2.24-2.19
(m, 1 H), 2.08-1.99 (m, 5 H), 1.92 (ddd, J = 4.4, 9.5, 13.4 Hz, 1 H), 1.58-1.25 (m, 40 H),
0.89 (t, J = 6.9 Hz, 3 H); 130 NMR (125 MHz, 0D01,) 6 174.0, 149.1, 134.5, 83.5, 83.1,
81.8, 792,776,748, 74.7, 73.8, 38.6, 35.1, 34.0, 32.1, 29.9, 298,297,296, 29.5, 29.4,
28.3, 27.6, 25.9, 25.3, 22.9, 19.4, 14.3; IR (thin film) 3378, 2919, 2848, 1751, 1467,
1319, 1029, 873, 794 cm"; HRMS (ES) calcd for C37H6607, 623.4887 m/z (M+H)*;

observed, 623.4874 m/z.

313

Preparation of Mosher’s ester derivatives:
General procedure

To a solution of methoxytrifluoromethylphenylacetic acid (21 mg, 0.09 mmol) in hexanes
(1 mL), oxalyl chloride (38 11L, 0.42 mmol) and DMF (7.5 11L, 0.09 mmol) were added at
room temperature. After 1 h, the reaction mixture was centrifuged to separate the solid
residues and supernatant clear liquid was concentrated under reduced pressure (using a

water aspirator) to afford methoxytrifluoromethylphenylacetyl chloride. The acid
chloride was dissolved in CHZCl2 (1 mL). To this was added a mixture of the alcohol
(0.02 mmol), DMAP (1.3 mg, 0.01 mmol) and triethyl amine (31 mL, 0.23 mmol) as a
solution in CHZCI2 (1 mL). After stirring overnight at room temperature, the reaction was

quenched by saturated NH4C1 (5 mL) solution and the aqueous layer was extracted with

CHZCI2 (3x5 mL). The combined organic layers were dried over NaZSO4, concentrated
and crude material was purified by column chromatography to afford the corresponding

Mosher’s ester (typical yields 85%-88%).

 

 

#1189 CF3
(FD-MTPA-Cl T880 010
TBSO OH . _
W08” from (S) aCId W08”
-. 12 — 9- 4
w . 4 DCC/DMAP ” ,
bras CH2C12, rt ores
IV-85 85% was
MeO ‘Ph
“ CI
F30
o
(FD-MPTA-Cl
from(S) acid

 

 

 

(S)-MTPA derivative IV-99 (18 mg, 85%) was obtained from alcohol IV-85 (15

mg, 17.7 umol) following the general procedure described above.

314

Partial data for 1.99; IH NMR (500 MHz, 0D01,) 6 7.68-7.67 (m, 2 H), 7.36-7.28 (m, 8

H), 5.45 (dt, J = 2.4, 8.8 Hz, 1 H), 5.30-5.19 (m, 2 H), 4.51 (s, 2 H), 4.36-4.34 (m, 1 H),
4.31-4.28 (m, 1 H), 3.85 (dd, J: 3.5, 9.1 Hz, 1 H), 3.65 (s, 3 H), 3.64—3.62 (m, 1 H), 3.46

(t, J = 7.1 Hz, 2 H), 2.00-1.83 (m, 6 H), 1.63-1.18 (m, 40 H), 0.92 (s, 9 H), 0.88 (t, J = 7.0
Hz, 3 H), 0.87 (s, 9 H), 0.10 (s, 3 H), 0.08 (s, 3 H), 0.05 (s, 3 H), 0.04 (s, 3 H); [30 NMR
(125 MHz, 0D01,) 6 166.3, 1390,1331, 130.1, 129.5, 128.6, 1284,1283, 127.8, 127.7,

127.6, 84.0, 79.5, 76.4, 73.6, 73.4, 73.1, 70.6, 55.9, 36.6, 32.4, 32.1, 31.0, 30.1, 29.9,

29.8, 29.7, 29.6, 29.5, 27.3, 26.4, 26.1, 26.0, 23.1, 22.9, 18.3, 18.1, 14.3, 1.2, -3.8, -4.1, -

 

 

4.3, 48.
Ph
MeO "- CF13
raso OH (S)-MTPA-Cl I
? o 08“ from (R) acid TBS? O O
12 __ 4 4' 0,, 12 _ 4 0811
‘5 ., DCC/DMAP 15 ,
’ores C112C12, rt bras
(v-35 87 /° lV-1OO
Meg Ph
F3C Cl
0
(S)-MPTA-Cl
trom(H) acid

 

 

 

(R)-MTPA derivative IV-100 (18.5 mg, 87%) was obtained from alcohol IV-85

(15 mg, 17.7 umol) following the general procedure described above.

Partial data for I-100: 1H NMR (500 MHz, CDCl,) 6 7.65—7.64 (m, 2 H), 7.34-7.25 (m, 8
H), 5.40-5.28 (m, 3 H), 4.50 (s, 2 H), 4.33-4.32 (m, 1 H), 4.25-4.23 (m, 1 H), 3.85 (dd, J
= 3.4, 9.0 Hz, 1 H), 3.59-3.57 (m, 1 H), 3.53 (s, 3 H), 3.46 (t, J = 6.6 Hz, 2 H), 2.15-1.88

(m, 5 H), 1.76-1.23 (m, 41 H), 0.91 (8,9 H), 0.88 (t, J = 7.0 Hz, 3 H), 0.86 (s, 9 H), 0.08

(8,3 H), 0.07 (s, 3 H), 0.03 (s, 6 H); ”0 NMR (125 MHz, 0D01,) 6 166.1, 139.0, 132.5,

315

130.9, 129.5, 128.5, 128.4, 128.3, 127.8, 127.8, 127.6, 83.3, 79.6, 76.7, 73.8, 73.2, 73.1,
70.6, 55.5, 36.3, 32.1, 31.9, 31.0, 30.0, 29.9, 29.8, 29.7, 29.6, 29.5, 27.4263, 26.1, 26.0,

23.5, 22.9, 18.3, 18.2, 14.3, 1.2, «3.9, -4.2, -4.8.

 

MeO
ratioF3
reso OH (m-MTPA-Cl T889 0 o
17? 5 0,, 0., OBn DCC/DMAP ‘7 16 0,, O,,_ 8 OBn
15 13' ' 8 5 *— ‘5 13 5
3, CH C. ,n 14‘;
14 ores 8802 2 ores
IV-98 lV-101

(S)-MTPA derivative IV-101 (19 mg, 88%) was obtained from alcohol IV-98 (15

mg, 17.4 umol) following the general procedure described above.

Partial data for 1.101: 1H NMR (500 MHz, 0D01,) 6 7.66-7.64 (m, 2 H), 7.41—7.27 (m, 8
H), 5.15-5.12 (m, l H), 4.49 (s, 2 H), 4.31—4.19 (m, 3 H), 406402 (111, l H), 3.72-3.65
(m, 2 H), 3.61 (s, 3 H), 3.41 (1,] = 6.7 Hz, 2 H), 2.1-1.81 (m, 4 H), 1.63-1.20 (m, 42 H),
0.90 (s, 9 H), 0.89 (t, J = 7.0 Hz, 3 H), 0.88 (s, 9 H), 0.08 (s, 6 H), 0.06 (s, 6 H); 13C
NMR (125 MHz, CDCl,) 6 166.6, 138.9, 132.7, 129.5, 128.6, 128.5, 127.9, 127.8, 127.7,

85.8, 80.4, 79.3, 79. 1, 78.3, 74.0, 73.6, 73.1, 70.5, 56.0, 36.4, 32.1, 31.9, 30.6, 30.1, 29.9,

29.8, 29.7, 29.6, 29.5, 28.7, 28.4, 26.5, 26.2, 26.1, 25.9, 25.0, 22.9, 18.3, 18.1, 14.3, -3.9,

 

~42, ~43, -4.8.
M90
Pith/:0}:3
TBSQ OH (S)-MTPA-Cl TBSQ O O
‘7: 6 0,, 0,, OBn DCC IDMAP ‘77 6 0,, 0,9 OBn
15 13' ' 8 5 4: 15 13 8 s
a, CH Cl , rt 1 "a
14 ores 860/20 2 4 ores
lV-98 IV-102

(R)-MTPA derivative lV-102 (18.5 mg, 86%) was obtained from alcohol IV-98

(15 mg, 17.4 umol) following the general procedure described above.

316

Partial data for 1.102: 1H NMR (500 MHz, 0D01,) 6 7.62—7.60 (m, 2 H), 7.43—7.21 (m, 8
H), 5.12 (m, 1 H), 4.50 (s, 2 H), 4.28422 (m, 2 H), 4.18413 (m, 1 H), 4.08—4.06 (m, 1
H), 3.74-3.68 (m, 1 H), 3.60 (dt, J = 3.5, 7.5 Hz, 1 H), 3.56 (s, 3 H), 3.46 (t, J = 6.7 Hz, 2

H), 1.87-1.21 (m, 46 H), 0.90 (s, 9 H), 0.89 (t, J = 7.0 Hz, 3 H), 0.88 (s, 9 H), 0.08 (s, 6

H), 0.06 (s, 6 H); [30 NMR (125 MHz, 0D01,) 6 166.5, 138.9, 132.5, 129.6, 128.6,

128.5, 127.9, 127.8, 127.7, 85.6, 80.7, 79.4, 78.5, 78.4, 74.2, 73.6, 73.1, 70.5, 55.8, 36.5,

32.2, 31.8, 30.1, 30.0, 29.9, 29.8, 29.7, 29.6, 29.5, 28.5, 27.5, 26.4, 26.3, 26.2, 25.9, 22.9,
18.4, 18. 1, 144, -3.9, -4.1, -4.4, -4.8.

 

MeO
PhiCFa
TBSQ QH (m-MTPA-Cl TBSQ Q 0
17" 5 0,, O 7 0811 DCC/ DMAP ‘71 6 0,, O I 0811
'2 CH Cl , 11 14>,
M ores 85°}: 2 ores
lV-104 IV-105

(S)-MTPA derivative IV-105 (18 mg, 85%) was obtained from alcohol IV-104

(15 mg, 17 .4 umol) following the general procedure described above.
Partial data for 1.105; 1H NMR (500 MHz, 0D01,) 6 7.60-7.59 (m, 2 H), 7.40-7.29 (m, 8

H), 5.31-5.02 (m, 1 H), 4.50 (s, 2 H), 4.29-4.24 (m, 1 H), 4.13-4.10 (m, 1 H), 4.09-4.06
(m, 1 H), 4.00-3.95 (m, 1 H), 3.77-3.73 (1 H), 3.58 (s, 3 H), 3.52 ((11, J = 3.0, 8.0, Hz, 1

H), 3.46 (t, J = 6.6 Hz, 2 H), 1.84-1.19 (m, 46 H), 0.90 (t, J = 7.0 Hz, 3 H), 0.88 (s, 9 H),
0.86 (s, 9 H), 0,08 (s, 3 H), 0.07 (s, 3 H), 0.06 (s, 3 H). 0.01 (s, 3 H); 130 NMR (125
MHz, 0D01,) 6 166.5, 138.9, 1329,1296, 128.5, 128.4, 127.8, 127.7, 127.6, 86.4, 80.9,

80.1, 78.7, 78.4, 74.2, 73.8, 73.1, 70.5, 55.9, 36.7, 32.1, 31.6, 30.5, 30.1, 29.9, 29.8, 29.7,

29.6, 29.5, 27.7, 27.4, 26.3, 26.2, 25.9, 22.9, 18.4, 18.1, 14.3, -3.8, -4.0, -45, -49.

317

M90 CF3

 

TBSQ QH (S)-MTPA-CI TBSQ o o
17" s 0., O i 0811 DCC/DMAP 17’ e 0., O 080
"a H , 4",
14 ores S7020]?- n 1 ores
lV-104 lV-106

(R)-MTPA derivative IV-106 ( 19 mg, 87%) was obtained from alcohol IV-104
( 15 mg, 17.4 umol) following the general procedure described above.
Partial data for I-106: 1H NMR (500 MHz, CDCl,) 6 7.62-7.61 (m, 2 H), 7.40-7.27 (m, 8
H), 5.31-5.07 (m, 1 H), 4.49 (s, 2 H), 4.31-4.27 (m, 2 H), 4.04—3.98 (m, 2 H), 3.74-3.72
(m, 1 H), 3.63—3.61 (m, 1 H), 3.61 (s, 3 H), 3.41 (t, J = 6.6 Hz, 2 H), 1.95-1.84 (m, 4 H),

1.82-1.26 (m, 42 H), 0.90 (t, J = 7.0 Hz, 3 H), 0.88 (s, 18 H), 0.09 (s, 3 H), 0.07 (s, 3 H),
0.06 (s, 3 H), 0.05 (s, 3 H); 130 NMR (125 MHz, 0D01,) 6 166.7, 138.9, 132.8, 129.6,

128.6, 128.5, 127.8, 127.7, 127.6, 86.3, 80.7, 80.1, 79.0, 78.5, 74.0, 73.8, 73.], 70.5, 56.0,

36.2, 32.1, 31.5, 30.7, 30.1, 30.0, 29.9, 29.8, 29.7, 29.6, 28. 1, 27.7, 26.5, 26.2, 26.1, 25.9,

25.0, 22.9, 18.3, 18.1, 14.3, -3.8, -41, -4.4, -4.9.

318

E. References

10.

ll.

12.

13.

14.

15.

16.

17.

18.

Trost, B. M.; Fleming, I., Eds. Comprehensive Organic Synthesis; lst ed.;
Pergamon Press: Oxford, 1991; Vol. 1,2.

Otera, J. Modern Carbonyl Chemistry; Wiley-VCH: Weinheim ; New York, 2000.

Wakefield, B. .l. Organomagnesium Methods in Organic Synthesis; Academic
Press: London ; San Diego, 1995.

Silverman, G. S.; Rakita, P. E. Handbook of Grignard Reagents; Marcel Dekker:
New York, 1996.

Wakefield, B. J. Organolithium Methods; Academic Press: London ; San Diego,
1988.

Kocienski, P. J. Protecting Groups; 2nd ed.; G. Thieme: Stuttgart, 2000.

Lipshutz, B. H.; Sengupta, S. Org. React. (N. Y.) 1992, 41, 135.
Weidmann, B.; Seebach, D. Angew. Chem, Int. Ed. Engl. 1983, 22, 31.

Simmons, H. E.; Cairns, T. L.; Vladuchick, A.; Hoiness, C. M. Org. React. (N. Y.)
1972, 20, 1.

Furstner, A. Synthesis 1989, 571.
Knochel, P.; Yeh, M. C. P.; Berk, S. C.; Talbert, .1. J. Org. Chem. 1988, 53, 2390.

Negishi, E. I. Acc. Chem. Res. 1982, 15, 340.
Knochel, P.; Singer, R. D. Chem. Rev. 1993, 93, 2117.
Knochel, P.; Perea, J. .1. A.; Jones, P. Tetrahedron 1998, 54, 8275.

Knochel, P.; Millot, N.; Rodriguez, A. L.; Tucker, C. E. Org. React. (N.Y.) 2001,
58, 417.

Hupe, E.; Calaza, M. 1.; Knochel, P. Tetrahedron lett. 2001, 42, 8829.

Langer, F.; Schwink, L.; Devasagayaraj, A.; Chavant, P. Y.; Knochel, P. J. Org.
Chem. 1996, 61, 8229.

Reetz, M. T.; Kesseler, K.; Schmidtberger, S.; Wenderoth, B.; Steinbach, R,
Angew. Chem, Int. Ed. Engl. 1983, 22, 989.

319

 

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

Reetz, M. T.; Kesseler, K.; Jung, A. Tetrahedron Lett. 1984, 25, 729.
Hoveyda, A. H.; Evans, D. A.; Fu, G. C. Chem. Rev. 1993, 93, 1307.

Rodefeld, L.; Tochtermann, W. Tetrahedron 1998, 54, 5893.

Couturier, M.; Dory, Y. L.; Rouillard, F.; Deslongchamps, P. Tetrahedron 1998,
54, 1529.

Anan, H.; Seki, N.; Noshiro, 0.; Honda, K.; Yasumuro, K.; Ozasa, T.; Fusetani,
N. Tetrahedron 1996, 52, 10849.

Yeh, M. C. P.; Knochel, P.; Santa, L. E. Tetrahedron Lett. 1988, 29, 3887.

Decamp, A. E.; Kawaguchi, A. T.; Volante, R. P.; Shinkai, I. Tetrahedron Lett.
1991, 32, 1867.

Tamaru, Y.; Nakamura, T.; Sakaguchi, M.; Ochiai, H.; Yoshida, Z. J. Chem. Soc.,
Chem. Commun. 1988, 610.

Ochiai, H.; Nishihara, T.; Tamaru, Y.; Yoshida, Z. J. Org. Chem. 1988, 53, 1343.
Rozema, M. J.; Sidduri, A.; Knochel, P. J. Org. Chem. 1992,57, 1956.

Rozema, M. J.; Eisenberg, C.; Lutjens, H.; Ostwald, R.; Belyk, K.; Knochel, P.
Tetrahedron Lett. 1993, 34, 3115.

Knochel, P.; Jones, R, Eds. Organozinc Reagents: A Practical Approach; Oxford
University Press: Oxford ; New York, 1999.

Hanessian, S.; Moitessier, N.; Wilmouth, S. Tetrahedron 2000, 56, 7643.
Ward, D. E.; Rhee, C. K. Tetrahedron Lett. 1991, 32, 7165.
Larcheveque, M.; Lalande, J. J. Chem. Soc., Chem. Commun. 1985, 83.
Overman, L. E.; McCready, R. J. Tetrahedron Lett. 1982, 23, 2355.
Nakata, T.; Tanaka, T.; Oishi, T. Tetrahedron Lett. 1983, 24, 2653.

Brown, H. C.; Krishnan, S. J. Am. Chem. Soc. 1972, 94, 7159.

Rueger, H.; Stutz, S.; Goschke, R.; Spindler, F.; Maibaum, J. Tetrahedron Lett.
2000, 41, 10085.

Yang, W. Q.; Kitahara, T. Tetrahedron 2000, 56, 1451.

320

39.

41.

42.

43.

45.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

Ley, S. V.; Meek, G. J. Chem. Soc., Perkin Trans. 1 1997, 1125.

Boger, D. L.; lchikawa, S.; Zhong, W. J. Am. Chem. Soc. 2001, 123, 4161.

Achyutharao, S.; Knochel, P. J. Org. Chem. 1991, 56, 4591.

Couturier, M.; Dory, Y. L.; Rouillard, F.; Deslongchamps, P. Tetrahedron 1998,
54, 1529.

Peng, S.; Okeley, N. M.; Tsai, A. L.; Wu, G.; Kulmacz, R. J.; van der Donk, W.
A. J. Am. Chem. Soc. 2002, 124, 10785.

Nicolaou, K. C.; Ritzen, A.; Namoto, K.; Buey, R. M.; Diaz, J. F.; Andreu, J. M.;
Wartmann, M.; Altmann, K. H.; O'Brate, A.; Giannakakou, P. Tetrahedron 2002,

58, 6413.
Reetz, M. T. Angew. Chem, Int. Ed. Engl. 1984, 23, 556.
Reetz, M. T. Acc. Chem. Res. 1993, 26, 462.

Still, W. C.; Schneider, J. A. Tetrahedron Lett. 1980, 21, 1035.

Still, W. C.; McDonald, J. H. Tetrahedron Lett. 1980, 21 , 1031.

Asami, M.; Kimura, R. Chem. Lett. 1985, 1221.
Koert, U.; Wagner, H.; Pidun, U. Chem. Ber. 1994, 127, .1447.

Amdt, S.; Emde, U.; Baurle, 5.; Friedrich, T.; Grubert, L.; Koert, U. Chem. Eur.
J. 2001, 7, 993.

Black, T. H. in Encyclopedia of Reagents for Organic Synthesis; Paquette, L. A.,
Ed.; Wiley: Chichester ; New York, 1995; Vol. 5, pp 3197-3199.

Black, T. H.; Mcderrnott, T. S.; Brown, G. A. Tetrahedron Lett. 1991, 32, 6501.

Indictor, N.; Brill, W. F. J. Org. Chem. 1965, 30, 2074.
Trost, B. M.; Verhoeven, T. R. J. Am. Chem. Soc. 1980, 102, 4743.

Fukuyama, T.; Vranesic, B.; Negri, D. P.; Kishi, Y. Tetrahedron Lett. 1978, 2741.

Hoffmann, R. W. Chem. Rev. 1989, 89, 1841.

Katsuki, T. in Catalytic Asymmetric Synthesis; 2nd ed.; Ojima, 1., Ed.; Wiley-
VCH: New York, 2000, pp 287-325.

321

59.

60.

61.

62.

63.

65 .

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

Tu, Y.; Wang, Z. X.; Shi, Y. J. Am. Chem. Soc. 1996, 118, 9806.

Wang, Z. X.; Shi, Y. A. J. Org. Chem. 1997, 62, 8622.

Tian, H. Q.; She, X. G.; Shu, L. H.; Yu, H. W.; Shi, Y. J. Am. Chem. Soc. 2000,
122, 11551.

Jacobsen, E. N.; Zhang, W.; Loebach, J. Abstr. Pap. Am. Chem. S. 1990, 199,
241.

Irie, R.; Noda, K.; Ito, Y.; Matsumoto, N .; Katsuki, T. Tetrahedron Lett. 1990, 31,
7345.

Kolb, H. C.; Sharpless, K. B. Tetrahedron 1992, 48, 10515.

Kolb, H. C.; Vannieuwenhze, M. S.; Sharpless, K. B. Chem. Rev. 1994, 94, 2483.

Johnson, R. A.; Sharpless, K. B. in Catalytic Asymmetric Synthesis; 2nd ed.;
Ojima, 1., Ed.; Wiley-VCH: New York, 2000, pp 231-280.

Baldwin, J. E. J. Chem. Soc., Chem. Commun. 1976, 734.

Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic Synthesis; 3rd ed.;
Wiley: New York, 1999.

Johnson, R. A.; Sharpless, K. B. in Catalytic Asymmetric Synthesis; 2nd ed.;
Ojima, 1., Ed.; Wiley-VCH: New York, 2000, pp 357-398.

Hale, K. J.; Manaviazar, 8.; Peak, S. A. Tetrahedron Lett. 1994, 35, 425.
Wang, L.; Sharpless, K. B. J. Am. Chem. Soc. 1992, 114, 7568.

Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512.

Barth, G.; Voelter, W.; Mosher, H. S.; Djerassi, C. J. Am. Chem. Soc. 1970, 92,
875.

Marshall, J. A.; Piettre, A.; Paige, M. A.; Valeriote, F. J. Org. Chem. 2003,68,
1771.

White, J. D.; Somers, T. C.; Reddy, G. N. J. Org. Chem. 1992, 57, 4991.

Shi, G.; Kozlowski, J. F.; Schwedler, J. T.; Wood, K. V.; MacDougal, J. M.;
McLaughlin, J. L. J. Org. Chem. 1996, 61, 7988.

Emde, U.; Koert, U. Eur. J. Org. Chem. 2000, 1889.

322

78.

79.

80.

81.

82.

Born, L.; Lieb, F.; Lorentzen, J. P.; Moeschler, H.; Nonfon, M.; Sollner, R.;
Wendisch, D. Planta Med. 1990, 56, 312.

Pretsch, E. Tables of Spectral Data for Structure Determination of Organic
Compounds; 2nd ed.; Springer-Verlag: Berlin ; New York, 1989.

Homann, R. W.; Kruger, J.; Bruckner, D. New J. Chem. 2001, 25, 102.
Takahashi, S.; Maeda, K.; Hirota, S.; Nakata, T. Org. Lett. 1999, I, 2025.

Shimojo, M.; Matsumoto, K.; Hatanaka, M. Tetrahedron 2000, 56, 9281.

323

APPENDIX

324

 

 

 

 

 

 

 

 

 

 

cum

 

 

 

325

p

cm

hlr_—pp

oh
_

_

OHH
__

b

p

h

ONH oma
_prthp

. Pill-i

»

l'lr

p

 

 

 

 

 

5

14 141111

 

 

 

 

 

Z

326

 

 

 

 

 

 

 

 

 

 

not...

mmeo a.
010/;

mwem

327

 

 

 

 

 

 

 

 

328

 

 

 

 

 

 

 

 

 

329

 

 

 

 

 

 

 

 

330

MG ..-:—

:0

E;
mzdo m.
:0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

now¢=

Io
\/\mq/0\—$/
$50 9

I0

331

Fll_

—

 

 

 

 

 

 

 

 

 

 

voF4=
mmeo

.9
m2do\)/\)/(44/

mmem

332

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

«ova:
mmeo

.wp
m260\)/\)/(x4/

mwem

333

 

 

 

 

 

 

 

 

 

 

334

 

 

 

mo 7.:

mwFO
m @—

IO\/\/\$/

$6

 

 

 

on on

 

 

 

 

OF 00

 

 

 

mo 7.:
wwHO
m or

Io\)/\)/(*4/

mm»@

335

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

we _.r___
mmErntV 9
N N
am 00/:

$5

336

uxwu om
__._.______.______________

H-;

av

11

D

ow

_______.________.__4b._.__.___

Ev

 

 

om OOH ONH ova owH

....—...—....__~_____.______.__h_

1' 1! 6 11h Ii )1 i

 

 

.J

L -

 

 

D II 1| tlbr
4 2‘ ll 1 1 14 1 [‘11

moF4=
mmeo

N u 0P
um 00/:

mwem

337

 

 

 

 

 

 

 

 

338

he 7.:

 

 

 

 

 

 

 

 

 

 

ha ......

O.

.. 6.me e

   
   

IO

339

 

 

 

 

 

 

 

 

 

 

 

 

_.N ..-:—
o<n~.

 

 

340

ON

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

341

 

 

 

 

 

 

 

 

Nm 7.:
mmHO.

cam

 

0%:

342

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

343

3.2
85..

 

and c on 8. 8 oo 2: cad ova
_»_____ulr»Ll_leL#~_L_..FL~____r»_F__»__~prr____._p.___..2_____bp__hu_r__wblplr___auup___rr

 

 

1 z - , _- _

 

 

 

 

 

 

344

 

 

 

 

 

 

 

 

 

 

 

Fmr>_

345

 

 

____._.___.__..—_______

 

 

 

 

 

 

 

 

 

 

 

 

Fmr>_

346

 

 

 

 

 

 

 

 

 

 

 

 

347

 

 

 

 

 

 

 

 

 

 

 

348

 

 

 

 

 

 

 

 

 

 

 

 

 

 

349

 

an e on
l_l_.___________._____

 

 

ov oo om OOH

....________..___..____~..___..______.._»_e__

 

 

 

ONH

.___u.____

 

 

 

ova ova

b__...__..._______.__

 

350

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

351

 

 

 

 

 

 

 

 

 

352

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

353

 

 

 

 

 

 

 

 

 

 

 

354

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

355

 

 

 

 

 

 

 

 

356

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

357

 

 

 

 

 

 

 

 

358

 

 

 

 

 

 

 

 

 

359

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

 

5030 cc 27>.
6.

360

 

lrELLLLLLIELlL

611.11)
1‘

OV

1“

O@

OO

iii?

OOH

_...—....__._____r__._.___..__...__._..____

I4]

 

ONH OVH OOH

P——______b___b__—______~____~__

 

 

 

l.— r . uir Hit .lt._1llrh

 

361

 

 

 

 

 

362

 

oNH

OVH

 

 

363

 

MllLHIGAN \‘IAIE ‘JNthRSlIV l
1 .v . 1.1”1111171‘1111
1 l ‘

. 11 .‘11 11 “ll

‘ ii, l‘ 1,111.11 11

3 1293 02551

   

1
6

B
l
.11

 

     

    

"1?.
11111 ‘

1111

380